set-wet: a wetland simulation model to optimize …
TRANSCRIPT
SET-WET: A WETLANDSIMULATION MODEL TO OPTIMIZE
NPS POLLUTION CONTROL
ERIK RYAN LEE
Thesis submitted to the Faculty of theVirginia Polytechnic Institute and State University
in partial fulfillment of the requirements for the degree of
Master of Sciencein
Biological Systems Engineering
Saied Mostaghimi, ChairTheo A. Dillaha
Raymond B. ReneauJohn V. Perumpral
September 15,1999Blacksburg, VA
Keywords: Wetlands, Model, Nonpoint Source Pollution,Biological, Nutrients
Copyright 1999, Erik R. Lee
SET-WET: A WETLAND SIMULATION MODEL TOOPTIMIZE NPS POLLUTION CONTROL
Erik Ryan Lee
(Abstract)
A dynamic, compartmental, continuously stirred tank reactor, simulation model (SET-
WET) was developed for design and evaluation of constructed wetlands in order to optimize
non-point source (NPS) pollution control measures. The model simulates the hydrologic,
nitrogen, carbon, dissolved oxygen, bacteria, vegetative, phosphorous and sediment cycles of a
wetland system. Written in Fortran 77, SET-WET models both free water surface (FWS) and
sub-surface flow (SSF) wetlands and is designed in a modular manner which gives the user the
flexibility to decide which cycles and processes to model. SET-WET differs from many existing
wetland models in that it uses a system’s approach, and limits the assumptions made concerning
the interactions of the various nutrient cycles in a wetland system. It accounts for carbon and
nitrogen interactions, as well as effect of oxygen levels upon microbial growth. It also directly
links microbial growth and death to the consumption and transformations of nutrients in the
wetland system. Many previous models have accounted for these interactions with zero and first
order rate equations that assume rates are dependent only on initial concentrations. The SET-
WET model is intended to be utilized with an existing NPS hydrologic simulation model, such as
ANSWERS or BASINS, but may also be used in situations where measured input data to the
wetland are available.
The model was calibrated and validated using limited data collected at Benton, Kentucky.
A non-parametric statistical analysis of the model's output indicated eight out of nine examined
outflow predictions were not statistically different from the measured observations. Linear
regression analysis showed that six out of nine examined parameters were statistically similar,
and that within the expected operating range, all of the examined outflow parameters (9) were
within the 95% confidence intervals of the regression lines. A sensitivity analysis showed the
most significant input parameters to the model were those which directly affect bacterial growth
and oxygen uptake and movement. The model was applied to a subwatershed in the Nomini
iii
Creek watershed located in Virginia. Two year simulations were completed for five separate
wetland designs, with reductions in percentage of BOD5 (4%-45%), TSS (85%-100%), total
nitrogen (42%-56%), and total phosphorous (38%-57%) comparable to levels reported by
previous research.
iv
Acknowledgements
I would first like to thank my advisor Professor Saied Mostaghimi, who gave me
countless advice and information on how to do proper and professional thesis work. To my
committee members Professor Theo Dillaha and Professor Ray Reneau, your advice and tutelage
were sage and wise. To our Department head, Professor John Perumpral, I would like to give
thanks for helping me adjust to Virginia and making me feel at home. Big thanks to Kevin
Brannan and Shreeram Inmandar, who knew that when I knocked on their door they were going
to be interrupted for an hour. To Theresa Wynn I give thanks for all the help on my model when
you were dog-tired and the advice about life and other important things.
My parents, Priscilla and Collin Wong, were very encouraging and I am glad that they
made me learn how to cook and clean, because I’ve seen plenty of very helpless people in
college. My grandmothers, Lin Kim Lennie Lee and Susie Lum have always been supportive
and understanding. To my brothers, Daryl, William, and Alex, I thank you for giving me the
motivation to study because I wanted to get better grades than you. To my aunts and cousins
who have sent me cookies through my college years, my roommates and I thank you. Of course,
even though I am about to graduate that tradition may continue.
I would also like to acknowledge every one in my family and all of my friends. Now that
I have my Master’s in Biological Systems Engineering, I hope that you can finally remember
what the title of my degree is.
v
Table of ContentsI. INTRODUCTION ............................................................................................................................................. 1
A. GOAL AND OBJECTIVES ................................................................................................................................. 2
II. LITERATURE REVIEW ................................................................................................................................. 3
A. NPS POLLUTION ............................................................................................................................................ 3B. BEST MANAGEMENT PRACTICES (BMPS) .................................................................................................... 5C. WETLANDS ..................................................................................................................................................... 7
1. Classification ............................................................................................................................................. 8a. Natural Wetlands.........................................................................................................................................................8b. Constructed Wetlands..................................................................................................................................................9
2. Constructed Wetland Design................................................................................................................... 103. Nitrogen Cycle in Wetlands..................................................................................................................... 15
a. Nitrogen Transformation Processes .........................................................................................................................17i. Mineralization (ammonification) ..........................................................................................................................17ii. Nitrification ..........................................................................................................................................................18iii. Denitrification.......................................................................................................................................................19iv. Nitrogen Fixation.................................................................................................................................................19v. Assimilation: Plant and Bacterial Uptake ............................................................................................................20
b. Other Nitrogen Fluxes ..............................................................................................................................................21i. Atmospheric Nitrogen Inputs ................................................................................................................................21ii. Ammonia Volatilization .........................................................................................................................................21iii. Adsorption ............................................................................................................................................................22iv. Burial of Organic Nitrogen ..................................................................................................................................22v. Biomass Decomposition.......................................................................................................................................22
4. Phosphorous Cycle in Wetlands ............................................................................................................. 22a. Importance of Sediment – Sorption/Desorption .......................................................................................................23b. Precipitation ..............................................................................................................................................................24c. Biomass: Growth, Death, Decomposition, Uptake and Storage ..............................................................................25
5. Bacteria in Wetlands ............................................................................................................................... 256. Vegetative/Carbon Cycle in Wetlands ..................................................................................................... 277. Modeling Wetland Processes .................................................................................................................. 28
a. General Modeling Practices......................................................................................................................................29b. Modeling of Specific Wetland Processes ..................................................................................................................31
i. Hydrology .............................................................................................................................................................31Overall Water Budget ...........................................................................................................................................31Surface Water Flow...............................................................................................................................................33Evapotranspiration ................................................................................................................................................35Groundwater Flow ................................................................................................................................................37
ii. Nitrogen ................................................................................................................................................................37iii. Phosphorous .........................................................................................................................................................41iv. Sediment................................................................................................................................................................43v. Vegetation .............................................................................................................................................................45
c. Selected Wetland Models...........................................................................................................................................46D. LITERATURE REVIEW SUMMARY..................................................................................................... 55
III: MODEL DEVELOPMENT............................................................................................................................ 58
A. MODEL OVERVIEW:............................................................................................................................... 581. FWS vs. SSF Modeling ........................................................................................................................... 60
B. MODEL COMPONENTS: ........................................................................................................................ 621. Wetland main program: .......................................................................................................................... 622. Base submodel: ....................................................................................................................................... 633. Hydrology submodel: .............................................................................................................................. 644. Vegetation Submodel: ............................................................................................................................. 695. Nitrogen/Carbon/DO/Bacteria relations: ............................................................................................... 726. Carbon submodel: ................................................................................................................................... 73
vi
7. Nitrogen submodel:................................................................................................................................. 808. Dissolved oxygen submodel: ................................................................................................................... 879. Bacteria submodel: ................................................................................................................................. 91
a. Autotrophic Dynamics...............................................................................................................................................91b. Heterotrophic bacteria ..............................................................................................................................................93
10. Sediment submodel: ................................................................................................................................ 9611. Phosphorous submodel:.......................................................................................................................... 9812. Deltaht submodel: ................................................................................................................................. 10113. SET-WET Flow Chart .......................................................................................................................... 102
C. MODEL DEVELOPMENT SUMMARY ............................................................................................... 102
IV. MODEL EVALUATION............................................................................................................................... 106
A. MODEL CALIBRATION AND VALIDATION .................................................................................................. 1061. Study Area ............................................................................................................................................. 1062. Model Calibration: ................................................................................................................................ 1073. Model Validation: .................................................................................................................................. 125
B. STATISTICAL ANALYSIS: ............................................................................................................................ 132C. SENSITIVITY ANALYSIS:............................................................................................................................. 136D. MODELING APPLICATION .......................................................................................................................... 141
1. Study/Application Area ......................................................................................................................... 1412. Simulation Runs.................................................................................................................................... 1433. Simulation Results ................................................................................................................................ 146
E. MODEL EVALUATION SUMMARY ............................................................................................................... 152
V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ............................................................... 154
VI. CITED WORK:............................................................................................................................................. 158
VII. APPENDICES........................................................................................................................................... 166
A. APPENDIX A: MODEL PARAMETERS......................................................................................................... 167B. APPENDIX B: DATA ENTRY TO MODEL..................................................................................................... 180C. APPENDIX C: MODEL FORTRAN CODE FOR THE SET-WET MODEL ...................................................... 190D. APPENDIX D: SYMBOL DESCRIPTION FOR FIGURES 8 THROUGH 22 ........................................................ 239E. APPENDIX E: REGRESSION GRAPHS ......................................................................................................... 240F. APPENDIX F: SENSITIVITY ANALYSIS TABLES.......................................................................................... 244
VIII. VITA.............................................................................................................................................................248
vii
List of Tables
TABLE 1: NUTRIENT REMOVAL RATES FOR NATURAL WETLAND SITES RECEIVING WASTEWATER INPUTS.9TABLE 2: GENERAL HYDROPERIOD TOLERANCE RANGES FOR SELECTED WETLAND PLANT
COMMUNITIES………………………………………………………………..………………………14TABLE 3: WETLAND DESIGN PARAMETERS .........................................................………………………..15TABLE 4: A PARTIAL LIST OF PREVIOUS WETLAND MODELS..................................................……………30TABLE 5: MEASURED INFLOW VALUES TO WETLAND CELL 2 IN BENTON, KENTUCKY USED FOR
VALIDATION AND CALIBRATION OF SET-WET MODEL……...………………………………..…....107
TABLE 6:INPUT PARAMETER VALUES AND SOURCES FOR CALIBRATION AND VALIDATION PERIODS…….109TABLE 7: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED
VALUES FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR
THE CALIBRATED, PREDICTED VALUES……………………………………………………………..123TABLE 8: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED
VALUES FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR
THE VALIDATED, PREDICTED VALUES………………………………………………………………130TABLE 9: P-VALUES AND RESULTS OF THE WILCOXON SIGNED RANK TEST PROCEDURE FOR
DIFFERENCES BETWEEN THE MEASURED AND VALIDATED, PREDICTED VALUES OF WETLAND
EFFLUENT IN BENTON, KENTUCKY………… …………………………………………………....133 TABLE 10: LINEAR REGRESSION DATA FOR OBSERVED (Y-AXIS) AND PREDICTED (X-AXIS) WETLAND
EFFLUENT………………………………………………………………………………………..….134TABLE 11: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 50% CHANGE IN BASE VALUES……… ……………………………………...137TABLE 12: INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL SIMULATION
RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF NOMINI CREEK
WATERSHED…………………………………………………………………………………………146TABLE 13: INFLUENT, EFFLUENT, AND % REDUCTION OF NUTRIENTS FOR VARIOUS NUTRIENTS FOR
2 YEAR PERIODS OF WETLAND SIMULATIONS FOR QN2 SUBWATERSHED DATA………………….....150TABLE 14: RANGE OF POLLUTANT REMOVAL EFFICIENCIES REPORTED FOR CONSTRUCTED WETLAND
SYSTEMS……………………………………………………………………………………………152TABLE F.1: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 10% CHANGE IN BASE VALUES………… …………………………………...244TABLE F.2: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 25% CHANGE IN BASE VALUES………..……………………………………...246
viii
List of Figures
FIGURE 1: BREAKDOWN OF NPS POLLUTION EMANATION FOR RIVERS IN VIRGINIA....................................4FIGURE 2: CROSS SECTION OF A FWS WETLAND.........................................................................................10FIGURE 3: CROSS SECTION OF A TYPICAL SUBSURFACE FLOW WETLAND. .................................................11FIGURE 4: NITROGEN TRANSFORMATIONS IN WETLANDS. .........................................................................17FIGURE 5: PHOSPHORUS TRANSFORMATIONS IN WETLANDS......................................................................24FIGURE 6: WETLAND DESCRIPTION FOR SET-WET MODEL WETLANDS.......................................................59FIGURE 7: RELATIONSHIP OF SET-WET MAIN CODE TO SET-WET SUBMODELS ...........................................63FIGURE 8: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE
SUBMODEL FOR FWS WETLANDS IN THE SET-WET MODEL..................................................................66FIGURE 9: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE
SUBMODEL FOR SSF WETLANDS IN THE SET-WET MODEL ...................................................................67FIGURE 10: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE VEGETATION CYCLE
SUBMODEL OF THE SET-WET MODEL...................................................................................................71FIGURE 11: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................74FIGURE 12: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE
SUBMODEL FOR SSF WETLANDS OF THE SET-WET MODEL...................................................................75FIGURE 13: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................81FIGURE 14: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGENCYCLE
SUBMODEL FOR SSF WETLANDS OF THE SET-WET MODEL...................................................................83FIGURE 15: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL…………………………………………..88FIGURE 16: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR SSF.............................................................................................................................89FIGURE 17: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER.........................................................................92FIGURE 18: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE …………………………………………..92FIGURE 19: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER.........................................................................94FIGURE 20: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE…………………………………………...94FIGURE 21: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE SEDIMENT CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL .................................................................97FIGURE 22: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE PHOSPHOROUS
CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL…………………………………100FIGURE 23: FLOW CHART FOR CALLING ORDER OF SET-WET MODEL FROM MAIN CODE THROUGH
SUBROUTINES....................................................................................................................................103FIGURE 24A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR
HYDROLOGIC OUTFLOW FROM THE WETLAND..................................................................................114FIGURE 24B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR
HYDROLOGIC OUTFLOW FROM THE WETLAND..................................................................................114FIGURE 25A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE
WETLAND………………………………………………...115FIGURE 25B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE
WETLAND………………………………………………...115
ix
FIGURE 26A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................116FIGURE 26B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................116FIGURE 27A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................117FIGURE 27B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................117FIGURE 28A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................118FIGURE 28B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................118FIGURE 29A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................119FIGURE 29B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................119FIGURE 30A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................120FIGURE 30B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................120FIGURE 31A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................121FIGURE 31B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................121FIGURE 32A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................122FIGURE 32B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................122FIGURE 33: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND ........................................................................................................125FIGURE 34: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................126FIGURE 35: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................126FIGURE 36: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. .......................................................127FIGURE 37: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND. ..........................................................127FIGURE 38: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR BOD5
EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................................................128FIGURE 39: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND..........................................128FIGURE 40: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................129FIGURE 41: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND. ................................................129FIGURE 42: SIMULATED AND OBSERVED VALUES FOR DISSOLVED PHOSPHOROUS CONCENTRATIONS,
PLOTTED WITH THE DETERMINED LINEAR REGRESSION, AND IDEAL 1:1 LINE. ................................135FIGURE 43: LOCATION OF THE NOMINI CREEK WATERSHED IN VIRGINIA WITH RESPECT TO
RICHMOND, VA AND THE CHESAPEAKE BAY. ..................................................................................142FIGURE 44: NOMINI CREEK WATERSHED (QN1) WITH SUBWATERSHED (QN2; SHADED) .......................142
x
FIGURE 45: LINEAR REGRESSION OF RECORDED TOTAL BOD5 AND HYDROLOGIC INFLOW TO QN2SUBWATERSHED OF NOMINI CREEK WATERSHED FOR MARCH 26, 1992 TO MARCH 25, 1994.........144
FIGURE E.1.: SIMULATED AND OBSERVED VALUES FOR OUTFLOW, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ........................................240FIGURE E.2.: SIMULATED AND OBSERVED VALUES FOR AMMONIUM CONCENTRATIONS, PLOTTED
BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL,AND IDEAL 1:1 LINE. .........................................................................................................................240
FIGURE E.3.: SIMULATED AND OBSERVED VALUES FOR NITRATE CONCENTRATION, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ...........241FIGURE E.4.: SIMULATED AND OBSERVED VALUES FOR ORGANIC NITROGEN CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................241FIGURE E.5.: SIMULATED AND OBSERVED VALUES FOR DISSOLVED OXYGEN CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................242FIGURE E.6.: SIMULATED AND OBSERVED VALUES FOR BOD5 CONCENTRATIONS, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE. ...........242FIGURE E.7.: SIMULATED AND OBSERVED VALUES FOR TOTAL SUSPENDED SOLID CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................243FIGURE E.8.: SIMULATED AND OBSERVED VALUES FOR TOTAL PHOSPHOROUS CONCENTRATIONS,
PLOTTED BESIDE THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND
IDEAL 1:1 LINE. .................................................................................................................................243
1
SET-WET: a wetland simulation model tooptimize NPS pollution control.
I. Introduction
Nonpoint Source (NPS) pollution accounts for more than 50% of the nation’s total water
quality problems (Novotny and Olem, 1981) and over 65% of the total pollutant load to inland
surface waters (USEPA, 1993). Therefore, developing practices for controlling NPS pollution is
of major importance to the health of humans and wildlife. Various types of best management
practices (BMPs) have been developed to address this acute problem, one of which is the use of
wetlands.
Wetlands filter out pollutants and act as sinks for nutrients through physical, chemical
and biochemical processes (Novotny and Olem, 1994). Unfortunately humans have not always
perceived wetlands to be beneficial, and wetlands have been converted to other uses such as
agriculture, mining and development at an alarming rate in the United States. The U.S. Fish and
Wildlife Service estimates that over 50 percent of U.S. wetlands have been destroyed during the
last two centuries (Environmental Law Institute, 1993). Iowa alone has lost 99% of its original
natural marshes, while California has had 91% of it’s wetlands converted to other uses (Tiner,
1984). Nonetheless, wetlands still comprise over 6% of the entire land based area on the planet
Earth (Novotny and Olem, 1994).
In an effort to restore converted wetlands, many Federal management agencies have
active programs to restore wetlands under their jurisdiction and are encouraging private
landowners and other agencies to do the same (Whitacker and Terrell, 1993). Legislation in
Florida requires any natural wetland removal to be replaced with constructed or restored wetland
sites that are at minimum, two times the amount of lost wetland area.
The use of wetlands to control NPS pollution is a relatively new concept (Raisin and
Mitchell, 1995; Teague et al., 1997). Wetland restoration has taken place in northeastern Illinois
(Hey et al. 1989), and constructed wetlands have been established in Massachusetts (Daukas et
al., 1989) with encouraging results, as significant nutrients and sediments have been retained by
these systems. Research has supported the use of wetlands to treat NPS pollution, but the
2
question is whether these wetlands are being properly designed to optimize a wetland’s ability to
decrease NPS pollution.
The design of wetlands for NPS pollution removal can be optimized with the use of
models that accurately represent wetland system’s processes. The ability to optimize wetland
design is beneficial for several reasons. Due to the “no net loss” policy developed at the
National Wetland Policy Forum in 1987, there should be no removal of wetlands without the
construction of replacement wetlands. There are no laws controlling the quality of these
replacement wetlands however, and many are poorly planned and constructed. To use
replacement wetlands effectively, there is a need to predict how effective these replacements will
be. It is pointless to replace an efficient waste removing wetland with a “pond” that
accomplishes little. Use of models allows comparisons among various designs, and
consequently improves the effectiveness of replacement wetland with respect to NPS pollution
control efforts.
A. Goal and Objectives
The overall goal of the study is to develop a simulation model that can be used as a
planning tool for the design of constructed wetlands for effective control and treatment of NPS
pollution. The specific objectives are to:
1) Develop a user-friendly, dynamic, long-term, lumped parameter model for the design of
constructed wetlands to optimize NPS pollution control measures.
2) Evaluate the proposed model by comparing its predictions with field data collected from
representative constructed wetland site(s).
3
II. Literature Review
In this section a basic overview of the problems associated with NPS pollution is
presented. It describes various best management practices (BMPs) that are utilized to minimize
NPS pollution, but focuses mainly upon the use of wetlands as a pollution controller. A
description detailing the biological and chemical processes in a wetland is also presented,
followed by a general overview of modeling. The concluding section presents specific
descriptions of models previously developed for constructed wetlands.
A. NPS Pollution
The definition of nonpoint source pollution is tied to the definition of point source
pollution. Today’s statutory definition of point sources of pollution is as follows (Water Quality
Act, Sec.502-14, U.S. Congress, 1987):
The term “point source” means any discernible, confined, and discrete conveyance, including but
not limited to any pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling
rock, concentrated animal feeding operation, or vessel to other floating craft from which pollutants
are or may be discharged. This term does not include agricultural stormwater and return flow
from irrigated agriculture.
Nonpoint sources are defined as “everything else” and can be characterized as follows (Novotny
and Olem, 1994):
• Nonpoint discharges enter the receiving water at intermittent intervals in a diffuse manner
and are highly correlated with the occurrence of meteorological events.
• Pollution arises from an extended area of land and is in transit overland before it reaches
receiving waters or infiltrates into shallow aquifers.
• Nonpoint sources lack a specific point of origin.
• Unlike point sources where treatment is the most effective method of pollution control,
prevention of NPS pollution focuses on land and runoff management practices.
• Waste emissions and discharges cannot be measured in terms of effluent limitations
4
• The extent of NPS pollution is related to certain uncontrollable climatic events (rain, floods,
hurricanes, etc.) as well as geographic and geologic conditions.
There are five major forms of NPS pollution: sediments, nutrients, toxic substances,
pathogens, and oxygen demanding substances. Sediments are soil particles carried by runoff into
streams, bays, lakes, and rivers. Nutrients such as nitrogen (N) and phosphorous (P) are
necessary for plant and animal growth, but their usefulness has a plateau after which all excess is
potentially detrimental to the environment. Toxic substances such as pesticides, formaldehydes,
household chemicals, and motor oil, among others could cause human and wildlife health
problems. Pathogens are disease causing microorganisms that are present in animal and human
waste. Oxygen demanding substances decrease dissolved oxygen (DO) concentrations in aquatic
environments through degradation of organic materials.
There are approximately 45 nonpoint sources of pollution identified in the
Commonwealth of Virginia (DCR, 1996). Rivers receive a vast majority of its NPS pollution
impact from farms (64%), urban areas (6%), forest land (6%), and construction areas (6%), as
presented in Figure 1. All other sources of NPS pollution account for only 18% of the total NPS
pollution impact. Therefore, to maximize the use of limited resources (money and people), NPS
pollution control efforts should be directed towards highly contributive areas such as farms,
forest land, urban areas, and construction areas.
Farms 64%
Other Sources 18%
Urban 6%
Forest Land 6%
Construction 6%
FIGURE 1: BREAKDOWN OF NPS POLLUTION EMANATION FOR RIVERS IN VIRGINIAAdapted from DCR (1998)
5
B. Best Management Practices (BMPs)
Methods, measures, or practices for preventing or reducing nonpoint source pollution to a
level compatible with water quality goals are termed BMPs (Novotny and Olem, 1994). By
definition, BMPs must be economically and technically feasible and can be categorized as
structural, vegetative, or management. Selection of BMPs is based on either controlling a known
or suspected type of pollution from reaching a particular source, or to prevent pollution from a
category of land-use activity (such as agricultural row crop farming) (Novotny and Olem, 1994).
Various BMPs exist, but selection of a BMP is dependent upon the particular pollutants
and the forms in which they are being transported. The following process can be used when
selecting which particular BMP to implement (USDA, Soil Conservation Service, 1988):
1) Identify the water quality problem (e.g., eutrophication in a lake).
2) Identify the pollutants contributing to the problem and their probable sources.
3) Determine how each pollutant is delivered to the water source (e.g., runoff from a feedlot).
4) Set a reasonable water quality goal for the resources and determine the level of treatment
needed to meet that goal.
5) Evaluate feasible BMPs for water quality effectiveness, effect on groundwater, economic
feasibility, and suitability of the practice to the site.
Various structural BMPs such as terraces and sediment basins have been developed.
Structural BMPs help control NPS pollution with changes to the landscape that either capture
and contain, or slow pollutant movement. A terrace is an earthen embankment, channel or a
combination of ridges and channels constructed across a slope to intercept runoff (Novotny and
Olem, 1994). Terraces decrease the effective slope of the land, which decreases runoff velocity.
A decreased runoff velocity allows soil particles and adsorbed pollutants to settle out, thus
preventing transport from the field to the receiving water source. Terraces can remove up to 95%
of sediment, up to 90% of sediment’s associated adsorbed nutrients, and between 30% to 70% of
dissolved nutrients (Novotny and Olem, 1994). Sediment basins, sediment control basins, and
detention-retention ponds are earthen embankments that are generally designed as large pools
that control water outflow. These structures retard water flow, allowing heavier particulates to
6
settle out. Sediment basins can remove 40%-87% of the incoming sediment, up to 30% of the
adsorbed N and 40% of the total P (Novotny and Olem, 1994). Detention-retention ponds are
generally more effective than sediment basins due to the uptake of nutrients by associated
vegetation.
Vegetative BMPs include cropping practices, and vegetative filter strips. Cropping
practices such as conservation tillage and cover crops stress maintenance of vegetative cover
during critical times (heavy rains and strong winds) of NPS pollution generation (Novotny and
Olem, 1994). Conservation tillage is any tillage practice that leaves at least 30% of the soil
surface covered with crop residue after planting. Cover crops are close growing legumes,
grasses, or small grain crops that cover the soil during critical erosion periods for the area. Both
practices reduce NPS pollution by reducing erosion through decreased soil detachment, which
also decreases adsorbed pesticide and nutrient movement. Cover crops also store nutrients that
would otherwise be lost during fallow periods. Conservation tillage has been found to be highly
effective in sediment reduction (30-90%), but has very little effect on controlling soluble
nutrients and pesticides (Novotny and Olem, 1994). Cover crops have been found to be 40-60%
effective in reducing sediment, and 30-50% in removing total P (Novotny and Olem, 1994).
Vegetative filter strips utilize strips of closely growing vegetation, such as bunch grasses, sod, or
small grain crops with the primary objective of water quality protection. They are generally
placed between the source of pollution and the receiving water body. Vegetative filter strips are
designed to slow water velocity from sheet runoff and allow sediment and adsorbed pollutants to
deposit. They are effective in removing sediment and sediment-bound N (about 35-90%) but
much less effective in removing P, fine sediment, and soluble nutrients (Novotny and Olem,
1994).
Management BMPs focus on the use of potential pollutants and include integrated pest
management (IPM) and nutrient management. The combination of practices to control crop
pests (insects, diseases, weeds) while minimizing pollution is termed IPM. It works primarily by
decreasing the amount of pesticide or crop-protection chemical available for runoff by choosing
resistant crop varieties, modified planting dates, and selection of the least toxic, least mobile and
least persistent chemicals (Novotny and Olem, 1994). By decreasing the available chemical
amounts, pollution potential is reduced. The effectiveness of IPM is still being debated, with
some estimates being extremely high and others low. Nutrient management works with the same
7
concept of decreasing availability of excess nutrients through improvements in timing,
application rates, and location/selection of fertilizer placement. A more precise application rate
minimizes the potential pollutant availability and has been shown to reduce N and P
concentrations by 20-90% (Novotny and Olem, 1994).
Wetlands are another BMP used for NPS pollution control. This approach is explained in
detail in the following section.
C. Wetlands
Wetlands provide many important ecological functions. Wetlands provide flood storage
and conveyance; stream flow modification; erosion reduction and sediment control; groundwater
recharge/discharge; wildlife habitat; recreation and enjoyment; and pollution control (Novotny
and Olem, 1994). In many aspects, wetlands are excellent BMPs because they provide so many
benefits to the environment and can also be appreciated by wildlife and humans alike. For the
purpose of this study however, the focus will be on wetland’s abilities towards pollution control.
Mitsch and Gosselink (1993) described wetlands as the “kidneys of the landscape.”
Wetlands filter out pollutants and act as sinks for nutrients by purifying the water through
physical (sedimentation, filtration), physical-chemical (adsorption on plants, soil, and organic
substrates), and biochemical processes (biochemical degradation, nitrification, denitrification,
decomposition, and plant uptake) (Novotny and Olem, 1994). The mild slopes of wetlands serve
to slow the velocity of water, which consequently allows sediment and absorbed nutrients to
settle; enhances bacterial die-off due to longer retention times; allows wetland vegetation to
uptake nutrients; and provides a carbon source for microbial action (Novotny and Olem, 1994).
A precise definition which satisfactorily describes all wetland types is not possible due to
the varying types of wetlands (Mitsch and Gosselink, 1993); however, the most comprehensive
definition for wetlands was advanced by the U.S. Fish and Wildlife Service (Cowardin et al.,
1979):
Wetlands are lands transitional between terrestrial and aquatic systems where the water table is usually
at or near the surface or the land is covered by shallow water. Wetlands must have one or more of the
following attributes: (1) at least periodically, the land supports predominately hydrophytes; (2) the
substrate is predominately undrained hydric soils; or (3) the substrate is nonsoil (organic matter) with
water or covered by shallow water at some time during the growing season each year.
8
As seen by this definition, the hydrology, soil type, and vegetation play significant roles
in determining the functionality and effectiveness of wetlands in retaining pollutants. This
significance will be explored more thoroughly in the section dealing with the design of
constructed wetlands.
1. Classification
There are various ways to classify wetlands but a consistent method has not been
developed to describe them. The easiest way to differentiate wetlands are to divide wetlands
between natural and constructed types, but beyond this simplistic categorization, a clear cut
classification scheme for wetlands does not exist. The confusion in terminology stems from the
vast diversity of wetland types that exist throughout the world and the lack of direct equivalent
translations between various languages (Mitsch and Gosselink, 1993).
The U.S. Fish and Wildlife Service (Shaw and Fredine, 1956) developed the first
classification scheme in 1956. In this classification, twenty types of wetlands were described
under the following four categories; 1) inland fresh areas, 2) inland saline areas, 3) coastal
freshwater areas, and 4) coastal saline areas. Presently, the classification scheme used in the
United States, as part of the National Wetlands Inventory (Cowardin et al., 1979) is very formal
and all encompassing, but very difficult to use. The classification system is based on a
taxonomic separation scheme, in which all wetland and deep-water habitats are divided into five
systems (marine, estuarine, riverine, lacustrine, and palustrine), and further subdivided into
various subsystems and classes. Mitsch and Gosselink (1993) divide wetland types into two
initial systems (coastal and inland) and then further subdivide these systems into seven separate
categories that encompass most, but not all wetland types.
a. Natural Wetlands
Natural wetlands originate in geological settings due to water movement and
accumulation. The major geological settings in which wetlands form are areas of 1) slope
discontinuity, 2) topographic depression, 3) stratigraphic features which inhibit infiltration, and
4) permafrost (Widener, 1995). Wetlands that are formed in lowland areas tend to be underlain
by glacial outwash, clay and silt, or alluvial outwash comprised of sand or a mixture of sand and
9
TABLE 1: NUTRIENT REMOVAL RATES FOR NATURAL WETLAND SITES RECEIVING WASTEWATER INPUTS
Loading Nutrient RemovalType of (Population (percent)Wetland Location /Hectare) Substrate Total N Total PNorthern PeatlandBog Wisconsin 30 O 98 78
Nontidal freshwatermarshCattail marsh Wisconsin 17 O 80 88Lacustrine marsh Ontario n/a n/a 38 24Deepwater marsh Florida 99 O n/a 97Lacustrine marsh Hungary n/a n/a 95 n/aRiverine swamp South Carolina n/a O n/a 50
Tidal freshwater marshDeepwater marsh Louisiana n/a O 51 53Complex marsh New Jersey 198 I 40 0
Tidal salt marshBrackish marsh Chesapeake bay n/a O/I 0 1.5Salt marsh Georgia Sludge O/I 50 n/aSalt marsh Massachusetts Sludge O/I 85 n/aSource: Compiled by Mitsch and Gosselink (1986)Note: O= organic substrate; I= inorganic substrate; n/a= information not availablea Load given in g/m2-year
gravel, while wetlands formed in upland areas tend to be underlain by bedrock and glacial till
(Baker, 1973). Mitsch and Gosselink (1986) compiled data on the performance of natural
wetlands for removal of nutrients. As indicated in Table 1, retention of nutrients varies greatly
among different areas. This variability complicates modeling of wetland processes as further
explained in the modeling section.
b. Constructed Wetlands
Constructed wetlands are man-made systems designed to imitate the functions of natural
wetland systems. There are two fundamental types of constructed wetlands, the free water
surface (FWS) system, and the subsurface flow system (SSF) (Novotny and Olem, 1994). The
FWS system usually consists of basins or channels with a natural or subsurface barrier of clay or
impervious geotechnical “lining” to prevent seepage (U.S. EPA, 1988). The basins are then
filled with soils to support the accompanying planted vegetation (Figure 2). The water level in a
10
FWS wetland is above the soil substrate with water flow occurring primarily above ground. A
SSF system consists of a trench or bed underlain with an impermeable layer of clay. The trench
is back filled with media that usually consists of crushed stone, rock fill, gravel, and different
soils. Water flows through the medium and is purified through filtration; absorption by
microorganisms; and adsorption onto soils, organic matters, and plant roots (U.S. EPA, 1988)
(Figure 3). Hence, the performance of the wetland depends on the detention time of incoming
pollutants, the loading rates, the biotic condition within the system, and oxygen availability.
2. Constructed Wetland Design
Hydrology is the most important wetland design variable. With proper hydrologic
conditions, the potential chemical and biological elements necessary for a properly functioning
wetland exist. Hydrologic conditions can directly modify or change physical and chemical
properties, such as soil salinity, pH, sediment properties, substrate anoxia, and nutrient
availability (Mitsch and Gosselink, 1993). Hydrology is less forgiving than other biological
components, and if improperly accounted for, can cause a constructed wetland to fail.
FIGURE 2: CROSS SECTION OF A FWS WETLAND.Adapted from Novotny and Olem (1994)
11
FIGURE 3: CROSS SECTION OF A TYPICAL SUBSURFACE FLOW WETLAND.Adapted from EPA (1988)
Ultimately, the hydrologic conditions determine success of a wetland system, for it determines
the depth, residence time, and hydroperiod. The hydraulic residence time is the average length
of time a volume of water is detained in a wetland before exiting the system (Novotny and Olem,
1994), and can be estimated as:
Q
VpHRT
*= (1)
where HRT is the hydraulic residence time for a FWS system (T); p is the porosity ((ratio of
water volume)/(total volume); 0.9-1.0 for FWS); V is the active volume of the wetland (L3); and
Q is the average flow rate (L3/T).
The hydroperiod is the seasonal pattern of water level in a wetland or the water depth
above or below wetland surface level over time (Mitsch and Gosselink, 1993). The hydroperiod
is the dominant factor controlling the plant community composition of wetlands (Duever, 1988).
When hydrologic conditions in a wetland change even slightly, the biota may respond with
massive changes in species richness, composition, and ecosystem productivity.
12
The hydrologic conditions for a wetland are affected by various inputs, outputs and
storage patterns. The general balance between water storage and the outflows and inflows can
best be expressed with the following equation (Kadlec, 1996):
AETPQQQQQQdt
dVgwbosmci )( −+−−−++= (2)
where A is the wetland surface area (L2); ET is the evapotranspiration rate (L/T); P is the
precipitation rate (L/T); Qb is the bank loss rate (L3/T); Qc is the catchment runoff rate (L3/T);
Qgw is the percolation to groundwater (L3/T); Qi is the input stream flow rate (L3/T); Qo is the
output stream flow rate (L3/T); Qsm is the snowmelt rate (L3/T); t is the time step (T); and V is
the volume of water storage in wetland (L3).
The underlying soil strata play a very important role in wetland development. It
functions both as the medium in which many of the wetland chemical transformations take place
and as the primary storage of available chemicals for wetland vegetation (Mitsch and Gosselink,
1993). The soil is often described as hydric, defined by the U.S. Soil Conservation Service
(1987) as “a soil that is saturated, flooded, or ponded long enough during the growing season to
develop anaerobic conditions in the upper part.” Wetland soils usually have very high organic
matter content. Highly permeable soils are not suitable for wetlands that are not fed by
groundwater because a high permeability does not allow sufficient water storage for hydric soil
conditions to establish. Permeability must be kept below a certain threshold value, which may
vary according to site-specific and geographic conditions (Novotny and Olem, 1994).
Wetlands plants may be characterized as “submersed” (i.e., completely submerged),
“emergent” (i.e., those plants with a root system and stem below the water, but which reaches to
or above the surface), or “terrestrial” (land based) (Dennison and Berry, 1993). Due to the
anoxia, wide salinity range, and water fluctuations characteristic of an environment that is
neither aquatic nor terrestrial, wetland conditions can be physiologically harsh. The constant
fluctuations in living environment can be taxing to organisms as the changing conditions requires
limited energy supplies to be directed toward growth, and more towards survival practices.
Aquatic organisms can not easily adjust to the periodic drying that occurs in many wetlands and
terrestrial organisms could become stressed by long periods of flooding (Mitsch and Gosselink,
13
1993). To deal with anoxia, wetland plants have developed aerenchyma, or air spaces that run
from the stems to the roots, allowing the diffusion of oxygen from the aerial portions of the
plants to the roots. This adaptation allows plants to generate the required energy needed for
survival (Mitsch and Gosselink, 1993). Other adaptations are used by the species of woody trees
(mangroves, cypress, tupelo, willow and a few others) that have successfully adapted to the
wetland environment. Many woody trees have developed adventitious roots above the anoxic
zone, which allow them to attain the necessary air diffusion requirements for biological
processes. A whole plant strategy adopted by many wetland plants concerns the timing of seed
production and transport. Seed production occurs in the nonflooding season and is accompanied
by either delayed or accelerated flowering (Bloom et al., 1990); the production of buoyant seeds
that float until they lodge on unflooded, higher ground; and seed germination while fruit is still
attached to the trees (Mitsch and Gosselink, 1993). All of these mechanisms increase the
probability of plant survival in a wetland environment. Table 2 lists the general depth and
hydroperiod for selected wetland plant communities.
TABLE 2: GENERAL HYDROPERIOD TOLERANCE RANGES FOR SELECTED WETLAND PLANT COMMUNITIESAverage Water Average
Wetland Type Typical Species Depth (m) Hydroperiod *Floating Deep Hyacinths, pennywort
Floating rooted Water lily, water dock, 0.5 -02 70-100 aquatic water shield
Submerged hydrills, egeria, water 0.5-3.0 80-100 aquatic millfoil, naiad
Emergent Cattails, pickelrelweed, 0.1-1 40-100 marsh bulrush, sedgem maidencane
Floodplain Red maple, black gum, cabbage, 0.2-0.3 10 to 50 palm, pond cypress, oaks, pines, bald cypress, ash
Swamp Forest Bald cypress, ash, black gum, 0.3-1.0 50-80 tupelo, gum, red
lCypress dome Pond cypress, red maple, black 0.1-0.3 50-75
gum, dahoon holly
Wet prairie St.Johns wort iris, sagittaria 0.1-0.2 20-50
* The average % of the year the wetland water surface is above wetland ground level.Source: Adapted from Novotny and Olem (1994)
14
Constructed wetlands, as compared with natural wetlands, provide a better chance for
management and control of NPS pollution for two reasons; 1) government regulations, and 2)
location. In the Unites States, natural wetlands are considered natural receiving surface-water
bodies like oceans and lakes; hence they are protected from excessive pollution discharges, and
any discharge requires a permit (Novotny and Olem, 1994). There are limits on how much
pollution can be released to a wetland and this consequently reduces its use for water treatment.
Unlike natural wetlands, constructed wetlands do not have these restrictions placed upon them
and can therefore receive higher pollutant loadings for treatment. Consequently, constructed
wetlands are used more often for water quality improvement. In addition, constructed wetlands
can be created wherever the proper hydrologic, chemical and biological requirements can be
established. This allows constructed wetland systems to be more flexible for NPS pollution
treatment for they can be created where water treatment is necessary.
Novotny and Olem (1994) have summarized the basic principles of wetland design:
1. Design the system for minimum maintenance, where the system of plants, animals, microbes,
substrate and water flows are self-maintaining.
2. Design a system that utilizes natural energies, such as gravity flow and the potential energy
of streams.
3. Consider the landscape for system design. Do not overengineer wetland design with
unnatural basin shape, structures, uniform depths, and regular morphology. Try to mimic
nature.
4. Design the entire system as an ecotone, including the use of buffer strips around the site.
5. Consider the surrounding lands and future land-use changes.
6. Hydrologic conditions are paramount. A detailed surface and groundwater study is
necessary.
7. Give the system time to develop. Wetlands are not created overnight.
8. Soil surveys should be conducted, as highly permeable soils do not support wetland systems.
Table 3 lists wetland design parameters for constructed wetlands and compares them to
natural systems.
15
TABLE 3: WETLAND DESIGN PARAMETERS
Constructed ConstructedFWS SFS Natural
Minimum Size
requirement 2 to 4 1.2 to 17 5 to 10
(ha/1000m3/d)
Hydraulic Loading 2.5 to 5 5.8 to 8.3 1 to 2
(cm/day)
Maximum water 50 water level below 50; depend on
depth (cm) ground surface native vegetation
Bed depth (cm) n/a 30 to 90 n/a
Minimum hydraulic
residence time (days) 5 to 10 5 to 10 14
Minimum aspect 2 to 1 n/a 1 to 4
ratio
Minimum Primary; secondary Primary Primary; secondary;
pretreatment is optional nitrification; TP
reduction
Configuration Multiple Cells in Multiple beds in multiple discharge
parallel and series parallel series
Distribution swale, perforated Inlet zone (0.5m) swale, perforated
pipe of large gravel pipe
Maximum Loading,
(kg/ha-day)
BOD5 100 to 110 80 to 120 4
Suspended Solids up to 150
TKN 10 to 60 10 to 60 3
Phosphorous ? ? 0.3 to 0.4
Additional Mosquito control Allow flooding Natural hydroperiod
Consideration with mosquitofish; capability for should be >50%; no
remove vegetation weed control vegetation harvest
Source: Novotny and Olem (1994).
3. Nitrogen Cycle in Wetlands
The transformations and interactions of the various forms of N in soils, sediment of
surface waters, and substrates of wetlands is very complex. The basic forms of N in soils and
sediments are ammonium ion (NH4+), nitrate (NO3
-), organic phytonitrogen in plants and plant
residues, and protein N in living and dead bacteria (Novotny and Olem, 1994). As a negatively
16
charge ion, NO3- is not subject to adsorption by negatively charged soil particles like the
positively charged NH4+ ion, and is thus more mobile in solution. In flooded soils and sediments,
the organic forms of N predominate, while NH4+ is the predominant inorganic N form (Reddy
and Patrick, 1984). Some researchers refer to N content in an area as either Total Kjeldahl N
(TKN) or as total N (TN). Total Kjeldahl N is a measure of reduced N equal to the sum of
organic N and NH4+-N (Kadlec and Knight, 1996). Total N is a measure of all organic and
inorganic forms and is essentially equal to the sum of TKN, NO3- and NO2-N (Kadlec and
Knight, 1996).
Sources of N that contribute to wetland sites include: a) precipitation on the surface of
flooded soils and sediments; b) N fixation in the water and the sediments; c) inputs from surface
and ground water infiltration/percolation; d) application of fertilizers; e) N release during
decomposition of dead aquatic vegetation and animal community inputs; and f) discharge of
waste water effluents (Reddy and Patrick, 1984).
A number of processes can transport or translocate N compounds from one point in a
wetland to another without molecular transformation. These transfer processes are physical in
nature and include: 1) particulate settling and resuspension, 2) diffusion of dissolved forms,
3) litterfall, 4) plant uptake and translocation, 5) NH3 volatilization, 6) sorption of soluble N on
substrates, 7) seed release, and 8) organism migrations (Kadlec and Knight, 1996).
Important processes that transform the basic forms of N in soils and sediments are
presented in Figure 4. These processes are mineralization (ammonification), nitrification,
denitrification, nitrogen (N2) fixation, and assimilation (plant and bacterial uptake).
Understanding the N transfer and transformation processes is very important to the design
of a wetland system. If these processes are not understood, the design of constructed wetland
systems will be negatively affected. The following sections describe the transformations and
transport processes of the N cycle in further detail.
17
FIGURE 4: NITROGEN TRANSFORMATIONS IN WETLANDS.SON =soluble organic nitrogen. Adapted from Mitsch and Gosselink (1993).
a. Nitrogen Transformation Processes
i. Mineralization (ammonification)
Mineralization is the biological transformation of organic N to NH4+ that occurs during
organic matter degradation (Gambrell and Patrick, 1978). Mineralization occurs through
microbial breakdown of organic tissues containing amino acids, hydrolysis of urea and uric acid,
and through excretion of ammonia directly by plants and animals (Kadlec and Knight, 1996).
Mineralization occurs under both anaerobic and aerobic conditions but proceeds at a slower rate
in anaerobic conditions due to the decreased efficiency of heterotrophic bacteria in these
environments (Reddy and Patrick, 1984).
The mineralization rate is affected by temperature, pH, carbon to nitrogen (C:N) ratio of
the substrate, available nutrients in the soil, and soil properties such as texture and structure
(Reddy and Patrick, 1988). The effect of these factors on mineralization in well-drained soils is
fairly well understood, but less is known about their effects in flooded soils. Reddy et al. (1979)
18
concluded that the rate of mineralization doubles with a temperature increase of 10 °C, while the
optimum temperature of mineralization was found to be between 40 to 60 °C (Reddy and
Patrick, 1984), a rare field condition. The optimal pH range for the mineralization process is
between 6.5 and 8.5 (Reddy and Patrick, 1984), a condition found under most flooded conditions
because the oxidation of organic material produces CO2, which buffers the system.
Measured mineralization rates in natural wetlands range from 0.3 to 35 mg N/m2/d
(annual average of 1.5 g/m2/yr)) in a swamp forest in central Minnesota (Zak and Grigal, 1991),
and 4.3 to 5.9 g/m2/yr in a Minnesota bog (Urban and Eisenrich, 1988). Higher rates were
reported in organic soils in Florida by Reddy (1982), with rates of 41 to 125 g/m2/yr.
ii. Nitrification
After NH4+ ions are formed through the mineralization process, it can take several
pathways. It can be absorbed by plant root systems or taken up by anaerobic microorganisms
and converted to organic matter; immobilized through ion exchange by soil particles; or it can
undergo nitrification (Mitsch and Gosselink, 1993).
Nitrification is the biological oxidation of ammonium-N to nitrate-N with nitrite-N
(NO2-) as an intermediate product. Nitrification is accomplished with the help of two groups of
chemoautotrophic bacteria that allow the oxidation process to occur. The first step (Mitsch and
Gosselink, 1993):
energyHOHNOONH +++→+ +−− 42232 2224 (3)
is accomplished with the Nitrosomonas sp. The second step:
energyNOONO +→+ −−322 22 (4)
is conducted by the Nitrobacter sp.
Anaerobic conditions in wetland soils limit the amount of nitrification that can occur, as
nitrification requires oxygen. In a wetland system, nitrification can occur in; 1) the water
column above wet soils (Reddy and Patrick, 1984), 2) the thin oxidized layer at the surface of
19
wetland soils, and 3) the oxidized rhizosphere of plants (Mitsch and Gosselink, 1993).
Nitrification can still occur at low levels of about 0.3 mg/L of DO (Reddy and Patrick, 1984).
iii. Denitrification
As stated before, NO3- is far more mobile in solution than NH4
+. If NO3- is not
assimilated by plants or microbes or lost to groundwater flow through rapid movement,
denitrification may occur. Denitrification is the biological reduction of NO3--N to gaseous N
forms such as molecular N2, NO, NO2 and N2O (Novotny and Olem, 1994). Under anaerobic
(oxygen free) conditions and in the presence of available organic (carbon) substrate, denitrifying
organisms such as bacillus, micrococcus, alcaligenes, and spirillum, can use NO3- as an electron
acceptor during respiration. These organisms oxidize a carbohydrate substrate by converting
NO3- to carbon dioxide, water, N gas and other gaseous oxides that can result from denitrification
as indicated above (Reddy and Patrick, 1984):
OHNCOHNOOCH 22232 725445 ++→++ + (5)
This chemical reaction is irreversible in natural conditions.
Several factors are known to influence the rate of denitrification including the absence of
O2; presence of readily available C; temperature; soil moisture; pH; presence of denitrifiers; soil
texture; and presence of overlying floodwater (Reddy and Patrick, 1984). Denitrification rate
has been shown to increase with temperature and researchers (Reddy and Patrick, 1984) have
concluded that a 1.5 to 2.0 fold increase will occur with a 10 °C rise in temperature.
iv. Nitrogen Fixation
Nitrogen fixation is the process by which atmospheric N2 gas diffuses into solution and is
reduced to organic N by autotrophic and heterotrophic bacteria, blue-green algae, and higher
plants (Kadlec and Knight, 1996). N fixation is an adaptive process that provides N for
organisms to grow in conditions that are otherwise depleted of N. N fixation is inhibited by high
concentrations of available N; and is generally not observed in N rich ecosystems.
20
In wetlands, N fixation can occur in overlying waters, in the anaerobic or aerobic soils
layers, in the oxidized rhizosphere of the plants and on the leaves and stem surface of plants
(Mitsch and Gosselink, 1993). Observations of N fixation values vary greatly from differing
wetland sites. Dierberg and Brezonik (1984) observed fixation rates ranging from 1.2 to 19.0
kg/ha/yr in a Florida cypress dome receiving municipal wastewater, but fixation was concluded
to be an insignificant contributor to total N loading.
v. Assimilation: Plant and Bacterial Uptake
Nitrogen assimilation refers to a variety of biological processes that convert inorganic N
forms into organic compounds that serve as building blocks for cells and tissues (Kadlec and
Knight, 1996). The two most commonly used forms of N are NH4+-N and NO3
--N. NH4+ is
more reduced energetically than NO3-, thus it is the more preferred source for assimilation by
plants and bacteria.
Depending upon the loading rate to the wetland, plant N assimilation can involve a
significant fraction of the total N load. Adcock et al. (1994) determined that a SSF treatment
wetland in Australia had 65% of the N load contained in macrophyte biomass due to its low N
loading rate (25 to 40 g/m2/yr). At sites with higher loading rates, the amount of N lost to
assimilation is a smaller overall percentage.
In temperate climates, plant assimilation is a spring-summer phenomenon. Depending on
location, plant species can either be sinks or sources of N. During the spring and summer when
growth is taking place, plants uptake N, but during the winter months when vegetation dies,
uptake ceases and decomposition occurs.
Microorganisms assimilate nutrients for growth, as NH4+ is readily incorporated into
amino acids by many autotrophs and microbial heterotrophs (Kadlec and Knight, 1996). The
amino acids are transformed into proteins, purines, and pyramidines that are used as energy. The
magnitude of the uptake process has not been quantified for treatment wetlands (Kadlec and
Knight, 1996).
21
b. Other Nitrogen Fluxes
There are numerous other pathways that N compounds can follow besides the previously
described molecular transformations. These processes may be important when designing
wetland systems and can contribute or subtract from the TN content of a wetland system. These
processes include (1) atmospheric N inputs through rainfall and dryfall, (2) NH3 volatilization,
(3) NH4+ adsorption, (4) burial of organic N, and (5) biomass decomposition (Kadlec and Knight,
1996). Brief descriptions of each process follow.
i. Atmospheric Nitrogen Inputs
Atmospheric deposition of N contributes measurable quantities of N to land areas. All
forms of N are involved including particulate, dissolved, inorganic and organic. Wetfall (rain or
snow) contributes more than dryfall, and rain contributes more than snow (Kadlec and Knight,
1996).
Nitrogen concentrations in rainfall are highly variable and dependent on atmospheric
conditions, air pollution and geographic location. A typical range of TN concentrations
associated with rainfall is 0.5 to 2.0 mg/L, with about 50% of this present as NO3- and NH3-N
(Kadlec and Knight, 1996). Atmospheric sources are usually negligible contributors to the
overall wetland N budget.
ii. Ammonia Volatilization
Un-ionized NH3 is relatively volatile and can be removed through mass transfer of NH3
from the water surface to the atmosphere (Kadlec and Knight, 1996). Volatilization has limited
importance for wetlands. Volatilization practically ceases if pH is at or below 7 (Novotny and
Olem, 1994). Typically, volatilization is an insignificant factor when discussing the N cycle in
wetlands. However, in wetlands with a high concentration of NH3-N (20mg/L) and a pH greater
than 8, volatilization can play a significant role (Kadlec and Knight, 1996).
22
iii. Adsorption
Adsorption is the adherence of chemical ions to the surface of a solid. NH4+ can be
removed from solution through a cation exchange adsorption reaction with inorganic sediments
and detritus (Kadlec and Knight, 1996). The adsorbed NH4+ is loosely bound to the substrate
and can be released when water chemistry conditions change. Most forms of N are very soluble
and do not attach to sediment and other particle types; therefore adsorption plays a limited role in
the overall N balance.
iv. Burial of Organic Nitrogen
A fraction of the organic N incorporated in detritus and plants may eventually become
unavailable for additional nutrient cycling due to burial and peat formation. Burial of N can be
important for light N loading conditions, but becomes insignificant for high N loads (Kadlec and
Knight, 1996). For example, Reddy et al. (1991) reported a N burial rate of 14 to 34 g N/m2/yr
for a lightly fertilized zone of wetland, while the N burial rate was 365 g N/m2/yr in a treatment
wetland.
v. Biomass Decomposition
The N that is assimilated by macrophytes, microflora, and microfauna is partially
released during decomposition. Turnover times for leaf litter can vary from several months to
over 2 years in colder climates, but decomposition rates during warmer months do not vary much
with geographical conditions (Kadlec and Knight, 1996). The decomposition process is typified
by a rapid initial weight loss that is followed by an exponential loss of the remaining weight to
an irreducible residual which contributes to sediment and soil building (Kadlec and Knight,
1996).
4. Phosphorous Cycle in Wetlands
Due to the general scarcity of P in the natural environment and the absence of significant
atmospheric inputs, natural ecosystems such as wetlands, have numerous adaptations to
23
sequester this element (Kadlec and Knight, 1996). P is rendered relatively unavailable to
microconsumers and plants when (Mitsch and Gosselink, 1993): a) insoluble phosphates
precipitate with ferric iron, calcium, and aluminum under aerobic conditions; b) chemical
sorption of phosphate to clay particles, organic peat, and other minerals occurs; and c) P
incorporates into the living biomass of wetland biota. Phosphorous is not particularly mobile in
soils and phosphate ions do not readily leach, thus P transport is mostly from plant uptake or
through soil transport (Novotny and Olem, 1994). Figure 5 details the basic transport modes and
reactions for P in a wetland.
Phosphorous occurs as insoluble and soluble complexes in both organic and inorganic
forms in wetland soils. The principal inorganic form is orthophosphate, which includes the ions
PO4-3, HPO4
=, and H2PO4- (Mitsch and Gosselink, 1993). The phosphorous cycle is sedimentary
rather than gaseous (i.e., N); therefore, commonly a major portion of a wetland’s P content is tied
up in organic peat and litter and in sediment (Mitsch and Gosselink, 1993). Removal efficiencies
range from 0 to 90% (Watson et al., 1989).
a. Importance of Sediment – Sorption/Desorption
As stated before, the P cycle is sedimentary-based, therefore, sediment movement plays a
vital role in determining P transport and concentrations. Dissolved P in both inorganic and
organic forms usually interacts with suspended and bed sediments. Many of these interactions
are heterogeneous in nature and it is therefore likely that the kinetics of the processes rather than
the chemical equilibrium determine the P division (Grobbelaar, and House, 1995). The nature of
specific interactions for many systems is still unknown, because (Grobbelaar and House, 1995):
• The wide range of affinities of P for sediments, combined with the uncertainties in
sedimentary materials composition makes it difficult to identify the key processes,
• Dissolution/precipitation, adsorption/desorption and biological uptake and release are
difficult to separate for measurement purposes, and
• The transformations of organic P to inorganic P are not well understood.
24
FIGURE 5: PHOSPHORUS TRANSFORMATIONS IN WETLANDS.SOP = soluble organic phosphorous. Adapted from Mitsch and Gosselink, 1993.
In many wetlands, P cycling tends to follow sediment deposition and resuspension. This
is due to the high sorption rates associated with P. However, there is a common misconception
that wetlands provide P removal only through sorption processes on settling sediments.
Although most sediments do have sorptive capacity for P, this storage will become saturated
under constant P loading rates (Kadlec and Knight, 1996).
b. Precipitation
Precipitation of P in wetland systems is very complicated and is highly dependent on pH
in the system. At higher pH values, the P precipitates mostly in combination with calcium.
Below a pH of 7, which is characteristic of soils with high clay and organic matter (such as
wetlands), P reacts predominately with the iron and aluminum ions in soils (Novotny and Olem,
1994). Depending on soil pH, the dissolved P concentrations may decrease to values of 0.01
mg/L or less.
25
c. Biomass: Growth, Death, Decomposition, Uptake and Storage
The amount of P sustainably removed by a wetland is usually much less than the P taken
up by plants during a growing season. All wetland biota undergo a constant cycle of growth,
death and partial decomposition. This results in the decay of plant life and the subsequent
release of assimilated P. Therefore, increases in biomass should not be counted towards the
long-term sustainable P removal capacity of wetlands (Kadlec and Knight, 1996). Although
plants may temporarily remove P from the wetland water and soils, in the long term, it provides
very little retention.
Determining P removal is dependent upon the accretion of biomass residuals and
minerals because this is the only sustainable storage mechanism for P removal (Kadlec and
Knight, 1996). Burial of material removes P from the plant growth/death cycle; therefore the
more plant growth/death cycles, the more chances for burial. Turnover rate is defined as the
number of times the above ground biomass is replaced per year. In northern climates the
turnover rate is lower than in southern climates because southern areas have a longer growing
season (Kadlec and Knight, 1996). Since turnover rate is higher in southern climates, there are
increased chances for accretion and a higher probability of nutrient retention.
5. Bacteria in Wetlands
Many nutrient transformations in wetlands are due to microbial metabolism and are
directly related to microbial growth (Tanji, 1982). There are theories that state that
decomposition and ammonification rates are linked to microbial energy requirements, the C:N
ratio of the organic matter and the growth rate of microbes in the substrate (Parnas, 1975; Fyock ,
1977; Patrick, 1982). Nitrogen and C are both necessary as a source of energy, while C is
required for building microbial biomass (Parnas, 1975). Growth rates of microbes are a function
of both the environmental conditions and substrate availability.
Energy is obtained by the transference of electrons from an electron donor to an electron
acceptor. Examples of electron donors would be complex organics and NH4+, while oxygen and
NO3- are acceptable electron acceptors (Gidley, 1995). Most of the treatment in wetlands is due
to heterotrophic and autotrophic bacteria (Mitsch and Jorgensen, 1989). Particulate and soluble
labile organics are used as a C source and electron donor by heterotrophic bacteria (Gidley,
26
1995). Equations 3, 4, and 5 show how the microbial transformations generate energy, whose
yield differs for each process. Aerobic degradation of organic materials yields more energy per
mass of electron donor, than either organics degradation or nitrification.
Microbes also utilize N and C to build cell mass. A common formula for microbes is
C5H7O2N (Parnas, 1975). Nitrogen comprises more than 12% of cells, while C accounts for
more than 50% of cell mass. Since microbes use C and N organics; growth of heterotrophs are
influenced by the C:N ratio of the materials they degrade ( Reddy and Patrick, 1983). Aerobic
heterotrophs require organics with a C:N ratio of about 23.5 (Parnas, 1975). Part of the C is used
as an electron donor, while the rest is incorporated into cell mass. Anaerobic decomposition is
not as efficient, therefore more C is required to generate equal amounts of energy. Anaerobic
heterotrophs optimize organic use at a C:N ratio of about 80. Consequently, ammonification is
greater under anaerobic conditions (Reddy and Patrick, 1983). If the C:N ratio is lower than 23.5
or 80, for aerobic and anaerobic conditions respectively, growth will be C limited and the excess
N is wasted as NH4+. If the C:N ratio is higher than these ratios, growth is N limited and the C:N
ratio of the organic materials increase as N is incorporated into cell mass. If the excess N is
NH4+, microbes utilize NH4
+ and the C:N ratio remain the same (Parnas, 1975).
Microbial growth rate is determined by the availability of electron donors and acceptors,
the amounts of C and N, and environmental conditions (temperature, pH, space, etc.) (Grady and
Lim, 1980; Reddy and Patrick, 1983). While heterotrophs are responsible for ammonification,
nitrification is inhibited when the DO concentrations drop below 2 mg/L (Bowmer, 1987).
Conversely, the rate of denitrification is reduced in the presence of oxygen.
Optimal conditions for bacterial growth are generally reported as being between a pH of
six and nine, and at temperatures ranging from 15 °C and 40°C (Fyock, 1977; Reddy and Patrick,
1983; Bruno and Tomasso, 1991). Growth of microbes still occurs outside of these ranges but
the rates are reduced (Broderick et al., 1988). The pH of submerged soils is generally neutral
because the oxidation of organic material produces CO2, which buffers the system (Reddy and
Patrick, 1983). When organic loading is high, heterotrophs out-compete autotrophs and
nitrification is reduced (Grady and Lim, 1980).
The N and C cycles in a wetland are not mutually exclusive as other bacteria may require
the byproducts of one microbial process. For example, heterotrophic bacteria obtain energy from
organics and produce NH4+, which is in turn used by autotrophs as an energy source. The NO3
-
27
formed by the aerobic heterotrophs is then used by anaerobic heterotrophs as an electron
acceptor (Gidley, 1995). Heterotrophs rely on plants to provide organic substrates and a suitable
environment for survival, while plants are dependent on microbial decomposition for nutrient
recycling (Reed and Brown, 1992). At the same time though, plants and microbes both compete
for nutrients during the growing season (Good and Patrick, 1986). All of these interactions form
a complex system that is difficult to manage, model and recreate (Gidley, 1995).
6. Vegetative/Carbon Cycle in Wetlands
The C cycle in wetlands is dominated by wetland’s plant life. Wetland plants follow a
cycle of growth and nutrient uptake, death, and lastly, decomposition, nutrient release, and soil
accretion (Gidley, 1995). Plant growth and death follow seasonal patterns, while processes such
as decomposition and soil accumulation may take years. Wetlands usually have a seasonal
pattern of nutrient retention in the summer, followed by a nutrient release in the fall and early
spring floods when lower temperatures reduce biological activity (Mitsch and Jorgensen, 1989;
Hantzsche, 1985).
During the summer, vegetation grows and uptakes nutrients. Boyd (1978) found the
mean C and N content were 45% and 1.01%, respectively, for Typha latifolia, and 48% and
1.36% for Juncus effusus, in a natural wetland. Tanner (1996) found the range of N and P
content were 1.5% to 3.2% and 0.13% to 0.34%, respectively, for eight emergent plant species in
a constructed wetland. The chemical composition in wetland biomass varies among plant parts
and changes seasonally; therefore, it is difficult to determine an average nutrient concentration.
Younger plants that have grown in nutrient enriched environments have the highest nutrient
content and aboveground parts usually have higher concentrations than below ground parts
(Heliotis and DeWitt, 1983; Mitsch and Jorgensen, 1989). Kadlec (1989) estimated the typical
N content of wetland biomass to be 2% of the dry weight. The primary productivity in wetlands
is greater than the best agricultural land (Hammer, 1986). The average above ground biomass
growth rates ranged from 0.003 m2-day at Porter Ranch peatland in Michigan (Hammer, 1984) to
0.04 m2-day of total growth of Typha Latifolia in a cultivated northern U.S. peatland on a dry
weight basis (DeBusk and Ryther, 1986).
Nutrients stored in wetland vegetation represent only a small fraction of nutrient input to
wetlands, despite the high primary productivity rates (Mitsch and Gosselink, 1993). The soil
28
nutrient stock is often 1-2 times higher than the biomass nutrient stock (Johnston, 1991).
Regardless, the small amount that is assimilated by biomass is returned to the wetland system at
the end of the growing season as emergent vegetation dies and is decomposed by microbes.
Over the winter and spring the standing dead plants fall to the ground and become litter.
There is an initial period during which nutrients leach from the dead vegetation (Heliotis and
DeWitt, 1983; Kadlec, 1986). Polunin (1982) observed a 13% initial weight loss during a
decomposition study in England, and attributed this degradation to physical factors since the loss
was unaffected by biological inhibitors.
Several researchers have described microbial decomposition as a two-stage process
(Kadlec, 1989b; Kulshrestha and Gopal, 1982). The first stage is an initial rapid weight loss over
the first 30-60 days that is caused by the biological utilization of starches and sugars. The loss of
C from litter is rapid and can exceed initial loss of total organic N and P (Morris and Bowden,
1986). The second stage is an exponential decomposition with the biological release of
additional nutrients. Litter degradation is dependent on particle refractability, microbial growth
and the availability of electron acceptors and nutrients (Heliotis and Dewitt, 1983).
A portion of the biomass is resistant to decomposition and accumulates as soil. Jansson
and Persson (1982) have described this soil organic matter as a heterogeneous mixture. There is
little information regarding the rate of this soil formation, although soil accretion rates for the
Houghton Lake wetland were measured at 2.5 mm/yr using carbon dating (Kadlec, 1989b).
7. Modeling Wetland Processes
Modeling wetland processes is relatively new as compared to other ecosystems (Mitsch et
al., 1988). As the interest in wetlands has increased, so has the interest to model the processes
that occur within a wetland. A comprehensive understanding of processes is desired, so that the
construction of replacement wetlands will be equivalent or even better than natural wetlands; be
that for NPS pollution control or the creation of wildlife habitat.
A tremendous amount of wetland modeling has been drawn from previous works on
lakes and other large water bodies. This section will explore general modeling approaches and
techniques, present the specific modeling processes for wetlands used by researchers, and
conclude with descriptions of a few existing wetland models.
29
a. General Modeling Practices
In general, there are certain approaches modelers adopt to represent a system. Simple
classification systems can designate a model as either empirical or theoretical; lumped or
distributed; and steady state or dynamic.
Empirical models are functional relationships defined in terms of statistical analysis of
observed data, while theoretical models are functional relationships defined from physical laws
and relationships (Heatwole, 1998). Empirical models are based on site-specific data and
therefore may not be applicable to different areas. Theoretical models can, hypothetically, be
used without calibration, but this is rarely the case. Although a single model representing a
single process can not be both empirical and theoretical at the same time, there can be many
relationships combined to form one large system model that contains both empirical and
theoretical relationships.
Distributed models consider parameters as functions of time and space. They seek to
represent the spatial differences and relationships of the physical system. Lumped models leave
parameters within a range of prescribed spatial locations and/or time. A lumped model is more
of a “conceptual” model and spatial relationships are not represented (Heatwole, 1998).
Distributed models are more difficult to create and implement; however, they are necessary for
larger non-uniform areas because too much generalization and grouping of a represented area
will detrimentally affect the results.
A system can also be represented as either steady state or dynamic. For a steady state
model, the variables defining the system are not dependent upon time. A dynamic model on the
other hand has variables defining the system being a function of time (Jorgensen, 1983).
Table 4 exhibits a partial list of existing wetland models. As shown, the approaches for
modeling wetland areas vary greatly. Each of these models vary in complexity (various input
requirements, differing modeled processes), subsequently some are more difficult to apply than
others. The common assumption would be that a more complex model would better represent
the data of a system because more input usually equates with better results, yet this is not
necessarily the case. Costanza and Sklar (1985) reviewed 87 mathematical wetland models and
rated them on their articulation, accuracy and effectiveness. Articulation measures the size and
30
complexity of the model, accuracy measures the goodness-of-fit (comparison of differences
between observed field data and model generated output), while effectiveness is a function of
articulation and accuracy. They concluded that there is a trade-off between the articulation and
accuracy of a model. The more accurate a model became, the less articulate it was (the model
described a few processes with very good detail). The less accurate a model was, the more
articulate it was seen to be (the model described many processes with little accuracy). This study
suggests that it is difficult to be both accurate and articulate, which may have been a result of
smaller computing capabilities in the past.
TABLE 4: A PARTIAL LIST OF PREVIOUS WETLAND MODELS
Author(s) Model Type !Wetland Type applicable to
Parameters Simulated * Model Purpose #
HYDROLOGY:
Feng & Molz (1997) DY, DI, T FWS wetlands H R, M
Guertin, Barten, & Brooks (1987)
DY, L, T FWS peatland H M
Hammer & Kadlec (1986) DY, DI, T FWS wetlands H R, M, DE
Walton et al. (1996) DY, DI, T FWS wetlands H R, M
NITROGEN:
Widener (1995) DY, DI, T FWS H, N R
PHOSPHOROUS:
Mitsch & Reeder (1991) DY, DI, M FWS coastal H, P, Sd R
Christensen, Mitsch, & Jorgensen (1994)
S, L, TFWS
constructedH, P, Sd DE
MULTIPLE NUTRIENT PARAMETERS:
Brown (1988) DY, L, MFWS bogs,
swampsH, N, P M, DE
Dorge et al. (1994) DY, L, MFWS
freshwaterH,N, P M, DE
Gidley (1995) DY, L, MSSF
constructedH, N, DO,
C, BAC, BODM, DE
Jorgensen et al. (1988) DY, DI, MFWS
reedswampH, N, P M
Kadlec & Hammer (1988) DY, DI, M FWS wetlands H, N, P,CL M, DE
! DY = dynamic, S = steady state, DI = distributed, L = lumped,E = empirical, T = theoretical, M = mixed (theoretical and empirical)
* H = hydrology, N = nitrogen, P = phosphorous, SD = sediment, Cl = chloride,VG = vegetation growth, VS = vegetation survival, CB = coliform bacteria,
BOD = biological oxygen demand, Con = contaminants, SS = suspended solids# M = management, R = research, DE = design
31
Ideally, a model would require minimal input; and give output which is very accurate and
detailed, yet this is usually not the case. To design a model that is so esoteric, to the point in
which only a few specialists could use it, would be of limited use. There are models with a wide
spectrum of complexity yet it is best to select a model appropriate to available input data
(Jorgensen, 1995).
b. Modeling of Specific Wetland Processes
There have been various approaches to describe wetland processes. Researchers have usually
focused on one or two of the following specific processes: hydrology, nutrient (N, P, and C)
transport and transformations, and vegetative growth patterns. Described in the following
sections are the more frequently used approaches to modeling specific wetland processes.
i. Hydrology
Overall Water Budget
Proper modeling of hydrologic processes and overall water balance is pertinent to
wetland modeling. The complex interactions between soil, water and biota are driven, first and
foremost, by the hydrology of the wetland (Hammer and Kadlec, 1986). Nutrient
transformations and transport, vegetative survival, and sediment transport and associated
reactions are all dependent upon proper hydrologic functioning.
The overall water budget (Equation 2) details the inputs and outputs for a wetland. The
input and output parameters must be represented in the overall hydrologic modeling balance, to
determine the total volume of water stored in a wetland. These data can be combined with
known topography and construction data to determine wetland depths, residence time and
hydroperiod.
Many researchers have used the structure of the overall water budget (Equation 2) to
determine wetland hydrology (Dorge, 1994; Mitsch and Reeder, 1991; Walton et al., 1996).
Different researchers may have included or omitted certain input or output processes but the
concept of the model having one state variable (wetland water volume) is consistent throughout
all approaches. The overall budget may be used in either a lumped or distributed approach. If a
32
distributed modeling approach is used, then either a link node method (Walton et al., 1996; Hales
et al., 1991) or finite-difference approach (McDonald and Harbaugh, 1988) is usually used. The
link node method divides the system into a series of finite volumes called “nodes” where the
stage is defined. Flows are defined along 1-d links between adjacent nodes that can be modified
to represent simple or complex geometry. Essentially the link node method is applied to
continuous segments of the modeled area and uses continuity and momentum equations for
surface water to confirm logic in the system. Detail on the continuity and momentum equations
is included in the surface water description.
The finite-difference scheme or “block-centered” scheme is used in the groundwater flow
model MODFLOW (Harbaugh and McDonald, 1996). In this approach, the nodes are associated
with known parameter values and equations are solved to obtain unknown values (Fetter, 1994).
Any model that is distributed and represents more than one vertical layer of hydrologic
representation usually uses this scheme.
Many of the parameters that are included in the overall water budget are not modeled but
are included as input values. Amounts of precipitation, river inflow, and river outflow are
measured or estimated from previously recorded data. These inputs are forcing functions or
external variables that influence the state of the system, which in this case is wetland volume.
An input that is not specified to a great extent is the catchment runoff. Many researchers
have used recorded input volumes or first order estimations to determine the amount of nutrients
associated with inflow from catchment runoff (Brown, 1988; Christensen, 1994). This approach
does not allow for considerations of changes in watershed activities that may increase or
decrease nutrient loads.
One method of estimating catchment runoff is the Soil Conservation Service (SCS)
runoff curve model; a method that has been developed to estimate rainfall excess without the
need to compute infiltration and surface storage separately (Novotny and Olem, 1994). In the
SCS method, excess rainfall, Q, depends on the volume of precipitation, PP, and the volume of
total storage, S. S includes both the initial abstraction and total infiltration Ia.
The relationship between the rainfall excess and total rainfall on a daily basis is (SCS,
1968):
33
)8.0(
)2.0( 2
SP
SPQ
P
P
+−
= (6)
and
25425400 −=
CNS (7)
where Q is the rainfall excess (mm); PP is the precipitation rate (mm); S is a storage parameter
(mm); and CN is the runoff curve number. The curve number depends upon a number of factors
that include: land-use type, land-use cover, hydrologic soil group, hydrologic conditions, and
watershed soil moisture conditions. Typical values for the curve number can be found in the
SCS (1968) reference.
Another method for estimating hydrologic and nutrient transport to an existing or
potential wetland is to use an existing comprehensive watershed model such as ANSWERS,
AGNPS, or BASINS for predictions. The output results can be used as input sources to a
wetland model. These predictions should be more accurate than using more generalized
methods; thereby, improving input data and consequently output predictions of the wetland
model.
Surface Water Flow
In a FWS wetland, surface water flow plays an integral part in determining the movement
of dissolved and particulate nutrients out of the system. This is because the velocity at which
water flows through a wetland affects the mixing and settling of the constituents in the water. In
addition, the faster the water exits the system, the less time the processes that transform nutrients
in the system have to perform, thereby decreasing a wetland’s nutrient retaining capabilities.
Hydraulic structures, uneven topography, and vegetation cause frictional drag that decreases
surface water velocity. Accounting for the loss in surface water velocity due to obstacles will
allow for a better accounting of predicted residence times and outflows, thereby increasing the
accuracy of predictions for nutrient and sediment effluent.
A simple manner in which to determine the surface water flow velocity is:
34
A
Qv = (8)
where v is the water velocity (m/d), Q is the discharge from the wetland (m3/d), and A is the
cross sectional area of the wetland (m2). As Equation 8 shows, if the water velocity in the
system is over or under predicted, the determined outflow of the system is affected. Since a
lumped model assesses the system as a whole, the flow velocity rate is the same throughout the
entire wetland.
The modeling of surface water flow in a distributed system is more complicated. The
nodes in the system allow different velocities to be determined through the wetland based on the
site data. To assess logic and reasoning within a distributed modeling system, the continuity and
momentum equations are used to evaluate water movement. Walton et al. (1996) used the
equations for conservation of momentum and continuity:
)(/( 2
fo SSgAx
ygA
x
AQ
t
Q−=
∂∂+
∂∂+
∂∂ β
(9)
and
∑=
+=∂∂ N
nin QQ
t
V
1
(10)
where Q is the flow rate (L3/T); β is a momentum correction factor; A is the cross sectional area
of the link (L2); g is the acceleration due to gravity (L/T2); y is water depth (L); So is the bed
slope; Sf is the friction slope; V is the nodal volume (L3); Qn is the flow rate in link “n”; N is the
number of channels entering nodes; t is time (T); x is the longitudinal distance along a link (L);
and Qi is all other inflows such as precipitation, bank inflows, etc. (L3/T). In each adjoining
node, the momentum and conservation equations must be satisfied. This process confirms that
water is flowing smoothly in the wetland model.
Kadlec (1990) presented a system that is based on a friction rate law equation:
35
'
**
d
SdKv
αβ
= (11)
where v is the water velocity through the wetland bed; K is a premultiplier constant; d is the
average wetland depth (L); S is the slope; β is a depth exponent; α is a slope exponent; and d’ is
the average depth of free water. This rate law is combined with mass conservation to determine
wetland surface water flow. The model takes into account the movement of surface water in
response to gradient and vegetation flow resistance.
Evapotranspiration
Evapotranspiration (ET) is the term used for the water lost to the atmosphere from both
evaporation and plant transpiration. It is difficult to differentiate the two processes when
measuring water loss from a system, thus they are combined for many wetland modeling
situations. It is mainly a function of climatic variables such as solar radiation, temperature,
vapor pressure deficit, and wind speed (Abtew, 1996).
Like any other system, there are simple and more complicated approaches to modeling
ET. One empirical method is the pan evaporation method in which measured values of
evaporation from a standard pan are multiplied by a coefficient to represent actual ET in a
system (Novotny and Olem, 1994).
Pierce (1992) recommends the use of the Thornthwaite method for calculating potential
ET. The Thornthwaite method assumes that the soil moisture is not limiting and that air
temperature is the primary controlling factor of ET. The Thornthwaite method equation for
potential evapotranspiration (PET) follows the form of (Thornthwaite and Mather, 1955):
a
I
TDPET )
*10(*)
12(*54.2= (12)
where PET is the potential ET in cm/day; D is the number of daylight hours; T is the average
monthly air temperature (°C); I is the heat index; and a is a coefficient. The heat index, I, and
coefficient, a, are calculated as follows:
36
514.1)5
( jj
Ti = j = 1…12, (13)
∑=j
jiI , and (14)
49239.0*)10(792.1(*)10(771.*)10(75.6 22537 ++−= −−− IIIa (15)
where Tj is the historical average monthly temperature (°C).
Abtew (1996) examined six different models to estimate ET rates in South Florida. He
concluded that the Penman-Monteith equation best estimated ET for cattail and mixed marsh
vegetation while the Penman combination equation was more suitable for open water/algae
systems. The Penman-Monteith equation follows the form (Abtew, 1996):
)1(
1*)(**)(
α
α
γ
ρ
r
r
reecGR
ETc
dApn
++∆
−+−∆= (16)
where ET is the evapotranspiration (kw/m2); Rn is the net radiation flux at surface (kw/m2); G is
the water heat flux (kw/m2); ρ is the atmospheric density (kw/m3); cp is the specific heat of moist
air (KJ/ kg °C); (ea-ed) is the vapor pressure deficit (kPa); rc is the canopy resistance (s/m); ra is
the aerodynamic resistance (s/m); ∆ is the slope of vapor pressure curve (kPa/°C); and γ is a
psychometric constant (kPa/°C). The Penman Combination method is of the form (Allen et al.,
1989):
γγ
λ +∆−++−∆
= )2 )((43.6*)(1 dawwn eeubaGRET (17)
where ET is the grass or alfalfa reference ET (m/d); u2 is the wind speed at 2 m height (m/s); aw
is an empirical coefficient for the study area; bw is an empirical coefficient for the study area; Rn
37
is the net radiation flux at surface (kw/m2); G is the water heat flux (kw/m2); (ea-ed) is the vapor
pressure deficit (kPa); and γ is a psychometric constant (kPa/°C). Both approaches require large
data input, thus their use is limited as such data are not readily available.
Groundwater Flow
Groundwater flow can convey water and potentially nutrients to a wetland site. Some
wetlands are entirely fed and rely on groundwater flow for most hydrologic input. A few models
of constructed wetlands assume that groundwater inflow and outflow are negligible, compared to
the through-flow water volume, and are therefore ignored (Niswander and Mitsch, 1993; Mitsch
and Reeder, 1991; Hammer and Kadlec, 1986). Other researchers have assumed a constant rate
of infiltration to groundwater (Christensen et al., 1994; Brown, 1988).
The modeling of horizontal groundwater flow usually entails the use of Darcy’s law for
saturated groundwater flow (Fetter, 1994):
)(**dl
dhAKQ hh −= (18)
where Qh is the horizontal flow (L3/T); Kh is the horizontal hydraulic conductivity (L/T); A is
the cross sectional area (L2); dh is the change in hydraulic head (L); and dl is the change in
distance (L). Vertical groundwater flow can be determined in the same manner with Kh being
replaced with Kv, the vertical hydraulic conductivity.
With these two calculations, the total potentiometric head may be calculated at each
subsurface node with the continuity equation (Equation 9). By taking the calculated heads at
various points in a wetland, the release of wetland water (infiltration) or input of groundwater
(percolation) may be deduced. Walton et al. (1996) and Dorge (1994) are among a few
researchers who have used this approach in their modeling.
ii. Nitrogen
The modeling of N processes usually include nitrification, denitrification, mineralization,
immobilization, and assimilation. Processes such as N fixation, volatilization, atmospheric
38
deposition, and adsorption are either concluded to be negligible or are only modeled for specific
situations. The approaches to modeling the differing processes are very similar. There are
certain approaches that are used with their only differences being in parameter declaration and
values.
A commonly used empirical model for wetland design assumes plug flow hydraulics and
first order BOD removal kinetics with an adjustment for temperature effects (Gidley, 1995). The
equation follows the form:
tKoe
TeCC −= (19)
where Co is the influent concentration (M/L3); Ce is the effluent concentration (M/L3); KT is the
rate constant (day–1); and t is the hydraulic retention time (days). The rate constant is a function
of temperature and is determined by:
2020
−Θ= TT KK (20)
where K20 is the rate constant at 20 degrees Celsius (day-1), and T is the design or operating
temperature (°C).
There are theoretical models that emulate the transformations within a wetland. If the
reaction is independent of the substrate concentration, it is considered to be zero order. The rate
equation for this type of reaction is:
[ ]k
dt
Sd =−(21)
where S is the substrate (M/L3), k is the rate constant (M/L3/T), and t is time.
The first order equation is similar to the zero order form except it accounts for substrate
concentration. This equation is of the form:
[ ] [ ]SKdt
Sd =−(22)
39
where K is the rate coefficient (day-1).
Another approach is the use of Monod kinetics to describe wetland reactions. The
Monod kinetics account for the microorganisms responsible for the transformation processes
using the growth rate of the microorganisms. The Monod equation follows the form of
(Snoeyink and Jenkins, 1980):
SK
SMM
s += max (23)
where M is the growth rate (L3/MT); Mmax is the maximum growth rate (M/L3T); Ks is the
substrate concentration when M= Mmax/2 half-saturation constant (M/L3); and S is the substrate
concentration (M/L3).
Very similar to the Monod kinetics approach is the Michaelis-Menten kinetics equation
(Dorge, 1994):
[ ] [ ][ ]SK
SV
dt
Sd
m += max (24)
where S is the substrate (M/L3); Vmax is the maximum rate of reaction (M/L3T); and Km is the
half-saturation constant (M/L3). The two approaches bear a striking resemblance, and are based
on the concept of conversion of a single substrate to a single product. They are fairly
interchangeable and are basically a matter of semantics; however, the Monod equation was
initially developed for bacterial growth, while the Michaelis-Menten was developed for enzyme
reactions.
Previous models usually ignore diffusion of dissolved N. There are two common choices
for modeling diffusion; Fick’s law and a law with no formal name that involves a mass transfer
coefficient.
Fick’s law is the most commonly cited in modeling diffusion. There are many forms of
Fick’s law, the basic form is (Cussler, 1997):
40
z
cADAjJ
∂∂
−== 111 (25)
where J1 is the one dimensional flux of a constituent (M); A is the cross sectional area across
which diffusion occurs(L2); j1 is the flux per unit area (M/L2 ); c1 is the concentration of the
constituent (M/L3); z is the distance (L); and D is the diffusion coefficient (L/T).
Mass transfer is an approximate engineering idea that often gives a simpler description of
diffusion. Analyzing diffusion with mass transfer coefficients requires the assumption that
changes in concentration are limited to the small part of the system’s volumes connecting
boundaries (Cussler, 1997). The common form used to describe mass transfer is (Cussler, 1997):
)( 111 cckN i −= (26)
where N1 is the flux at the interface (M/L2T); k is the mass transfer coefficient (L/T); c1 is the
concentration in bulk solution 1 (M/L3); and c1i is the concentration at interface (M/L3). The
concentration of c1i is in the bulk concentration of c1 but is usually in equilibrium with the
concentration across the interface in a second adjacent fluid. The product of the flux and cross
sectional area will result in the amount of mass that has been transferred.
As stated earlier, nitrification, denitrification, mineralization, assimilation and
immobilization are modeled with similar approaches, the only difference being parameter
declarations and the type of substrate being examined. For example, if nitrification was being
modeled with a first order equation, the substrate substance would be NH4+ with an associated
rate constant. If the process examination concerns denitrification, the substrate examined would
be NO3-. For mineralization, immobilization, and assimilation, the respective substrates to be
examined would be organic N, NH3, and both NH3 and NO3-.
These approaches do not cover all attempts taken to model the N cycle within a wetland,
yet they are the most often used. In the selected models description sections, differing
approaches will be more fully examined.
41
iii. Phosphorous
Modeling of the P cycle would seem to be simpler than the N cycle because of the fewer
processes involved, yet this is not the case since processes in the P cycle are not entirely
understood.
For simplification, a high percentage of models simulate total P rather than its component
forms. A mass balance is implemented to account for inputs, outputs, and retention of P
amounts:
retoutinm PPP
dt
dP−−= (27)
where Pm is the mass of phosphorous per unit wetland volume (M/L3); Pin is the influent
phosphorous (M/L3); Pout is the effluent phosphorous (M/L3); and Pret is the retention of
phosphorous (M/L3)
Various modeling approaches differ with respect to how they represent the mass balance
parameters. Mitsch et al. (1995) developed a simple model that accounted for retention by
incorporating the effect of the hydrologic loading on the wetland. The retention was calculated
as:
Thret PaLkP )1( += (28)
where Lh is the hydrologic loading of the wetland (L/T); k is the retention coefficient for P (1/L3;
a is the coefficient reflecting the magnitude of added hydrologic effect (T/L); and PT is the total P
in the system (M). Input is the phosphorous concentration of inflow and output is the
concentration value of wetland outflow. The retention coefficients were determined through
calibration.
Mitsch and Reeder (1991) used the same mass balance concept with more detail. The
model accounted for macrophyte and plankton phosphorous uptake and release, sedimentation
and resuspension velocities, and phosphorous concentrations in sediments. Christensen et al.
(1994) accounted for pools of dissolved TP in the water column, particulate TP in water, bottom
42
soil TP, and macrophyte uptake and release. Mineralization, sedimentation and inflows were all
modeled using first order equations.
Both models use simple adsorption rates that were determined from field data.
When adsorption is calculated, it is usually modeled using the Langmuir expression, which is
valid for monolayer adsorption (Jorgensen, 1988). The Langmuir model is of the form (Novotny
and Olem, 1994):
e
eo
bC
bCQr
+=
1(29)
where Qo is the adsorption maximum at a fixed temperature (µg/g); b is a constant related to the
net energy of net enthalpy of adsorption (1/µg); r is the adsorbed concentration of the
contaminate (µg/g); and Ce is the dissolved (free) concentration of the contaminant water (µg/1).
Jorgensen (1988) uses the Langmuir model to determine adsorption amounts while Dorge (1996)
uses a temperature-dependent Langmuir adsorption model.
There are other models to determine adsorption of P to particles. One approach is use of
the Freundlich equation (Novotny and Olem, 1994):
neCKr /1*= (30)
where K and n are constants. The constants K and n are obtained by plotting r versus Ce on log-
log graph paper, where the logarithmic intercept is K and the logarithmic slope equals 1/n. The
Freundlich isotherm is useful when the energy term in the Langmuir isotherm varies as a
function of surface coverage (Novotny and Olem, 1994).
For low concentrations of contaminants, the Freundlich and Langmuir isotherms can be
simplified to a linear isotherm:
∏= eCr (31)
where Π is the partition coefficient (1/g).
43
Every model examined in the wetland review did not model the amounts of P
transformed through precipitation. This is due to the negligible amounts that are lost through
precipitation and also the difficulty involved in modeling the process.
iv. Sediment
Wetland ecosystems are effective sediment traps, generally retaining more suspended
sediments than they export. Sediment deposition is important in improving a wetland’s water
quality not only because sediments themselves are a water contaminant, but also because
sediments concentrate many toxic species and nutrients through sorption processes (Fennessey et
al., 1994). Sediment particles may possibly improve water quality as they settle and scavenge
suspended and dissolved contaminants from the water column (Hart, 1982), and can represent
permanent removal of contaminants through chemical breakdown, and burial in the bottom
(Johnson et al., 1984). Although intuitively wetland vegetation would augment sediment
deposition through filtration of the wetland waters, this is not necessarily the case. Hosokawa
and Horie (1992) observed no differences in TSS removal efficiency of vegetated and
unvegetated cells, suggesting that vegetation does not increase TSS removal.
Sediment deposition is one of the most difficult parameters to measure in wetland
ecosystems (Fennessey et al., 1994), because sedimentary processes are not well understood
(Mitsch and Gosselink, 1993). A constructed wetland in Florida retained over 80% of the
incoming total suspended solids (Knight, 1987), and a wetland bordering Lake Erie retained 55%
of the annual sediment inflow in a drought year (Mitsch and Reeder, 1992). The values reported
in the literature indicate sediment removal efficiencies between 80-95% (Daukas et al., 1989).
Modeling of sediment deposition entails determination of sedimentation velocities,
resuspension velocities, and the amount of available sediment material. Sedimentation and
resuspension velocities are usually modeled with zero order rate equations. The only differences
in amounts of sedimentation and resuspension is determined by the pools of available sediment
material which can increase and decrease due to catchment runoff and biological death. Mitsch
and Reeder (1991) included sediment coefficients for a plankton pool and for a macrophyte pool.
Christensen et al. (1994) used a mass balance approach to determine sediments in the water
44
column and sediments located on the wetland bottom. Sedimentation was modeled with a first
order equation.
One method used to increase the accuracy of the prediction of resuspension of material
from wetland bottoms is the determination of a critical velocity. The critical velocity is the
minimum velocity at which resuspension occurs. The theory of plain sedimentation predicts that
as the water velocity in a wetland increases, there is a point at which the shear stress will detach
particles from the wetland bottom. These stresses can be reflected with a Manning’s coefficient
value (Kadlec and Hammer, 1996). Wetland flows simplify this determination because generally
flow is in the laminar regime. The critical velocity, u, can be determined with (Kadlec and
Hammer, 1996):
≤
3/2
6/13/1
*2.7nd
Hwu (32)
where d is the particle diameter (m); H is the water depth (m); n is the Manning’s coefficient for
open channel (s/m1/3); u is the water velocity (m/d); and w is the particle settling velocity (m/s).
According to this theory, if the water velocity is less than the right-hand side of Equation 32,
resuspension will not occur. Since this result is based on laminar theory, it is necessary that two
Reynold’s number criteria are satisfied:
≤=
gn
Hud 6/1
1
*2**Re
µρ
(33)
1**
Re2 ≤
=
µρ wd
(34)
where µ is the water viscosity (kg/m⋅s ~ .001 at 20 °C), and ρ is the water density (kg/m3 ~1000
at 20°C). Although this theory places a constraint on when resuspension will and will not occur,
it does not predict how much will resuspend.
45
An approach used to determine the amount of sediment outflow from a sedimentation
basin may also be applied to wetlands. The classic “overflow rate” theory of settling tank design
determines whether a particle is removed from the system (Novotny and Olem, 1994):
wA
QOR ≤= (35)
where OR is the overflow rate (m/day), Q is the inflow to the basin (m3/day), A is the water
surface area (m2), and w is the particle settling velocity (m/day). Utilizing the “ideal settling
basin” assumption, all particles with settling velocities that are greater than the OR will be
trapped in the wetland. The theory assumes that all particles of a particular size will either be
completely retained (100%), or not at all (0%), by the wetland system. This may affect the
accuracy of the results.
Particle settling velocity may be estimated from Stoke’s equation (Novotny and Olem,
1994):
)1(18
2
−= s
gdw ρ
ν(36)
where d is the particle diameter (m), g is gravity acceleration (m/sec2), ρs is the specific gravity
of the particle with respect to water, and ν is the kinematic viscosity (m2/sec).
v. Vegetation
According to Kadlec and Hammer (1988) a simple model of nutrient uptake is sufficient
for describing wetland processes, and highly sophisticated models are not warranted for
simulation of overall wetland features. This is beneficial as finely detailed models may require
parameter inputs for rarely collected data on leaf resistance to CO2, light intensity variations, leaf
mortality rate, and stem mortality rate (Dixon, 1974).
Various researchers employ the idea of a simple vegetation model. Dorge (1994) based
plant uptake on a ratio of daily to yearly irradiance multiplied by an empirical parameter for
46
yearly plant uptake. He assumed that over a yearly scale, plant uptake is equivalent to plant
death. Brown et al. (1988) used empirical data to fit a growth curve to the growing season for
both N and P. He assumed that there was a breakdown of 75% uptake from the soil
concentration and 25% from the water column. Christensen et al. (1994) used a simple mass
balance on macrophyte biomass which determines the change in biomass based on the
macrophyte death rate subtracted from the macrophyte growth rate. Macrophyte rates were
modeled using first order equations.
Output that describes hydroperiod and depth of inundation can be used to determine
proper vegetation for a specific wetland site. If the hydrologic characteristics for a wetland are
known, then based on known growth patterns for vegetation and climate, a user can determine
what would be the most appropriate vegetation to include in a wetland.
c. Selected Wetland Models
The following wetland model descriptions are included to present an overview of
approaches taken to model the hydrologic and nutrient dynamics in wetland systems. Far from
being comprehensive, a basic overview of the models, results of their application, ease of use,
and their deficiencies will be discussed, to present the scope of previous research.
Hammer and Kadlec (1986) developed a distributed, hydrologic model for the overland
flow of a wetland. The model takes into account the movement of surface water in response to
the hydrologic gradient and vegetation flow resistance. It employs the rate law equation
(Equation 11), combined with a mass balance to determine surface water flow in the system.
Input information includes site topography, porosities, ET, stream flows, groundwater
recharge/discharge, and pumped additions. Depths and flow rates are calculated as functions of
position and time. Evapotranspiration is estimated using the Thornthwaite method (Equations
12, 13, 14, and 15). The model accounts for snowmelt and freezing of the wetland water. The
model uses a Crank-Nicholson finite difference approximation technique to solve the partial
difference equations, along with a volume-differencing procedure to reduce cumulative errors.
The model was applied to data from the Porter Ranch site in Houghton, Michigan.
Simulation data from 1978 to 1979 were presented to detail the surface water elevation
predictions in the wetland. Visual analysis showed a very good match between predicted and
observed values. Input to the model is specific and may be difficult to obtain for it requires very
47
precise descriptions of the shape of the wetland and soil porosities and conductivities of the area.
The set of parameters chosen for the model is site-specific to the Porter Ranch system, and
generalization of parameters is necessary.
Feng and Molz (1997) developed a 2-dimensional, diffusion based wetland flow model
(WETFLOW) which allows the incorporation of spatial variations in flow resistance and
vegetation. Continuity equations are combined with a modified Manning’s equation and solved
with a Picard-iteration scheme. The two basic parameter distributions required by WETFLOW
are land surface topography and flow resistance. Since the model is two-dimensional, the
amount of input needed to properly describe the site is large.
The model was applied to two situations: a laboratory experiment and a wetland pond.
The researchers related that the values were similar through visual analysis of experimental and
predicted data, and that WETFLOW performed well in a numerical sense. A decrease in grid
size significantly affects the representation of the land surface and hydrologic representation,
with smaller more detailed entries resulting in more accurate model predictions. For the
simulation concerning the wetland pond, 105,600 nodes was required to describe the site.
Guertin, Barten and Brooks (1987) developed a hydrologic model (PHIM) for wetlands
typical of those found in the Great Lakes region of the United States. PHIM is physically based
with input limited to climactic data (precipitation, maximum and minimum daily air temperature)
and wetland description information (e.g. vegetation type, canopy cover and soil depth). The
model accounts for three different types of wetlands; natural peatlands, mined peatlands, and
mineral soil upland peatlands. Inputs differ for each respective wetland system, but the basic
hydrologic inputs include net precipitation, snowmelt, upland inflow, groundwater inflow,
percolation, and ET.
All three wetland types were simulated with data from watersheds in northern Minnesota.
The short-term predictions for all three simulations were within one standard deviation of the
observed values. Long term evaluations were not completed.
Walton et al. (1996) developed the Wetlands Dynamic Water Budget Model, named
thusly because it provides magnitudes for the water budget components as well as water depths,
discharges and flow velocities throughout the entire wetland. There are three dynamically linked
modules which allow the model to represent the surface water, vertical processes and horizontal
groundwater movement in the wetland. The surface water module combines the equations for
48
conservation of momentum and volume (Equations 9 and 10). The vertical processes simulated
in the model are canopy interception and drainage, infiltration, surface water evaporation, soil
water evaporation, and transpiration. Horizontal movement of groundwater is based on Darcy’s
law (Equation 18). The model uses a link-node method that divides a wetland system into a
series of finite volumes where the water level is defined. The scheme is flexible because it can
easily represent simple or complex geometry, while the one dimensional flows used for the links
are simple to program and efficiently solve.
The model was applied to the Black Swamps wetland, located on the Cache River in
eastern Arkansas. A link node grid consisting of 66 nodes and 115 links were used to describe
the wetland area. The model was calibrated with data from the water year 1990 and validated
with data from the water years 1988 to 1991. The researchers presented the validation data
which visually matched the observed data points, but no statistical or sensitivity analyses were
presented. Although the model presents data that is accurate, the amount of input necessary to
generate the results are fairly extensive.
While all of the presented hydrologic models provide fairly, accurate predictions, their
use is limited as the number of input values is quite extensive. The amount of energy and
resources needed to collect the required input values make their use impractical. Although they
can be used to simulate the hydrologic functions of a wetland and the potential problems with
these designs, a wetland would not be designed specifically for water retention. If combined
with nutrient modeling, their use would be greatly enhanced.
Widener (1995) developed a mathematical model of the N cycle for FWS constructed
wetlands. The model is designed to represent the organic N, NH4+ and NO3
- concentrations in a
wetland system to a very fine degree, supported by the entry of spatial examination in
centimeters. Ionic exchange of NH4+ with soil particles, nitrification, mineralization, and
diffusion of ammonium NH4+-N affect ammonium NH4
+-N concentrations in the sediment.
Nitrification, denitrification, and diffusion processes affect NO3- concentration. Mineralization,
nitrification, and denitrification are modeled using first-order equations with temperature
dependent effects. Oxygen is an examined parameter, but not explicitly modeled. The model
assumes that there is active transport of oxygen to the wetland soil. The transport aerates the
wetland soil to a user-defined thickness, which the model assumes is the extent of root zone
penetration. The oxic zone is where nitrification occurs and where denitrification is inhibited.
49
The partial differential equations were solved using center-divided differences with
Taylor Series expansion. Inputs to the model are not extensive, and should be easily obtained
except for the determination of the wetland sediment aerobic zone. The model was not applied
to its intended target of NPS reduction due to lack of available data, but six scenarios were
examined to determine if model predictions are comparable to qualitative theory. The scenarios
tested the spatial distribution of the calculations, the temperature effects and the flow condition
effects. Widener (1995) concluded that the model predictions were sound from analyzing graphs
of the ammonium and nitrate movement.
The extensive time required for simulations during the study period was limiting, as the
computation to real time simulation ratio was 30:1. This eliminated any long term simulations as
it would have required 12.2 computing days to simulate one year of real world processes. This
problem may have been alleviated due to recent increases in computing capabilities.
Christensen, Mitsch, and Jorgensen (1994) developed an ecosystem model to examine the
retention of sediment and P at the Des Plaines, Illinois experimental wetlands. The hydrologic
model is a black box model and uses a simple water budget to account for runoff and pumped
inflow, runoff and pumped outflow, direct precipitation, and ET amounts. Two pools for
sediment are simulated, amounts located in the water and the bottom. There are three pools for
P; dissolved and particulate P in the water column, and bottom total P. The mass balance for
sediment in water is affected by influent amounts, net sedimentation of solids, sediment outflow,
and sediment production in the water which consists of macrophyte and algal production.
Sediments on the bottom are accounted for with additions from macrophyte death and decay of
organic bottom sediments. The dissolved P balance accounts for mineralization from the bottom
and particulate P pools, inflows, outflows, and sedimentation. Mineralization, and sedimentation
are modeled with first order rate equations. Particulate P balances inflows, outflows, and
sedimentation, while bottom P accounts for sedimentation, remineralization and contributions
from macrophyte death. Macrophyte biomass is modeled with a simple zero order growth and
death rate. Primary productivity is modeled with a dependence upon the solar radiation and
water temperature.
The model was calibrated with data collected at the Des Plaines wetlands, but not
validated with an independent data set. Calibration was determined through visual analysis of
the measured and predicted values, minimizing the differences between the two. The model
50
appeared to be the most sensitive to the sedimentation velocity in the wetland system. It does not
account for precipitation of P in the wetland system.
Mitsch and Reeder (1991) developed a model for a coastal wetland of Lake Erie called
Old Woman Creek, which included submodels for hydrology, productivity and P. The hydrology
submodel is capable of running independently from the other submodels. Factors affecting the
only hydrologic state variable, volume of water in the marsh, include rainfall, inflow, ET and
outflow. The productivity model simulates macrophyte cover as a function of water level, which
was determined through empirical observations. Plankton biomass is based on the gross primary
productivity, solar radiation and a zero order sedimentation rate. Incoming P concentrations were
based on empirical observations from similar areas and are a function of flow. The model
utilized two P storages, wetland waters and sediments. Sedimentation and resuspension
pathways are based on an average settling velocity and a function based on wetland depth. The
model differs from similar models in that it accounts for plankton biomass in the overall P
balance and determines the amount of P using a simple ratio between the P and growth rate base
on solar radiation.
The model assumes that there is no P limitation in the wetland, therefore no attempt was
made to divide the P in the water column into available and unavailable forms. Assumptions
were also made concerning the macrophyte growth, which was limited to the dominant
vegetation of Nelumbo lutea. The model uses a fourth order Runge-Kutta technique with and
integration interval of 0.1 day.
The model was calibrated with data collected from March, 1988 to November, 1988 in a
step-wise manner where first hydrology; then primary productivity, macrophyte biomass, and
chlorophyl data; and finally P were examined. Subsequent simulations were made for various
combinations of flow from the watershed to the wetland and varying water levels for Old Woman
Creek to determine optimal conditions of P retention and to test the stability and range of the
model. P retention varied from 17% to 52%, with highest retention of P by the wetland
occurring when high inflows from the watershed are coupled with high water levels in the
wetlands.
According to the researchers, results for the hydrologic and nutrient modeling were fairly
accurate. The hydrology submodel predictions were usually within 0.2 m of recorded field data,
but did not reflect dramatic changes in water level. Field measurements of gross primary
51
productivity were within one standard deviation of the field measurements 70% of the time.
Predicted average P concentrations followed the observed field data trends.
Brown (1988) developed a model in BASIC, for the hydrology and nutrient dynamics of
a wetland typical of those found in north and central Florida. The model predicts water depth,
recharge, and surface outflow as well as concentrations of P (total P) and three N species (NOX,
NH3, and organic N). The hydrology submodel may be run independently from the nutrient
dynamics and accounts for runoff, precipitation, evaporation, transpiration, and surface and
groundwater outflow. Evaporation is estimated as the average pan evaporation with adjustments
made for rain, temperature and humidity. Transpiration is described as the product of a growth
coefficient and solar radiation. The N cycle accounts for atmospheric, plant decomposition,
sedimentation, groundwater, and runoff inputs and for transformations such as nitrification,
denitrification , and mineralization. Transformations of N were modeled with zero order rate
equations based upon measured rate coefficients. Hydroperiod and depth of inundation are
calculated, which allow appropriate vegetation choices to be planted to optimize plant growth
and nutrient uptake. Nutrient uptake by vegetation is based upon the product of tissue nutrient
concentration and growth rate. The output consists of water levels and nutrient concentrations at
a daily time step.
Required input to the model are readily available or fairly easy to estimate. This is
because the model is lumped and requires mostly weekly, monthly and yearly, but not daily
inputs. Comparisons of field data to model predictions were not presented, instead simulation
results for hypothetical additions of treated effluent were presented. The simulation results
concluded that the addition of treated effluent increased the depth and duration of flooding in the
wetland. The results also showed a predicted decrease in the effluent concentration of the N and
P species. The decrease in concentration may stem from the extra water volume which enters
and subsequently leaves the wetland, but could also mean that the additional treated effluent may
have had an initial lower concentration of contaminants.
Drawbacks to the model include the failure to model diffusion processes, and the
assumption that the amount of organic C is non-limiting in the denitrification process. This
limits the application of the model to areas where these assumptions apply. In addition, Brown’s
model does not allow for seasonal parameter changes, which may vary considerably.
52
Jorgensen et al. (1988) modeled the nutrient removal capabilities of a wet meadow and
natural reedswamp in Denmark. The hydrology, N, and P submodels were created using the
language CSMP. The hydrology submodel accounts for precipitation, ET, inflows, outflows,
water uptake by plants, and the modeling of vertical and horizontal flows with Darcy’s equation.
The N submodel included N input from inflows and rainwater, ammonification, nitrification,
denitrification, adsorption, and plant assimilation processes. Ammonification and nitrification
are described with first order functions dependent on moisture content and temperature.
Volatilization is described as a first order equation that is dependent on moisture content,
temperature and pH. Denitrification is modeled using a Michaelis-Menton equation (Equation
24). Adsorption is modeled using a Langmuir expression (Equation 30), which is valid for
monolayer adsorption. Plant uptake is dependent on the root zone nutrient gradients and the
maximum concentration of nutrients within the vegetation. Plant decomposition is described
with a temperature dependent first order equation that allows for complete plant degradation
between autumn and the following springtime. The P submodel is described similar to its
parallel N processes with different parameters, being the only distinction.
The presented model was not calibrated or validated due to a lack of available data;
therefore no statistical or sensitivity analyses were presented. Since the model covers so many
processes, model input is fairly difficult to obtain. Although the model may be very
encompassing, the accuracy of the model is unknown.
Kadlec and Hammer (1988) developed a dynamic, compartmental, simulation model to
evaluate the effects of wastewater additions to a natural wetland. Flow rates and water levels are
calculated using a one-dimensional model for surface flow through vegetation (Hammer and
Kadlec, 1986). The ecosystem model took a different approach than other models by containing
nine compartments that accounted for annual and woody biomass; roots; standing dead; annual
and woody litter; and top, mid and deep soil layers. Solids exchange rates were computed as a
constant depletion rate multiplied by the peak pool amount. Summer decomposition and nutrient
transformation rates were 10 times the winter rates. The annual biomass pool is assumed to be
zero on the first day of the growing season, producing at a constant composition until it reaches
peak biomass on the last day of the growing season. Growth rates are a function of soil nutrient
limits and are constrained by the limiting nutrient. Mass balances were applied to chloride, N
and P amounts in the wetland.
53
The model was applied to the Porter Ranch wastewater treatment facility at Houghton
Lake, Michigan. Results were reported for simulations comparing the first two years of the
system’s operation (1978 and 1979). The author’s determined that the model predicted surface
water solute concentrations, as well as biomass, litter, and soil dynamics accurately based on a
visual analysis of recorded effluent and model predictions. No statistical or sensitivity analyses
were provided.
The model requires a number of plant growth and death parameters for input that may be
difficult to attain. In addition, the model does not allow for seasonal changes in parameter rate
values.
Dorge (1994, 1996) developed a simulation model describing the hydrology, N turnover,
and P turnover in freshwater wetlands. The WET model describes four components in the water
phase (NH4-N, NO3-N, PO4-P and Particulate P) and five attached to the sediment/peat (plant
biomass, labile organic matter, stabile organic matter, immobile N and adsorbed N). The N cycle
is dynamically modeled while the P cycle is modeled as a black box, i.e. as an inexhaustible pool
of P. The hydrologic submodel accounts for inflow, precipitation, evaporation, outflow, and
infiltration. Nitrification and denitrification are described with temperature dependent
Michaelis-Menton equations. Adsorption is determined with the Langmuir equation (Equation
30), while mineralization is a temperature dependent process coupled with the decomposition of
organic material. The plant uptake component assumes that there is no N limitation and that
growth is only dependent on the actual radiance intensity at the wetland surface. The model
accounts for microbial immobilization with a temperature dependent, first order assimilation rate
and assumes the subsequent release is based on a first order decay rate. Diffusion is modeled
based upon the hydrologic gradients in the system, where transport of dissolved constituents
travels between the surface water and peat water depending on the flow of the water. Diffusion
is not based on concentration gradients in the wetland. The P cycle accounts for mineralization,
immobilization and plant uptake of PO4-P. The particulate-P component accounts for
sedimentation and resuspension. Sedimentation has an empirically determined maximum and
minimum rate and is dependent on plant biomass in the wetland. Resuspension is an empirically
determined constant that is input as grams per meter squared.
An early version of the model that modeled only hydrologic and N processes was applied
to three wetlands in Denmark. The researchers reported results as being fairly accurate based on
54
visual analysis, with no statistical analysis. The model had difficulty in modeling extreme
weather events such as the colder winter months and seasonal flooding. The model is interfaced
with a Windows based system, allowing for easy entry of data. Input entries are minimal and
should be fairly easy to determine.
Gidley (1995) developed a mechanistic, compartmental simulation model for SSF
constructed wetlands, written in STELLA. The model consists of six submodels, one each for
water volume, N, C, dissolved oxygen, heterotrophic and autotrophic bacteria. The N cycle
includes pools for dissolved, immobilized and particulate organic N, ammonium and nitrate.
Processes modeled include both particulate and dissolved organic N ammonification, N uptake,
incorporation by biomass and microbes, nitrification and denitrification. The C cycle models the
balances of particulate and dissolved C, by accounting for contributions made from plant death,
leaching, decomposition, wastewater BOD, and microbial death; and removals due to
heterotrophic growth and peat accumulation. Bacteria are modeled using dual substrate Monod
kinetics with limitations due to electron donors and acceptors, along with temperature
considerations. The DO balance is computed assuming plant translocation and losses due to
microbial respiration. The water balance uses Darcy’s law to determine outflow, and
Thornthwaite’s method to estimate ET.
Gidley’s model took a different approach to the typical modeling of wetland systems. It
accounted for the C and N interaction, and the effect of DO levels upon microbial growth. It also
directly linked microbial growth and death to the consumption of nutrients in the system.
Previous models accounted for these interactions with zero to first order rate equations that
assume rates are dependent only on initial concentrations.
The model was tested on a constructed wetland site in Maryland by examining model
predictions of DO, BOD5, and N concentrations against actual effluent values. Model results
reproduced seasonal trends and treatment efficiency well based on visual analysis of the output
data. A statistical comparison of measured and predicted values were not completed; however a
sensitivity analysis determined that the model was more sensitive to parameters affecting
microbial growth and substrate use directly.
55
D. LITERATURE REVIEW SUMMARY
A prevalent problem today is NPS pollution. Wetlands are a BMP implemented in an
attempt to alleviate the impact of NPS pollution. Constructed wetlands are an attractive BMP
because in addition to controlling NPS pollutants, they provide other beneficial functions to the
environment such as wildlife habitat and recreation, while also being relatively low maintenance
and cost effective.
Many processes affect the N and P cycles in a wetland. To complicate matters, many of
these processes are poorly understood and are affected by a complex web of dependent
interactions. The hydrologic cycle controls all of the wetlands functions, but factors which affect
the N cycle include the vegetative, C, bacteria and dissolved oxygen cycles. The N cycle is
affected by transformations such mineralization, nitrification, denitrification, N fixation, and
assimilation; and transfer processes such as particulate settling and resuspension, diffusion,
litterfall, plant uptake and translocation, ammonia volatilization, adsorption, seed release, and
organism migrations. The P cycle is a sedimentary cycle and is affected by adsorption,
precipitation, desorption, mineralization, sedimentation, and resuspension.
One of the more effective ways to enhance the design and construction of wetlands is
with the use of models. Models provide an ability to make comparisons among alternative
designs and management strategies, allowing a wetland to be optimally utilized for its intended
purpose, be that for pollution control or wildlife interactions.
The problems with most existing models is that model development was geared towards
certain objectives and goals to fit the specific needs and functions of a certain wetland type. This
is exemplified by assumptions made during model development by Brown (1988), and Mitsch
and Reeder (1991). Assumptions initially made to ease model development restrict the
transferability of the model from one area to another, as assumptions for one location may not
hold in another. To develop a model that is applicable to many situations, generalizations will
need to be made. Generalizations usually decrease accuracy; nonetheless, these model types are
necessary so that an easy to use, widely applicable program can be developed which will not
require constant modifications. From experience, it is noted that a person can operate a model
better if they have a chance to utilize the model on a frequent basis. Therefore, a model that can
56
provide the ability to examine many different wetland designs at one time, would be the most
beneficial for a wetland designer.
Individually each model presented in the model review works for the intended purpose of
the described research. However, the question is whether others would use the model to actually
design and model wetlands. If a proposed wetland fits the assumptions made by previous
models, they might be able to, otherwise the alternative is to find a model that is similar to the
desired system and alter the modeling code.
The ability to model many designs is a key point missed by many existing models. Most
simulate either FWS or SSF wetlands, but usually never both. There is an inability to change
values of parameters after initial input, which does not allow the modeling of changes due to
seasonal variability. Most models require all components of a model (hydrologic, N and P) to be
run, even if only one cycle is desired. This can also apply to processes in individual cycles such
as N fixation or atmospheric deposition in the N cycle. Additionally, there is a lack of
documentation on how to operate most existing models.
Another overall problem with previous models is that the input requirements are
restricting. Most models are not linked to existing models that will accurately quantify input
values to the wetland. Known daily input values would be ideal as rapid changes in weather can
effect incoming loads; however, when predicting input values for a possible wetland these values
are unknown. The ability to predict incoming concentrations with one model or the ability to
input values directly from an existing model would be beneficial, as better input values usually
results in more accurate effluent predictions.
Additionally, one would like the ability to determine input values on a more theoretical
basis, because less laboratory and research efforts would be needed to determine how wetland
designs would function. For example, most wetland models have lumped the effects of the N
processes with first order (Widener, 1995) rate equations, which do not account for the
interactions of the bacterial, C, and DO cycles on these processes. Attempts were made by
Dorge (1994, 1996) and Jorgensen (1988) to use Michaelis-Menten equations to account for
bacterial affects, and Gidley (1995) directly incorporates actual bacterial growth, in addition to
the C and DO cycles. These approaches are a movement in a new direction towards modeling
wetland processes and relations.
57
From the review, it can be seen that there is a need to develop a model that is general
enough to be applied to various situations, but accurate enough to be useful and implemented. In
this study, attempts were made to develop such a model.
58
III: Model Development
A. MODEL OVERVIEW:
The developed simulation model, SET-WET, models both FWS and SSF wetlands. The
SET-WET model simulates the hydrologic, vegetative, N, C, DO, bacteria, sediment, and P
cycles of the wetland. SET-WET is designed so that the cycles may be modeled independently
with some constraints. The N, C, bacteria, and DO cycles are linked and may not be run
independently. These cycles will be referred to as the NCOB cycle to denote this dependence.
The P cycle is dependent on the sediment cycle, and all of the cycles except the hydrologic,
require that vegetative growth also be simulated. The sediment and P cycles may not be modeled
in conjunction with the SSF wetland option. In addition to the choice of which cycles may be
simulated, there are options within each modeling cycle that allow certain processes to be
simulated. More details will be given with each individual cycle description.
The SET-WET model simulates the wetland as a continuously stirred tank reactor. The
model user must enter the average wetland length, average wetland width, wetland substrate
bottom (HB, m), wetland substrate top (HT, m); the top of the wetland cell (HO, m) and the
initial water height (HI, m) to describe the physical description and initial hydrologic conditions
in the wetland system. The wetland levels (HB, HI, HO, HT) are all relative to a user-defined
datum; however, it is suggested that the user designate the substrate bottom as the reference level
(datum). Figure 6 represents how the FWS and SSF wetlands are dimensionally represented in
the SET-WET model.
The SET-WET model is designed as a flexible tool and could be applied to many
situations. Every wetland type could not be covered, yet there are many options in the model
that should allow the design and modeling of various new and old wetland systems. The SET-
WET model is also designed to be user-friendly. One feature of SET-WET which increases the
ease of model use is the lack of formatted model input. This style was chosen to simplify
inputting parameter values because it alleviates problems associated with incorrect value entries
due to differences in spacing and decimal places. It is suggested that inputs be entered similar to
the format the parameters are placed in the Fortran code (usually six entries per line) to facilitate
the comparisons of input parameter readings. Appendix A lists all possible entries to the model
59
W is free lying surface water; B is wetland substrate water Striped area represents wetland volume to which water level affects Shading represents wetland substrate
FIGURE 6: WETLAND DESCRIPTION FOR SET-WET MODEL WETLANDS
along with the required units. Appendix B details the order and entry of inputs to the model for
each respective submodel. Appendix C lists the SET-WET FORTRAN code.
The SET-WET model is designed to accept two forms of data inputs. The first option
requires hydrologic, sediment and nutrient inputs which consist of daily input values for various
hydrologic and nutrient parameters (daily inflow (m3/d); NO3-, NH4
+, DO etc., (mg/L)). These
values may have been collected or been generated by a NPS model such as ANSWERS or
BASINS. Output from the existing NPS models would be used as input to the SET-WET model.
The second option for data input is based on the SCS curve method (Equations 6 and 7). The
hydrologic inflow that is determined by the SCS method is combined with a runoff concentration
to determine the constituent’s inflow to the wetland system in the following manner:
RUNCONWATINPUTCONIN *= (37)
where CONIN is the constituent input to wetland (g); WATINPUT is the water input due to
catchment runoff (m3); and RUNCON is the concentration in the runoff for the individual
constituent (mg/L). Runoff concentration coefficients may be determined from empirical
analysis of site data or estimated from similar watershed situations.
60
Two different models (Thornthwaite method or pan method) may be used to model ET.
Six different options to describe the wetland outlet may be used to simulate the hydrologic
outflow from the wetland system, including five types for FWS systems, and one for SSF
wetlands. For the N cycle, the atmospheric deposition inputs, N fixation and volatilization may
be simulated. These components are not mandatory inputs as various wetland sites may or may
not be significantly affected by these processes. For the P cycle, adsorption may be modeled in
two different manners, using the Freundlich isotherm, which may be simplified to a linear
isotherm through choice of parameters (Equations 30 and 31).
The model is designed to control time management with a season and time period format.
Each season period consists of a user-defined number of time periods, with each time period
representing one day. To conserve memory, the season periods are programmed for a maximum
length of 500 days; however, this length can easily be changed in the program code. The
season/time period format minimizes data entry by allowing stable parameter values to be
entered on fewer occasions. Instead of entering data for individual days, values that stay the
same for longer time frames need only be entered once every season period. However, a change
in season periods allows the user to input new values for these parameters. For example, four
season periods could be used to simulate the four seasons of the year; plant growth and microbial
parameters may differ for these time periods and can be changed accordingly. In between season
periods, certain flagged entries will determine whether parameter values need to be changed; a
‘1’ entry signifying changes will be made, while a ‘0’ entry signifies no changes are necessary.
1. FWS vs. SSF Modeling
Due to the free lying water of the FWS system and the lack thereof for the SSF, there are
two pools of water simulated for FWS systems and one for SSF systems. To ease the transition
between coding for both systems, the same base parameter names have been used for both
wetland types soil/peat systems. For processes or pools that occur in both the surface water and
peat of the FWS system, the difference is signified in the programming code with a ‘W’ to
represent surface water, and a ‘B’ for wetland soil/peat bottoms, at the end of individual
parameter code name. Figure 6 shows how these respective pools are represented. For example,
the parameter code HETEROW is used to account for the number of heterotrophic bacteria in the
61
FWS wetland surface water, while HETEROB accounts for those in the wetland soil/peat of both
wetland systems.
The principal differences that exist between the modeling of FWS and SSF wetlands are
the inclusion of settling and resuspension rates for particulates, and the use of mass transfer rates
for dissolved constituents, to model movement in a FWS wetland. The fall rates of particles such
as particulate organic N, particulate organic C and sediment particles is based on the comparison
of the fall rate with the height of the water differential in the FWS surface (wetland water surface
minus wetland soil/peat top). The particles are considered to be homogeneously mixed through
the water surface volume and the settling amount is considered 100% if the fall rate of the
particles is greater than the difference between the wetland water surface and the top of the
wetland soil/peat layer. If the fall rate is less than the difference between the water surface and
the wetland soil then a proportion of the constituent, based on the outflow and water volume, is
assumed to settle.
Resuspension of sediment is based on a critical velocity, and is modeled using Equation
32. When the critical velocity is attained, a certain thickness of the soil/peat layer is affected.
From this thickness a user-defined percentage of the analyzed constituent will re-enter the water
column and be resuspended from the surface bottom. For these processes, the resuspension is
modeled as:
HBHT
RESTHCCONRESCONBCONRE
jjj −
=−
−1
1 ** (38)
where CONRE is the constituent resuspension amount (g constituents); CONB is the total
constituent amount in the wetland peat bottom (g constituents); CONRES is the percentage of
constituent that resuspends from the critical resuspension area (g resuspended cons./ g cons.
total); RESTHC equals the thickness of wetland soil to which the velocity resuspends
particles(m); HT is the top of the wetland soil/peat surface (m); and HB is the bottom of the
wetland soil/peat surface (m). The CONRES and RESTHC parameter values are determined
through calibration.
Mass transfer rates are used to model the transfer of constituents from the surface water
to the peat water. Mass transfer is modeled as:
62
ACCKMASSTC wsmt *)(**864 −= (39)
where MASSTC is the mass transfer amount for a constituent (g/day); Kmt is the mass transfer
coefficient (cm/sec); Cs is the concentration of the constituent in surface water (mg/L); Cw is the
concentration of the constituent in the peat water (mg/L); and A is the wetland area (m2).
B. MODEL COMPONENTS:
The following sections will describe the various submodels; their purpose and the
reasoning behind the equations and assumptions used in their development. For clarity and
conciseness, while explaining individual processes in the submodel descriptions, the ‘W’ and ‘B’
indicators at the end of each parameter code name will be omitted from the parameter names.
However, to differentiate which variables have two pools for the FWS system, a parenthesized
‘W/B’ will be placed after the initial parameter description. If the process occurs in only one of
the FWS pools they will be distinguished with a parenthesized W or B. The described processes
will occur in the SSF system, unless otherwise stated.
Figure 7 shows the relationships between the main code and the various submodels for
SET-WET. As Figure 7 shows, the main program calls every submodel a various number of
times. The number of times and whether a sub-model is called are dependent on the length of
the simulation and which cycles are modeled.
Figures 8 through 22 represent the relationships between modeled processes affecting the
respective cycles in SET-WET. The figures represent the various stocks, parameters and flows
that are accounted for by SET-WET during a simulation run. Appendix D explains the symbols
used in the respective figures.
1. Wetland Main Program:
The wetland main program is the bookkeeper and accountant of the model. It determines
when and if each submodel should be called, transfers data between time and season periods,
opens input and output files, and writes to the files. There are no inputs to the main program, but
it is the code that links all of the submodel components together.
63
BASE ( ONE TIME )!
DELTAH ( T )*,#
HYDROLOGY ( S + T )!
PHOSPHOROUS ( S + T )#
VEGETATION ( S + T )*,#
MAIN CODE
SEDIMENT ( S + T )#
CARBON ( S + T )*
DISSOLVED OXYGEN (S + T)*
BACTERIA ( S + T )*
NITROGEN ( S + T )*
Number of times the MAIN CODE calls each respective submodel for an entire simulation run.! Submodel is called whenever SET-WET runs a simulation* If the NCOB cycle is simulated, then submodel is called by the main code# If the phosphorous cycle is simulated, then submodel is called by the main code One Time - submodel is called only once for the entire simulation run T – submodel is called once every time period S + T – submodel is called by main code once every season and time period
FIGURE 7: RELATIONSHIP OF SET-WET MAIN CODE TO SET-WET SUBMODELS
2. Base Submodel:
The base submodel describes the basic design of the wetland system and determines
which nutrient cycles will be modeled. As stated before, the input values are not in a fixed
format; however, it is suggested that the values are entered in the same manner of the Fortran
code, or similar to the manner that it is displayed in Appendix B.
64
3. Hydrology Submodel:
The hydrology submodel is active for every simulation. The model represents two pools
of water for the FWS simulations and one for the SSF wetland simulations (Figures 8 and 9).
The model assumes that the FWS wetland always has water lying upon the surface and is never
completely exhausted of free lying surface water. It further assumes that the water flow for a
SSF wetland is always beneath surface level. For FWS simulations, the soil and peat area is
assumed to be saturated at all times. A valid assumption for many constructed wetlands would
be that they are lined, thus infiltration/percolation is minimal; however, there is a percolation and
infiltration option that allows a daily set amount of water to be released or enter the system.
Water input could be derived from four sources; 1) watershed catchment runoff, 2) direct
precipitation input, 3) percolation additions, and 4) point source additions. Although initially
designed for constructed wetlands, the model may analyze any natural wetlands that follow these
conditions.
The total amount of water in the system is determined using a simple mass balance
similar to Equation 2. The water budget for the model may be written as:
AETPIPSDVBMVQQQdt
dVoutpc *)( −−+∆+∆+−+= (40)
where dV/dt is the change in surface storage (m3/day); Qc is the runoff from the catchment
(m3/day); Qp are the water flow from possible point source additions (m3/day); Qout is the outflow
rate (m3/day); ∆BMV is the change in living biomass volume in the surface water (m3/day);
∆SDV is the change in standing dead volume in surface water (m3/day); P is the daily
precipitation rate (m/day); PI is the percolation/infiltration rate (m/day); ET is the
evapotranspiration rate (m/day); and A is the wetland surface area (m2), determined from the
input wetland length and width.
As stated earlier, there are two options available for determining runoff from the
watershed; the daily option and the SCS method. The daily option requires input of Qc on a daily
basis, where the amount is either known or determined with an existing NPS model. If the SCS
65
method is used, runoff will be determined from the daily precipitation and the SCS curve
number.
The outflow from the system is a function of the water volume, and the type of outflow
device associated with the wetland, while ET is a function of daily temperature and local climate.
There are six different options that can be chosen to represent the wetland outlet, five for FWS
wetlands and one for SSF wetlands.
Outlet option 1 is for rectangular weirs. A modified Francis equation (Benefield et. al,
1984) is used to describe the head-discharge relationship for weirs of this type and is determined
by:
2
3
)2.0(843.1 HHLQ ××−×= (41)
where Q is the discharge (m3/s); L is the weir length (m); and H is the head on the weir (m). The
head on the weir is the difference between the wetland water height (HI) and the bottom level of
the outflow weir (HOUT, m). The Francis equation allows both suppressed and contracted weir
outflows to be determined.
If the water height in the system is lower than HOUT, then it is assumed that outflow
from the system is zero. If the water height in the system is higher than the HOVER (water
height completely above the weir) mark, outflow is the addition of the overflow through the weir
and the water that spills over the weir height.
Outlet option 2 is for fully contracted v-notch weirs of any angle between 25 and 100
degrees (Kulin and Compton, 1975). A modified Kindsvater-Shen relationship is used to
determine outflow:
2/51*)
2tan(**363.2 ee hCQ
θ= (42)
where Q is the discharge over the weir (m3/s); Ce is the effective discharge coefficient; θ is the
angle of the v-notch (degrees); h1 is the head on the weir (m); kh is the head correction factor (m,
function of θ); and h1e is equivalent to the sum of h1 and kh.
66
WATER VOLUME SURFACE
WATER VOLUME BOTTOM
Catchment runoff inflow
Precipitation
Daily Precipitation
Point Flow
Outflow
Evapotranspiration
Evap=1;Panevap
Length
Width
Surface Area
HI
Outlet=1;Hout,Outw idth,Hover,Ncont
Outlet=2;Angvnot,Diseffc,Khcoef,Hout
Outlet=3;Hout,Areapipe,Contc
Outlet=4;Flow out,Toppump
Outlet=5;Alpha,Beta,Kcoeff,Hout
Evap=0;Air Temp.,Daylength,
HT
Peat Porosity
HB
Evap=0;Evap. Coef.,Heat Index
Percolation/Infiltration
Percolation/Infiltration Rate
Percolation/Infiltration Flag
Point Flow Flag
Hydtype=0,Water input
Hydtype=1,Watershed Area,Length, Width,SC Curve Number
Hydtype=1, Daily Precipitation
Biomass Volume
Standing Dead Volume
% Biomass Underw ater
% Standing deadUnderw ater
FIGURE 8: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE SUBMODEL FOR FWS WETLANDS IN THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
67
WATER VOLUME BOTTOM
Outflow
Evapotranspiration
Surface Area
HT
Evap=0;Air Temp.,Daylength,
Point Flow Flag
Outf low
Evapotranspiration
Surface Area
HB
Catchment runoff inf low
Precipitation
Daily Precipitation
Point Flow
Outflow
Evapotranspiration
Evap=1;Panevap
LengthWidthSurface Area
HI
Hout PorosityBed slope
Hydraulic Conductivity
Peat Porosity
Evap=0;Evap. Coef.Heat Index
Hydtype=0,Water input
Hydtype=1,Watershed Area,Length, Width,SC Curve Number
Hydtype=1, Daily Precipitation
Percolation/Infiltration Rate
Percolation/Infiltration Flag
Percolation/Infiltration
FIGURE 9: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HYDROLOGIC CYCLE SUBMODEL FOR SSF WETLANDS IN THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
68
Outlet option 3 is for a fixed non-perforated pipe that removes water from the wetland
according to the hydraulic pressure on the outlet opening.
)(*81.9*2** HOUTHIAREAPIPECONTCOUTFLOW −= (43)
where CONTC is the contraction coefficient for the pipe; AREAPIPE is the area of opening for
the outflow pipe (m2); HI is the water level in the wetland relative to a zero datum (m); and
HOUT is the height (m) of the outflow opening relative to a zero datum.
Outlet option 4 represents pumped discharge that removes a constant amount of water
from the system per day. The input values are TOPPUMP and FLOWOUT and represent
respectively the height necessary in the wetland for water to flow from the system (m, with
respect to zero datum) and the amount removed per day when outflow is continuos through the
day (m3). Flow is continuos when the wetland water level is above the TOPPUMP location.
For each of these outlet options, the water velocity in the wetland system is determined
after the outflow in the system is determined. Water velocity is calculated by dividing the
outflow through the system by the wetland cross-sectional area (Equation 8).
For option 5, water velocity is determined first and is then used to calculate outflow.
Water velocity is determined using the rate law equation (Equation 11). In the rate law equation
when flow is in the turbulent range, α (slope component) is approximately 0.5, and if flow is in
the laminar range, α=1.0. The β (depth component) value is usually in the range of 2≤β≤4, but is
dependent on microtopography as well as the stem-density depth distribution of the wetland
(Hammer and Kadlec, 1986). The calculated velocity in the wetland system is multiplied by the
wetland cross-sectional area to determine outflow.
Outlet option 6 is for SSF wetlands. To determine the outflow from the SSF wetland the
water flow through the wetland substrate must be determined. Darcy’s equation (Equation 18) is
used to model this type of flow (Kadlec, 1989; Jorgensen et. al, 1988). Although flow through
the SSF wetland may be unsteady due to factors such as precipitation and ET, it is assumed that
it is steady for short time periods, which is a requirement for Darcy’s law. The system porosity
is considered constant throughout each season period. Changes in porosity may be accounted for
by the user during changes in season periods; however, the effects of peat accumulation on
hydraulic conductivity and porosity are not determined by the SET-WET model. The hydraulic
69
gradient in the system is assumed to be the greater value between the difference in elevation
between the wetland water surface and the outflow pipe height, or the bed slope.
The model runs on a daily time step, but to improve the accuracy of the model, the
outflow data is calculated on an hourly basis. The amounts of individual water inputs and
outputs (except outflow) are first divided by twenty-four to give an hourly, instead of daily rate.
These values are used to determine outflow for a one-hour time period and a new wetland water
surface height (HI) is determined. The new water surface is then used in the following outflow
calculation and continued until 24 one-hour time steps are conducted. Outflows for each hourly
time step are then added to determine the total daily outflow from the system. No other
component of the model is run on a time step shorter than a day.
Evaporation and transpiration losses are difficult to determine separately, and thus are
combined as an ET parameter. Various models are available for estimating ET. Thornthwaite’s
method (Equations 12, 13, 14 and 15) and the pan method were used in SET-WET because their
required input data are readily available. For the Thornthwaite method, daily ET approximations
are determined by using the daily average air temperature instead of the monthly average
temperature and then dividing by 30 days:
=
30
**
*10*
12
LW
HINDEX
ATDLET
a
(44)
where DL is the daylength (hours daylight/12 ); AT is the daily air temperature (°C); HINDEX is
the heat index based on the historical average monthly temperatures (unitless); a is a determined
coefficient based on the heat index (unitless); W is the average wetland width (m); and L is the
average wetland length (m). ET is considered zero when temperatures are below freezing.
4. Vegetation Submodel:
Solar radiation, temperature, water, and available nutrients are a few of the factors
affecting plant growth. Solar radiation data is not readily available, therefore, a simple model of
nutrient uptake was used in SET-WET to model vegetative growth (Figure 10). A simple
wetland model is sufficient for overall wetland features (Hammer and Kadlec, 1988) which is
70
important because most proposed plant models are very complex. The model is similar in
structure and form to the one previously described by Gidley (1995).
There are two components to the vegetation submodel, BIOMASST and STANDDT.
The total amount of living plant biomass in the wetland system is defined as BIOMASST, while
STANNDT refers to the dead aboveground plant parts before they have had a chance to fall to
the ground and become litter.
The total amount of plant biomass in the system is determined by:
BIOMASSTJ = BIOMASSTJ-1 + (BIOMGROWJ
- BIOMDTHJ - BIOMOUTJ)*dt (45)
where BIOMASST is the amount of live plant biomass in wetland (g biomass); BIOMGROW is
the amount of plant growth in the system (g biomass/day); BIOMDTH is the amount of biomass
death in the system (g biomass/day); BIOMOUT is the amount of biomass removal from system
(g biomass/day); J equals the time period number (day from start of season period), and dt is the
daily time step (day).
The amount of standing dead in the system is accounted for by:
STANDDTJ = STANDDTJ-1 + (BIOMDTHJ
+ PHYDEGJ - SDEADOUTJ)*dt (46)
where STANDDT is the amount of standing dead biomass in wetland (g biomass); PHYDSEG is
the amount of physical degradation of standing dead (g biomass/day); and SDEADOUT is the
amount of standing dead removal from the wetland (g biomass/day).
Biomass growth is simulated with a simple zero order equation (Equation 21) that
multiplies the wetland area by a biomass growth rate. At the beginning of the growing season,
biomass of an assumed constant composition throughout the wetland system, grows at a linearly
increasing rate for a user-specified number of days, until the maximum growth rate is reached.
This growth continues throughout the growing season until winter, at which time the growth rate
linearly decreases to zero at a user-defined rate. The growth rates are ramped to represent the
gradual changes within the system and to reduce model instabilities. The start and end of the
71
TOTAL BIOMASS
TOTAL STANDING
DEAD
Biomass growth
Biomass growth
Winter flag
Day w inter turnsTime period
Point biomass decays to Days degradation occurs over
WINKCOLENGTH WIDTH
Biomass Death
Living biomass removal
Biomass removal flag
Amount removal of BIOMASST
Day removal of BIOMASSTStanding dead removal
STANDD removal flag
Amount of STANDD
Day removal of STANDD
Time period
Physical Degradation
Biomass degradation rate
FIGURE 10: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE VEGETATION CYCLE SUBMODEL OF THE
SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
72
growing season will differ by location; but a general rule would be that the start and end of the
growing seasons corresponds with the first and last frosts in the area. The time periods over
which plant growth increases and decreases are estimated, as literature values are unavailable. It
is assumed that biomass growth is not limited by lack of nutrients or water supply. Since little is
known about the dynamics of below ground parts of plants, it is not modeled.
Biomass growth and death are assumed to be dependent and can not occur
simultaneously. Biomass death only occurs in the winter and is simulated with a simple first
order equation (Equation 22). When biomass death begins it becomes standing dead plant
material at an exponential rate. The percentage of which biomass degrades (DEGBIO) and the
length of time (DAYSDEG) over which biomass degradation occurs is input by the model user.
The standing dead physically degrades to other sources at an exponential rate over the
course of one year. Since this plant mass is not lying on the ground, degradation occurs due to
physical processes such as rain and wind rather than microbial processes.
An option to account for removal of plant material from the system is available. This
option can be used to represent anthropomorphic removal, or any other removal from the system.
The day from the start of the season period, and the amount of removal are input by the user and
removed from the system accordingly.
5. Nitrogen/Carbon/DO/Bacteria Relations:
As examined in the literature review, the N cycle is not an independent process, but one
whose reactions are dependent upon many factors. The C, DO, and microbiological cycles all
play a role in the rates and reactions of N processes which occur within the wetland system.
Each cycle was modeled separately, and the relationships between the respective cycles are
expressed in Figures 11 through 22. Each cycle will be examined individually, but the
relationships between them must always be accounted for. The work in the N, C, DO, and
bacteria relations are based on the work by Gidley (1995) and Parnas (1975).
73
6. Carbon Submodel:
The C cycle consists of either five (SSF) or seven (FWS) state variables (Figures 11 and
12). The state variables are; biomass C (W), standing dead C (W), particulate organic C (POC,
W/B), dissolved organic C (DOC, W/B), and refractory C.
Biomass C and standing dead C components are intertwined heavily with the vegetative
submodel as the biomass C is calculated as:
BIOMASSJ = BIOMASSJ-1 + ((BIOMASSTJ
– BIOMASSTJ-1) * BIOCCONT)*dt (47)
where BIOMASS is the total plant C biomass (g C), and BIOCCONT is the fraction of C in
plants (g C/g biomass). The growth, death and removal of biomass C have already been
accounted for by the BIOMASST term from the vegetative cycle. The standing dead C values
are determined similarly:
STANDDCJ = STANDDCJ-1 + ((STANDDTJ – STANDDTJ-1)
* BIOCCONT) – DOCLEACHJ - PHYSDEGCJ)*dt (48)
where STANDDC is the dead biomass that has not become litter (g C); DOCLEACH is the
leaching of standing dead to DOC (g C/day); and PHYSDEGC is the conversion of standing
dead C to homogeneous litter (POC) (g C/day). Standing dead physically degrades to other
sources at an exponential rate over the course of one year. Physical degradation is an input
source to the POC pool, which Jansson and Persson (1982) define as a homogenous organic
mixture of decaying plant litter, sloughed microbial cells and particulate influent wastewater
BOD. Over the first 30 days of winter, 15% of the plant nutrients are quickly lost by leaching to
DOC and DON (Heliotis and DeWitt, 1983; Kadlec, 1986).
74
DONB
DOCB
DO Bottom Water Bottom
WATER VOLUME BOTTOM
Water Surface Water Surface
POCBBIOMASS CARBON DOCW
TOTAL BIOMASS
STANDING DEAD CARBON DON SURFACE
DO SURFACEPOCW
WATER VOLUME SURFACE
PONW
REFC
Outf low
DOC outf low
Soluble BOD inf lowBOD C fraction
BOD runoff inf luent
BOD particulate fraction
Runoff inflow
Point source inf low
Point flow BOD conc.
DOC leaching
Leaching Rate
Biomass C content
Physical Deg. Carbon
Biomass deg. rate
Microbial death W
Microbial C content
HT death WNS death W
Particulate BOD inf low
BOD C fraction
HT=0; BOD runoff
inf luent conc.;HT=1; BOD runoff coef.
BOD particulate fraction
Point source inf low
Point flow BOD conc.Runoff Inflow
Peat accumulation
Peat C content
Peat accumulation rate W
DOC min\imob W
POC min\imobw
TOCW
TONW
HT grow th W
HT yield W
Microbe Total C:N W
POC outf low
POC resuspension
POC settlling
POC falling rate
HT
HI
POC size
POC resuspension crit.
MANNC
HB
Water velocity
DOC mass transfer
DOC min\imob B
TOCB
TONB
HT grow th B
HT yield B
Microbe Total C:N B
POC min\imob B
Microbial death B
Microbial C content
HT death B NS death B
DOC MT coef.
Length
Width
POC resusp. rate
POC resusp. thick
Percolation/Inf iltration Total DOC percolation
DOC Percolation/Inf iltration Total
FIGURE 11: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET
MODEL
(See Appendix D for explanation of figure’s symbols)
75
TOTAL BIOMASS
Water Volume Bottom
BIOMASS CARBON DOCB
STANDING DEAD CARBON DON BOTTOM
DO BOTTOMPOCB
WATER VOLUME BOTTOM
PONB
REFC
Outf low
DOC outf low
Soluble BOD inf low
BOD C f raction
BOD runof f
inf luent conc.
BOD particulate f raction
Runof f inf low
Point source inf low
Point f low BOD conc.
DOC leaching
Leaching Rate
Biomass C content
Physical Deg. Carbon
Biomass deg. rate
Microbial death B
Microbial C content
HT death B
NS death B
Particulate BOD inf low
BOD C f raction
HT=0; BOD runof f inf luent conc.;HT=1; BOD runof f coef .
BOD particulate f raction
Point source inf low
Point f low BOD conc.Runof f Inf low
Peat accumulation
Peat C content
Peat accumulation rate B
DOC min\imob B
POC min\imob B
TOCB
TONB
HT grow th B
HT yield B
Microbe Total C:N B
Percolation/Inf iltration Total
DOC percolation conc.
DOC Percolation/Inf iltration Total
Length
Width
FIGURE 12: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE CARBON CYCLE SUBMODEL FOR SSF WETLANDS OF THE SET-WET
MODEL
(See Appendix D for explanation of figure’s symbols)
76
The mass balance for POC depends on which type of wetland is being examined. For
FWS wetlands the mass balances for the two pools are:
POCWJ = POCWJ-1 + (PBODINJ + MICRODWJ
- PEATCACWJ – POCMINIWJ – POCSETJ
+ POCREJ + PHYSDEGCJ – POCOUTJ)*dt (49)
and
POCBJ = POCBJ-1 + (POCSETJ + MICRODBJ
- POCMINIBJ – PEATCACBJ -POCREJ)*dt (50)
while for SSF wetlands the mass balance of POC is
POCBJ = POCBJ-1 + (PHYSDEGCJ + POCSETJ + PBODINJ
+ MICRODBJ - PEATCACBJ - POCREJ)*dt (51)
where POC(X) is the amount of POC in water volume (g); PBODIN is the particulate BOD
influx from catchment runoff (g/day); MICROD(X) is the sloughing of dead microbial cells to
POC (g/day); PEATCAC(X) is the accumulation of refractory solids from surface water (g/day);
POCMINI(X) is the conversion of POC to microbial biomass and carbon dioxide (g/day);
POCSET is the amount of POC that settles from POCW (g/day); POCRE is the amount of POC
that resuspends from POCB (g/day); POCOUT is the effluent POC amount (g/day); and
PHYSDEGC is the conversion of standing dead to POC (g/day).
The mass balance of DOC also differs for the two wetland types. For FWS wetlands the
mass balance pools are:
DOCWJ = DOCWJ-1 + (SOBODINJ – DOCOUTJ
- DOCMINIWJ + DOCMTJ)*dt (52)
and
DOCBJ = DOCBJ-1 + (DOCLEACHJ – DOCMINIBJ
- DOCMTJ - DOCPERCJ)*dt (53)
77
while for SSF wetlands, the mass balance is:
DOCBJ = DOCBJ-1 + (SOBODINJ + DOCLEACHJ
- DOCMINIBJ – DOCOUTJ)*dt (54)
where DOC(X) is the amount of DOC in the water volume (g); SOBODIN is the soluble BOD
influx from catchment runoff (g/day); DOCOUT is the effluent DOC (g/day); DOCMINI(X) is
the conversion of DOC to microbial biomass and carbon dioxide (g/day); DOCMT is the amount
of DOC that diffuses at the interface (g/day); DOCLEACH is the leaching of DOC from standing
dead (g/day); and DOCPERC equals the amount of DOC lost/gained from infiltration/percolation
(g/day).
Since the model is designed for NPS pollutants, a general assumption of the chemical
composition of the material coming into the wetland system is difficult to determine. This value
is site-specific and can be estimated with the approach presented by McCarty (1975).
Microbial sloughing contributes to the POC content and is the product of the microbial C
content and the sum of heterotroph and autotroph death. Microbes are not considered to be a part
of POC until they die and are accounted for in the bacteria counts. It is assumed that microbial
death contributes only to POC, not DOC, since most microbes in wetland systems are associated
with plant litter and soil organic matter. Particulate BOD inflow is determined as:
PBODINJ = 1.4 * BODPFRAC * BODCFRAC * (BODINFCO
* WATINPUTJ + APFLOW * BODCONC) (55)
where PBODIN is the particulate BOD flux from waste additions (g C/day); BODPFRAC is the
fraction of influent BOD that is in particulate form (g part. BOD5/g total BOD5); BODCFRAC is
the fraction of C in influent BOD (g C/g BOD5); BODINFCO is the BOD5 concentration in
catchment runoff (mg BOD5/l); WATINPUT is the amount of catchment runoff (m3); APFLOW
is the amount of point flow contributions (m3); and BODCONC is the BOD5 concentration in
point source (mg BOD5/l). Five day BOD is converted to ultimate BOD assuming k=0.25
(Tchobangolous and Burton, 1991).
78
For SSF wetlands, removal of particulate constituents such as PON and POC is assumed
to be 100%. This simplification is necessary for there are no mechanistic models available for
estimating particulate removal in porous substrates (Gidley, 1995). SSF wetland effluent TSS
values are generally low, thus minimal error should be introduced. For FWS wetlands,
particulate constituents of this type are subject to settling and resuspension like any other
particle. Settling and resuspension of POC is modeled as described by Equation 38.
The submodel of C and N immobilization, mineralization, and ammonification is based
on the work of Gidley (1995) and Parnas (1975). The total amount of C that is required for
heterotroph growth (HT) is calculated as heterotrophic growth (HTGROW) divided by the
heterotrophic yield (HTYIELD) where HTGROW has units of g microbes/day and HTYIELD
has units of g cells/g C. Heterotrophic yield is a function of DO concentration (Henze et al.,
1986; Tchobongolous and Burton, 1991).
Whether fractions of DOC and POC (or DON, PON, and NH4+
to be discussed later), are
utilized for microbial growth and energy are determined by the required substrate C:N ratio, as
defined by the microbe total C:N (MICTCN). The MICTCN is directly related to the DO
concentration (DOXYC) as (Parnas, 1975):
MICTCN = 80. - 10. * DOXYC (56a)
(For DOXYC > 0.0 and DOXYC < 5.0)
MICTCN = 30. (56b)
(For DOXYC > 5.0)
The increase in the required MICTCN is the result of anaerobic conditions, where
reactions are less efficient thereby requiring more C for equivalent amounts of cell growth
(Gidley, 1995). Combined with the actual amounts of C and N in the system, the MICTCN ratio
can be used to determine if the processes in the wetland are either N or C limited. The linear
change in MICTCN is assumed to reflect the anaerobic microsites within aerobic environments
and the presence of aerobic rootzone sites at low DO concentrations.
POC, DOC, PON, DON, and NH4+ are all available for microbial growth. The wetland
C:N ratio is calculated as total organic C (TOC) divided by total organic N (TON) where
TOC=POC+DOC and TON=PON+DON. If the ratio of TOC/TON is greater than MICTCN,
79
microbial growth is N limited and the amount of PON and POC utilized depends on the relative
fraction of PON to other N sources (PON/(TON +NH4)). The fraction of required C and N
derived from dissolved organic materials depends on the ratio of DON and TON + NH4. If
TOC/TON is less than MICTCN, microbial growth is C limited and the amount of PON and
DON that is utilized will depend on the relative proportions of POC and TOC. The fraction of
required C and N derived from dissolved components is proportional to DOC divided by TOC.
The respective C fraction utilized is proportional to the organic N utilization. The amount of
each N source that is immobilized depends on its relative fraction, with the excess organic N
(ON) being converted to NH4+ by ammonification. These guidelines help establish the following
relationships for DOC mineralization/immobilization (DOC min/imob) and POC
mineralization/immobilization (POC min/imob):
DOC min/imob = DON/TON * HTGROW/HTYIELD (57a)
(when TOC/TON>MICTCN)
DOC min/imob = DOC/TOC * HTGROW/HTYIELD (57b)
(when TOC/TON<MICTCN)
and
POC min/imob = PON/TON * HTGROW/HTYIELD (58a)
(when TOC/TON>MICTCN)
POC min/imob = POC/TOC * HTGROW/HTYIELD (58b)
(when TOC/TON<MICTCN)
A part of POC is considered refractory and is thus permanently lost to the soil through
peat accumulation. Peat accumulation is the product of the peat accumulation rate and the peat C
content. DOC is assumed to be 100 percent labile (Gidley, 1995). DOC accumulation occurs
because of DOC leaching, and soluble BOD inflow. Soluble BOD inflow is the contributions of
TOC that are not associated with POC and can be determined by changing the BODPFRAC
parameter in Equation 55 to (1-BODPFRAC). Microbes utilize DOC for energy and cell growth.
Outflow of DOC is calculated by multiplying the DOC concentration by the outflow.
80
7. Nitrogen Submodel:
Processes that are always modeled by the N submodel include ammonification (W/B),
immobilization (W/B), nitrification (W/B), denitrification (W/B), and peat accumulation (W/B).
As stated before, there are flags that signify whether ammonia volatilization (W), atmospheric
deposition (W) and N fixation (W) will also be included in the overall N cycle modeling. These
are user-defined options, whereby all or none of the processes have to be included in the
simulation. Sorption of NH4+ is not modeled as it is assumed that sorbed NH4
+ is still available
to attached microbes. The state variables include dissolved organic N (DON; W/B), particulate
organic N (PON; W/B), NH3 and NH4+-N (NH4; W/B), nitrate N (NO3; W/B), immobilized N
(W/B), and refractory N (Figures 13 and 14).
Influent DON enters the wetland through catchment runoff, point source inflow, percolation,
and atmospheric deposition (dry and wet):
DONINJ = (1 – ONPARTF) * WATINPUTJ * ORGNINCJ
+ APFLOW * DONCONC (59)
where DONIN is the incoming DON to wetland (g N/day); ONPARTF is the fraction of organic
N in particulate form (g PON/g TON); ORGNINC is the influent ON from catchment runoff (mg
ON-N/L); and DONCONC is the ON concentration from point sources (mg ON-N/L).
If the atmospheric deposition option is chosen and there is no precipitation, then the total
dry deposition for that day is the dry deposition rate (g DON/ m2) multiplied by the wetland
surface area. If precipitation does occur, then the input to the wetland is the total precipitation
amount multiplied by the concentration of DON in the rain.
Dissolved organic N also accumulates by leaching of N from the standing dead, due to
the physical degradation of standing dead biomass. It is assumed that accumulation occurs based
on a constant proportion of C to N in the plant biomass. Dissolved organic N leaching is
calculated as the product of the DOC leaching rate and the biomass C:N ratio (BIOMASS C:N).
Dissolved organic N outflow is determined by multiplying the wetland DON concentration by
the outflow. As previously described, a part of DON is reincorporated into microbial biomass
during the degradation of organic C; while the rest is wasted as ammonium. These relations are
as follows:
81
WATER VOLUME BOTTOM
NO3 SURFACE
IMMOBILIZED N SURFACE
NH4 SURFACE
NO3 BOTTOMWater Surface
Water Bottom
WATER VOLUME SURFACE
DOC SURFACE
NH4 BOTTOM
DON SURFACE
DON BOTTOM
Water Surface
Water Bottom
NH4 SURFACE DON SURFACEIMMOBLIZED N
SURFACEPON SURFACE
AnaHT NO3 yield B
Outflow
NO3 inf luent
HT=0; NO3 runof f
influent conc.;HT=1; NO3 runof f
coef f.
Nitrif ication B
NS Yield B
NH4 influent
DON influent
HT=0; Inf luent runof f
ON conc.;
HT=1; ON runoff coef f.
ON particulate fraction
runoff inflow
Point source inflow
HT=0; inf luent
runoff NH4 conc.HT=1; NH4 runoff
coef f.
Point NH4 conc.
Point NO3 conc
Dry A.D. NO3
Wet A.D. NO3
Length
Width
Direct precip.
A. D. flag
Point ON conc.
Wet A.D.
Dry A.D.
Wet A.D. DONDry A.D. DON
HT yield B
N leaching
Biomass C:N
DOC leaching
NO3 outf low
Volatization
pH factorpH
Nitrogen Fixation
Nitrogen fixation rate
Width
Length
Denitif ication B
Anaerobic HT grow th W
HT NO3 yield W
Nitrif ication W
NS Yield W
biomass grow thNS grow th W
Biomass C:N
H
nitrate uptake W
ammonium uptake W
PON
Microbial N content
HT NH4 immob. W
HT g
Plant
Plant up division
HT yield W
TON W
DON immobilization W
TOCW
Microbe Total C:N W
HT grow th W TONWTONW & NH4W
Microbial N Content
HT NH4 immob. W
DON ammoniafication W
DON ammoniafication B
TON B
Outflow
DON outflow
NH4 outflow
Volatilization rate
NO3 mass
transfer
NO3 MT coef.
Surface
Area
NH4 MT coef .NH4 MT
Surface area
DON mass transfer
DON MT coef .
DON Perc./Inf. Conc.
DON Per
Perc./Inf. Total
NH4 PeInf. Co
Perc./Inf
FIGURE 13: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
82
WATER VOLUME BOTTOM
POC SURFACE
IMMOBILIZED N SURFACE
PON SURFACE
POC SURFACE
NO3 BOTTOM
POC BOTTOM
IMMOBILIZED N BOTTOM
NH4 BOTTOM PON BOTTOM REFRACTORY N
DOC BOTTOMPOC BOTTOM
DON BOTTOM
Water Surface
Water Bottom
NH4 BOTTOM DON BOTTOM
PON BOTTOM
PON inf luent
Peat accumulation rate W
runoff
ON Particulate fraction
HT=0; Inf luent runoff ON conc.;HT=1; ON runoff coeff .
Point source flow
Point ON conc.
Anaerobic HT grow th BO3 yield B
Nitrif ication B
NS Yield B
biomass grow thNS grow th B
Biomass C:N
HT death B NS death B physical degradation C
HT grow th B
nitrate uptake B
ammonium uptake B
d B
DON immobilization B
TOCB
Microbe Total C:N B
PON immobilization B
PON ammonif ication B
Microbial N content
HT grow th B
death BTONB & NH4B
A. D. flag Dry A.D. PON
Wet A.D. PON
Length
Width
Direct precip.
HT NH4 immob. B
TOCBTONB
Microbe Total C:N B
HT yield
HT grow th B TONBTONB & NH4B
Microbial N Content
HT NH4 immob. B
fication B
Biomass C:N
HT death W NS death W physical degradation C
HT grow th WPON immobilization
PON ammonif ication W
death TONW & NH4W
mmob. W
TOCBTONW
Microbe Total C:N
HT yield WHT grow th W
Plant up division
tion B
Peat N content
Peat accumulation rate B
PON resuspension
PON falling rate
HT
HI
PON size
PON resusp. crit. vel.
MANNC
HB
Water velocity
PON settling
PON resusp. thick.
PON resusp rate
oef.
DON Perc./Inf. Conc.
DON Perc./Inf il.
Perc./Inf. Total
NH4 Perc./Inf. Conc.
NH4 Perc./Inf il. Perc./Inf. Total
NO3 Perc./Inf. Conc.
NO3 Perc./Inf il.
Perc./Inf. Total
FIGURE 13 (CONT.): RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN
CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
83
WATER BOTTOM
WATER VOLUME BOTTOM
NO3 BOTTOM
IMMOBILIZED N BOTTOM
NH4 BOTTOM
WATER BOTTOM
DOC BOTTOM
DON BOTTOM
NH4 BOTTOMWATER BOTTOM
Denitrification B
Anaerobic HT growth BHT NO3 yield B
Outflow
NO3 influentHT=0; NO3 runoff influent conc.;HT=1; NO3 runoff coeff.
Nitrification B
NS Yield B
NH4 influent
DON influent
HT=0; Influent runoff ON conc.;HT=1; ON runoff coeff.
ON particulate fraction
runoff inflow
Point source inflow
HT=0; influent runoff NH4 conc.HT=1; NH4 runoff coeff.
Point NH4 conc.
biomass growthNS growth B
Biomass C:N
HT
nitrate uptake B
ammonium uptake B
Point NO3 conc
Dry A.D. NO3
Wet A.D. NO3
Length
Width
Direct
A. D. flag
Point ON conc.
Wet A.D. NH4
Dry A.D. NH4
Wet A.D. DON
Dry A.D. DON
DON ammonification
HT yield B
NH4 outflow
DON outflow
DON immobilization B
TON BTOCB
Microbe Total C:N B
N leaching
Biomass C:N
DOC leaching
PON
Microbial N content
NO3 outflow
HT NH4 immob. B
HT growth B T
HT NH4 immob. B
Volatization
Volatization rate
pH factor
pH
Nitrogen Fixation
Nitrogen fixation rate
Width
Length
Plant up division
Perc./Inf. Total
NO3 Percolation/Infiltration Conc.
NO3 Perc./Infil.
Perc./Inf. Total
NH4 Perc./Inf. Conc.
NH4 Perc./Infil.
Perc./Inf. Total
DON Perc./Infil.
DON Perc./Inf. Conc.
FIGURE 14: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE SUBMODEL FOR SSF WETLANDS OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
84
WATER VOLUME BOTTOM
POC BOTTOM
IMMOBILIZED N BOTTOM
PON BOTTOM REFRACTORY N
POC BOTTOM
DOC BOTTOM
WATER VOLUME BOTTOM
NH4 BOTTOM DON BOTTOM
PON BOTTOM
PON influent
Peat N content
Peat accumulation rate B
catchment runoff inflow ON Particulate fraction
HT=0; Influent runoff ON conc.;HT=1; ON runoff coeff.
Point source flow
Point ON conc.
HT growth B
w
biomass growthBiomass C:N
HT death B NS death B physical degradation C
HT growth B
ake B
DON outflow
DON immobilization B
Outflow
be Total C:N B
PON immobilization B
PON ammonification B
bial N content
HT growth B
death BTONB & NH4B
A. D. flag
Dry A.D. PON
Wet A.D. PON
Length
Width
Direct precip.
HT NH4 immob. B
TOCBTONB
Microbe Total C:N B
HT yield B
T growth BTONBTONB & NH4B
Microbial N Content
HT NH4 immob. B
Perc./Inf. Total
Perc./Inf. Total
FIGURE 14 (CONT.): RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE NITROGEN CYCLE SUBMODEL FOR SSF WETLANDS
OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
85
DONIM = DON/ (TON+NH4) * MICRONC *HTGROW (60a)
(when TOC/TON>MICTCN)
DONIM = DOC/TOC * (MICRONC * HTGROW - HTNH4IM) (60b)
(when TOC/TON<MICTCN)
PONIM = PON/(TON+NH4)*MICRONC*HTGROW (61a)
(when TOC/TON>MICTCN)
PONIM = POC/TOC *(MICRONC * HTGROW - HTNH4IM) (61b)
(when TOC/TON<MICTCN)
HTNH4IM = MICRONC *HTGROW *NH4 / (TON +NH4) (62)
DONAM = 0 (63a)
(when TOC/TON > MICTCN or DOC < 0.1)
DONAM = DOC/TOC * HTGROW/HTYIELD (63b)
*DON/DOC – DONIM
(when TOC/TON < MICTCN)
PONAM = 0 (64a)
(when TOC/TON >MICTCN or POC < 0.1)
PONAM = POC/TOC * HTGROW/HTYIELD * PON/POC (64b)
-POC/TOC * (MICRONC * HTGROW - HTNH4IM)
(when TOC/TON < MICTCN)
where DONIM is DON immobilization (g N/day); DONAM is DON ammonification (g N/day);
MICRONC is the microbial N content (g N/g microbes); HTGROW equals the heterotrophic
(HT) growth rate (g microbes/day); HTYIELD is the yield of HT bacteria (g microbes/g C
degraded); HTNH4IM is the NH4 utilized by HT bacteria during organics degradation (g NH4+-
N); PONIM is PON immobilization (g N/day); and PONAM is PON ammonification (g N/day).
86
Another source of DON to the wetland system is N fixation. N fixation is assumed to
occur at a constant user provided rate. Mass transfer of DON between the surface and bottom of
the wetland is modeled as discussed previously (Equation 39).
Particulate organic N dynamics are similar to those for DON, except that PON is
assumed to also accumulate in peat as refractory N. Peat N accumulation is the product of the
peat accumulation rate and the peat N content. Particulate organic N outflow is zero for SSF
wetlands and is calculated as the product of the wetland concentration and the outflow.
NH4+ and NH3 are referred to as NH4 only because at neutral pH, the predominate form is
NH4+. NH4
+ accumulates through NH4+ influent (catchment, point, percolation, and
atmospheric), and PON and DON ammonification. NH4+ influent is modeled similar to DON
(Equation 60), while PON and DON ammonification are modeled as previously discussed
(Equations 64 and 65). Plant and microbial uptake, NH4+ effluent, infiltration and nitrification
all decrease NH4+ amounts in the wetland. NH4
+ uptake is the sum of the plant uptake (minus
nitrate uptake), and microbial uptake (the sum of HTNH4IM and the product of Nitrosomonas
bacteria growth and microbial N content). Biomass is assumed to prefer NO3- rather than NH4
+
as a N source, thus NH4+ utilization only occurs when NO3
- amounts are insufficient. NH4+ is
used by autotrophic bacteria as an electron source converting it to nitrate through nitrification,
which is modeled as the quotient of Nitrosomonas growth and Nitrosomonas yield.
Volatilization is an optional model component that affects ammonium concentrations.
Volatilization is modeled as a first order equation with an included pH factor. Mass transfer
occurs as a function of the concentration gradient (Equation 39).
Immobilized N is the sum of DON and PON immobilization, NO3- uptake and NH4
+
uptake. Immobilized N eventually returns to PON through the death of microbes and biomass.
These values are the product of the respective death rates and N contents as discussed in the
microbial and C submodels.
The last N pool is NO3- (NO3). Influent concentrations and mass transfer are modeled in
the same manner as other dissolved N pools. Nitrate pool increases are due to nitrification and
influent contributions, and is decreased by plant uptake and denitrification. Nitrate uptake is
modeled as the product of biomass growth and biomass C:N ratio. Nitrate outflow is the product
of the wetland outflow and wetland nitrate concentration. Anaerobic heterotrophs use NO3- as an
electron acceptor to convert the NO3- to N2 gas, which is eventually lost to the atmosphere. It is
87
assumed that the movement of the N2 gas from the soil system is instantaneous, as the diffusion
of the product is not modeled. Denitrification is modeled as the quotient of anaerobic
heterotroph growth and heterotroph NO3- yield. Other intermediates such as NO and N2O are
produced by microbes, but are not considered in the simulation.
8. Dissolved Oxygen Submodel:
The oxygen budget (Figures 15 and 16) consists of the single state variable, DO (W/B).
Oxygen is added to the wetland through influent runoff, influent point sources, DO
concentrations in precipitation, percolation, by plants and re-aeration. Re-aeration with the
atmosphere is modeled only for FWS wetlands as it is assumed that oxygen transfer with the
atmosphere is negligible in SSF systems because the water surface is below the substrate. The
mass balance for DO in the FWS wetland surface is:
DOXYWJ = DOXYWJ-1 + (DOINFJ + MTDOXYJ
+ MTFWSJ – HTRESPWJ – NSRESPWJ - DOOUTJ)*dt (65)
for FWS wetland bottoms:
DOXYBJ = DOXYBJ-1 + (BIOFLUXB – DOPERCJ
-HTRESPBJ – NSRESPBJ - MTDOXYJ)*dt (66)
and for SSF wetlands:
DOXYBJ = DOXYBJ-1 + (DOINFJ – DOPERCJ
BIOFLUXB – HTRESPBJ - NSRESPBJ)*dt (67)
where DOXY(X) is the amount of DO in water volume (g); DOINF is the oxygen additions from
runoff and point inflow (g/day); MTDOXY is the diffusion of DO from one pool to the other
(g/day); MTFWS is the DO re-aeration from atmosphere (g/day); HTRESP(X) is the oxygen loss
by heterotroph growth (g/day); NSRESP(X) is the oxygen loss due to autotroph growth (g/day);
DOOUT is the effluent DO (g/day); DOPERC are additions from percolation (g/day); and
BIOFLUXB is the oxygen flux from rootzone aeration by plants (g/day).
88
DISSOLVEDOXYGENSURFACE
WATERVOLUME
SURFACE
WATERVOLUMEBOTTOM
DISSOLVEDOXYGENBOTTOM
HT respiration W NS respiration W
Outf low
Aerobic HT
growth W
HT DO
y ield WNS DO
y ield W
NS growth W
DO outf low
Runoff inf low
Point f low DO conc.
Point source inf low
Precipitation input
At. DO conc.
Inf luent DO
Surf ace area
Biomass oxy genation rate
Biomass f lux
HT respiration B NS respiration B
Aerobic HT
growth B
HT DO
y ield WNS DO
y ield B
NS growth B
DO mass transf er
Reairation
Parameters;
Reairc,Reairm,
Reairn,
Doxy csat
HT=0, inf luent DO conc.;
HT=1, DO runoff conc.
DO MT coef .
DO Percolation/
Inf iltration
DO Percolation/
Inf iltration Conc.
Percolation/
Inf iltration Total
FIGURE 15: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE
SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
89
DISSOLVEDOXYGENBOTTOM
WATERVOLUMEBOTTOM
Surface areaBiomass oxygenation rate
Biomass flux
HT respiration W NS respiration W
OutflowAerobic HTgrowth W
HT DOyield W
NS DOyield W
NS grow th W
DO outflow
HT=0, influent DO conc.;HT=1, DO runoff conc.
Runoff inflow
Point f low DO conc.
Point source inf low
Precipitation input
At. DO conc.
Influent DO
DO Percolation/Infiltration
DO Percolation/Infiltration Conc.
Percolation/Infiltration Total
FIGURE 16: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE OXYGEN CYCLE SUBMODEL FOR SSF
WETLANDS OF THE SET-WET MODEL (SEE APPENDIX D FOR EXPLANATION OF FIGURE’S SYMBOLS)
90
Re-aeration to the wetland is modeled using a general two-film theory that is based on
mass transfer. The mass transfer of oxygen from air to water can be presented as (Jorgensen,
1983):
( )CCV
AK
dt
dCS
L −= * (68)
where KL is the mass transfer coefficient (m/day); A is the surface area through which diffusion
takes place (m2); V is the volume of water being re-aerated (m3); Cs is the oxygen saturation
concentration (g/m3); and C is the oxygen concentration in surface water volume (g/m3).
Biomass flux is the product of the biomass oxygenation rate and the wetland surface area.
It is assumed that there is a uniform vegetation stand throughout the wetland system and that
plants transport of oxygen to the wetland bottom occurs at a constant rate through the growing
season. This assumption is based on the theory that processes such as Knudsen diffusion are
responsible for rootzone aeration (Grosse, 1989). The biomass oxygenation rate is linearly
increased from zero to the maximum biomass oxygenation rate starting from the first day of the
growing season and decreasing linearly to zero at the end of the growing season. Although not
linked through any proportionality ratios, the oxygenation rate is ramped to correspond with the
vegetative growth model.
Chemical oxidation of reduced iron and manganese, and autotrophic and heterotrophic
respiration are two of the processes that consume dissolved oxygen in a wetland (Reddy and
Patrick, 1983). Heterotroph (HTRESP) and Nitrosomonas (NSRESP) respiration are modeled as
proportional to microbial growth:
HTDOY
AEHTGRWHTRESP = (69)
NSDOY
ANHTGRWNSRESP = (70)
where HTRESP is the oxygen loss due to heterotrophic growth (g O2/day); AEHTGRW is the
growth of aerobic heterotrophs (g microbes/day); HTDOY is the oxygen yield of aerobic
91
heterotrophs (g microbes/g O2); NSRESP is the oxygen loss due to Nitrosomonas growth (g
O2/day); ANHTGRW is the growth of anaerobic autotrophs (g microbes/day); and NSDOY is
the oxygen yield of Nitrosomonas bacteria (g microbes/g O2).
Most constructed wetlands have gravel substrates and secondary wastewater additions are
generally aerobic; therefore reduced iron, manganese and sulfide concentrations are considered
negligible. The dissolved oxygen outflow from the system is the product of the wetland DO
concentration and the effluent volume from the system.
9. Bacteria Submodel:
The bacteria submodel accounts for the microbial growth and associated activities in the
model. To simplify the description, the model is described according to the natural breakdown
between autotrophic (AT) and heterotrophic (HT) bacteria.
a. Autotrophic Dynamics
The state variable NITROSO (NS, W/B) represents the AT population within the
wetland. Although there are other AT species that carry out nitrification, Nitrosomonas and
Nitrobacter are the dominant species responsible for the rate limited reaction (Wheaton et al.,
1991). Changes in the population of NS are due to NS growth and NS death (Figures 17 and 18).
Growth rate of NS is described using Monod dual substrate limitation kinetics (Gidley,
1995):
+
+
=DONH KDO
DO
KNH
NH**
44
4maxµµ (71)
where µ is the actual NS specific growth rate (day-1); µmax is the NS maximum growth rate
(day-1); NH4+
is the ammonium concentration (mg/L); KNH4 is a NS NH4+ half saturation
constant (mg/L); DO is the wetland dissolved oxygen concentration (mg/L); and KDO is the NS
DO half saturation constant. This expression modifies the NS growth rate when oxygen (the
electron acceptor) or ammonium (the electron donor) is limiting.
92
NITROSOMONAS SURFACE
NH4 SURFACE
WATER VOLUME SURFACE
DISSOLVED OXYGEN SURFACE
NS grow th W NS death W
NS Temp. Factor W
NS max. grow thrate W NS DO
half sat. con. W
NS NH4 half sat.con. W
Water Temp. W
NS death rate W
FIGURE 17: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC BACTERIA
CYCLE IN FWS WETLAND SURFACE WATER
(See Appendix D for explanation of figure’s symbols)
NITROSOMONAS BOTTOM
NH4 BOTTOM
WATER VOLUME BOTTOM
DISSOLVED OXYGEN BOTTOM
NS grow th B NS death B
NS Temp. Factor B
NS max. grow thrate B NS DO
half sat. con. B
NS NH4 half sat.con. B
Water Temp. B
NS death rate B
FIGURE 18: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE AUTOTROPHIC BACTERIA
CYCLE IN FWS AND SSF WETLAND SUBSTRATE
(See Appendix D for explanation of figure’s symbols)
93
Monod models are steady state and do not describe microbial growth well in unsteady
conditions (Boyd, 1978). Benefield and Molz (1984) used the Monod model to develop a
biofilm model for an idealized plate assuming pseudo steady state conditions. Gidley (1995)
also assumed pseudo steady state conditions were adequate for Monod kinetics in wetlands.
Besides substrate limitations, temperature and pH also limit microbial growth. The
optimum temperature range for NS is from 15 to 35 °C (Reddy and Patrick, 1983; Wheaton et
al., 1991), and while growth still occurs outside of this range, it is at a reduced rate. To account
for these temperature effects, the NS temperature factor is included. In the optimum temperature
range, the temperature factor equals 1.0. The factor value linearly decreases to 0 °C (the lower
limit) and to 40 °C (upper limit), once outside of the optimum range. Fyock (1977), and Reddy
and Patrick (1983) have reported that the optimum pH range for NS growth is 6-9. Wetlands
tend to drive pH toward neutrality (pH = 7), therefore a pH factor is not included. NS growth is
calculated by:
1** −= jj SNITRSOMONANSTEMPFNSGROW µ (72)
where NSGROW is the amount of NS growth (g microbes); NSTEMPF is the NS temperature
factor; and NITROSOMONAS is the amount of NS microbes (g microbes). Since there is little
information concerning the factors controlling microbial die-off, it is modeled as a first order
reaction (Equation 22).
b. Heterotrophic Dynamics
Heterotrophic (W/B) dynamics is also modeled in the bacteria submodel (Figures 19 and
20). Monod kinetics is used to model the changes in microbial growth rate caused by substrate
limitations. For HT bacteria, the electron acceptor is either oxygen or NO3- while the electron
donors are the C compounds in the system. Heterotrophs can use electron acceptors other than
oxygen, including NO3-, sulfate, iron and manganese (Gidley, 1995); however, only NO3
- is
modeled and is assumed to be the only other available electron acceptor when DO concentrations
drop below 1-2 mg/L.
94
HETEROTROPHS SURFACE
NO3 SURFACE
WATER VOLUME SURFACE
DISSOLVED OXYGEN SURFACE
HT grow th W HT death W
Aerobic HT grow th W
HT temp. fac. W
Water temp. W
Aerobic max.grow th rate W
HT organics half sat. con. W
Anaerobic HT grow th W
HT DO half sat constant W
Anaerobe fraction W
Anaerobic max. grow th rate W
HT NO3 halfsat. con. W
HT death rate W
TOC
FIGURE 19: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS WETLAND SURFACE WATER
(See Appendix D for explanation of figure’s symbols)
HETEROTROPHS BOTTOM
NO3 BOTTOM
WATER VOLUME BOTTOM
DISSOLVED OXYGEN BOTTOM
HT grow th B HT death B
Aerobic HT grow th B
HT temp. fac.
Water temp.
Aerobic max.grow th rate B
HT organics half sat. con. B
Anaerobic HT grow th B
HT DO half sat constant B
Anaerobe fraction B
Anaerobic max. grow th rate B
HT NO3 halfsat. con. B
HT death rate B
TOC
FIGURE 20: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE HETEROTROPHIC
BACTERIA CYCLE IN FWS AND SSF WETLAND SUBSTRATE
(See Appendix D for explanation of figure’s symbols)
95
HT bacteria are facultative, meaning they can survive under both aerobic or anaerobic
conditions. SET-WET determines the fraction of the HT bacteria which utilize either aerobic or
anaerobic conditions based on the DO concentrations in the system (Gidley, 1995):
)75.0*(*8.0 DOXYCANFRAC = (73a)
(when 0.0 < DOXYC < 8.0)
ANFRAC = 0.2 (73b)
(when DOXYC > 8.0)
where ANFRAC is the anaerobic fraction of HT bacteria; and DOXYC is the DO concentration
(mg/L). The ANFRAC value is multiplied by the total HT bacteria amounts in the system to
determine the amount of aerobic and anaerobic bacteria that are present in the wetland. Bacteria
which utilize aerobic or anaerobic conditions always exist due to the presence of anaerobic
microsites under anaerobic conditions and aerobic microsites in the root zones under anaerobic
conditions (Gidley, 1995). Denitrification can even occur in aerobic wetland systems as
anaerobic microsites are present, and aerobes receive the necessary oxygen to survive due to
plant root diffusion.
The Monod models used for the HT bacteria are:
+
+
=DOTOC
AHAH KDO
DO
KTOC
TOC**max,µµ (74)
+
+
=DO
DO
TOCANHANH KDO
K
KTOC
TOC**max,µµ (75)
where µmaxis the maximum growth rate (day-1); TOC is the total organic C concentration (mg/L);
KTOC is the HT organics half saturation constant (mg/L); DO is the dissolved oxygen
concentration (mg/L); KDO is the HT DO half saturation constant (mg/L); AH are aerobic
heterotrophs; and ANH are anaerobic heterotrophs.
The same optimum temperature and pH rules apply to HT bacteria, as they did for the AT
bacteria. Total HT growth is the sum of the aerobic HT and anaerobic HT growth. The HT
96
growth and death are modeled in the exact manner as the AT bacteria, with the differing
respective parameter values.
10. Sediment Submodel:
The sediment cycle simulates sediment particles and any other desired suspended solid in
the wetland system (Figure 21). As stated earlier, the sediment model does not work in
conjunction with the SSF wetland simulation. For the SSF system, it is assumed that the
retention of particulates in the system is 100%, yet there is no account to the loss in porosity for
the wetland gravel substrate. Between season periods, there is an ability to change the porosity
values in the SSF system. If the system did have high loading of sediment to the system, the
model may have difficulty in modeling the NCOB cycle in the SSF system due to the clogging of
the wetland substrate. It is advised that the SSF system be used in conjunction with another
BMP designed to lower sediment concentrations, such as a detention basin, if using a SSF
system to control NPS pollution, to prevent clogging of the SSF wetland substrate.
The sediment model has been designed to accept up to five particle size categories. Each
category requires inputs for sediment diameter (SEDSIZE, mm), fall rate (SEDFALL, m/s),
initial amount in the water surface (SEDINITW, g), initial amount in the wetland bottom
(SEDINITB, g), and the percentage of the total incoming amount based on weight (SEDPER).
The final input category may be used to represent the organic suspended solids in the wetland
system. All organic material must be represented in this single category, as there is an associated
decomposition component for the final category. The decomposition component in the sediment
submodel does not run concurrently with the NCOB cycle. Decomposition is not modeled based
on the bacterial consumption modeled in the NCOB cycle, but with a simple user defined first
order rate equation. This was a necessary step to allow SET-WET the ability to simulate the
nutrient components independently.
Total incoming sediment amounts are either directly input or determined with a
combination of the SCS curve method and runoff concentrations (Equation 37). The total
incoming amount from catchment runoff and point flow is multiplied by SEDPER to determine
the incoming amounts for each respective category (SEDINW).
The balances for sediment in the wetland are (adapted from Christensen et al., 1994):
97
WATER VOLUME SURFACE
SEDIMENT SURFACE, X
SEDIMENT BOTTOM, X
Sed. X influent to surface
HT=1; Sed. X runoff coeff.
HT=1; runoff inflow
Sed. X influent from runoff
Point source inflow
Sed. Total point conc.
Sed. X % of Sed. Total
Outf low
Sed. X outflow
HI
HT
Outlet
Outlet = 4Pump outlet
Outlet = 1,2,3,5Outf low height
Sed. X, deposition
Sed. X falling rate
Water velocity
Sed. X, Resuspension
Sed. X, Reynold's number 2 check
Sed. X, particle diameter
Sed. X, falling rate
HI HT
Sed. X Critical Velocity
Manning's coeff. Sed. X, Reynold's number 1 check
Sed. Reynold's number 1b check
HB
Sed. Resusp. thick
Sed. Resusp. rate
Sediment Decomposition
Decomposition Rate
Physical Degradation
FIGURE 21: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE SEDIMENT CYCLE SUBMODEL FOR FWS WETLANDS OF THE SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
98
SEDQTYWC,J = SEDQTYWC,J-1 + (SEDINWC,J
+ RESUSPC,J – SEDOUTC,J – SEDDEPC,J)*dt (76)
and
SEDQTYBC,J = SEDQTYBC,J-1 + (SEDDEPC,J
+SDECOMPSEDCAT,J – RESUSPC,J)*dt (77)
where SEDQTYW is the amount of sediment in the surface water (g), SEDQTYB is the amount
of sediment in the surface bottom (g), SEDINW is the amount of incoming sediment (g/day);
RESUSP is the amount of sediment resuspended from the surface bottom (g/day); SEDOUT is
the sediment outflow (g/day); SEDDEP is sediment deposition (g/day); C is the sediment
category; and SEDCAT is the total number of sediment categories.
Resuspension and settling of sediment particles are determined with Equation 38.
Sediment outflow is based on the fall rate and is treated like other particle constituents in the
wetland. If the fall rate exceeds the free water surface height, then sediment outflow is
considered to be zero, while for rates less than the free water surface height, a ratio of removal is
based on outflow and surface water volume.
11. Phosphorous Submodel:
The P cycle (Figure 22) is dependent on the sediment cycle, and does not operate if the
sediment cycle is not included in the simulation. The basis of the model is that all of the
suspended sediment particles provide surface area to which P can be attached and consequently,
settled, resuspended, or transformed. There are four pools for the P cycle; particulate and
dissolved for both the surface and bottom of the wetland. The budgets for the pools are as
follows (adapted from Christensen et al., 1994):
For surface water dissolved P:
DTPHOSWJ = DTPHOSWJ-1 + (DISPHOSIJ
+ PMINPPTJ – DPHOSOUTJ + PHOSMTJ)*dt (78)
99
for soil/peat water dissolved P,
DTPHOSBJ = DTPHOSBJ-1 + (PRMINBPTJ
+ PHOSPERCJ – PHOSMTJ - PHOSUPWJ)*dt (79)
for particulate P in the surface water
PPHOSC,J = PPHOSC,J-1 + (PPHOSINC,J + PPHOSRESC,J
- PMINPPC,J – PPHOSSETC,J –PPHOSOUTC,J)*dt (80)
and for bottom particulate P
BTPHOSC,J = BTPHOSC,J-1 + (PPHOSSETC,J
+ DEADPHOSC,J – PRMINBPTC,J –PPHOSRESC,J)*dt (81)
where DTPHOSW is the total dissolved P in surface water (g P); DTPHOSB is the total
dissolved P in wetland bottom (g P); PPHOS is the particulate P in surface water (g P); BTPHOS
is the particulate P in the wetland bottom (g P); DISPHOSI is the influent dissolved P (g P/day);
PMINPPT is the P mineralization from the particulate pool (g P/day); MASSTP is the mass
transfer of dissolved P (g P/day); DPHOSOUT is the effluent dissolved P outflow (g P/day);
PRMINBPT is the P remineralization from bottom particulate pool (g P/day); PPHOSRES is the
resuspension of particulate P from bottom (g P)/day; PPHOSOUT is the particulate P effluent
outflow (g P/day); PPHOSSET is the particulate P settling from surface (g P/day); and
DEADPHOS is the contribution of P from physical degradation of standing dead (g P/day).
Incoming particulate P amounts are modeled using either the Freundlich isotherm
(Equation 30), the Linear isotherm (Equation 31) or as direct input. If using either the Linear or
Freundlich isotherms, the dissolved P concentration is used to determine the particulate P
concentration. Knowing the weight of sediment, the surface area for each particle class is
determined and the particulate P is assumed to be partitioned among each particle class
100
DISSOLVED PHOSPHOROUS (DP) SURFACE
PARTICULATE PHOSPHOROUS
(PPS) SURFACE; XSEDIMENT
SURFACE, X
WATER VOLUME SURFACE
SEDIMENT SURFACE, X
SEDIMENT BOTTOM, X
WATER VOLUME BOTTOM
DISSOLVED PHOSPOROUS (DP) BOTTOM
PARTICULATE PHOSPHOROUS
(PPB) BOTTOM; X
Point source inf low
Runof f inf low
HT=0, runof f DP conc.HT=1;DP runof f coef f .
Point DP conc.
DP inf luent
PPS mineralization coef f .
P minerlization f rom Sed. X
P min. total f rom Sed. Total
P reminerlization f rom Sed. X
PPB remineralization coef f .
P remin. total f rom Sed. Total
PPS,Xsettling
Sed. X., deposition
PP,X resuspensionSed. X Resuspension
Biomass Grow thBiomass Plant/Phos ratio
Plant DP uptake
DP Mass transfer
# of sed. categoriesDP mass transfer coef f .
Physical degradation
PPB f rom plant deg.
PPB ratio
Sed. X outf low
PPS outf low
PPS, x inf luent
PPS attachment ratio
runof f inf low
PPS inf luent total
point source inf low
PPS conc. In point
PPS conc. In runof f
AD=0, Freundlich K coef .
AD=0, Freundlich N coef .
AD=1, Linear partition coef .
Adsorption (AD) f lag
AD=2, PPS inf luent
Sed. X part. volume
Sed. X part surface area
Sed. X total surface area
Sed. Total total surface area
Sed. X particle #
Sed. X inf luent volumeSed. X inf luent mass
Sed. specif ic gravity
Length
Width
DP Percolation/Inf iltration Rate
Percolation/Inf iltration Total
DP Percolation/Inf iltration
FIGURE 22: RELATIONSHIPS BETWEEN MODELED PROCESSES THAT AFFECT THE PHOSPHOROUS CYCLE SUBMODEL FOR FWS WETLANDS OF THE
SET-WET MODEL
(See Appendix D for explanation of figure’s symbols)
101
according to the incoming surface area ratio for the particle class, divided by the total of all
particle surface areas.
Mass transfer of dissolved P is modeled in the form of Equation 39. Mineralization and
remineralization are modeled as first order equations for each particular category of particles.
Resuspension and settling of P corresponds to the amount of sediment particles and is related to
the ratio of each category that resuspends and settles. This also applies to the attached P for P
outflow. The contributions of P made by physical degradation (DEADPHOS) is based on the
plant:phosphorous ratio (BIOMPP). The quotient of physical degradation and BIOMPP
determines DEADPHOS. As the suspended particles are broken into categories, the addition of
each category amount allows the determination of the total amounts in each pool.
12. Deltaht Submodel:
The DELTAHT submodel determines the change in height, respective to the zero datum,
of the wetland surface bottom. This applies only to FWS wetlands and is activated whenever the
sediment or N cycles are being simulated.
If only the N cycle is being modeled the changes in the wetland bottom surface height
(HT) are determined by:
( )WIDTHLENGTH
PEATDENS
PEATACRB
DELTAHT
J
*
1000*
= (82)
where DELTAHT is the change in height of the soil top layer (m); PEATACRB is the bottom
peat accumulation rate (g/day); and PEATDENS is the peat density (kg/m3). If only the sediment
cycle is being modeled the changes in HT are determined by:
( )WIDTHLENGTH
SEDSPG
SEDTOTAL
DELTAHT*
1000000*
= (83)
102
where SEDSPG is the sediment specific gravity (kg/m3), and SEDTOTAL is the total change in
sediment mass of the wetland bottom (g). If both options are activated then DELTAHT is
determined by:
WIDTHLENGTH
SEDSPG
SEDTOTAL
PEATDENS
PEATACRBJ
DELTAHT**1000
*1000
+
= (84)
13. SET-WET Flow Chart
Figure 23 shows the flow chart of the SET-WET model. However, the main program was
removed from the flow chart for sake of clarity. As Figure 7 indicates, the main program calls
every submodel and thus the number of connections required to represent these interactions
would make Figure 23 impossible to view or understand. Instead, the main program is not
shown in Figure 23, and the connections shown represent the flows from one submodel to
another without referring to the actual flow from the main program for each section.
C. MODEL DEVELOPMENT SUMMARY
The developed model, entitled SET-WET, simulates the hydrologic, N, C, bacteria, DO,
vegetative, P and sediment cycles of a wetland system. The N, C, DO, and bacteria cycles are
linked and cannot be run independently. These cycles are referred to as the NCOB cycle to
denote their dependence. The P cycle is dependent on the sediment cycle, and neither the
sediment nor P cycles may be simulated with the SSF wetland option. Any simulation run that
includes modeling the NCOB, sediment or P cycles requires that vegetative growth be simulated.
The SET-WET model is designed so that the hydrologic component may be run independently or
concurrently with any or all of the NCOB, sediment and P cycles.
The SET-WET model development was based on the work of Gidley (1995), Parnas
(1975) and Christensen et al. (1994). The SET-WET model is written in Fortran 77 to facilitate
linking with existing NPS models, such as ANSWERS. The SET-WET model is based on
options. The program is developed such that a main program calls and manages the cycles and
options that need to be simulated, based on input values. SET-WET may use two different forms
103
START OF SIMULATION
BASE
HYDSTR
VEGSTR
CARSTR SEDSTR
BACSTR
NITSTR PHOSSTR
OXYSTR
HYDTIME
VEGTIME
CARTIME SEDTIME
BACTIME
NITTIME PHOSTIME
OXYTIME
DELTAHT
END OF SIMULATION
PCYCLE=1 and SEDCYCLE=0 or, WETTYPE=1 and (PCYCLE=1 or SEDCYCLE=1)
NITCYCLE =1
PCYCLE=1
NITCYCLE=0 & PCYCLE=0 & SEDCYCLE=0
NITCYCLE=1 or SEDCYCLE=1 or PCYCLE=1
SEDCYCLE=1
PCYCLE=0
SEDCYCLE=0
NITCYCLE =1
PCYCLE=1
NITCYCLE=0 & PCYCLE=0 & SEDCYCLE=0
NITCYCLE=1 or SEDCYCLE=1 or PCYCLE=1
SEDCYCLE=1
PCYCLE=0
SEDCYCLE=0
J=NUMTMPER
J<NUMTMPER
M=NUMSTPER
M<NUMSTPER
WETTYPE=0
NITCYCLE=0
NITCYCLE=0
FIGURE 23: FLOW CHART FOR CALLING ORDER OF SET-WET MODEL FROM MAIN CODE THROUGH
SUBROUTINES
of input: 1) input on a daily basis with known values or, 2) input estimated with a combination
of the SCS runoff curve method and runoff concentration coefficients. There are options to
include modeling of groundwater interactions, and point source additions for the hydrologic
cycle, and also N fixation, volatilization and atmospheric deposition for the N cycle. Time steps
are managed with season and time period loops, where each season period can be designated a
104
certain number of daily time periods. Designed to act like a continuously stirred tank reactor, the
model assumes all incoming nutrients are evenly mixed throughout the entire volume. Dissolved
nutrient effluent concentrations from the wetland system are determined with a ratio of the
hydrologic outflow and the total amount of water volume in the system. Particulate effluent
concentrations are determined with the ratio of the fall rate and water depth in the wetland, in
conjunction with the ratio of the hydrologic outflow and the total amount of water in the wetland.
The model allows for both free water surface (FWS) and sub-surface flow (SSF) wetlands to be
modeled and, although initially developed to help with the design of constructed wetlands, SET-
WET may also be applied to model the functions of natural wetlands.
SET-WET differs from many existing wetland models in that it uses a system’s approach
and links the many interactions of the various nutrient cycles in a wetland system. It accounts
for C and N interactions, as well as for effect of DO levels upon microbial growth. It also
directly links microbial growth and death to the consumption and transformations of nutrients in
the wetland system. Many previous models have accounted for these interactions with zero and
first order rate equations, which assume that rates are dependent only on initial concentrations.
SET-WET was developed to be as general as possible, but there were assumptions made
during model development which might not apply to all wetland locations. The model assumes
that the FWS system always has free lying surface water, and that the SSF system never floods.
This may limit its ability to model extreme climactic occurrences during drought or flood
periods. It does not account for snow or snow melt, which may limit its use to warmer climate
areas. The model assumes that the dissolved and particulate concentrations in both the surface
water and substrate water are evenly mixed through the respective volumes, which is known to
not physically be the case.
Plant growth is assumed to be of constant composition throughout the wetland, and is not
limited by lack of nutrients or water supply. In addition, growth and death cannot occur at the
same time, therefore, the model does not account for turnover of plant material during the year.
Oxygen transfer by plants is considered to be constant through the growing season, and it is also
assumed that biomass and bacteria prefer NO3- rather than NH4
+. The use of a simple plant
model will affect the release rate of nutrients to the system from plant decomposition, but more
complicated plant growth model’s input requirements are too data intensive to be used.
105
Modeling of adsorption of NH4+ is not conducted as it is assumed that the amount that is
adsorbed is still available for microbial use. The model assumes that P is attached to all of the
particulates in the system based on the surface area of the particles. In addition, SSF wetland
aeration is assumed to be negligible because the water surface is below the substrate, and SSF
wetland PON and POC removal is assumed to be 100%.
In addition to assumptions made directly concerning wetland interactions and processes,
there are assumptions made by previous researchers that were incorporated into the model. The
Thornthwaite method for estimating ET assumes that the soil moisture in the area is not limiting
and that air temperature is the primary controller of ET. Another assumption is induced by using
mass transfer coefficients to model diffusion, which assumes that changes in concentrations
between the two pools (surface and substrate water) are limited to the small part of the system’s
volume connecting boundaries.
Although SET-WET does incorporate a few assumptions into model development, it still
offers the flexibility of modeling various designs for a desired wetland. However, these
assumptions are noted for the benefit of the user.
106
IV. Model Evaluation
Another objective of this study was to “evaluate the proposed model by comparing its
predictions with field data collected from representative constructed wetland site(s).” This
objective was difficult to satisfy because there are a lack of field data related to the control of
NPS pollution using FWS constructed wetlands. Wetlands have been created (Silverman, 1989;
Daukas et al., 1989; Teague et al., 1997) whose intent is to control NPS pollution, but data
collection has not included all relative parameters (NH4+, BOD5, DO, etc) for substantial time
periods (1-2 years) that would allow the use of data for calibration and validation of SET-WET.
The SET-WET model’s performance in simulating the functions of FWS wetlands was
evaluated with data collected at a wetland site located in Benton, Kentucky. This was not an
ideal data set because the Benton wetlands treat municipal waste, but there were no available
NPS pollution wetland data sets. The model’s performance for SSF wetlands was not evaluated;
however, previous research by Gidley (1995) examined SET-WET’s basic approach towards
modeling SSF wetlands. Model evaluation procedures included the calibration and validation of
the model, performing two types of statistical analyses, conducting a sensitivity analysis, and
using SET-WET to demonstrate its application for design of wetlands. Overall the model
performed well, even though the data used for evaluation were somewhat limited.
A. Model Calibration and Validation
1. Study Area
The SET-WET model was calibrated and validated with data collected from a constructed
wetland located in Benton, Kentucky. As described by Choate et al. (1990) and summarized by
Kadlec and Knight (1996), the Benton wetlands were designed to upgrade and polish municipal
effluent from an existing lagoon for 5000 people. There are three parallel wetland cells (two
FWS, one SSF) of equivalent size (333 m by 45m) located in Benton, but only one of the FWS
cells (cell 2) was used to examine SET-WET’s performance.
The wetland cell consists of substrate made of native clay and impermeable clay lining of
3 to 4.5 m thickness, which eliminated infiltration or percolation effects. It was planted with
107
Scripus cyperinus (L.) Kunth (woolgrass bulrush), Scripus validius Vahl. (softstem bulrush), and
Typha latiofolia L. (cattail). Influent and effluent samples were collected either monthly or
bimonthly for a variety of parameters. Measurements for most of the parameters were conducted
from March 30, 1998 until September 6, 1990; however, total suspended solids (TSS), and
dissolved and total P were discontinued after April 26, 1989. Listed in Table 5 is the data set
used as input for the calibration and validation of the SET-WET model.
Incoming input daily water flow, nutrient, and sediment values were determined by linear
interpolation between the monthly data points. This was the case for all daily parameters,
excluding the daily weather data such as precipitation and air and water temperature.
Precipitation and air temperature were obtained from NOAA (National Oceanographic and
Atmospheric Administration) records at a site located in Padukah, Kentucky; approximately 25
miles northeast of Benton (NOAA, 1988: NOAA, 1989). Water temperature was estimated from
a linear regression of the air temperature and measured water temperature. With the known air
temperature, the daily water temperature was determined.
2. Model Calibration:
Data for one year were used to calibrate and validate the SET-WET model. This time
period was chosen because the nutrient and hydrologic input data needed for SET-WET were
available. During this time period, only thirteen data points for the hydrologic and nutrient
TABLE 5: MEASURED INFLOW VALUES TO WETLAND CELL 2 IN BENTON, KENTUCKY USED FOR
VALIDATION AND CALIBRATION OF SET-WET MODEL.Flow DO BOD5 NH3-N NO3-NO2 Org-N TKN TSS Dis-P Tot-P
Date (m^3/d) mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l4/27/88 213.2 2.4 30 12.0 0.02 5 17 30 3 4.55/25/88 567.9 8.1 34 12.0 0.02 8 20 49 5.5 6.86/29/88 339.0 2 18 5.2 0.02 11 16.2 63 6.8 7.87/27/88 350.4 8.3 19 3.1 0.01 12.3 15.4 110 4.1 5.68/30/88 472.5 16.8 * 20 0.1 0.09 9.9 10 5.4 9.79/28/88 403.6 2.4 14 4.1 0.01 9.6 13.7 59 5.6 5.7
10/25/88 478.0 6.2 23 9.8 0.36 7.1 16.9 54 6.611/29/88 1263.1 10.8 22 3.8 0.74 6.5 10.3 31 2.3 3.412/13/88 523.9 19.1 * 38 4.8 0.4 6.3 11.1 54 2.9 3.31/24/89 877.1 13 24 3.1 0.51 1.5 4.6 31 0.7 3.72/22/89 1816.1 10 23 1 0.57 4.3 5.3 9 0.9 1.43/28/89 1016.4 8.3 24 2.0 0.25 6 8 29 1.1 2.64/26/89 633.0 9 26 3.0 0.24 12 15 53 2.3 2.9
(Choate et al., 1990)* Data removed; physically impossible
108
conditions were collected, which may have limited the model’s predictions as explained later.
For a couple of the parameters (TSS and DP), data collection was missing, and a few of the DO
measurements were removed for values that were physically impossible (See Table 5). The data
were divided into three groups, two (4/27/88 to 7/27/88 and 1/24/89 to 4/26/89) of which were
used for calibration, while the final time frame (7/27/88 to 1/24/89) was used to validate the
model. The calibration periods were selected to evaluate how the model performed during a
warmer and colder time period.
Listed in Table 6 are the input values used for the calibration and validation periods for
the simulations performed at the Benton wetland site. The initial parameter values used in the
model simulations, as well as the values used for the calibrations and validation are listed in
Table 6. Model input parameters were estimated from domestic wastewater, wetland and
microbiology literature. It was assumed that the make-up of the water entering the system was
similar to that of domestic wastewater since the wetland cells were designed for municipal
treatment. More information on the determination of the bacterial parameters may be found in
Gidley (1995). The hydrologic component of SET-WET was calibrated first, followed by the
NCOB, the sediment and then the P cycles.
The design of the wetland as described by Choate et al. (1990) allowed the basic design
and hydrology of the wetland system to be easily described in the model. No detailed
description of the wetland outflow system was available, except it was known to consist of an
outflow pipe. The size and contraction of the pipe was determined through calibration of the
hydrologic cycle.
To estimate the current solids content of the wetland substrate, it was assumed that 5% of
the pore space was already filled and that the accumulated material had the same bulk density as
peat (0.11 Mg/m3) (Reed et al., 1994; Johnston, 1991). This gives a total mass of accumulated
solids of 19,320 kg. These solids were then broken into labile and refractory fractions with 10%
assumed to be labile, and the remaining 90% to be refractory and unavailable to microbial
decomposition. Each of these fractions were divided into C (0.80) and N contents (0.025)
respectively (Gidley, 1995). These values were used as the initial values of POC, PON, REFC,
and REFN. The BOD particulate fraction was estimated based on the ratio of TSS to BOD5 site
influent data.
109
TABLE 6: INPUT PARAMETER VALUES AND SOURCES FOR CALIBRATION AND VALIDATION PERIODS.Initial Value Value Value Typical
Parameter Value 1st. Cal. 2nd Cal. Validation Range Reference
BASELENGTH 333.5 333.5 333.5 333.5 Site Data WIDTH 43.9 43.9 43.9 43.9 Site DataHO 2 2 2 2 Site DataHB 0 0 0 0 Site DataHTI 0.6 0.6 0.6 0.6 Site DataHII 1.115 1.127 1.01 Site DataSO 0.001 0.001 0.001 0.001 .01-.0001 Reed , 1988
HYDROLOGYPORPEAT 0.3 0.3 0.3 0.3 .2-.4 Brune and Tomasso, 1991
Boyd, 1991OUTLET 3 3 3 3 Site DataHOUT 1.056 1.057 1.056 CalibrationAREAPIPE 0.0105 0.0095 0.01 CalibrationCONTC 0.8 0.9 0.85 CalibrationWATINPUT Site DataPRECIPR NOAA, 1988; NOAA 1989AIRTEMP NOAA, 1988; NOAA 1989DAYLEN .81-1.27 Site Data
BIOMASSBIOINIT 3,800,000 5,000 8,635,000 See discussionSTANDIN 8,000,000 8,000,000 6,000,000 "PEATACRW 300 300 300 300 Johnston, 1991PEATACRB 2,000 2,000 2,000 2,000 Johnston, 1991PEATDENS 110 32 32 32 110 Reed et al., 1994
Johnston, 1991PRATEUP 0.5 0.7 0.7 0.7 CalibrationBIODENS 50 50 50 50 Kadlec and Knight, 1996STDDENS 50 50 50 50 Kadlec and Knight, 1996PBIOUW 0.3 0.4 0.4 0.4 CalibrationPSTDUW 0.3 0.2 0.2 0.2 CalibrationDEGBIO 0.99 0.99 0.99 See discussionDAYSDEG 20 20 20 See discussionBIOMGRR 10 3 3 3 5.-15. Kadlec, 1991
Hammer, 1984
BACTERIANITROSIW 18,300 6,000 1,000 1,000 See discussionNITROSIB 183,000 18,000 2,400 2,400 "HETEROIW 5,300 40,000 10,000 25,000 "HETEROIB 53,000 250,000 230,000 240,000 "NDRATEW/NDRATEB .1 / .1 .002 / .002 .002 / .002 .002 / .002 .05-.15 Henze et al., 1986
Brune and Tommasso, 1991NDOHSATW/NDOHSATB 1. / 1. 1. / 1. 1. / 1. 1. / 1. .4-1.3 Henze et al., 1986
Fritz et al., 1979NMAXGRW/NMAXGRB 1 .005 / .005 .005 / .005 .005 / .005 .3-2.0 Henze et al., 1986NNH4HSCW/NNH4HSCB 1 1 1 1 .2-5.0 Henze et al., 1986
Brune and Tommasso, 1991Grady and Lim, 1980
110
Table 6 (cont.): Input parameter values and sources for calibration and validation periods.Initial Value Value Value Typical
Parameter Value 1st. Cal. 2nd Cal. Validation Range ReferenceAEMAXGRW/AEMAXGRB 6 .05 / .01 .05 / .01 .05 / .01 3.-12. Grady and Lim, 1980ANMAXGRW/ANMAXGRB 4 .05 / .01 .05 / .01 .05 / .01 2.-8. Grady and Lim, 1980HTDRW/HTDRB 0.05 0.001 0.001 0.001 .02-.08 Henze et al., 1986
Brune and Tommasso, 1991Fritz et al., 1979
HTDOHSCW/HTDOHSCB 1 0.5 / 0.5 0.5 / 0.5 0.5 / 0.5 .1-2.0 Henze et al., 1986Strand et al., 1985
HNO3HSCW/HNO3HSCB 0.15 0.15 / 0.15 0.15 / 0.15 0.15 / 0.15 .1-.2 Henze et al., 1986Brune and Tommasso, 1991
HORGHSCW/HORGHSCB 50 50 / 50 50 / 50 50 / 50 15-100 Henze et al., 1986Grady and Lim, 1980Brune and Tommasso, 1991
ANFRACW/ANFRACB .1-.9 .2-.8 .2-.8 .2-.8 0.0-0.4 Henze et al., 1986HTTEMPFW/HTTEMPFB 0.0-1.0 0.0-1.0 0.0-1.0 0.0-1.0 - See discussionWATTEMPW Site DataWATTEMPB Site Data
CARBONREFCINIT 13,900,000 13,900,000 20,000,000 16,000,000 See DiscussionDOCINITW * 30,000 20,000 50,000 "DOCINITB * 40,000 30,000 60,000 "POCINITW * 107,000 75,000 200,000 "POCINITB 1,545,000 2,000,000 3,000,000 2,500,000 "BIOCCONT 0.47 0.47 0.47 0.47 Boyd, 1978BODCFRAC 0.8 0.8 0.8 0.8 StoichiometryBODPFRAC 0.5 0.5 0.5 0.5 StoichiometryLEACHR 0.01 0.01 0.01 Kulshretha and Gopal, 1982
Polunin, 1982MICROBEC 0.53 0.53 0.53 0.53 StoichiometryPEATCC 0.8 0.8 0.8 0.8 Brady, 1984POCFALL 0.7 0.45 0.45 0.45POCRES 0.1 0.001 0.001 0.001 CalibrationMANNC 2 2 2 2RESTHC 0.02 0.01 0.01 0.01 CalibrationPOCSIZE 0.2 0.25 0.15 0.2MTCDOC 0.00005 0.00004 0.00004 0.00004 Cussler, 1997POCCOUT 1 0.75 0.3 0.3 CalibrationBODINFCO Site Data
NITROGENDONINITW * 14,000 5,500 5,000 See DiscussionDONINITB * 14,000 10,000 10,000 "IMMINITW * 50,000 700,000 700,000 "IMMINITB * 2,000,000 7,500,000 9,900,000 "NH4INITW * 27,000 20,250 30,000 "NH4INITB * 20,000 10,000 55,000 "NO3INITW * 2,100 6,000 400 "NO3INITB * 5,000 11,000 4,000 "PONINITW * 16,250 10,000 15,000 " PONINITB 48,000 250,000 100,000 300,000 "REFNINIT 435,000 435,000 565,000 500,000 "BIOMCN 23.5 23.5 23.5 23.5 Boyd, 1978BIOMPN 95 95 95 95 Boyd, 1978HTNO3YW/HTNO3YB 3.29 / 3.29 3.29 / 3.29 3.29 / 3.29 3.29 / 3.29 StoichiometryMICRONC 0.125 0.125 0.125 0.125 Stoichiometry
111
Table 6 (cont.): Input parameter values and sources for calibration and validation periods.Initial Value Value Value Typical
Parameter Value 1st. Cal. 2nd Cal. Validation Range ReferenceNSYIELDW/NSYIELDB 0.3 / 0.3 0.3 / 0.3 0.3 / 0.3 .1-.34 Henze et al., 1986
Tchobangolous and Burton, 1991ONPARTF 0.6 0.6 0.6 Site DataPEATNC 0.025 0.025 0.025 Gidley, 1995PONRES 0.01 0.01 0.01 CalibrationPONFALL 0.25 0.25 0.25MTCDON 0.00002 0.00002 0.00002 Cussler, 1997MTCNH4 0.00006 0.00006 0.00006 Cussler, 1997MTCNO3 0.00006 0.00006 Cussler, 1997PONSIZE 0.05 0.05 0.05RESTHN 0.01 0.01 0.01 CalibrationPONCOUT 0.375 0.375 0.375 CalibrationORGNINC Site DataNH4INC Site DataNO3INC Site Data
DISSOLVED OXYGENDOINITW * 60,000 22,000 15,000 See discussionDOINITB * 15,000 15,000 15,000 "HTDOYW/HTDOYB 0.81 .15 / .15 .15 / .15 .15 / .15 0.81 StoichiometryNSDOYW/NSDOYB 0.084 .02 / .02 .02 / .02 .02 / .02 0.084 StoichiometryDOCONCP 0.001 0.001 0.001MTDOX 0.0001 0.0001 0.0001 Cussler, 1997MTFWSDOC 0.00008 0.00008 0.00008 Cussler, 1997DOXYCSAT 8.5 8.5 8.5 Cussler, 1997BIOOXRB Site DataDOCONCIN Site Data
SEDIMENTSEDCAT 3 2 2 2 See DiscussionSEDRES 1 0.1 0.1 0.1 "SEDSIZE (SEDCLASS) .25-.5 .25-.5 .25-.5 .25-.5 "SEDFALL (SEDCLASS) .3-.8 .3-.7 .3-.6 .3-.7 "
SEDINITW (SEDCLASS) 80,000-300,000
10,000-20,000
10,000-190,000 site data
SEDINITB (SEDCLASS) 1,000,000-23,000,000
12,000,000-28,000,000
1,100,000-26,000,000 site data
SEDSPG (SEDCLASS) 2.65 2.65 2.65 2.65SEDPER (SEDCLASS) 1 1 1 1 Site DataRESTHICK 0.1 0.0015 0.0005 0.001 CalibrationMANNC 2 2.5 2.5 2.5DECOMPR 0.1 0.1 0.1 0.1 CalibrationPSEDDEP 0.5 0.5 0.5 0.5 CalibrationSEDINT . Site Data
PHOSPHOROUSDTPHOSIW 21,000 10,000 38,000 See DiscussionDTPHOSIB 25,000 15,000 40,000 " BTPHOSI 250,000 250,000 250,000 "PPHOSI 10,000 20,000 7,150 "PMINPPC 0.05 0.05 0.05 CalibrationPRMINBPC 0.001 0.0005 0.00075 CalibrationADSORP 2 2 2 Site DataMTCPHOS 0.00006 0.00006 0.00006 Cussler, 1997BIOMPP 300 300 300 Kadlec and Knight, 1996PHOSCON 0 0 0 Site DataDISPHOSC Site DataPPHOSCAL Site Data
112
The initial amount of biomass in the system was determined using the time period (days)
between the start of the growing season and the start of the simulation, multiplied by the
assumed constant growth rate of the system. There were no collected estimates on biomass
amounts in the system, therefore a conservative growth estimate of 3 g/m2-day was assumed
(Kadlec, 1991). The initial amount of standing dead material in the system was similarly
determined using a decay rate of .0126 day –1 multiplied by the length of time between the start
of simulation and the end of the previous growing season (Gidley, 1995).
The initial values of DOC, DON, NH4+, NO3
-, DO, BOD5, DP, TP, and TSS were
determined using the initial water volume and the site effluent concentrations at the start of each
simulation period. The water volume in the system was multiplied by the respective effluent
concentrations to determine the nutrient loads in each system. Since no site data were collected
concerning the concentrations of the nutrients in the wetland substrate, these amounts were
determined through calibration.
The initial amount of heterotroph and autotroph cells in the wetland substrate were
estimated based on values given in Stevenson (1982). NS populations are in the range of 106 –
107 cells/gram of soil. It was assumed that NS were present at a density of 106 cells/gram peat.
Generally there are a larger number of heterotrophs present than autotrophs due to their higher
growth rate, thus a density of 108 heterotroph cells/gram peat was assumed (Stevenson, 1982).
An average cell diameter of 1 µm and a density of 1 g/cm3 were used to determine the mass of
bacteria cells. It was assumed that microbes had roughly the same density as water and that they
were spherical. To convert from the number of cells per soil mass to an estimate of local initial
mass of cells, an average bulk density (1.25 Mg/m3) was used (Brady, 1984). To estimate the
number of bacteria in the wetland surface water it was assumed that the number of bacteria in the
water was 10% of the amount in the substrate. It has been shown that up to 30% of the amount
of bacteria in the substrate, may be located in the surface water (Hatano et al., 1994) for
wastewater constructed wetlands.
There was no site data describing the particle size distribution of the incoming sediment
to the wetland cell; however, since the influent to the wetland system initially passed through one
treatment system, it was assumed that many of the larger particulate particles and aggregates had
113
already settled. Since the breakdown or size of the particles was unknown, only one small
particle category (0.2 mm) was used to describe sediment.
The model was determined to be “calibrated” through visual analysis of the graphical
output (Figures 24a to 32b) of the predicted and observed values for the hydrology and nutrients.
Since SET-WET models the interactions between various nutrient cycles, it can be difficult to
simultaneously calibrate all of the cycles. This is because a parameter change for one cycle can
affect up to eight other cycles. For example, the rate of mineralization will affect the amount of
organic N and NH4+ in the system. In addition, this rate will affect the ratio of TON to TOC in
the system, which in combination with the MICTCN ratio determines if the system is N or C
limited. These effects will be further explored with the sensitivity analysis.
Table 7 lists the differences between the observed and predicted values for the Benton
site. The data shows that the predicted values of SET-WET mostly follows the trends for the two
calibration periods. From visual analysis, the hydrology (Figures 24a and 24b), BOD5 (Figures
29a and 29b), dissolved P (Figures 31a and 31b), and total P (Figures 32a and 32b) predictions
values followed the observed values trends the best.
Ammonium concentrations were over-predicted for the first calibration period, while the
NO3-, organic N, and DO effluent concentration predictions did not completely match the trends
for the observed values. The objective of the calibration process was to examine the effects of
both warmer and colder seasons in the study area. Since there were limited data for each
calibration period (4 data points each), each calibration period was modeled as one season
period, eliminating the ability to change parameter values during the simulation. The high NH4+
predictions may be attributable to the extremely high influent concentrations on April 27, 1988
and May 25, 1988 for the Benton wetlands. These two data points were averaged over the entire
time period, meaning for a one-month period, the model was receiving input concentrations of 12
mg/L for NH3-N. These measurements may have been spikes in the observed measurements,
which would contribute to the high NH4+ predictions by the model. Any errors in the input
measurements will be amplified for all nutrients because so few data points were available.
For the second calibration period (1/24/88 to 4/26/89), the DO concentration predictions
were over predicted. Once again this can probably be attributed to the lack of sufficient data
points used for input. The measured incoming DO counts, for the entire second calibration
period were at high levels. Another reason DO values may have been over-predicted is due to
114
0
100
200
300
400
500
600
700
800
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Out
flow
(m
3 /d)
Predicted Hydrologic Outflow
Measured Hydrologic Outflow
FIGURE 24A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
0
500
1000
1500
2000
2500
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Out
flow
(m
3 /d)
Predicted Hydrologic Outflow
Measured Hydrologic Outflow
FIGURE 24B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
115
0
2
4
6
8
10
12
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Amm
oniu
m C
once
ntra
tion
(mg/
l)
Predicted Ammonium
Measured Ammonium
FIGURE 25A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
2
4
6
8
10
12
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Amm
oniu
m C
once
ntra
tion
(mg/
l) Predicted Ammonium
Measured Ammonium
FIGURE 25B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
116
0
0.1
0.2
0.3
0.4
0.5
0.6
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Nitr
ate
Conc
entr
atio
n (m
g/l)
Predicted Nitrate
Measured Nitrate
FIGURE 26A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Nitr
ate
Conc
entr
atio
n (m
g/l)
Predicted Nitrate
Measured Nitrate
FIGURE 26B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR NITRATE
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
117
0
1
2
3
4
5
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Org
anic
Nitr
ogen
Conc
entr
atio
n (m
g/l)
Predicted Organic Nitrogen
Measured Organic Nitrogen
FIGURE 27A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Org
anic
Nitr
ogen
Co
ncen
trat
ion
(mg/
l)
Predicted Organic Nitrogen
Measured Organic Nitrogen
FIGURE 27B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
118
0
3
6
9
12
15
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Dis
solv
ed O
xyge
n Co
ncen
trat
ion
(mg/
l)
Predicted Dissolved Oxygen
Measured Dissolved Oxygen
FIGURE 28A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
1
2
3
4
5
6
7
8
9
10
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Dis
solv
ed O
xyge
n Co
ncen
trat
ion
(mg/
l)
Predicted Dissolved Oxygen
Measured Dissolved Oxygen
FIGURE 28B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
119
0
2
4
6
8
10
12
14
16
18
20
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
BOD
5 Co
ncen
trat
ions
(m
g/l)
Predicted BOD5
Measured BOD5
FIGURE 29A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
0
2
4
6
8
10
12
14
16
18
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
BOD
5 Co
ncen
trat
ions
(m
g/l)
Predicted BOD5
Measured BOD5
FIGURE 29B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
120
0
10
20
30
40
50
60
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Tota
l Sus
pend
ed S
olid
s Co
ncen
trat
ion
(mg/
l)
Predicted TSS
Measured TSS
FIGURE 30A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
5
10
15
20
25
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Tota
l Sus
pend
ed S
olid
s Co
ncen
trat
ion
(mg/
l)
Predicted TSS
Measured TSS
FIGURE 30B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
SUSPENDED SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
121
0
1
2
3
4
5
6
7
8
9
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Dis
solv
ed P
hosp
horo
us
Conc
entr
atio
n (m
g/l)
Predicted Dissolved Phosphorous
Measured Dissolved Phosphorous
FIGURE 31A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Dis
solv
ed P
hosp
horo
us
Conc
entr
atio
n (m
g/l)
Predicted Dissolved Phosphorous
Measured Dissolved Phosphorous
FIGURE 31B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
122
0
1
2
3
4
5
6
7
8
4/24/88 5/4/88 5/14/88 5/24/88 6/3/88 6/13/88 6/23/88 7/3/88 7/13/88 7/23/88
Date
Efflu
ent
Tota
l Pho
spho
rous
Co
ncen
trat
ions
(m
g/l)
Predicted Total Phosphorous
Measured Total Phosphorous
FIGURE 32A: OBSERVED AND CALIBRATED PREDICTED VALUES (4/27/88 TO 7/27/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
1
2
3
4
5
1/12/89 2/1/89 2/21/89 3/13/89 4/2/89 4/22/89 5/12/89
Date
Efflu
ent
Tota
l Pho
spho
rous
Co
ncen
trat
ion
(mg/
l)
Predicted Total Phosphorous
Measured Total Phosphorous
FIGURE 32B: OBSERVED AND CALIBRATED PREDICTED VALUES (1/24/88 TO 4/26/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
123
TABLE 7: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED VALUES
FOR THE HYDROLOGY, AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR THE CALIBRATED,PREDICTED VALUES.
Measured Predicted Difference Measured Predicted
Hydrologic Hydrologic in NH4 NH4
Outflow Outflow Hydrologic Effluent Effluent DifferenceDate (m^3) (m^3) Outflow Conc. (mg/l) Conc. (mg/l) NH4
4/27/88 657.52 696.22 38.70 5 3.58 -1.425/25/88 621.05 518.56 -102.49 3.4 9.92 6.526/29/88 318.32 190.90 -127.42 0.4 10.57 10.177/27/88 305.97 332.33 26.36 5.3 7.72 2.42
1/24/89 864.97 858.29 -6.68 3.1 2.62 -0.482/22/89 1853.03 1815.93 -37.10 2.7 1.36 -1.343/28/89 1071.55 816.99 -254.56 2 1.86 -0.144/26/89 598.78 530.25 -68.53 11 2.54 -8.46
Measured Predicted Measured PredictedNO3 NO3 Organic-N Organic-N
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) NO3 Conc. (mg/l) Conc. (mg/l) Organic-N
4/27/88 0.37 0.279 -0.092 3.45 2.67 -0.785/25/88 0.54 0.013 -0.527 4.5 2.96 -1.546/29/88 0.01 0.013 0.003 4.2 3.87 -0.337/27/88 0 0.013 0.013 1.5 4.34 2.84
1/24/89 0.82 0.778 -0.042 1.5 1.20 -0.302/22/89 0.34 0.537 0.197 1 1.74 0.743/28/89 0.15 0.323 0.173 3 2.55 -0.454/26/89 0.01 0.039 0.029 4.2 4.27 0.07
Measured Predicted Measured PredictedDO DO BOD5 BOD5
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) DO Conc. (mg/l) Conc. (mg/l) BOD5
4/27/88 8.75 7.10 -1.65 16.50 11.16 -5.345/25/88 11.50 5.01 -6.49 18.00 15.11 -2.896/29/88 0.30 4.13 3.83 6.00 15.04 9.047/27/88 0.20 4.63 4.43 19.00 11.07 -7.93
1/24/89 4.80 5.14 0.34 9.00 4.92 -4.082/22/89 3.80 7.82 4.02 9.00 11.85 2.853/28/89 1.20 5.77 4.57 12.00 12.84 0.844/26/89 0.20 6.48 6.28 16.00 13.33 -2.67
Measured Predicted Measured PredictedTSS TSS Dis.-P Dis.-P
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) TSS Conc. (mg/l) Conc. (mg/l) Dis.-P
4/27/88 40.00 39.79 -0.21 3.20 4.61 1.415/25/88 53.00 22.07 -30.93 5.80 6.25 0.456/29/88 23.00 21.36 -1.64 5.40 5.96 0.567/27/88 27.00 21.45 -5.55 3.10 2.79 -0.31
1/24/89 2.00 2.59 0.59 1.40 1.30 -0.102/22/89 2.00 6.14 4.14 1.20 2.83 1.633/28/89 20.00 6.61 -13.39 3.70 2.74 -0.964/26/89 14.00 3.34 -10.66 4.10 3.34 -0.76
124
the estimation of water temperature in the system. Water temperature was estimated from the air
temperature, with the assumption that the water temperature in the surface and substrate water
were the same. These values may have been incorrect, due to the limited data points used in the
regression. In addition, air temperature fluctuates appreciably more than water temperature.
Water temperature is a factor in bacterial growth. If the water temperature is below 0 °C,
bacterial growth will cease, limiting the amount of oxygen consumption in the wetland. There
were no bacterial counts made in the system thus it is unknown if bacteria counts are close to
actual values.
Additionally, matching the predicted and simulated values for sediment was difficult for
the calibration periods. The hydrologic outflow from the system is used to determine the water
velocity in the wetland. The wetland water velocity is then used to determine if resuspension in
the system occurs based on the water velocity being greater than the critical velocity (Equation
32). There is no function controlling the amount of sediment that is lifted up (resuspended) from
the wetland substrate based on the speed above the critical velocity. A set amount is lifted up
whenever the critical velocity is exceeded; this makes the predictions less accurate.
The calibrated values input did not greatly deviate from the values previously
determined through literature, except for parameters directly involving bacterial growth and
oxygen transfer. For the microbial parameters, the most dramatic changes involved the
maximum growth rates, which were changed by two order of magnitudes (from 6.0 and 4.0. to
0.05 and 0.05 for HT bacteria; 1.0 to 0.005 for AT bacteria). The literature values are typical for
conventional wastewater treatment systems, therefore it is not unreasonable for microbial growth
to be slower in a less controlled environment. Additionally, the heterotroph and autotroph yields
and death rates were each decreased significantly. These values would correspond with the
changes associated with the lowering of the growth parameters. If the death parameter were not
lowered to correspond to the decrease in microbial growth parameters, then the bacteria in the
system would completely die out. Furthermore, the microbial yields needed to be changed as the
growth of the microbes is decreased with the smaller growth parameter; therefore, more oxygen
per bacteria is consumed since there is less competition.
125
3. Model Validation:
SET-WET was validated by comparing the model’s predicted results with the observed
data for the period 7/27/88 to 1/24/89. In the validation process, the initial amounts of each
respective nutrient in the system were determined in the same manner as the calibration
procedure, by multiplying the effluent concentrations with the initial water volume in the system.
The input parameter values for the validation period were determined by taking the average of
the two calibration period parameter values (Table 6). On 08/30/88 (16.8 mg/L) and 12/13/88
(19.1 mg/L), the measured DO concentrations were above the maximum DO solubility under
atmospheric conditions (14.62 mg/L at 0 °C; Reed et al., 1994). These values were removed
from the validation process. Figures 33-41 display the model predictions and the observed data
for the outflow and the effluents of NH4+, NO3
-, organic N, DO, BOD5, TSS, DP and TP
concentrations for the validation period. Table 8 lists the measured values, predicted values, and
the difference between the two values for the hydrology outflow, and the NH4+, NO3
-, organic N,
DO, BOD5, TSS, and DP effluent concentrations.
0
200
400
600
800
1000
1200
1400
1600
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Out
flow
(m
3 /d)
Predicted Hydrologic Outflow
Measured Hydrologic Outflow
FIGURE 33: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR HYDROLOGIC
OUTFLOW FROM THE WETLAND
126
0
1
2
3
4
5
6
7
8
9
10
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Amm
oniu
m C
once
ntra
tion
(mg/
l)
Predicted Ammonium
Measured Ammonium
FIGURE 34: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR AMMONIUM
EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
0.2
0.4
0.6
0.8
1
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
D a t e
Efflu
ent
Nitr
ate
conc
entr
atio
n (m
g/l) Predicted Nitrate
Measured Nitrate
FIGURE 35: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR NITRATE EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
127
0
1
2
3
4
5
6
7
8
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Org
anic
Nitr
ogen
Con
cent
ratio
n (m
g/l)
Predicted Organic Nitrogen
Measured Organic Nitrogen
FIGURE 36: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR ORGANIC
NITROGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
1
2
3
4
5
6
7
8
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Dis
solv
ed O
xyge
n Co
ncen
trat
ion
(mg/
l)
Predicted Dissolved Oxygen
Measured Dissolved Oxygen
FIGURE 37: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
OXYGEN EFFLUENT CONCENTRATIONS FROM THE WETLAND.
128
0
4
8
12
16
20
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
BOD
5 co
ncen
trat
ions
(m
g/l)
Predicted BOD5
Measured BOD5
FIGURE 38: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR BOD5 EFFLUENT
CONCENTRATIONS FROM THE WETLAND.
0
5
10
15
20
25
30
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Tota
l Sus
pend
ed S
olid
s Co
ncen
trat
ion
(mg/
l)
Predicted TSS
Measured TSS
FIGURE 39: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL SUSPENDED
SOLIDS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
129
0
1
2
3
4
5
6
7
8
9
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Dis
solv
ed P
hosp
horo
us C
once
ntra
tion
(mg/
l)
Predicted Dissolved Phosphorous
Measured Dissolved Phosphorous
FIGURE 40: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR DISSOLVED
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
0
1
2
3
4
5
6
7
8
9
10
7/16/88 8/5/88 8/25/88 9/14/88 10/4/88 10/24/88 11/13/88 12/3/88 12/23/88 1/12/89 2/1/89
Date
Efflu
ent
Tota
l Pho
spho
rous
Co
ncen
trat
ion
(mg/
l)
Predicted Total Phosphorous
Measured Total Phosphorous
FIGURE 41: OBSERVED AND VALIDATED PREDICTED VALUES (7/27/88 TO 1/24/89) FOR TOTAL
PHOSPHOROUS EFFLUENT CONCENTRATIONS FROM THE WETLAND.
130
TABLE 8: MEASURED, PREDICTED, AND DIFFERENCE BETWEEN THE MEASURED AND PREDICTED VALUES
FOR THE HYDROLOGY AND VARIOUS WETLAND EFFLUENT CONCENTRATIONS FOR THE VALIDATED,PREDICTED VALUES.
Measured Predicted Difference Measured PredictedHydrologic Hydrologic in NH4 NH4
Outflow Outflow Hydrologic Effluent Effluent DifferenceDate (m^3) (m^3) Outflow Conc. (mg/l) Conc. (mg/l) NH4
7/27/88 332.33 345.64 -13.31 5.30 4.36 0.948/30/88 473.75 424.41 49.34 8.60 3.54 5.069/28/88 332.33 371.21 -38.88 8.40 2.83 5.5710/25/88 560.24 492.28 67.96 9.00 6.22 2.7811/29/88 1355.81 1274.87 80.94 4.60 5.53 -0.9312/13/88 864.91 631.60 233.31 3.80 5.02 -1.221/24/89 858.29 856.48 1.81 3.10 4.99 -1.89
Measured Predicted Measured PredictedNO3 NO3 Organic-N Organic-N
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) NO3 Conc. (mg/l) Conc. (mg/l) Organic-N
7/27/88 0.01 0.06 -0.05 1.50 1.54 -0.048/30/88 0.01 0.01 0.00 4.90 4.28 0.629/28/88 0.02 0.00 0.02 7.20 3.95 3.2510/25/88 0.17 0.03 0.14 2.40 3.26 -0.8611/29/88 0.55 0.54 0.01 1.00 2.85 -1.8512/13/88 0.76 0.54 0.22 2.40 2.77 -0.371/24/89 0.82 0.46 0.36 1.50 1.22 0.28
Measured Predicted Measured PredictedDO DO BOD5 BOD5
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) DO Conc. (mg/l) Conc. (mg/l) BOD5
7/27/88 0.20 3.94 -3.74 19.00 14.28 4.738/30/88 1.40 7.83 -6.43 11.00 11.85 -0.859/28/88 3.60 5.04 -1.44 9.00 9.24 -0.2410/25/88 2.50 5.62 -3.12 5.00 9.48 -4.4811/29/88 4.00 6.59 -2.59 8.00 11.78 -3.7812/13/88 3.30 9.34 -6.04 7.00 15.62 -8.621/24/89 4.80 7.42 -2.62 9.00 14.19 -5.19
Measured Predicted Measured PredictedTSS TSS Dis.-P Dis.-P
Effluent Effluent Difference Effluent Effluent DifferenceDate Conc. (mg/l) Conc. (mg/l) TSS Conc. (mg/l) Conc. (mg/l) Dis.-P
7/27/88 27.00 27.62 -0.62 5.40 5.52 -0.128/30/88 11.00 2.47 8.53 5.30 6.36 -1.069/28/88 9.00 2.53 6.47 5.80 5.62 0.1810/25/88 2.00 2.60 -0.60 4.30 4.59 -0.2911/29/88 4.00 4.15 -0.15 2.80 2.56 0.2412/13/88 3.00 3.89 -0.89 2.80 2.64 0.161/24/89 2.00 3.34 -1.34 1.40 1.24 0.16
131
The data set used to calibrate and validate the SET-WET model was not ideal, as the use
of monthly linear interpolation between the observed data points contradicts the principle idea of
NPS pollution’s randomness and dependence upon climatic occurrences. Since the Benton
wetland site is developed for municipal waste treatment, however, this problem may have been
mitigated. The ideal data set would consist of daily data points for all concerned hydrologic and
nutrient parameters over a minimum of two-years, but no such data were available. This type of
data set would allow the seasonal as well as long term comparison of model performances.
As evident in Figures 33-41, the predicted trends for the validated period match the
observed values fairly well for various parameters, except for the NH4+ (Figure 34) and DO
(Figure 37) predictions. The trend for predicted NH4+ concentrations does not follow the
measured values for the period of middle of August 1988 to the end of October 1988. During
this period, the model’s predictions are low, thus implying that the conversion of organic N
through mineralization is not sufficiently handled by the model for this period. However when
one examines the organic N concentration (Figure 36), there is not an overestimation of values;
therefore, the model may be under-predicting organic N amounts due to a lack of accounting for
biomass degradation and possible additions from N fixation. Another possible explanation for
the extremely low NH4+ predictions is the very low influent NH4
+ concentrations during this time
frame. The values are comparably low and as stated earlier, since the data points are so few, any
major errors in measurements during data collection will have a significant effect on model
input.
The inability of the model to match DO concentrations is a cause for concern. This
concern stems from the fact that many of the bacterial growth and bacteria uptake rates are
directly and indirectly affected by oxygen concentrations in the system. It is possible that the
assumption of a constant oxygen transfer rate by plants during the growing season may be
incorrect, as the DO concentrations are much higher than observed.
The dissolved concentrations associated with the total P effluent (Figures 40 and 41)
comprise a significant percentage of the total P. These values suggest that there is a significant
amount of particulate P buried in the system, which is in accordance with findings of previous
researchers (Kadlec and Knight, 1996).
To confirm that the model is performing proper mathematical calculations, a mass
balance of various hydrologic and nutrient pools of the model’s predictions were performed.
132
The cumulative inflows and outflows from the system were determined for every pool in the
system. The difference between these two values was compared with the change in storage in
the system, which is the difference between the final and initial amounts for each pool. The
largest percent difference between these values was 1%, indicating that the model sufficiently
accounted for the interactions of each pool.
From visual analysis of the data points it appears that the data matches very well
considering the data limitations. Further support will be lent by the statistical analyses.
B. Statistical Analysis:
Statistical analyses were performed on the differences between the observed
measurements and model’s predicted values for the hydrology, and nutrient parameters. The
SAS software package “proc univariate’ procedure was used to perform the Wilcoxon signed
rank test on the data (Ott, 1993). In the test, the null hypothesis (H0) was that there are no
differences between the observed and predicted values, while the alternative hypotheses (H1) is
that there are significant differences between the values. Using a two-sided test, with an alpha
value of 0.10, it was determined whether the model predictions were similar to the measured
values.
Ordinarily, the objective of a statistical test is to try to reject the null hypothesis, giving
fairly strong support to the alternative hypothesis. The Wilcoxon test is not designed in this
manner, therefore, we do not want to reject the null hypothesis. Failure to reject the hypothesis
means acceptance of H0, indicating that the observed and predicted values are not statistically
different from each other. A p-value that is greater than the alpha cut-off value of 0.10 would
result in the failure to reject the null hypothesis. The confidence level in this comparison
increases as the reported p-value approaches the maximum value of 1.0.
Table 9 presents the p-values for the differences between the observed measurements and
predicted model values for selected outflow parameters. For the hydrology, NO3-, NH4
+, organic
N, BOD5, TSS, dissolved and total P outflows, we failed to reject the null hypothesis, as their p-
values are greater than 0.10 indicating that the measured and predicted values were statistically
similar. The only parameter for which the null hypothesis was rejected was DO, whose p-value
was 0.016. These results must be examined closely however, as there were only seven values
133
TABLE 9: P-VALUES AND RESULTS OF THE WILCOXON SIGNED RANK TEST PROCEDURE FOR DIFFERENCES
BETWEEN THE MEASURED AND VALIDATED, PREDICTED VALUES OF WETLAND EFFLUENT IN BENTON,KENTUCKY.
Outflow Parameter P-value H0
Hydrology 0.156 AcceptNitrate 0.375 Accept
Ammonium 0.297 AcceptOrganic Nitrogen 0.938 Accept
Dissolved Oxygen 0.016 Not AcceptBOD 5 0.109 Accept
Total Suspended Solids 0.938 AcceptDissolved Phosphorous 0.999 Accept
Total Phosphorous 0.688 Accept
in the data set used for each comparison. The Wilcoxon signed rank test is more reliable when
more observations are used. For the BOD5 (p=0.109), and hydrology (p=0.156) comparisons,
the p-values are barely above the cutoff value, meaning that caution should be used in failing to
reject the null hypothesis. On the other hand, the p-values for the TSS (p=0.938), ON (p=.0938),
and DP (p=0.999) are close to one, indicating a higher confidence in the acceptance of the null
hypothesis because the values are symmetrical around the observed values. Results of the
statistical analyses indicate that the SET-WET model predicted values that are similar to the
observed data.
In addition to the Wilcoxon signed rank test, linear regression was also performed on the
output values. The regression analysis was used to compare the measured and predicted values
for the nutrients and hydrology of the wetland. For this analysis, our null hypothesis (H0) was
that the slope coefficient between the observed and predicted values is zero, while the alternative
hypothesis (H1) was that the slope coefficient is not equal to zero. The ideal situation, which is
referred to as the 1:1 line, would be a line that crosses the y intercept (B0) at zero and has a slope
(B1) of one. This line represents the situation in which the observed and predicted values are
exactly the same.
Table 10 lists the results of the linear regression analysis. The values for B0 and B1 are the
linear regression of the simulated (independent variable) and observed (dependent variable) data
points, respectively. Because SET-WET is a deterministic model, the simulated values are used
as the independent values since the model will always predict these values with the same data
input. There is variability and uncertainty in the observed values since there are differences in
the observed measurements depending on how, when and who takes and analyses the samples
134
(Brannan, 1999). The human error introduced into the observed values due to the collection and
analysis of a sample dictate that the observed values are stochastic and therefore should be the
dependent variable.
Analysis of the linear regression results is not as straight forward as the non-parametric
analysis. P-values of less than 0.05 indicate that the analysis of the observed and simulated
results do not have a slope coefficient equivalent to zero. Independently, these results tell us
little, but if analyzed in conjunction with the slope coefficient and R2 value, we may be able to
distinguish which parameters were statistically similar. If the slope value is near 1.0 (idealized),
and the R2 value is close to 1.0, indicating a good fit of observed values to the regressed line, we
can be more confident in the statistical equivalence of the observed and simulated values. Table
10 indicates that the parameters which follow these constraints are hydrology, NO3-, DO, TSS,
DP, and total P. The other parameters either have p-values above 0.05 (Org-N, NH4+, BOD5 ),
slope coefficients distant from 1.0 (NH4+, organic N), or have low R2 values (NH4
+, BOD5).
Another conclusion that may be drawn from the linear regression analysis, stems from the
95% confidence intervals (CI). In the expected operating range (real world values), CI
predictions for all nine parameters, overlap zero for B0 and one for B1. This indicates that SET-
WET has shown a good agreement between the predicted and observed values, since the range of
CI values include the idealized line of equivalent predictions. However, there are large amounts
of variance in the data sets for each parameter, indicated by the large range in CI predictions;
therefore, results need to be accepted with caution.
TABLE 10: LINEAR REGRESSION DATA FOR OBSERVED (Y-AXIS) AND PREDICTED (X-AXIS) WETLAND
EFFLUENT.Lower 95% Upper 95% Lower 95% Upper 95%
B0 B1 R2p-value C.I. B0 C.I. B0 C. I. B1 C. I. B1
Hydrology 9.820 1.070 0.945 0.001 -199.30 218.93 0.77 1.37
NH4+ 9.45 -0.710 0.105 0.480 -2.05 20.94 -3.10 1.68
NO3- 0.03 1.300 0.881 0.002 -0.16 0.22 0.75 1.86
Org-N -1.23 1.480 0.560 0.053 -5.81 3.36 -0.03 3.00
DO -2.72 0.933 0.574 0.048 -8.32 2.88 0.01 1.86
BOD5 1.69 0.650 0.126 0.434 -22.96 26.35 -1.32 2.61
TSS 2.478 0.873 0.812 0.006 -2.77 7.73 0.39 1.36
DP 0.535 0.843 0.956 0.001 -0.39 1.46 0.63 1.05
TP 0.693 0.869 0.887 0.002 -0.90 2.28 0.51 1.23
B0 = y intercept, B1 = slope, R2 = r squared value, C.I. = confidence interval
135
Figure 42 depicts an example of the observed and predicted values for dissolved P plotted
against each other, along with the linear regression and the 1:1 line. As the figure depicts, the
prediction interval does contain the 1:1 line indicating that there is strong agreement between the
model predictions and observed data. Since the p-value (.001) is also low, the R2 value is high
(0.956), and the C.I.s cover both B0=0 and B1=1, there is strong agreement between the observed
and predicted values. The error bars represent the 95% prediction interval band for the linearly
regressed line. Appendix E contains the plots for the hydrology, ammonium, nitrate, organic N,
dissolved oxygen, BOD5, TSS, and total P concentrations.
FIGURE 42: SIMULATED AND OBSERVED VALUES FOR DISSOLVED PHOSPHOROUS CONCENTRATIONS,PLOTTED WITH THE DETERMINED LINEAR REGRESSION, AND IDEAL 1:1 LINE.
136
C. Sensitivity Analysis:
A sensitivity analysis was conducted to determine which parameters would require the
most scrutiny in future simulations. The relative sensitivity (RS) of each parameter was
determined with the following equation (Heatwole, 1998):
−−
=b
b
b
bR O
P
PP
OOS * (85)
where SR is the relative sensitivity; O is the model output variable of interest; P is the parameter
value; and b is a subscript which represents the parameter values and output of the base scenario.
Each examined parameter was adjusted a total of six times, with changes of (+/-) 10%,
(+/-) 25%, and (+/-) 50%. The cumulative average value for the base scenario was compared
with the average change in model response for the hydrologic and nutrient components. If the
absolute value of the unitless RS is equal to one, then the same percent change occurs in the
output and input; if the RS is less then one, the model response is damped; while an RS value
greater than one indicates the model response is inflated. If the RS value is equal to zero, then
the selected parameter has no effect on the outcome of the component’s predicted values.
Additionally, if the RS value is positive or negative, it can be determined if there is an indirect (-)
or direct (+) relationship between the parameter and the output (Heatwole, 1998).
Table 11 presents the relative sensitivity analysis results for changes of (+/-) 50% to the
base value. Appendix F contains the tables for the relative sensitivity changes of (+/-) 10% and
(+/-) 25%. The data is presented for the hydrologic and nutrient components of SET-WET. The
results of the (+) 50% change are presented first followed by the results of the (–) 50% change.
The (+/-) sign designates whether the RS value relation is direct or indirect, while the letters
represent the one of seven categories into which the RS value falls. A note is made of the ‘NC’
category, which indicates that the simulation run was incomplete due to the generation of
unintelligible results. Parameter values could not be changed (+) 50% for certain situations in
which the parameters were fractions. A (+) 50% change resulted in a fractional value greater
than one, which is impossible. For these cases (PEATCC, BODCFRAC, and PONCOUT), Table
11 presents RS values for a change of +25% from the base. For SET-WET, the results of the
137
TABLE 11: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-) 50% CHANGE IN BASE VALUESP a r a m e te r B a s e V a lu e N H 4 N O 3 D O N P O N D O B O D 5 T S S D P T PB A C T E R I A
A E M A X G R B 0 . 0 1 (N C )/ (+ E ) (N C )/ (-D ) (N C )/ (-C ) (N C )/ (-C ) (N C )/ (-E ) (N C )/ (-C ) A / A A / (-C ) (A / -C )A E M A X G R W 0 . 0 5 (+ E )/ (+ D ) (-C )/ (-C ) (-D )/ (-C ) (-C )/ (-C ) (-D )/ (-D ) (-D )/ (-C ) A / A A / A A / AA N M A X G R B 0 . 0 1 (+ D )/ (+ D ) (-D )/ (-D ) (-B )/ (-B ) (-B )/ (-B ) (-C )/ (-C ) (-B )/ (-B ) A / A A / A A / AA N M A X G R W 0 . 0 5 (+ B )/ (+ B ) (-C )/ (-C ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) A / A A / A A / A
H E T R O I B 2 4 0 0 0 0 (N C )/ (+ E ) (N C )/ (-E ) (N C )/ (-C ) (N C )/ (-C ) (N C )/ (-E ) (N C )/ (-C ) A / A A / A A / AH E T E R O I W 2 5 0 0 0 (+ D )/ (+ D ) (-C )/ (-C ) (-C )/ (-C ) (-B )/ (-B ) (-D )/ (-D ) (-C )/ (-C ) A / A A / A A / AH N O 3 H S C B 0 . 1 5 (-C )/ (-D ) (+ D )/ (+ D ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B ) A / A A / A A / AH N O 3 H S C W 0 . 1 5 (-B )/ (-B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) A / A A / A A / AH O R G H S C B 5 0 (-C )/ (-D ) (+ C )/ (+ C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / AH O R G H S C W 5 0 (-C )/ (N C ) (+ C )/ (N C ) (+ C )/ (N C ) (+ B )/ (N C ) (+ D )/ (N C ) (+ C )/ (N C ) A / A A / A A / AH T D O H S C B 0 . 1 5 (-D )/ (N C ) (-C )/ (N C ) (+ C )/ (N C ) (+ C )/ (N C ) (+ D )/ (N C ) (+ C )/ (N C ) A / A A / A A / AH T D O H S C W 0 . 5 (-C )/ (-C ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / A
H T D R B 0 . 0 0 1 2 5 (-D )/ (-E ) (+ D )/ (+ D ) (+ B )/ (+ C ) (+ B )/ (+ C ) (+ D )/ (+ D ) (+ B )/ (+ C ) A / A A / A A / AH T D R W 0 . 0 0 1 (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / A
N D O H S A T B 1 (-C )/ (-C ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / AN D O H S A T W 1 (-B )/ (-B ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / A
N D R A T E B 0 . 0 0 2 (-C )/ (-C ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A A / A A / AN D R A T E W 0 . 0 0 2 (-B )/ (-B ) (-C )/ (-C ) (-B )/ (-B ) (-B )/ (-B ) (+ C )/ (+ B ) (-B )/ (-B ) A / A A / A A / AN I T R O S I B 2 4 0 0 (+ D )/ (+ D ) (+ D )/ (+ D ) (-B )/ (-B ) (-C )/ (-B ) (-D )/ (-D ) (-B )/ (-B ) A / A A / A A / AN I T R O S I W 1 0 0 0 (+ C )/ (+ C ) (+ D )/ (+ D ) (-B )/ (-B ) (-B )/ (-B ) (-C )/ (-C ) (-B )/ (-B ) A / A A / A A / AN M A X G R B 0 . 0 0 5 (-D )/ (N C ) (-F )/ (N C ) (+ E )/ (N C ) (-D )/ (N C ) (+ D )/ (N C ) (-E )/ (N C ) A / A A / A A / AN M A X G R W 0 . 0 0 5 (+ C )/ (+ C ) (+ D )/ (+ D ) (-B )/ (-B ) (-B )/ (-B ) (-D )/ (-D ) (-B )/ (-B ) A / A A / A A / AN N H 4 H S C B 1 (-C )/ (-C ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ (C ) (+ B )/ (+ B ) A / A A / A A / AN N H 4 H S C W 1 (-B )/ (-B ) (-C )/ (-C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ C )/ (+ C ) (+ B )/ (+ B ) A / A (-C )/ (+ C ) (-C )/ (+ C )
N I T R O G E NB I O M C N 2 3 .5 (+ D )/ (-D ) (+ F )/ (-B ) (-C )/ (-C ) (-D )/ (-E ) (-C )/ + B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )B I O M P N 9 5 (+ E )/ (+ E ) (+ D )/ (+ D ) (-B )/ (-B ) (-C )/ (+ C ) (-C )/ (-D ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
D O N I N I T B 1 0 0 0 0 (+ B )/ (+ C ) (-B )/ (+ B ) (+ D )/ (+ D ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )D O N I N I T W 5 0 0 0 (-B )/ (+ C ) (-B )/ (+ B ) (+ D )/ (+ D ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )H T N O 3 Y B 3 . 2 9 (+ C )/ (+ D ) (+ D )/ (+ D ) (-B )/ (-B ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )H T N O 3 Y W 3 . 2 9 (-B )/ (+ B ) (+ C )/ (+ C ) (-B )/ (+ B ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )I M M IN I T B 9 9 0 0 0 0 0 (-B )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-C )/ (+ D ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )I M M IN I T W 7 0 0 0 0 0 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )M I C R O N C 0 . 1 2 5 (-C )/ (-D ) (-B )/ (-B ) (-B )/ (+ B ) (+ D )/ (+ D ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )M T C D O N 0 . 0 0 0 0 2 (-B )/ (+ B ) (-B )/ (+ B ) (+ C )/ (+ C ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )M T C N H 4 0 . 0 0 0 0 6 (-C )/ (-C ) (-B )/ (-B ) (-B )/ (-B ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )M T C N O 3 0 . 0 0 0 0 6 (+ C )/ (+ C ) (+ C )/ (+ D ) (-B )/ (-B ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )N H 4 IN I T B 5 5 0 0 0 (+ E )/ (+ E ) (+ C )/ (+ C ) (-B )/ (-B ) (-C )/ (+ C ) (-C )/ (-C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )N H 4 IN I T W 3 0 0 0 0 (+ D )/ (+ D ) (+ B )/ (+ C ) (-B )/ (-B ) (-C )/ (+ C ) (-C )/ (-C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )N O 3 IN I T B 4 0 0 0 (+ C )/ (-D ) (+ D )/ (-F ) (-B )/ (+ C ) (-C )/ (+ C ) (-B )/ (+ C ) (-B )/ (-+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
138
TABLE 11 (CONT.) SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-)50% CHANGE IN BASE VALUES
P a r a m e te r B a s e V a lu e N H 4 N O 3 D O N P O N D O B O D 5 T S S D P T PN S Y IE L D B 0 . 3 (-B )/ (+ C ) (-D )/ (-D ) (+ B )/ (+ B ) (-C )/ (+ C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )N S Y IE L D W 0 . 3 (-C )/ (+ C ) (-D )/ (-D ) (+ B )/ (+ B ) (-C )/ (+ C ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )O N P A R T F 0 . 6 (+ D )/ (+ D ) (+ B )/ (+ B ) (-F )/ (-F ) (+ E )/ (+ E ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O N C O U T 0 . 7 5 (-C )/ (-C ) (-B )/ (-B ) (-B )/ (+ B ) (+ E )/ (+ E ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O N F A L L 0 . 2 5 (+ C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-E )/ (-F ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O N I N I T B 3 0 0 0 0 0 (+ E )/ (-D ) (+ F )/ (-F ) (-B )/ (+ B ) (+ C )/ (+ D ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) A / A A / AP O N I N I T W 1 5 0 0 0 (+ D )/ (-D ) (+ F )/ (-F ) (-B )/ (+ B ) (+ D )/ (+ D ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
P O N R E S 0 . 0 1 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (+ D )/ (+ D ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O N S IZ E 0 . 0 5 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )R E F N IN IT 5 0 0 0 0 0 (+ D )/ (-D ) (+ F )/ (-F ) (-B )/ (+ B ) (-C )/ (+ C ) (-C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B / (+ B )
R E S T H N (N IT ) 0 . 0 1 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (+ D )/ (+ D ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B / (+ B ) (-B )/ (+ B )
V E G E T A T I O NB I O D E N S 5 0 0 0 (+ B )/ (-B ) A / (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B )B I O IN IT 8 6 3 5 0 0 0 (-B )/ (-B ) (-B )/ (-B ) (+ B )/ (+ B ) (+ D )/ (+ D ) (-C )/ (-C ) (+ D )/ (+ D ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B )
P B I O U W 0 . 4 (+ B )/ (+ B ) (+ B )/ (-B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B )P E A T A C R B 2 0 0 0 (+ C )/ (+ C ) (+ B )/ (+ B ) (+ B )/ (+ B ) (-B )/ (-B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (-B )/ (-B ) (+ B )/ (+ B ) (+ B )/ (+ B )P E A T A C R W 3 0 0 (+ B )/ (+ B ) (+ B )/ (+ B ) (-B )/ (-B ) (-C )/ (-C ) (+ B )/ (+ B ) (-B )/ (-B ) A / A A / A A / AP E A T D E N S 0 . 7 (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (+ B )/ (+ C ) (-B )/ (-B ) (-B )/ (-B )P R A T E U P 0 . 7 (-D )/ (+ B ) (+ E )/ (+ D ) (-B )/ (-B ) (-B )/ (-B ) (+ B )/ (+ B ) (-B )/ (-B ) A / A (+ B )/ (+ B ) (+ B )/ (+ B )P S T D U W 0 . 4 (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B )S T A N D I N 6 0 0 0 0 0 0 (-D )/ (-D ) (-C )/ (-C ) (+ B )/ (+ B ) (+ D )/ (+ D ) (-C )/ (-C ) (+ D )/ (+ D ) (+ B )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B )S T D D E N S 5 0 0 0 (+ B )/ (+ B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B ) (-B )/ (-B )C A R B O N
B I O C C O N T 0 . 4 7 (-D )/ (-D ) (-C )/ (-C ) (+ B )/ (+ B ) (+ D )/ (+ E ) (-C )/ (-C ) (+ D )/ (+ D ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )B O D C F R A C * 0 . 8 (-C )/ (-C ) (-C )/ (-C ) (+ B )/ (+ B ) (-D )/ (+ C ) (-D )/ (-D ) (-E )/ (-E ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )B O D P F R A C 0 . 5 (-D )/ (-D ) (+ C )/ (+ C ) (-B )/ (-B ) (-C )/ (+ C ) (+ D )/ (+ C ) (-E )/ (-E ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )D O C I N I T B 6 0 0 0 0 (-B )/ (+ B ) (-B )/ (-B ) (+ B )/ (+ B ) (-C )/ (+ C ) (-C )/ (-C ) (+ D )/ (+ D ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )D O C I N I T W 5 0 0 0 0 (-B )/ (+ C ) (-B )/ (-B ) (+ B )/ (+ B ) (-C )/ (+ C ) (-C )/ (-C ) (+ D )/ (+ D ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
L E A C H R 0 . 0 1 (-B )/ (+ B ) (-B )/ (+ B ) (+ C )/ (+ C ) (-C )/ (+ C ) (-B )/ (-B ) (+ D )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )M I C R O B E C 0 . 5 3 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-C )/ (+ C ) (-B )/ (-B ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
M T C D O C 0 . 0 0 0 0 4 (-B )/ (+ B ) (-B )/ (-B ) (-B )/ (-B ) (-C )/ (+ C ) (-B )/ (-B ) (+ C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P E A T C C * 0 . 8 (-C )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-D )/ (+ C ) (-B )/ (-B ) (+ B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
P O C C O U T * 0 . 3 (+ B )/ (+ C ) (+ B )/ (+ B ) (-B )/ (-B ) (-C )/ (+ C ) (+ B )/ (+ B ) (+ E )/ (+ E ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O C F A L L 0 . 4 5 (-C )/ (-C ) (+ B )/ (+ C ) (+ B )/ (-B ) (-C )/ (+ C ) (+ B )/ (+ D ) (-E )/ (-E ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O C I N I T B 3 0 0 0 0 0 0 (-D )/ (-D ) (-C )/ (-C ) (+ B )/ (+ C ) (-C )/ (+ D ) (-C )/ (-C ) (+ B )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O C I N I T W 2 0 0 0 0 0 (-C )/ (-C ) (-B )/ (-B ) (+ B )/ (+ B ) (-C )/ (+ C ) (-C )/ (-C ) (+ C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
P O C R E S 0 . 0 0 1 (-C )/ (+ B ) (+ B )/ (+ B ) (+ B )/ (+ B ) (-C )/ (+ C ) (+ B )/ (-B ) (-E )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )P O C S IZ E 0 . 2 (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-C )/ (+ C ) (-B )/ (-B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )R E F C IN IT 1 5 0 0 0 0 0 0 (+ B )/ (+ B ) (+ B )/ (+ B ) (-B )/ (+ B ) (+ C )/ (+ C ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B ) (-B )/ (+ B )
139
TABLE 11 (CONT.) SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON WETLAND FOR (+/-)50% CHANGE IN BASE VALUES
P a r a m e t e r B a s e V a lu e N H 4 N O 3 D O N P O N D O B O D 5 T S S D P T PD I S S O L V E D O X Y G E N
D O C O N C P 0 . 0 0 1 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B / (+ B ) ( -C ) / (+ C ) (+ B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )D O I N I T B 1 5 0 0 0 (-D ) / (-D ) (+ C )/ (+ C ) (+ C )/ (+ C ) ( -C ) / (+ D ) (+ D )/ (+ D ) (+ C )/ (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )D O I N I T W 1 5 0 0 0 (-C ) / (-C ) (+ C )/ (+ C ) (+ B ) / (+ C ) ( -C ) / (+ D ) (+ D )/ (+ D ) (+ C )/ (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )
D O X Y C S A T 8 . 5 (-E ) / (N C ) (+ D )/ (N C ) (+ C )/ (N C ) ( -C ) / (N C ) (+ E ) / (N C ) (+ C )/ (N C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )H T D O Y B 0 . 1 5 (-E ) / (N C ) (+ E ) / (N C ) (+ C )/ (N C ) ( -C ) / (N C ) (+ E ) / (N C ) (+ C )/ (N C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )H T D O Y W 0 . 1 5 (-C ) / (N C ) (+ C )/ (N C ) (+ B ) / (N C ) ( -C ) / (N C ) (+ D )/ (N C ) (+ B ) / (N C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )M T D O X 0 . 0 0 0 1 (-D ) / (N C ) (+ D )/ (N C ) (+ C )/ (N C ) ( -C ) / (N C ) ( -C ) / (N C ) (+ B ) / (N C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )
M T F W S D O C 0 . 0 0 0 0 8 (-D ) / (N C ) (+ D )/ (N C ) (+ C )/ (N C ) ( -C ) / (N C ) ( -D ) / (N C ) (+ C )/ (N C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )N S D O Y B 0 . 2 (-C ) / (-E ) (+ C )/ (+ D ) (+ B ) / (+ C ) ( -C ) / (+ D ) (+ D )/ (+ D ) (+ B ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )N S D O Y W 0 . 2 (-C ) / (-D ) (+ B ) / (+ C ) (+ B ) / (+ B ) ( -C ) / (+ C ) (+ C )/ (+ D ) (+ B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B )
P H O S P H O R O U SB I O M P P 3 0 0 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ C )/ (+ D ) (+ C )/ (+ D )
B T P H O S I 2 5 0 0 0 0 (+ C )/ (+ C ) (+ B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ D ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ C )/ (+ C ) (+ C )/ (+ C )D T P H O S I B 4 0 0 0 0 (+ B ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ E ) / (+ E ) (+ E ) / (+ E )D T P H O S I W 3 8 0 0 0 (-B ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ E ) / (+ E ) (+ E ) / (+ E )M T C P H O S 0 . 0 0 0 0 6 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ C )/ (+ C ) (+ C )/ (+ C )P M I N P P C 0 . 0 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ C )/ (+ C ) (+ C )/ (+ C )
P R M I N B P C 0 . 0 0 0 0 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ C )/ (+ C ) (+ C )/ (+ C )P S E D D E P 0 . 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -E ) / (-F ) ( -C ) / (-C ) ( -C )( -C )
S E D I M E N TD E C O M P R 0 . 1 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (-B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (-B ) ( -B ) / (-B )
M A N N C (S E D ) 2 . 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (+ E ) ( -B )) / (+ C ) ( -B ) / (+ C )P S E D D E P 0 . 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -E ) / (-F ) ( -C ) / (-C ) ( -C ) / (-C )
R E S T H I C K (S E D ) 0 . 0 0 1 1 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) (+ E ) / (+ E ) (+ C )/ (+ C ) (+ C )/ (+ C )S E D F A L L 1 0 . 3 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -E ) / (-F ) ( -C ) / (-C ) ( -C ) / (-C )S E D F A L L 2 0 . 7 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (-F ) ( -B ) / (-B ) ( -B ) / (-B )S E D I N I T B 1 2 6 0 0 0 0 0 0 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ E ) / (+ E ) ( -B ) / (+ B ) ( -B ) / (+ B )S E D I N I T B 2 1 0 0 0 0 0 0 0 (-B ) / (+ B ) ( -B ) / (-B ) ( -B ) / (-B ) ( -C ) / (+ C ) ( -B ) / (-B ) ( -B ) / (-B ) (+ B ) / (+ B ) ( -B ) / (-B ) ( -B ) / (-B )S E D I N I T W 1 1 9 0 0 0 0 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) (+ B ) / (+ B ) ( -B ) / (+ B ) (+ E ) / (+ E ) ( -B ) / (+ B ) (+ B ) / (+ B )S E D I N I T W 2 1 0 0 0 0 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / ((+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (-B ) ( -B ) / (-B )
S E D R E S 0 . 1 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / ((+ B ) ( -B ) / (+ B ) (+ E ) / (+ E ) (+ C )/ (+ C ) (+ C )/ (+ C )S E D S I Z E 1 0 . 2 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ E ) ( -B ) / (+ B ) ( -B ) / (+ B )S E D S I Z E 2 0 . 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ B ) / (+ B ) (+ B ) / (+ B )S E D S P G 1 1 . 1 (-B ) / (+ B ) ( -B ) / (+ B ) (+ B ) / (+ B ) ( -C ) / (+ C ) (+ B ) / (+ B ) (+ B ) / (+ B ) ( -B ) / (-B ) ( -B ) / (+ B ) ( -B ) / (-B )S E D S P G 2 2 . 6 5 (-B ) / (+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) ( -C ) / (+ C ) ( -B ) / ((+ B ) ( -B ) / (+ B ) ( -B ) / (+ B ) (+ B ) / (+ B ) (+ B ) / (+ B )
- Results presented as the RS values for the +50% change in base value followed by the –50% change.- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible results)- * RS values presented for +25% due to fraction value greater than one.
140
sensitivity analysis indicated that most of the RS values approached zero and thus model
response is extremely damped, implying that the effect of most parameters on model predictions
will be minimal.
Generally, the NCOB cycle is insensitive to changes to a single parameter due to the
complexity and number of interactions in the wetland system; however, there are exceptions.
The NCOB cycle is most sensitive to changes that affect microbial growth and oxygen; which is
exemplified by the situations in which the model was unable to complete the simulations
(HTDOYB, AEMAXGRB, HETROIB, MTDOX, NMAXGRB, HORGHSCW, MTDOX,
MTFWSDOC, HTDOHSCB, and HTDOYW). Since SET-WET is driven by the interactions
and processes of the bacteria cycle, it makes sense that the model is most sensitive to the
parameters affecting the growth and number of bacteria in the system. The faster the growth of
bacteria in the system, the quicker organics degradation, ammonification, nitrification and
denitrification occurs. Changing bacterial parameter values can considerably change the
bacterial growth, rates of nutrient uptake, and transformation rates determined by SET-WET in
the wetland system. As the changes accumulate, the modeling system may be unable to properly
account for the values, producing incomprehensible results, which include any negative values
for a stock variable. Since, the DO concentration interacts greatly with the bacteria cycle inSET-
WET, it is not surprising that the model is sensitive to changes in the DO parameter values.
Bacterial growth is affected by the DO concentrations because various rates in the bacteria cycle
that determine bacterial nutrient utilization, the determination of the anaerobic fraction
(ANFRAC), and growth, are a function of DO. The bacteria in turn consume the oxygen,
forming a carefully balanced cycle.
Trends in which parameters were the most significant did materialize. The BOD5
simulated values were sensitive to changes involving the C cycle parameters, while the NH4+ and
NO3- output were sensitive to changes associated with bacteria and NO3
- parameters. Any
attempt to vary the fall rate (SEDFALL) or resuspension percentage of particles, significantly
affected particle outflow. These results are expected as changes in these parameters directly
affect the respective cycles. In addition, parameters associated with the bottom of the wetland
system have a greater effect on predictions, which is attributable to the more prevalent number of
bacteria in the wetland bottom. Since the NCOB, and sediment and P cycles were designed to
run independently, changes in associated parameter values had little effect upon the predictions
141
of the other independent cycles. The model is generally insensitive to changes affecting
sediment or P parameters.
D. Modeling Application
SET-WET was developed to assist with the design of constructed wetlands for optimizing
the control of NPS pollution. To examine the potential use of SET-WET, it was applied to a
hypothetical situation where a constructed wetland might be desired. The objective of this
analysis was to observe the potential use of SET-WET for long-term simulations, to analyze the
design capabilities of SET-WET, and determine its strengths and weaknesses. The wetland
designs emphasized the capture of NPS pollutants from a watershed area, with the goal of
optimizing pollutant removal. This is a simplification of the wetland design process because
many factors must be accounted for. Besides the need to control pollution; ecological,
economics, and landowner concerns must also be addressed. The designer must determine the
proper balance between these factors.
1. Study/Application Area
Data that was collected by Mostaghimi et al. (1998) in the Nomini Creek watershed was
used as input values for the simulation. This site was selected because long term field data that
contained most of the required input to the SET-WET model was collected from the watershed.
Any other data set that contains the hydrologic and nutrient values from a watershed area could
be used for simulation with the SET-WET model.
The Nomini Creek watershed is located in Westmoreland County, Virginia, and is 80-km
northeast of Richmond (Figure 43). In its entirety, the watershed is 1463 hectares large, but for
the purpose of this analysis a subwatershed, entitled QN2, was used (Figure 44). QN2 is 214
hectares in size. The watershed has typical Coastal Plain land use characteristics as 49 % is used
for cropland, 47% woodland, and 4% for homesteads and roads, with no significant point sources
in the area (Mostaghimi et. al, 1998).
The climate in the area is temperate and humid. Average annual precipitation is 101.6
cm, with a large percentage of the rainfall occurring between April and September; while the
142
FIGURE 43: LOCATION OF THE NOMINI CREEK WATERSHED IN VIRGINIA WITH RESPECT TO RICHMOND,VA AND THE CHESAPEAKE BAY.
FIGURE 44: NOMINI CREEK WATERSHED (QN1) WITH SUBWATERSHED (QN2; SHADED)
average annual snowfall is about 10 cm. The average summer and winter temperatures are 25 °C
and 3 °C, respectively (Mostaghimi, et al., 1998).
Data collection in the watershed occurred from 1986 to 1997, and various hydrologic
(streamflow, precipitation), nutrient (ammonium, nitrate, DON, PON, COD dissolved ortho P,
dissolved organic P, Particulate P, TSS) and climactic (air temperature) parameters were
143
collected. Specific BMPs were implemented in the watershed including: strip cropping;
vegetative filter strips; no-till cropping systems; critical area stabilization; drop structures;
diversion structures; nutrient management; and IPM (Mostaghimi et al., 1998). The use of
BMPs should theoretically decrease the amount of nutrients that leave the watershed outlet,
thereby leaving smaller amounts of pollutants for the hypothetical wetland to treat.
Data for a two-year period (March 26, 1992 to March 25, 1994) were used for these
simulations. This time period was chosen because data collection during this period was
complete, with very few missing data points. Data collection of the hydrologic and nutrient
parameters occurred on a weekly and storm-event basis, excluding the water temperature and
influent BOD5. These values were extended to daily input points by linear interpolation between
the weekly and storm values. Water temperature was estimated from the average daily
temperature in the system by subtracting 2 °C from the average daily air temperature, while BOD
influent was estimated from the periodically taken COD measurements. The determination of
the daily BOD influent required an estimate drawn from the COD measurements in the
watershed. It was assumed that the ultimate BOD/COD ratio approached one (Perrich, 1981).
Fifty-five measurements of COD were taken over the eleven-year research period. These fifty-
five measurements were plotted with the recorded influent flow as shown in Figure 45. A linear
regression was calculated from these points which allowed a daily estimation of BOD5
(dependent variable; y) to be made based upon the hydrologic influent measurements
(independent variable x) for the simulation period. Although the R2 value of the regression line
was not that high, it was the best estimate for the site. Data for DO was not collected, therefore
it was assumed that the concentration of DO entering the site was 5.7 mg/L.
2. Simulation Runs
Simulation runs were divided into eight season periods with each season period
corresponding to the seasonal changes in the area. The growing season was designated as the
time period between the average first and last freeze of the year. For the Nomini Creek
watershed, data from Richmond and Norfolk, Virginia (Wood, 1996) were averaged to determine
the growing season. There is a mean freeze free period of 237 days in the area meaning 128 days
could be designated as ‘winter.’ The ‘winter’ time period was from Nov. 17th to March 25th.
144
FIGURE 45: LINEAR REGRESSION OF RECORDED TOTAL BOD5 AND HYDROLOGIC INFLOW TO QN2SUBWATERSHED OF NOMINI CREEK WATERSHED FOR MARCH 26, 1992 TO MARCH 25, 1994.
‘Spring,’ occurred from March 26th to June 21st, ‘summer,’ from June 22nd to September 21st, and
‘fall,’ from September 22nd to November 16th. The Thornthwaite method was used to determine
ET, with average monthly temperature values being estimated from data collected at Chatham,
Virginia (Ruffner, 1980).
It was assumed that the wetland would be placed at the outlet of the QN2 subwatershed.
It was placed at the outlet since this would allow the highest capture rate of pollutants from the
area, as all surface flows must leave from this area. Since the watershed outlet consists of a
perennial stream, and groundwater flow is prevalent in the area, it is assumed that the wetland is
lined. Water may flow from the watershed to the wetland, but due to the lining, direct
groundwater interactions with the wetland water cycle is assumed negligible. It is assumed that
there are no point source additions to the wetland and that it is fully established for the
simulation periods, where full plant and bacteria colonies exist.
145
The first step in determining the design of a wetland is to size the wetland for proper
hydrologic outflow and properties, because improper hydrologic design will not allow a
wetland’s vegetation to establish, and the system will become a detention basin rather than a
wetland. The Water Pollution Control Federation (1990; Table 3) suggests that the desired
minimum hydraulic residence time in the wetland system should be 5 days with a maximum
water depth of 50 cm. The residence time is suggested to allow the various nutrient processes
enough time to complete their respective reactions, while the maximum depth is suggested for
human safety reasons. The average hydraulic flow into the system is 2700 m3/day, which would
require that the wetland area be at least 27,000 m2 (2700m3/day*5days/.5m) or 2.7 hectares. The
second step in the wetland design would be to decide on the configuration of the wetland. Since
the Water Pollution Control Federation suggests a desired length to width ratio of at least 2 to 1,
the wetland system’s length was input as 270 meters with an input width of 100 meters.
Technically, since the wetland is modeled as a continuously stirred tank reactor (CSTR), any
length to width ratio (1:1, 1:2, 5:1) whose product was a total wetland area of 2.7 hectares, would
have resulted in the same output values. This is because the SET-WET model accounts for the
volume and mass relations, and not space relations.
Once the proper hydrologic design was designated, SET-WET was used to model the
NCOB, sediment and P cycles. It was assumed that the input parameters were similar to those at
the Benton wetland site, as no other information was available for comparison. Similar input
values were necessary to complete model simulations; however, since the model has not been
calibrated with proper NPS pollution data, it is not certain if the parameter values are correct.
Table 12 lists the initial input parameters to the SET-WET model for the simulation run. This
simulation is referred to as the “2.7 hectare,” wetland from this point on. Occasionally,
parameter values were changed between season periods through the simulations to better
represent the processes occurring in the wetland. These changes were made to the more sensitive
parameters, mainly those that affect microbial growth rates, and oxygen transfer.
Another suggestion by the Water Pollution Control Federation, for the design of a FWS
constructed wetland is to have the wetland be at least 2 hectares in size for every 1000m3 of
inflow per day. This would require the wetland design to be at least 5.4 hectares in size (2700
m3/day * 2 ha/1000 m3/day). The wetland for this shape was designated as 540 meters in length
146
with a width of 100 meters. Once again this design is to the discretion of the model user. This
simulation is referred to as the “5.4 hectare,” wetland in Table 12.
The objective of the model application was to determine the proper design of a wetland to
optimize wetland NPS pollution control. Other considerations include cost and the size of the
wetland. A smaller wetland would be less expensive to build and would obviously take up less
area. For this reason a smaller area wetland with a higher water depth design was also examined.
This wetland was designated as the “Smaller,” site in Table 12.
Another design parameter examined was the amount of biomass in the wetland system.
Plants absorb nutrients in the wetland system; therefore if the plant biomass in the system is
doubled, theoretically more nutrients will be assimilated. However, when decomposition of the
plant material occurs there will also be twice as much to decompose. This simulation will allow
plant growth to be examined and is referred to as “Plants,” in Table 12.
Substrate depth also plays an important role in wetland processes. In the wetland
substrate, many of the transformation processes for the various nutrient cycles occurs. If the
amount of substrate in the bottom does not have a large affect on the total retention of nutrients
in the system, less humic soil would need to be transported to a potential wetland site. The
smaller wetland substrate would also increase the concentration of nutrients in this compartment.
This condition was examined and is referred to as “Substrate,” in Table 12.
In this analysis, the biomass growth rate was the same for all simulation runs, except for
the "Plant" analysis where the value was doubled. Due to the difference in wetland water depth,
it is unlikely that the same species of vegetation would be able to survive in the differing wetland
constructions, but for these simulations, they were considered to be the same.
3. Simulation Results
Table 13 lists the results of the model simulations for the five separate events. Listed are
the influent total, effluent total, and reduction efficiency for the NH4+, NO3
-, DON, PON, TKN,
TN, TSS, DP and TP for the respective wetland designs for simulation runs of two years. The
influent and effluent totals are based on the entire two-year time frame, with the reduction
efficiency based on the retention of nutrients over this period. In addition, Table 13 lists the
147
TABLE 12: INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL SIMULATION RUNS
FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF NOMINI CREEK
WATERSHED.2.7 5.4
P a ra m e te r He cta re He cta re S m a lle r P la nts S ubstra teBAS ELE NGTH 270 540 200 200 200W IDTH 100 100 55 55 55HO 2.2 2.2 2.3 2.3 2.3HB 0 0 0 0 0.5HTI 1 1 1 1 1HII 1.4 1.4 1.7 1.7 1.7S O 0.001 0.001 0.001 0.001 0.001
HYDROLOGYP ORP E A T 0.32 0.32 0.32 0.32 0.32OUTLE T 2 2 2 2 2A NGV NOT 60 60 60 60 60DIS E FFC 0.175 0.175 0.175 0.175 0.175K HCOE F 0.0012 0.0012 0.0012 0.0012 0.0012HOUT 1.25 1.25 1.6 1.6 1.6HOV E R 2 2 2 2 2
BIOM AS SB IOINIT 20000 10000 4000 8000 4000S TA NDIN 35000000 25000000 15000000 20000000 15000000P E A TA CRW 300 300 300 300 300P E A TA CRB 2000 2000 2000 2000 2000P E A TDE NS 110 110 110 110 110P RA TE UP 0.3 0.3 0.6 0.6 0.6B IODE NS 75 75 75 75 75S TDDE NS 75 75 75 75 75P B IOUW 0.4 0.4 0.4 0.4 0.4P S TDUW 0.4 0.4 0.4 0.4 0.4DE GB IO 0.99 0.99 0.99 0.99 0.99DA Y S DE G 20 20 20 20 20(M A XIM UM ) B IOM GRR 5 5 5 10 5
BACTERIANITROS IW 10000 5000 2000 2000 2000NITROS IB 50000 25000 10000 10000 10000HE TE ROIW 15000 7500 3000 3000 3000HE TE ROIB 1000000 500000 200000 200000 100000NDRA TE W /NDRA TE B .001/.001 .001/.001 .001/.001 .001/.001 .001/.001NDOHS A TW /NDOHS A TB 1./1. 1./1. 1./1. 1./1. 1./1.NM A XGRW /NM A XGRB .008/.008 .008/.008 .008/.008 .008/.008 .008/.008NNH4HS CW /NNH4HS CB 1./1. 1./1. 1./1. 1./1. 1./1.A E M A XGRW /A E M A XGRB .08/.01 .08/.01 .08/.01 .08/.01 .1/.03A NM A XGRW /A NM A XGRB .2/.04 .2/.04 .2/.04 .2/.04 .2/.04HTDRW /HTDRB .0035/.0035 .0035/.0035 .0035/.0035 .0035/.0035 .0035/.0035HTDOHS CW /HTDOHS CB .5/.05 0.5/0.5 .5/.5 .5/.5 .5/.5
148
TABLE 12 (CONT.): INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL
SIMULATION RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF
NOMINI CREEK WATERSHED.2.7 5.4
Parameter Hectare Hectare Sm aller Plants Substra teHNO3HSCW /HNO3HSCB .15/.15 0.15/0.15 .15/.15 .15/.15 .15/.15HORGHSCW /HORGHSCB 50/50 50/50 50/50 50/50 50/50ANFRACW /ANFRACB .2-.8 .2-.8 .2-.8 .2-.8 .2-.8HTTEMPFW /HTTEMPFB 0.0-1.0 0.0-1.0 0.0-1.0 0.0-1.0 0.0-1.0
CARBONREFCINIT 1400000 700000 340000 340000 340000DOCINITW 250000 125000 50000 50000 50000DOCINITB 50000 25000 10500 10500 10500POCINITW 200000 100000 40000 40000 40000POCINITB 1000000 500000 2000000 2000000 2000000BIOCCONT 0.47 0.47 0.47 0.47 0.47BODCFRAC 0.8 0.8 0.8 0.8 0.8BODPFRAC 0.5 0.5 0.5 0.5 0.5LEACHR 0.1 0.1 0.1 0.1 0.1MICROBEC 0.53 0.53 0.53 0.53 0.53PEATCC 0.8 0.8 0.8 0.8 0.8POCFALL 0.6 0.6 0.6 0.6 0.6POCRES 0.001 0.001 0.001 0.001 0.001MANNC 2 2 2 2 2RESTHC 0.005 0.005 0.005 0.005 0.005POCSIZE 0.2 0.2 0.2 0.2 0.2MTCDOC 0.00005 0.00005 0.00005 0.00005 0.00005POCCOUT 0.2 0.2 0.2 0.2 0.2
NITROGENDONINITW 15000 7500 3000 3000 3000DONINITB 25000 12500 5000 5000 5000IMMINITW 50000 25000 10000 10000 10000IMMINITB 25000000 12500000 5000000 5000000 5000000NH4INITW 25000 1250 500 500 500NH4INITB 15000 7500 3000 3000 3000NO3INITW 15000 7500 3000 3000 3000NO3INITB 25000 12500 5000 5000 8000PONINITW 75000 37500 15000 15000 15000PONINITB 1500000 750000 300000 3000000 3000000REFNINIT 50000 250000 105000 105000 105000BIOMCN 23.5 23.5 23.5 25 23.5BIOMPN 95 95 95 95 95HTNO3YW /HTNO3YB .2/1 .2/1 .2/1 .2/1 .5/1.0MICRONC 0.125 0.125 0.125 0.0125 0.125NSYIELDW /NSYIELDB .2/.15 .2/.15 .2/.15 .2/.2 .3/.3ONPARTF 0.6 0.6 0.6 0.6 0.6PEATNC 0.025 0.025 0.025 0.025 0.025PONRES 0.01 0.01 0.01 0.01 0.01PONFALL 0.5 0.5 0.5 0.5 0.5MTCDON 0.00004 0.00004 0.00004 0.00004 0.00004
149
TABLE 12 (CONT.): INITIAL INPUT PARAMETERS TO SET-WET MODEL FOR FIVE HYPOTHETICAL
SIMULATION RUNS FOR POTENTIAL FWS CONSTRUCTED WETLAND IN QN2 SUBWATERSHED OF
NOMINI CREEK WATERSHED.2.7 5.4
Parameter Hectare Hectare Smaller Plants SubstratePONSIZE 0.05 0.05 0.05 0.05 0.05RESTHN 0.01 0.01 0.01 0.01 0.01PONCOUT 0.375 0.375 0.375 0.375 0.375
DISSOLVED OXYGENDOINITW 140000 70000 28000 28000 28000DOINITB 20000 10000 4000 4000 4000HTDOYW/HTDOYB .1/.15 .1/.15 .5/.5 .1/.15 .4/.4NSDOYW/NSDOYB .02/.05 .02/.05 .08/.08 .02/.05 .05/05DOCONCP 0.001 0.001 0.001 0.001 0.001MTDOX 0.0003 0.0003 0.0003 0.0003 0.0006MTFWSDOC 0.00008 0.00008 0.00008 0.00008 0.00008DOXYCSAT 8.5 8.5 8.5 8.5 8.5BIOOXRB 0/.18 0-.18 0-.18 .0-.18 0-.18DOCONCIN 5.7 5.7 5.7 5.7 5.7
SEDIMENTSEDCAT 3 3 3 3 3SEDRES 0.001 0.001 0.001 0.001 0.001SEDSIZE (SEDCLASS) .074-.2 .074/.2 .074-.2 .074-.2 .074-.2SEDFALL (SEDCLASS) .7/1.2 .7/1.2 .7-1.2 .7-1.2 .7-1.2SEDINITW (SEDCLASS) 5000-120000 2500-60000 1000-24000 1000-24000 1000-24000
SEDINITB (SEDCLASS)2500000-
59400000001250000-
2970000000500000-
1210000000500000-
1210000000500000-
605000000SEDSPG (SEDCLASS) 2.65 2.65 2.65 2.65 2.65SEDPER (SEDCLASS) .8/.2 .8/.2 .8/.2 .8/.2 .8/.2RESTHICK 0.0001 0.0001 0.0001 0.0001 0.0001MANNC 0.5 0.5 0.5 0.5 0.5DECOMPR 0.1 0.1 0.1 0.1 0.1PSEDDEP 1 1 1 1 1
PHOSPHOROUSDTPHOSIW 10000 5000 2000 2000 2000DTPHOSIB 50000 25000 10000 10000 10000BTPHOSI 5400000 2700000 900000 900000 900000PPHOSI 75000 37500 15000 15000 15000PMINPPC 0.005 0.005 0.005 0.005 0.005PRMINBPC 0.00005 0.00005 0.00005 0.00005 0.00005ADSORP 2 2 2 2 2MTCPHOS 0.00006 0.00006 0.00006 0.00006 0.00006BIOMPP 300 300 300 300 300PHOSCON 0 0 0 0 0
150
TABLE 13: INFLUENT, EFFLUENT, AND % REDUCTION OF NUTRIENTS FOR VARIOUS NUTRIENTS FOR 2-YEAR PERIODS OF WETLAND SIMULATIONS
FPOR QN2 SUBWATERSHED DATA.
HRT HLR Water2.7 hectare NH4 NO3 DON PON TKN TN BOD5 TSS DP TP (days) (cm/d) Depth (cm)
Total In 1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06Total Out 2.31E+05 3.07E+06 2.00E+06 2.16E+05 2.44E+06 5.52E+06 6.01E+07 1.14E+06 7.88E+05 7.88E+05 4.96 10.92 0.52% Reduction -65.79% 18.48% -9.38% 95.97% 66.66% 50.29% 44.62% 99.78% 16.43% 46.33%
HRT HLR Water5.4 hectare NH4 NO3 DON PON TKN TN BOD5 TSS DP TP (days) (cm/d) Depth (cm)
Total In 1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06Total Out 1.96E+05 2.46E+06 2.06E+06 1.22E+05 2.38E+06 4.84E+06 5.91E+07 5.91E+05 6.39E+05 6.39E+05 9.84 5.49 0.53% Reduction -41.08% 34.73% -12.95% 97.72% 67.52% 56.38% 45.50% 99.88% 32.24% 56.48%
HRT HLR WaterSmaller ! NH4 NO3 DON PON TKN TN BOD5 TSS DP TP (days) (cm/d) Depth (cm)
Total In 1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06Total Out 6.10E+05 3.38E+06 1.95E+06 5.09E+05 3.07E+06 6.45E+06 6.40E+07 7.61E+07 9.00E+05 9.11E+05 3.46 26.72 0.87% Reduction -338.15% 10.32% -6.81% 90.51% 58.14% 41.89% 40.98% 85.09% 4.54% 37.94%
HRT HLR WaterPlant * NH4 NO3 DON PON TKN TN BOD5 TSS DP TP (days) (cm/d) Depth (cm)
Total In 1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06Total Out 5.36E+05 3.20E+06 1.95E+06 5.28E+05 3.01E+06 6.21E+06 6.51E+07 7.90E+07 8.39E+05 8.49E+05 3.40 26.72 0.87% Reduction -285.62% 15.20% -6.70% 90.16% 58.90% 44.05% 39.94% 84.52% 11.03% 42.12%
HRT HLR WaterSubstrate # NH4 NO3 DON PON TKN TN BOD5 TSS DP TP (days) (cm/d) Depth (cm)
Total In 1.39E+05 3.77E+06 1.82E+06 5.36E+06 7.33E+06 1.11E+07 1.08E+08 5.10E+08 9.43E+05 1.47E+06Total Out 3.05E+05 3.44E+06 1.95E+06 5.63E+05 2.82E+06 6.26E+06 6.42E+07 7.61E+07 9.03E+05 9.14E+05 3.46 26.72 0.87% Reduction -119.51% 8.78% -7.05% 89.49% 61.48% 43.58% 40.83% 85.08% 4.21% 37.73%
! "Smaller" is 1.1 hectares in size.* "Plant" has 2 times the biomass amount of "Smaller".# "Substrate" has 50% the substrate thickness of "Smaller".
151
average hydraulic residence times (HRT), hydraulic loading rate (HLR), and water depth for the
two-year simulations.
For all simulations, the NH4+ concentration (-41.1% to –338.2%) considerable increased
after traveling through the wetland. This is not an unreasonable result because the loading rates
of NH4+ to the system were minute when compared to the loading rates of the organic N and
NO3- to the system. The wetland acts as a source for NH4
+ because a relatively large amount of
organic N goes through the mineralization process, especially during the decomposition stage of
winter, and the NH4+ does not have enough time to be converted to NO3
- before leaving the
system.
The largest wetland (5.4 hectares) retained more nutrients than the other wetland designs
because in the "5.4 hectare" wetland there are more bacteria to carry out processes, more area for
transformation processes to occur, and a longer time frame for these processes to occur.
Retention of TSS in the "5.4 hectare" and "2.7 hectare" sites were nearly 100%, as compared to
the 85% retention for the other three designs, which is attributable to the smaller average water
depth in the systems (.52 m to .87 m, respectively). The system with more vegetation (Plants)
retained more nutrients when compared to the "Smaller" wetland; however, the extra amount
retained seem negligible (3% to 5% greater for most parameters) as only a slightly higher
percentage is retained by the wetland system. Another interesting note is that the thickness of
the wetland substrate seemed negligible to the amount of nutrients retained by the wetland
system. In fact, the retention in "Substrate" was greater for NH4+, TKN, and TN. Thus, it seems
that depending on the location of the wetland material, the designer has flexibility with the
amount of substrate that needs to be transported to or from a construction location.
All of the designed wetlands reduced the percentage of BOD5 (39.9%-45.5%), TSS
(84.5%-99.9%), TN (41.9%-56.4%), and total P (37.7%-56.5%) to levels reported by previous
research (Table 14). Depending on the retention requirements of the wetland, an ‘optimal’
wetland can be designed for the area by using the information gained from the five simulations.
If an increase in TN retention was needed, it can be seen from Table 13 that the retention
increased with an increase in wetland size, plant amounts, and a decrease in substrate thickness.
These design parameters can be modified until a desirable retention is attained. Conversely, if
the goal is to retain a high percentage of TSS, it is apparent that a larger wetland with a smaller
water depth is more effective for sediment retention.
152
TABLE 14: RANGE OF POLLUTANT REMOVAL EFFICIENCIES REPORTED FOR CONSTRUCTED WETLAND
SYSTEMS
P a r a m e te r R e m o v a l (% )
B O D 5 5 0 -9 0
S S 4 0 -9 4
N it ro g e n 3 0 -9 8
P h o s p h o ro u s 2 0 -9 0
(Bastian and Hammer, 1993)
An optimal design is constrained not only by the nutrient retention of the wetland but also
by other constraints, such as area available to construct the wetland and the cost of the system.
For example, it is useless to design a wetland for 5 hectares which costs $50,000 to build, when
there are only 3 hectares to construct on and $35,000 to spend. Working with known constraints
allows a designer to set up the ‘optimal’ design by accounting for all of the factors that affect
wetland design and construction.
A drawback of modeling the system with a continuously stirred tank reactor design is that
one can not determine the effect of the systems shape, length and width on total retention by the
system. This problem can be addressed with a distributed modeling style, but it is unknown if
the increase in data input requirements make this style feasible since the model would be very
difficult to calibrate, validate and apply.
E. Model Evaluation Summary
Due to the lack of data collected for NPS pollution control with FWS wetlands, the SET-
WET model was evaluated with data collected from a wetland site built for municipal
wastewater treatment located in Benton, Kentucky. Model evaluation included the calibration
and validation of the model by using two statistical analyses, performing a sensitivity analysis
and using SET-WET to demonstrate its application for design of wetlands.
A non-parametric Wilcoxon Rank Sum statistical analysis indicated eight out of nine
examined outflow predictions were not statistically different from the measured observations.
Linear regression analysis showed that six out of nine examined parameters were statistically
similar, and that within the expected operating range, all of the examined outflow parameters
153
were within the 95% confidence intervals of the regression lines. A sensitivity analysis showed
the most significant input parameters to the model were those which directly affect bacterial
growth and oxygen uptake and movement. SET-WET was applied to a hypothetical simulation
for the QN2 subwatershed in the Nomini Creek watershed located in Virginia. Various designs
were applied to the area to determine which design parameters had the largest control upon
sediment and nutrient retention at the site. An optimal site design can be made based on the
SET-WET output and constraints from the site. Selecting one ‘optimal’ wetland design depends
on many factors, but when considering nutrient retention only, SET-WET may assist in the
optimization of wetland design.
154
V. Summary, Conclusions, and Recommendations
Constructed wetlands are used as a BMP to alleviate the impact of NPS pollution
problems. Constructed wetlands are an attractive BMP because in addition to controlling NPS
pollutants, they provide other beneficial functions to the environment such as wildlife habitat and
recreation, among others.
One of the more effective ways to enhance the design and implementation of wetlands is
to use computer models. Models provide an ability to make comparisons among alternative
designs and management strategies, thus allowing a wetland to be optimally utilized for its
intended purpose. A model, SET-WET, has been developed which allows these comparisons to
be made.
The SET-WET model is a user-friendly, dynamic, simulation model for design and
evaluation of constructed wetlands in order to optimize NPS pollution control measures.
SET-WET is written in Fortran 77 to facilitate linking with existing NPS models. The model
simulates the hydrologic, N, C, bacteria, DO, vegetative, P and sediment cycles within a wetland
system. The model allows for either free water surface (FWS) or subsurface flow (SSF) wetland
simulations, and is designed in a modular manner; thus it gives the user the flexibility to
concentrate on simulation of specific cycles and processes. The season/time period breakdown
accounts for seasonal variation by allowing the user to change parameter values in the middle of
a simulation run. It allows two forms of data input, one based on measured daily values and
another based on estimates from the SCS curve method in conjunction with runoff concentration
coefficients. Designed as a continuously stirred tank reactor, the model assumes that all
incoming constituents are evenly mixed throughout its entire volume.
The model was calibrated and validated with limited data collected from a constructed
wetland located in Benton, Kentucky. This data was not ideal, as monthly data points needed to
be interpolated to daily values, and the site was designed for wastewater treatment, not NPS
pollution control. Parameter input values were based on previous research and site data. Non-
parametric statistical analyses of the validated results indicated that the predicted hydrologic,
NO3-, NH4
+, organic N, BOD5, TSS, dissolved and total P concentrations were not statistically
different from the measured observations. The Wilcoxon-Rank statistical analyses indicate that
the confidence in the organic N, TSS, and DP predictions were high as the respective p-values
155
approached 1.0. Predictions for DO were consistently higher than observed values, which are a
cause for concern as the DO concentration directly affects many of the bacteria processes.
Linear regression analysis showed that six out of nine examined parameters were statistically
similar, and that within the expected operating range, all of the nine examined outflow
parameters were within the 95% confidence intervals of the regression lines. Due to the limited
amount of data points used for the statistical analyses, results need to be used with caution.
The sensitivity analysis showed that the model was most sensitive to parameters that
directly or indirectly affect bacterial growth or oxygen concentrations in the system. The model
was very sensitive to parameters such as the heterotrophic DO yield and the anaerobic maximum
growth rate in the soil, as values became incomprehensible. In general, however, the NCOB,
sediment and P cycles are relatively insensitive to changes to any single parameter due to the
complexity of the wetland system.
Hypothetical simulations were conducted for design of a possible wetland located in the
QN2 subwatershed of the Nomini Creek watershed in Virginia. The model made predictions on
the nutrient retention of five wetland designs (2.7 hectare, 5.4 hectare, 1.1 hectare, 2 * Plants,
50% Substrate) which when used in conjunction with other constraints (land, ecological,
monetary), a designer would be able to select an optimum design for the wetland. The designed
wetlands reduced the percentage of BOD5 (39.9%-45.5%), TSS (84.5%-99.9%), total N (41.9%-
56.4%), and total P (37.7%-56.5%) to levels reported by previous researchers.
SET-WET differs from many existing wetland models in that it uses a system’s approach,
and limits the assumptions made concerning the interactions of the various nutrient cycles in a
wetland system. It accounts for C and N interactions, as well as for effect of oxygen levels upon
microbial growth. It also directly links microbial growth and death to the consumption and
transformations of nutrients in the wetland system. Many previous models have accounted for
these interactions with zero and first order rate equations that assume rates are dependent only on
initial concentrations.
SET-WET was developed to be as general as possible, but there were assumptions made
during model development which might not apply to all wetland locations. The model assumes
that the FWS system always has free lying surface water, and that the SSF system never floods.
It does not account for snow or snow melt and assumes that the dissolved and particulate
concentrations in both the surface water and substrate water are evenly mixed through the
156
respective volumes. Plant growth is assumed to be of constant composition throughout the
wetland, and is not limited by lack of nutrients or water supply. In addition, growth and death
cannot occur at the same time, and the model does not account for turnover of plant material
during the year. Oxygen transfer by plants is considered to be constant through the growing
season, and it is also assumed that biomass and bacteria prefer nitrate rather than ammonia.
Modeling of adsorption of NH4+ is not conducted as it is assumed that the amount that is
adsorbed is still available for microbial use. The model assumes that P is attached to all of the
particulates in the system based on the surface area of the particles. In addition, SSF wetland
aeration is assumed to be negligible because the water surface is below the substrate, and SSF
wetland PON and POC removal is assumed to be 100% since there are no simple models
available to handle this process.
Through the analysis of the SET-WET model, there are a few suggestions made to
improve the model's predictions. These suggestions concern both laboratory work and improved
model development:
1) There is a need to incorporate a function for sediment resuspension based on the
difference between wetland water velocity and the critical velocity. Sediment
resuspension from the wetland substrate is currently a constant amount whenever the
critical velocity is attained. Another possibility is the use of the ‘overflow rate’ equation
to determine sediment outflow.
2) There is a need for an improved dissolved oxygen model that considers the biomass
oxygenation rate as a function of time. Oxygen is a very sensitive parameter in the SET-
WET model; therefore the assumption of constant rates over the entire growing season
may be limiting the model’s functions.
3) Improve the capabilites of the plant model. There is a need to account for the turnover
ratio of plant life throughout the growing season and not have all death occur only during
a ‘winter’ period.
4) Examine the validity of the assumption that ammonium adsorption is negligible.
157
5) Examine if the modeling of diffusion in the system is quantified properly. Examine if the
diffusion coefficient needs to be a function of time and concentration, instead of a
uniform rate through each season period.
6) Better quantify the initial microbial biomass for HT and AT populations, since the model
is sensitive to these values. In addition, quantify growth parameters in situations that are
more applicable to NPS conditions.
7) Link the model to an existing NPS model, such as ANSWERS, to improve incoming
nutrient flows when the input data is unknown.
8) Extend the model to a series of continuously stirred tank reactors. This would allow the
analysis of wetland length to width ratios, and should allow more accurate predictions of
hydrologic and nutrient effluent.
The SET-WET model takes a novel approach in its attempt to model the constructed
wetland system. It varies from similar models due to its modularity, extreme generality, time
management approach, and the fact that it models both FWS and SSF wetlands. In many ways
this flexibility is the model’s strongest asset, because it provides the model user the ability to
attempt a number of designs for constructed wetlands.
In closing, the SET-WET model is a promising start to modeling FWS constructed
wetlands in a manner that incorporates most of the interactions in a wetland system. There is a
need to further test the model's capabilities, but a useful design tool for both FWS and SSF
constructed wetlands has been developed.
158
VI. Cited Work:
Abtew, W. (1996) Evapotranspiration measurements and modeling for three wetland systems insouth Florida. Water Resources Bulletin 32(3): 465-473
Allen, R. G., M. E. Jensen, J. L. Wright, and R. D. Burman (1989) Operational estimates ofreference evapotranspiration. Agronomy Journal 81: 650-662.
ASTM (1995) Standard practice for evaluating mathematical models for the environmental fateof chemicals. Standard E 978-92, American Society for Testing and Materials, Philadelphia.
Baker, L. A. (1993) Introduction to nonpoint source pollution and wetland mitigation. In:Richard K. Olson (editor), Created and natural wetlands for controlling nonpoint sourcepollution/Office of Wetlands, Oceans, and Watersheds. U.S. Environmental ProtectionAgency, 7-41.
Baker, J. A. (1973) Wetland hydrology. In M. W. Lefor and W. C. Kennard (editors), Proceedingsof the first annual wetlands conference, Storrs, Conn, U Conn, 84-95.
Bastian, R. K., and D. A. Hammer (1993) The use of constructed wetlands for wastewatertreatment and recycling. In: Gerald A. Moshiri (editor), Constructed wetlands for waterquality improvement. Boca Raton: Lewis Publishers, 59-67.
Benefield, L., and F. Molz (1984) Mathematical simulation of a biofilm process. Biotechnologyand Bioengineering 27: 921-931.
Benefield, L. D., J. F. Judkins Jr., and A. D. Parr (1984) Treatment plant hydraulics forenvironmental engineers. Prentice-Hall, Inc., Englewood Cliffs, New Jersey.
Blom, C. W., G. M. Bogemann, P. Laan, A. J. M. Candersman, H. M. vandersteeg, and L. A.Vosenek. (1990) Adaptations to flooding in plants from river areas. Aquatic Botany 38: 29-47.
Bowmer, K. H. (1987) Nutrient removal from effluents by and artificial wetland: Influence ofrhizosphere aeration and preferential flow studied using bromide and dye tracers. WaterResources 21(5): 591-599.
Boyd, C. E. (1978) Chemical composition of wetland plants. In: R. E. Good, D. F. Whigham, andR. L. Simpson (editors). Freshwater wetlands: ecological processes and managementpotential. New York: Academic Press, 155-167.
Brannan, K. (1999) Personal communication: Blacksburg, VA: July, 1999
Brezonik, P. L. (1972) Nitrogen: sources and transformations in natural waters. In: H. E. Allenand J. R. Kramer (editors), Nutrients in natural waters. Wiley –Interface, New York, 1-50.
159
Burns, R. C. and R. W. F. Hardy (1975) Nitrogen fixation in bacteria and higher plants. Springer-Velag, New York, New York.
Chapra, S. C. (1997) Surface water-quality modeling. McGraw-Hill, New York, New York.
Choate, K. D., J. T. Watson, and G. R. Steiner (1990) Demonstration of constructed wetlands fortreatment of municipal wastewaters, monitoring report for the period of March 1998 toOctober 1989. TVA/WR/WQ—90/11, Tennessee Valley Authority.
Christensen, N., W. J. Mitsch, and S. E. Jorgensen (1994) A first generation ecosystem model ofthe Des Plains river experimental wetlands. Ecological Engineering 3: 495-521.
Cowardin, L.M., V. Carter, F. C. Golet, and E. T. Laroe (1979) Classification of wetlands anddeepwater habitats in the United States, U.S. Fish and Wildlife Service pub. FWS/OBS-79/31, Washington, DC.
Costanza, R., and F. H. Sklar (1985) Articulation, accuracy and effectiveness of mathematicalmodels: A review of freshwater wetland applications. Ecological Modelling 27: 45-68.
Cussler, E. L. (1997) Diffusion: mass transfer in fluid systems. Cambridge University Press,United Kingdom.
Daukas, P., D. Lowry, and W. W. Walker Jr. (1989) Design of wet detention basins andconstructed wetland for treatment of stormwater runoff from a regional shopping mall inMassachusetts. In: Donald A. Hammer (editor), Constructed wetlands for wastewatertreatment: municipal, industrial, and agricultural, Lewis Publishers, Inc., Chelsea, Michigan,686-694.
Dennison, M. S., and J. F. Berry (1993) Wetlands: guide to science, law, and technology. NoyesPublications. Park Ridge, New Jersey.
Department of Conservation and Recreation (1996) NPS: nonpoint source pollution and you(brochure). Virginia Department of Conservation and Recreation, Richmond, VA.
Detenback, N. E., and P. L. Brezonik (1991) Phosphorous sorption by sediments from a soft-water seepage lake. 1. An evaluation of kinetic and equilibrium models. EnvironmentalScience and Technology, 25: 395-403.
Dorge, J. (1994) Modeling nitrogen transformations in freshwater wetlands. Estimating nitrogenretention and removal in natural wetlands in relation to their hydrology and nutrient loadings.Ecological Modeling, 75/76:409-420.
Dorge, J. (1998) Personal communication: Blacksburg, VA: September, 1998.
160
Dierberg, F. E., and P. L. Brezonik (1984) Nitrogen and phosphorous mass balances in a cypressdome receiving wastewater. In: K. C. Ewel and H. T. Odum (editors), Cypress swamps.Gainesville, University of Florida Press, 112-118.
Dillaha, T. A. (1990) Role of best management practices in restoring the health of theChesapeake Bay. In: Perspectives on the Chesapeake Bay, 1990: Advances in EstuarineSciences. Chesapeake Bay Program, USEPA, Washington, DC CBP/TRS41/90.
Duever, M. J. (1988) Hydrologic processes for models of freshwater wetlands. In : W.J. Mitsch,M. Straskraba and S. E. Jorgensen (editors), Wetland Modelling, Elsevier, Amsterdam, 9-39.
Environmental Law Institute, Wetlands deskbook. (Washington, D.C.: author) pg. 18.
EPA (1988) Constructed wetlands and aquatic plant systems for municipal wastewater treatment:Design manual EPA/625/1-88/022. Washington, DC: Office of research and development.
Fennessy, M. S., C. C. Brueske, and W. J. Mitsch (1994) Sediment deposition patterns in restoredfreshwater wetlands using sediment traps. Ecological Engineering 3:409-428.
Fetter, C. W. (1994). Applied hydrogeology. Prentice Hall, Upper Saddle River, New Jersey.
Fields, S. (1993) Regulations and policies relating to the use of wetlands for nonpoint sourcecontrol. In: Richard K. Olson (editor), Created and natural wetlands for controlling nonpointsource pollution/Office of Wetlands, Oceans, and Watersheds. U.S. Environmental ProtectionAgency, 151-158.
French, R. H. (1984) Lake modeling: state of the art. Critical Reviews in EnvironmentalChemistry. 13(4): 311-357.
Fyock, O. L. (1977) Nitrification requirements of water reuse systems for rainbow trout.Colorado Division of Wildlife, Fisheries Research Section, No. 41; February 1977.
Gambrell, R. P. and W. H. Patrick Jr. (1978) Chemical and microbiological properties ofanaerobic soils and sediments. In: D.D. Hook and R. M. M. Crawford (editors), Plant Life inAerobic Environments, Ann Arbor Sci. Pub. Inc., Ann Arbor Michigan, 375-423.
Gidley, T. M. (1995) Development of a constructed subsurface flow wetland simulation model.MS Thesis, North Carolina State, pp. 139.
Good, R. E., D. F. Whigham, and R. L. Simpson. (1978) Freshwater wetlands: ecologicalprocesses and management potential. Academic Press, New York, New York.
Grady Jr., C. P. L. and H. C. Lim. (1980) Biological wastewater treatment – theory andapplications. New York: Marcel Dekker Inc.
161
Guardo, M., and R. S. Tomasello (1995) Hydrodynamic simulations of a constructed wetland inSouth Florida. Water Resources Bulletin, 31(4): 687-701.
Hales, L. Z., S. L. Bird, B. A. Ebersole, and R. Walton (1990) Bolsa Bay, California, proposedocean entrance system. Tidal circulation and transport computer simulation and waterquality assessment. U.S. Army Engineer Waterways Experiment Station. Vicksburg, MS,USA. Report 3.
Hammer, D. H., and R. H. Kadlec (1986) A model for wetland surface water dynamics. WaterResources Research, 22(13) 1951-1958.
Hart, B. T. (1982) Uptake of trace metals by sediment and suspended particles: a review.Hydrbiologia, 91:299-313.
Hatano, K, D. J. Frederick, and J.A. Moore (1994) Microbial ecology of constructed wetlandsused for treating pulp mill wastewater. Water Science and Technology, 29(4) 233-239.
Heatwole, C. D. (1988) Class Notes from BSE 5354: NPS modeling. Virginia PolytechnicInstitute and State University, BSE Department.
Henze, M., C. P. L. Grady Jr., W. Gujer, G, v, R, Marias, and T. Matsuo (1986) Activated sludgemodel no. 1. In: IAWPRC scientific and technical reports No. 1. IAWPRC task group onmathematical modeling for design and operation of biological wastewater treatment.
Hey D. L., M. A. Cardamone, J. H. Sather, and W. J. Mitsch (1989) In: W. J. Mitsch and S. E.Jorgensen (editors), Ecological engineering, an introduction to ecotechnology, John Wiley &Sons, New York, 159-183.
Hosokawa, Y. and K. Furukawa (1994) Surface flow and particle settling in a coastal reed field.Water Science and Technology, 29(4) 45-53.
Hosokowa, Y, and T Horie (1992) Flow and particulate nutrient removal by wetland withemergent macrophyte. Science and total environment, supplement; 1271-1282.
Jansson, S. L. and J. Persson (1982) Mineralization and immobilization of soil nitrogen. In: F. J.Stevenson (editor). Nitrogen in agricultural soils. Madison, WI: American Society ofAgronomy, Inc., Crop Science Society of America, Inc., and Soil Science Society of AmericaInc., 229-252.
Johnson, C. A., G. D. Bubezner, G. B. Lee, F. W. Madison, and K. R. McHenry (1984) Nutrienttrapping by sediment deposition in a seasonally flooded lakeside wetland. A Landscapeapproach. Biogeochemistry, 10: 105-141.
Johnston, C. A. (1991) Sediment and nutrient retention by freshwater wetlands: effects onsurface water quality. Critical reviews in environmental control 21(5-6), 491-565.
162
Jorgensen, S. E. (1983) Application of ecological modelling in environmental management, partA. Elsevier Scientific Publishing Company. New York, New York.
Jorgensen, S. E. (1989) Sedimentation. In S.E. Jorgensen and M.J. Gromiec (editors),Mathematical submodels in water quality systems. Elsevier, Amsterdam.
Jorgensen, S. E. (1995) State-of-the-art management models for lakes and reservoirs. Lakes andreservoirs: research and management. 1:79-87.
Jorgensen, S. E., C. C. Hoffman, and W. J. Mitsch (1988) Modeling nutrient retention by areedswamp and wet meadow in Denmark. . In : W.J. Mitsch, M. Straskraba and S. E.Jorgensen (editors), Wetland modelling, Elsevier, Amsterdam, 133-151.
Kadlec, R.H., and D. E. Hammer (1988) Modeling nutrient behavior in wetlands. EcologicalModelling, 40:37-66.
Kadlec, R. H. (1989a) Hydrologic factors in wetland water treatment. In: Donald A. Hammer(editor), Constructed wetlands for wastewater treatment: municipal, industrial, andagricultural, Lewis Publishers, Inc., Chelsea, Michigan, 21-40.
Kadlec, R. H. (1989b) Decomposition in wastewater wetlands. In: Donald A. Hammer (editor),Constructed wetlands for wastewater treatment: municipal, industrial, and agricultural, LewisPublishers, Inc., Chelsea, Michigan, 459-468.
Kadlec, R. H. (1990) Overland flow in wetlands: vegetation resistance. Journal of hydraulicengineering, 116(5):691-706
Kadlec, R.H. and R.L. Knight. (1996). Treatment wetlands. CRC Lewis Publishers. Boca Raton,Florida.
Kadlec, R. H. (1997) An autobiotic wetland phosphorous model. Ecological Engineering 8: 145-172.
Kulin, G., and P. R. Compton (1975) A guide to methods and standards for the measurement ofwater flow: Special publication 421, National Bureau of Standards.
Loague, K. and R. E. Green (1991) Statistical and graphical methods for evaluating solutetransport models: Overview and application. Journal of Contaminant Hydrology 7: 51-73.
McCarty, P.L. (1975) Stoichiometry of biological reactions. Progress in Water Technology 7:157-172.
Mitsch, W.J. (1988) Productivity-hydrology nutrient models of forested wetlands. In: W.J.Mitsch, M. Straskraba and S. E. Jorgensen (editors), Wetland modelling, Elsevier,Amsterdam, 115-132.
163
Mitsch, W. J., and J. G. Gosselink (1986) Wetlands. Van Nostrand Reinhold. New York, NewYork.
Mitsch, W. J., and J. G. Gosselink (1993) Wetlands. Van Nostrand Reinhold. New York, NewYork.
Mitsch, W. J., J. K. Cronk, X. Wu, R. W. Nairn, and D. L. Hey (1995) Phosphorous retention inconstructed freshwater riparian marshes. Ecological Applications, 5(3):830-845.
Mitsch, W.J., and B. C. Reeder. (1991) Modelling nutrient retention of a freshwater coastalwetland: estimating the roles of primary productivity, sedimentation, resuspension, andhydrology. Ecological modelling 54:151-187.
Mostaghimi, S., S. Shukla, and P. W. McClellan (1998) BMP Impacts on nitrate and pesticidetransport to groundwater in the Nomini creek watershed: Report No. NC-0298.
Niswander S. F., and W. J. Mitsch (1995) Functional analysis of a two-year old created in-streamwetland: Hydrology, phosphorus retention, and vegetation survival and growth. Wetlands,15(3):212-225.
Novotny, V. and G. Chesters (1981) Handbook of nonpoint source pollution sources andmanagement. Van Nostrand Reinhold Company. New York, New York.
Novotny, V., and H. Olem (1994) Water quality: prevention, identification and management ofdiffuse pollution. Van Nostrand Reinhold. New York, New York.
Ott, R. L. (1993) An introduction to statistical methods and data analysis. Duxbury Press.Belmont, CA.
Perrich, J. R. (1981) Activated carbon adsorption for wastewater treatment. CRC Press, Inc.Boca Raton, FL.
Raisin, G. W. and D. S. Mitchell (1995) The use of wetlands for the control of non-point sourcepollution. Water, Source and Technology, 32(3): 177-186.
Reddy, K. R., Khaleel, R., Overcash, M. R., and P. W. Westerman (1979) A nonpoint sourcemodel for land areas receiving animal wastes. I: Mineralization of organic nitrogen.Transactions of the ASAE 22: 863.
Reddy, K. R. (1982) Mineralization of nitrogen in organic soils. Soil Sci. Soc. Am. J., 46(3):561-566.
Reddy, K. R. and W. H. Patrick (1984) Nitrogen transformations and loss in flooded soils andsediments. CRC Critical Reviews in Environmental Control, 13(4): 273-309.
164
Reed, S. C. (1994) Design of subsurface flow constructed wetland for wastewater treatment. In:S. C. Reed, E. J. Middlebrooks, and R. W. Crites (editors). Natural systems for wastemanagement and treatment, 2nd Edition. New York, McGraw and Hill.
Ruffner, J. A. (1980) Climates of the states, Volume 2. Gale Research, Detroit, Michigan.
Salvesen, D. (1994) Wetlands: mitigating and regulating development impacts. Second edition.Washington D.C.: ULI – the Urban Land Institute.
Shaw, S. P., and C. G. Fredine (1956) Wetlands of the United States, their extent, and their valuefor waterfowl and other wildlife. U.S. Department of Interior, Fish and Wildlife Service,Circular 39, Washington D.C.
Silverman, G. S. (1989) Treatment of nonpoint source pollutants – urban runoff and agriculturalwastes: Development of urban runoff treatment wetlands in Fremont, California. In: DonaldA. Hammer (editor), Constructed wetlands for wastewater treatment: municipal, industrial,and agricultural, Lewis Publishers, Inc., Chelsea, Michigan, 669-676.
Soil Conservation Service (1968) Hydrology, supplement A to sec.4, Engineering Handbook,USDA-SCS, Washington D.C.
Snoeyink, V. L., and D. Jenkins (1980) Water chemistry. John Wiley & Sons, New York, NewYork.
Tanji K. K. (1982) Modeling of the soil nitrogen cycle. In: F. J. Stevenson (editor). Nitrogen inagricultural soils. Madison, WI: American Society of Agronomy, Inc., Crop Science Societyof America, Inc., and Soil Science Society of America Inc., 721-772.
Tanner, C. C. (1996) Plants for constructed wetland treatment systems – A comparison of thegrowth and nutrient uptake of eight emergent species., Ecological Engineering (7) 59-83.
Tchobangolous, G. and F. L. Burton (1991) Wastewater engineering: treatment, disposal andreuse. Third edition. New York: Metcalf and Eddy, Inc. McGraw-Hill, Inc.
Teague L. L., A. L. Kenimer, B. J. Lesker, and F. L. Mitchell (1997) Constructed wetlands for thecontrol of agricultural nonpoint source pollution. ASAE Paper No. 972037
Tiner Jr., R. W. (1984) Wetlands of the United States: current status and recent trends, NationalWetlands Inventory Project. U.S. Fish and Wildlife Service.
USDA, Soil Conservation Service (1988) National handbook of conservation practices.Washington, D. C.
USEPA (1993) Created and natural wetlands for controlling nonpoint source pollution. BocaRaton: Smoley C.K.
165
USSCS (1975) Soil taxonomy: A basic system of soil classification for making and interpretingsoil surveys. U. S. Soil Conservation Service Agricultural Handbook 436. Washington, D.C.
Urban, N. R., and S. J. Eisenreich (1988) Nitrogen cycling in a forested Minnesota bog.Canadian Journal of Botany, 66: 435-449.
Walton, R., R. S. Chapman, and J. E. Davis (1996) Development and application of the wetlandsdynamic water budget model. Wetlands, 16(3): 347-357.
Watson, J. T., Sherwood, C. R., Kadlec, R. H., Knight, R. L., and A. E. Whitehouse (1989)Performance expectation and loading rates for constructed wetlands. In: Donald A. Hammer(editor), Constructed wetlands for wastewater treatment: municipal, industrial, andagricultural, Lewis Publishers, Inc., Chelsea, Michigan, 319-351.
Wheaton, F., J. Hochheimer, and G.E. Kaiser (1991) Fixed film nitrification filters foraquaculture. In: Brune, D.E. and J. R. Tomasso (editors), Aquaculture and Water Quality.Baton Rouge, LA: The World Aquaculture Society, 272-303.
Whitaker, G., and C. R. Terrell (1993) Federal programs for wetland restoration and use ofwetlands for nonpoint source pollution control. In: Richard K. Olson (editor), Created andnatural wetlands for controlling nonpoint source pollution/Office of Wetlands, Oceans, andWatersheds. U.S. Environmental Protection Agency, 201-215.
Widener, A. S. (1995) A mathematical model of the nitrogen cycle in a constructed wetland. MSThesis, Virginia Polytechnic Institute and State University. pp. 145.
Wood. R. A. (1996) The weather almanac. Gale research, Detroit, MI.
Zak, D. R. and D. F. Grigal (1991) Nitrogen mineralization, nitrification, and denitrification in
upland and wetland ecosystems. Oegologia, 88: 189-196.
167
Appendix A: Model Parameters
List of all model parameters for SET-WET. In the following table, inputs have their first lettercapitalized, stocks are entirely capitalized, while any flows or parameters which are determinedby SET-WET are entirely lower case.
MODELCOMPONENT DESCRIPTION UNITS
Abioremd Amount of standing dead biological mass removal g biomass/dayAbioreml Amount of living biological mass removal g biomass/day
Adsorp Type of adsorption model used -
AemaxgrbAerobic heterotroph maximum growth rate at optimal conditions in peat water day-1
AemaxgrwAerobic heterotroph maximum growth rate at optimal conditions in surface water day-1
aerohtgb growth of bottom aerobic heterotrophs g microbes/dayaerohtgw growth of surface aerobic heterotrophs g microbes/dayAirtemp Ambient Air Temperature degrees CAlpha Kadlec equation (slope exponent) -
ammupb plant ammonium uptake from wetland bottom g N/ dayammupw plant ammonium uptake from wetland surface g N/ day
anfracbfraction of heterotrophs that are anaerobes - function of DO conc. in bottom
g anaerobic HT/ g total HT
anfracwfraction of heterotrophs that are anaerobes - function of DO conc. in surface
g anaerobic HT/ g total HT
Angvnot Angle of weir notch (outlet=2) degrees anhtgb growth of bottom anaerobic heterotrophs g microbes/dayanhtgw growth of surface anaerobic heterotrophs g microbes/day
AnmaxgrbAnaerobic heterotroph maximum growth rate at optimal conditions in peat water day-1
AnmaxgrwAnaerobic heterotroph maximum growth rate at optimal conditions in surface water day-1
Apflow amount of point flow contribution (point=1) m3/day
Areapipe Area of outflow pump (outlet=3) m2
Atdep Atmospheric Deposition (0=not include, 1=include)Beta Kadlec equation (depth exponent)
Bioccont Fraction of carbon in plants g C/ g biomass
biodegrstanding dead degradation rate - set for 99% over the course of a year - converted to POC day-1
Biodens Living biomass density kg/m3
biofluxb oxygen flux from rootzone aeration by plants g O2/day
Bioinit Initial total plant mass g Plant BIOMASS total plant carbon mass g C
BIOMASST total biomass in the wetland system g biomassbiomassv volume of biomass g biomassBiomcn Ratio of carbon to nitrogen in plants g C/g Nbiomdth biomass death g biomass/day
biomgrow biomass growth g biomass/day
Biomgrr Rate at which plants grow
g biomass/
m2-daybiomout amount of biomass removal for external reasons g biomass/day
168
MODELCOMPONENT DESCRIPTION UNITS
Biompn Ratio of plant mass to nitrogen in plants g plant mass/g N
Biompp Biomass plant/phosphorous ratiog plant/g
phosphorous
Biooxrbrate at which oxygen is added to the rootzone by plants gO2 /m2*day
BioremdRemoval of dead biomass biological material (0=no removal for stress period, 1=removal) -
BioremlRemoval of live biomass biological material (0=no removal for stress period, 1=removal) -
BocconcInfluent BOD5 concentration in point source additions mg BOD5/L
Bocdinfco Influent BOD5 concentration in water additions mg BOD5/LBodcfrac Fraction of carbon in influent BOD g C/g BOD5
Bodpfrac Fraction of influent BOD which is in particulate formg. part. BOD5/ g
total BOD5Bodrc BOD runoff coefficient mg BOD5/L
btphos amount of bottom particulate phosphorous attached to each respective sediment category g PP
BtphosiInitial amount of total particulate phosphorous in wetland bottom water pool g P-phos
BTPHOST total particulate phosphorous in wetland bottom g PP
check (1-7)Flag to determine entry of data between changing stress periods -
Coefk Kadlec equation (premultiplier constant) -Contc Pipe contraction coefficient (outlet=3) -
Daddon Dry atmospheric deposition rate for DON g DON/m2
Dadnh4 Dry atmospheric deposition rate for NH4 g NH4/m2
Dadno3 Dry atmospheric deposition rate for NO3 g NO3/m2
Dadpon Dry atmospheric deposition rate for PON g PON/m2
Daylen fraction of 24 hour day between sunrise and sunset hours daylight/12
Dayremd Removal date of material day of season
period
Dayreml Removal date of material day of season
periodDaysdeg Time frame over which plant biomass degrades days
Daywins Day when winter starts in time periodday of season
period
Daywine Day when winter ends in season periodday of season
period
deadphosamount of PP entering system from plant physical degradation per category g PP/day
deadphottotal amount of PP entering system from plant physical degradation g PP/day
deathbtransfer of immobilized nitrogen to bottom PON due to death of bottom microbes or plants g N/day
deathwtransfer of immobilized nitrogen to surface PON due to death of surface microbes or plants g N /day
Decompdecomposition rate of organic material in wetland bottom
g deg. SS/ g SS
Degbio Point to which biomass will degrade % degradation
169
MODELCOMPONENT DESCRIPTION UNITS
deltahtchange in wetland bottom height due to peat accumulation, sediment resuspension/settling m
denitbconversion of bottom nitate to N2 gas by bottom anaerobic heterotrophs g N/ day
denitwconversion of surface nitate to N2 gas by surface anaerobic heterotrophs g N/ day
Diseffc Effective discharge coefficient mdisphosc DP watershed runoff inflow concentration mg DP/L
DisphoscDissolved phosphorous inflow concentrations from catchment mg/l
disphosi total amount of DP entering wetland system g DP/dayDOCB amount of dissolved organic carbon in bottom g Cdoccb dissolved organic carbon concentration in bottom mg DOC/Ldoccw dissolved organic carbon concentration in surface mg DOC/L
Docininc DOC concentration for percolating sources mg C/L
DocinitbInitial amount of dissolved organic carbon in peat water g C
DocinitwInitial amount of dissolved organic carbon in surface water g C
docleach leaching of DOC from standing dead to DOC g C/ day
docminibconversion of bottom DOC to microbial biomass and carbon dioxide g C/day
docminiwconversion of surface DOC to microbial biomass and carbon dioxide g C/day
docmt DOC mass transfer g C/dayDoconcin Dissolved oxygen concentration in water input mg O2/LDoconcp Oxygen concentration in precipitation mg O2/L
docout effluent DOC flux g C/daydocperc DOC percolation/infiltration to/from wetland bottom g C/dayDOCW amount of dissolved organic carbon in surface g Cdoinf dissolved oxygen additions from watershed runoff g O2/day
Doinitb Total mass of oxygen in the peat water g O2Doinitw Total mass of oxygen in the surface water g O2
donammbconversion of excess DON to ammonium by bottom heterotrophic bacteria during organics degradation g N/ day
donammwconversion of excess DON to ammonium by surface heterotrophic bacteria during organics degradation g N/ day
donat DON contributed by atmospheric deposition g DON/ dayDONB amount of dissolved organic nitrogen in bottom g DONdoncb dissolved organic nitrogen concentration in bottom mg DON/L
Donconc DON concentration in point flow mg DON/Ldoncw dissolved organic nitrogen concentration in surface mg DON/Ldonec DON effleunt concentration g N/l
donimmbincorporation of DON by bottom heterotrophic microbes during organics decomposition g N /day
donimmwincorporation of DON by surface heterotrophic microbes during organics decomposition g N /day
donin incoming DON from watershed runoff additions g N/dayDonininc DON concentration from percolating source mg DON/L
170
MODELCOMPONENT DESCRIPTION UNITS
Doninitb Initial dissolved organic nitrogen in peat water g NDoninitw Initial dissolved organic nitrogen in surface water g N
donminb
conversion of excess bottom DON to ammonium by bottom heterotrophic bacteria during organics degradation g N/day
donminw
conversion of excess surface DON to ammonium by surface heterotrophic bacteria during organics degradation g N/day
donmt mass transfer of DON between surface and bottom g DON/daydonout effluent DON flux g N/day
donperc DON percolation/infiltration from bottom g DON/dayDONW amount of dissolved organic nitrogen in surface g DON/daydoout effluent disolved oxygen flux g O2/day
doperc percolation/infiltration of DO from wetland bottom g O2/day
DOXYB amount of dissolved oxygen in wetland bottom g O2Doxycsat DO saturation constant mg O2/L
DOXYW amount of dissolved oxygen in wetland surface g O2
dphosec Effluent dissolved phosphorous concentration mg DP/Ldphosout effluent dissolved phosphorous flux g DP/dayDTPHOSB total dissolved phosphorous in wetland bottom g DPdtphoscb Dissolved phosphorous concentration for bottom mg DP/Ldtphoscw Dissolved phosphorous concentration for surface mg DP/L
DtphosibInitial amount of dissolved phosphorous in wetland bottom water pool g DP
DtphosiwInitial amount of dissolved phosphorous in wetland surface water pool g DP
DTPHOSW total dissolved phosphorous in wetland surface g DPEvap Evaporation model (0=Thornthwaite, 1=pan) -
evapcoefcalculated climate parameter for Thornthwaite's ET method -
Evapt Amount of evaporation from wetland cell m3/day
evapt water loss due to evapotranspiration m3
evapttamount of water loss due to evapotranspiration for hourly time step m3/hour
fixnitamount of atmospheric nitrogen converted to DON by microbes g ON/day
Flowout Amount of water removed (outlet=4) m3/day
flowratecalculated parameter for SCS curve method - amount of surface runoff mm
Freundk Freundlich isotherm constant -Freundn Freundlich isotherm constant -
Hb Bottom of peat component in wetland cell m
heatijcalculated climate parameter for Thornthwaite's ET method -
heatindxcalculated climate parameter for Thornthwaite's ET method -
HETEROB amount of heterotrophs in bottom g microbes
171
MODELCOMPONENT DESCRIPTION UNITS
HeteroibInitial total mass of heterotrophic bacteria in peat water g microbes
HeteroiwInitial total mass of heterotrophic bacteria in surface water g microbes
HETEROW amount of heterotrophs in surface g microbes Hii Initial height of water in wetland cell relative to (z=0) mhit wetland surface water height for hourly time step m
hnh4immb
bottom ammonium nitrogen utilized by bottom heterotrophic bacteria during organics degradation g NH4+-N/day
hnh4immw
surface ammonium nitrogen utilized by surface heterotrophic bacteria during organics degradation g NH4+-N/day
Hno3hscbAnaerobic heterotroph nitrate half saturation constant for peat water mg NO3--/L
Hno3hscwAnaerobic heterotroph nitrate half saturation constant for surface water mg NO3--/L
Ho Top of wetland cell m
HorghscbHeterotroph organics half saturation constant for peatwater mg C/L
HorghscwHeterotroph organics half saturation constant for surface water mg C/L
HoutHeight of overflow for outlet option (outlet=1, 2 , 3 ,4, or 6) m
Hover Height for water overflow of weir (outlet=1) mhrt hydraulic residence time days
htdeathb death of bottom heterotrophs g microbes/dayhtdeathw death of surface heterotrophs g microbes/day
HtdohscbAerobic heterotroph dissolved oxygen half saturation constant in surface water mg O2/L
HtdohscwAerobic heterotroph dissolved oxygen half saturation constant in peat water mg O2/L
Htdoyb Oxygen yield of aerobic heterotrophs in peat water g microbes/g O2
HtdoywOxygen yield of aerobic heterotrophs in surface water g microbes/g O2
Htdrb Heterotroph death rate in peat water g microbes/dayHtdrw Heterotroph death rate in surface water g microbes/day
htgrowb growth of bottom heterotrophs g microbes/dayhtgroww growth of surface heterotrophs g microbes/day
Hti Top of peat component in wetland cell m
Htno3y Nitrate yield of anaerobic heterotrophsg microbes/
g NO3--N
htrespb bottom oxygen loss by bottom heterotroph growth g O2/day
htrespw surface oxygen loss by surface heterotroph growth g O2/day
httempfbHeterotroph bottom temperature factor - reduces growth rate at nonoptimal conditions -
httempfwHeterotroph surface temperature factor - reduces growth rate at nonoptimal conditions -
htyieldb yield of bottom heterotrophic microbesg microbes/ g C
degraded
htyieldw yield of surface heterotrophic microbesg microbes/ g C
degraded
172
MODELCOMPONENT DESCRIPTION UNITS
Hydconp Hydraulic conductivity of wetland gravel bed m/day
HydtypeHydrologic input (0 = NPS pollution runoff model, 1=SCS runoff curve method) -
Imminitb Nitrogen in plants and microbes in peat water g NImminitw Nitrogen in plants and microbes in surface water g N
IMMNB nitrogen in bottom plants and microbes g N
immnout Removal of immoblized nitrogen in living biomass due to external removals g N/day
immnoutdRemoval of immoblized nitrogen in standing dead due to external removals g N/day
IMMNW nitrogen in surface plants and microbes g Nj equivalent to TIMPER # time periodk equivalent to number of time periods # of time periods
Khcoef head correction factor (outlet=2) mLeachr Rate at which biomass leaches DOC g C/g NLength Average length of wetland cell m
Linpartc Linear isotherm partition coefficient l/gm equivalent to STRPER # stress period
Mannc Manning's coefficient s/m1/3
Mannc Manning's coefficient s/m1/3
Microbec Fraction of cell mass that is carbon g C/g microbesmicrodb sloughing of dead bottom microbial cells to POC g C/daymicrodw sloughing of dead surface microbial cells to POC g C/dayMicronc Fraction of nitrogen in microbial cells g N/g microbesmictcnb ratio of carbon to nitrogen in bottom microbial cells g C/g Nmictcnw ratio of carbon to nitrogen in surface microbial cells g C/g NMtcdoc Mass transfer coefficient for DOC cm/sMtcdon Mass Transfer coefficient for DON cm/s
Mtchphos Mass transfer coefficient for DP cm/sMtcnh4 Mass transfer coefficient for NH4 cm/s
Mtcno3 Mass transfer coefficient for NO3 cm/s
Mtdox Mass transfer coefficient for oxygen cm/smtdoxy mass transfer of DO between surface and bottom g O2/day
mtfws
Re-aeration of DO between water surface and atmosphere g O2/day
Mtfwsdoc Re-aeration mass transfer coefficient cm/sndeathb death of Nitrosomonas in bottom g microbes/dayndeathw death of Nitrosomonas in surface g microbes/day
NdohsatbNitrosomonas dissolved oxygen half saturation constant in peat water mg O2/L
NdohsatwNitrosomonas dissolved oxygen half saturation constant in surface water mg O2/L
Ndrateb Nitrosomonas death rate in peat water g microbes/dayNdratew Nitrosomonas death rate in surface water g microbes/day
nh4at NH4 additions due to atmospheric deposition g N/dayNH4B amount of ammonium in wetland bottom g N
nh4cb ammonium concentration in wetland bottom mg NH4+-/L
Nh4conc NH4 concentration in point flow mg NH4+-/L
173
MODELCOMPONENT DESCRIPTION UNITS
nh4cw ammonium concentration in wetland surface mg NH4+-/L
nh4ec NH4 effluent concentration mg NH4+-/L
nh4in ammonium flux from watershed runoff additions -Nh4inc NH4 influent concentration mg NH4+-/L
Nh4ininc NH4 concentration from percolating source mg NH4+-/L
Nh4initbInitial ammonia and ammonium nitrogen in peat water g NH4
Nh4initwInitial ammonia and ammonium nitrogen in surface water g NH4
nh4mtmass transfer og ammonium between surface and bottom g N/ day
nh4out effluent NH4 flux g N/daynh4perc NH4 percolation/infilitration from wetland bottom g N/day
Nh4rc NH4 runoff coefficient mg NH4+-/L
NH4W amount of ammonium in wetland surface g NNitcycle Nitrogen Cycle (0=not included, 1=included) -
Nitfix Nitrogen fixation (0=not included, 1=included) -
Nitfrate Nitrogen fixation rate g DON/ (day*m2)
nitrifbconversion of ammonium to nitrate by bottom autotrophs g N /day
nitrifwconversion of ammonium to nitrate by surface autrotrophs g N /day
NitrosibInitial total mass of Nitrosomonas sp. Autotrophic bacteria in peat water g microbes
NitrosiwInitital total mass of Nitrosomonas sp. Autotrophic bacteria in surface water g microbes
NITROSOB amount of Nitrosomonas in bottom g microbes NITROSOW amount of Nitrosomonas in surface g microbes
nitupb plant nitrate uptake from wetland bottom g N/ daynitupw plant nitrate uptake from wetland surface g N/ daynleach nitrogen leached from standing dead to DON g N/day
NmaxgrbNitrosomonas maximum growth rate at optimal conditions in peat water day-1
NmaxgrwNitrosomonas maximum growth rate at optimal conditions in surface water day-1
Nnh4hscbNitrosomonas ammonium half saturation constant for peat water mg NH4+-/L
Nnh4hscwNitrosomonas ammonium half saturation constant for surface water mg NH4+-/L
no3at NO3 additions due to atmospheric deposition g N/dayNO3B amounf of nitrate in wetland bottom g N
no3cb nitrate concentration in wetland bottom mg NO3--/L
No3conc NO3 concentration in point flow mg NO3--/L
no3cw nitrate concentration in wetland surface mg NO3--/L
no3ec effluent nitrate concentration mg/lno3in influent nitrate flux from watershed runoff g N/day
No3inc NO3 influent concentration mg NO3--/L
No3ininc NO3 concentration from percolating source mg NO3--/LNo3initb Initial oxidized nitrogen in peat water g NO3
174
MODELCOMPONENT DESCRIPTION UNITS
No3initw Initial oxidized nitrogen in surface water g NO3
no3mt mass transfer of NO3 between surface and water g N/day
no3out effluent nitrate flux mg NO3--/L
no3perc NO3 percolation/infiltration from wetland bottom g N/dayNo3rc NO3 runoff coefficient mg NO3--/L
NO3W amount of nitrate in wetland surface g N
NsdoybOxygen yield of Nitosomonas bacteria in surface water g microbes/g O2
NsdoywOxygen yield of Nitosomonas bacteria in surface water g microbes/g O2
nsgrowb growth of Nitrosomona in bottom g microbes/daynsgroww growth of Nitrosomonas in surface g microbes/daynsrespb bottom oxygen loss by bottom autotroph growth g O2/day
nsrespw surface oxygen loss by surface autotroph growth g O2/day
nstempfbNitrosomonas bottom temperature factor - reduces growth rate at nonoptimal conditions -
nstempfwNitrosomonas surface temperature factor - reduces growth rate at nonoptimal conditions -
Nsyieldb Yield of Nitrosomas bacteria in peat waterg microbes/ g
NH4+-N
Nsyieldw Yield of Nitrosomas bacteria in surface waterg microbes/ g
NH4+-N
Numstper Number of stress periods in simulation run # of periodsNumtmper Number of time periods for respective stress period # time steps
Onpartf Fraction of organic nitrogen in particulate form g PON/g TONOrgininc Organic nitrogen influent concentration mg ON/LOrgnrc Organic nitrogen runoff coefficient mg ON/L
Outflow effluent flowrate m3/day
outflow effluent water flux m3/day
outflowt effleunt water flux for hourly time step m3/hourOutlet Outlet type (6 options) -
Outwidth Width of outflow weir (outlet=1) mOxyconc Oxygen concentration from point flow mg O2/L
Oxyrc Runoff coefficient for oxygen mg O2/L
Panevap Pan Evaporation Rate m/day
Pbiouw Percentage of living biomass which is underwater% mass
underwater pbodin particulate BOD flux from watershed runoff additions g C/dayPcycle Phosphorus cycle (0=not included, 1=included) -
PeatacrbRate at which refractory solids accumulate in the peat water g solids/day
PeatacrwRate at which refractory solids accumulate in the surface water g solids/day
peatcacb rate at which bottom refractory solids accumulate g C/daypeatcacw rate at which surface refractory solids accumulate g C/day
Peatcc Fraction of carbon in refractory solids g C/g solids
Peatdens Density of peat material kg/m3
peatnacb nitrogen in accumulating bottom refractory solids g N /day
175
MODELCOMPONENT DESCRIPTION UNITS
peatnacw nitrogen in accumulating surface refractory solids g N /dayPeatnc Fraction of nitrogen in accumulated solids g N/g solids
Percinf Percolation or Infiltration additions (0=no,1=yes)Percinfa Percolation/Infiltration rate for stress period m/day
percinftpercolation/infiltration gains/losses to/from wetland bottom m3
percittamount of percolation/infiltration changes for hourly time step m3/hour
Ph pH of water in wetland system pHphdep pH factor for volatilization -
PhosconIncoming amount of dissolved phosphorous from point source additions g DP/day
phosmt mass transfer of DP from surface and bottom g DP/day
phosperattachment ratio for incoming phosphorous to respective sediment categories -
phosperbratio of phosphorous attachment upon bottom sediment -
phosperbattachment ratio for bottom phosphorous to respective sediment categories -
phosperwattachment ratio for surface phosphorous to respective sediment categories -
phosupb plant uptake of DP from bottom g DP/dayphosupw plant uptake of DP from surface g DP/dayphysdeg degradation of standing dead to refractory material g biomass/dayphysdegc conversion of standing dead to plants to POC g C/day
PinincDissolved phosphorous concentration for percolating water source mg DP/L
pminpp phosphorous remineralization from surface particulate pool for individual sediment category g DP/day
Pminppc Mineralization rate from surface particulate pool day-1
pminppttotal amount of remineralization from surface particulate pool g PP/day
POCB amount of POC in bottom g C/daypoccb particulate organic carbon concentration in bottom mg POC/Lpoccw particulate organic carbonconcentration in surface mg POC/LPocfall POC Fallling rate m/day
PocinitbInitial amount of particulate organic carbon in surface water g C
PocinitwInitial amount of particulate organic carbon in peat water g C
pocminibconversion of bottom POC to microbial biomass and carbon dioxide g C/day
pocminiwconversion of surface POC to microbial biomass and carbon dioxide g C/day
pocout effluent POC flux g C/daypocre POC resuspension form wetland bottom g C/day
Pocres POC resuspension rate ratio
% (g POC resuspended/g POC in peat)
pocset settling of surface POC to wetland bottom g C/dayPocsize Average size of particulate organic carbon mm
176
MODELCOMPONENT DESCRIPTION UNITS
POCW amount of POC in surface g C/dayPoint Point flow addition (0=no, 1=yes) -
ponammb
conversion of excess bottom PON to ammonium by bottom heterotrophic bacteria during organics degradation g N/day
ponammw
conversion of excess surface PON to ammonium by surface heterotrophic bacteria during organics degradation g N/day
ponat PON additions due to atmospheric deposition g N/dayPONB amount of PON in wetland bottom g N/dayponcb particulate organic nitrogen concentration in bottom mg PON/L
Ponconc PON concentration in point flow mg PON/Lponcw particulate organic nitrogen concentration in surface mg PON/Lponec effluent PON concentration mg PON/LPonfall PON fall rate m/d
ponimmbincorporation of bottom PON by bottom heterotrophic bacteria during organics degradation g N/day
ponimmwincorporation of surface PON by surface heterotrophic bacteria during organics degradation g N/day
ponin particulate organic nitrogen flux from watershed g N/dayPoninitb Initial particulate organic nitrogen in peat water g NPoninitw Initial particulate organic nitrogen in surface water g N
ponminb
conversion of excess bottom PON to ammonium by bottom heterotrophic bacteria during organics degradation g N/day
ponminw
conversion of excess surface PON to ammonium by surface heterotrophic bacteria during organics degradation g N/day
ponout effluent PON flux g N/dayponre resuspension of PON from wetland bottom g N/day
Ponres PON resuspension rate ratio
g PON resuspended/ g
PON in peatponset settling of PON from wetland surface g N/day
Ponsize PON diameter mmPONW amount of PON in wetland surface g N/day
Porosity ratio of pore volume to total gravel bed volume m3/m3
Porpeat ratio of pore volume to total peat soil volume m3/m3
PPHOSamount of surface particulate phosphorous attached to each respective sediment category g PP
Pphoscal Amount of incoming P-phos to wetland g P-Phospphoscw PP concentration in wetland surface mg PP/L
pphosinPP inflow from watershed for individual sediment categories g PP/day
pphosinc PP watershed runoff inflow concentration mg PP/Lpphosinp PP additions from point source flow g PP/daypphosint total amount of PP entering wetland system g PP/daypphosout effluent particulate phosphorous flux g PP/daypphosres PP resuspension from wetland bottom g PP/daypphosset PP settling from wetland surface g PP/day
177
MODELCOMPONENT DESCRIPTION UNITS
PPHOST total amount of particulate phosphorous in surface g PP
PpphosiInitial amount of total particulate phosphorous in wetland surface water pool mg PP/L
Pra te upDivis ion of plant nutrient uptake from either the bottom or surface pools
% taken from bottom
precip amount of direc t prec ipitation to wetland m3
Pre cipr precipitation rate m/day
prm inbp phosphorous remineralization from bottom particulate pool for individual sediment category g DP/day
Prm inbpc Remineralization rate from bottom particulate day-1
prm inbpttotal amount of remineralization from bottom particulate pool g PP/day
Pse ddep Ratio of sediment which settles from surface water
sediment depos ition/total
surface sediment
Pstduw Ratio of s tanding dead material that is underwaterRatio mass underwater
re 1 Reynold's number criteria for critical veloc ity determ ination -
re 1bReynold's number criteria for critical veloc ity determ ination -
re 2Reynold's number criteria for critical veloc ity determ ination -
REFC accumulated refrac tory carbon g CRe fcinit Initial accumulated refrac tory carbon g C
REFN nitrogen in refrac tory accumulated solids g NRe fninit Initial nitrogen in rerfactory accumulated solids g N
Re sthcThe thickness to which particulate carbon is resuspended from peat bottom m
Re sthick Thickness of peat affec ted by resuspens ion m
Re sthnThe thickness to which particulate nitrogen is resuspended from peat bottom m
re susp amount of sediment resuspended from wetland bottom for partic le category g sed./day
re suspv resuspens ion critical velocity m/dayre suspv crit ical veloc ity for partic le resuspension m/dSccurve SC runoff curve number (hydtype=1) -
sde a doutamount of s tanding dead removal for ex ternal reasons g biomass/day
sde com pgrams of suspended solids that decompose from organic material g SS/day
sedbsa total surface area of one sediment partic le mm2
sedbv total volume of one sediment partic le mm3
Se dca t The number of sediment categories # categoriesSe dcons Sediment concentration in point flow sources mg sed./LSe dcycle Sediment cyc le (0=not inc luded, 1= inc luded) -sedde lta change in sediment quantity from wetland bottom g sediment/day
seddepsediment deposition of partic le category from wetland surface g sed./day
Se dfa ll Sediment Fall rate m/d
178
MODELCOMPONENT DESCRIPTION UNITS
sedincvincoming total volume of sediment for particle category mm3
Sedinitb Inititial amount of sediment weight in peat water g sedimentSedinitw Initial amount of sediment weight in surface water g sediment
sedint total amount of sediment runoff from watershed g sed./daySedint Total incoming sediment from watershed area g sed./day
sedinwamount of incoming sediment from all sources for particle category g sed./day
sedout effluent sediment flux for particle category g sed./daysedpart amount of sediment particles for each particle class # of particles
Sedper Ratio of sediment by weight from incoming sourcesmass sed. cat./ mass sed. Total
SEDQTYB amount of sediment in bottom for particle category g sed.SEDQTYW amount of sediment in surface for particle category g sed.
Sedrc Sediment runoff coeffcient mg sediment/l
SedresPercentage of sediment that resuspends from possibly resuspended material % sediment
Sedsize Sediment particle size mmSedspg Sediment specific gravity -
sedtotaltotal change in sediment quantity from wetland bottom g sediment/day
sedtsabtotal surface area for sediment particle category in wetland bottom mm2
sedtsaptotal surface area for incoming sediment particle category mm2
sedtsatb total surface area for sediment in wetland bottom mm2
sedtsatp total surface area for all incoming sediment mm2
sedtsatw total surface area for sediment in wetland surface mm2
sedtsawtotal surface area for sediment particle category in wetland surface mm2
So Wetland bottom surface gradient m/msobodin soluble BOD flux from watershed runoff additions g C/ day
STANDDCdead plant carbon biomass that has not become litter g C
STANDDT total standing dead in the wetland system g biomassstanddv volume of standing dead g biomassStandin Initial dead biomass that has not become litter g Plant
Stddens Standing dead material density kg/m3
storparm calculated parameter for SCS curve method -strper counter for the number of season period # season periodtime counter for outflow calculation -
timper counter for the number of time period # time period
TjHistorical average monthly temperature; for Thorthwaite's ET model degrees Celsius
TOCB amount of total organic carbon in bottom g TOCtoccb total organic carbon concentration in bottom mg TOC/Ltoccw total organic carbon concentration in surface mg TOC/LTOCW amount of total organic carbon in surface g TOCTONB Total organic nitrogen in wetland bottom g ONtoncb total organic nitrogen concentration in bottom mg TON/L
179
MODELCOMPONENT DESCRIPTION UNITS
toncw total organic nitrogen concentration in surface mg TON/LTONW Total organic nitrogen in wetland surface g ON
Toppump Height of outflow top (outlet=4) mtphosec Total phosphorous effluent concentration mg P/L
Volat Volatilization (0=not included, 1=included) -volatiz conversion of ammonium ion to ammonium gas g N/day
Volatr Volatilization rate day-1
Waddon Wet atmospheric deposition rate for DON mg DON/LWadnh4 Wet atmospheric deposition rate for NH4 mg NH4+-/L
Wadpon Wet atmospheric deposition rate for PON mg PON/LWasno3 Wet atmospheric deposition rate for NO3 mg NO3--/L
WATERB amount of water in wetland bottom water pool m3
watervel wetland surface water velocity m/day
WATERVOL amount of water in wetland surface water pool m3
Watinput water input to system from watershed m3/day
watint Total water additions to wetland m3/day
watintt amount of water additions for hourly time step m3/hourWattemp Influent water temperature degrees CWettype Wetland type (0=FWS, 1=SSF) -Width Average width of wetland cell m
winkcorate at which living biomass degrades once it starts to die -
winter flag for when winter is in progress -
Wtrshdar Watershed area of adjoining area m2
180
Appendix B: Data entry to Model
A procedural listing of when and how to enter all input parameters to the SET-WETmodel. If the input is mandatory for submodel use, parameter names will be listed with noconstraints. If input is optional and dependent on other values, these constraints will be stated.All input values are capitilized. When there are both season and time period inputs, seasonparameters will be explained first, and then time periods. The number of input values dependson the number of season periods and time periods that are declared. M is the season periodnumber and J is the time period number.
BASE submodel:
WETTYPE, HYDTYPE, POINT, NITCYCLE,SEDCYCLE, PCYCLE, EVAP, PERCINF
If HYDTYPE = 0 then,NUMSTPER, LENGTH, WIDTH, HO, HB, HTI, HII, SO
Else if HYDTYPE = 1 then,NUMSTPER, LENGTH, WIDTH, HO, HB, HTI, HII, SO, WTRSHDAR, SCCURVE
End if
If EVAP = 0 then,TJ1, TJ2, TJ3, TJ4, TJ5, TJ6, TJ7, TJ8, TJ9, TJ10, TJ11, TJ12
HYDROLOGIC submodel:-HYDSTR (hydrologic season period)
If M = 1 and WETTYPE = 0 thenNUMTMPER, PORPEAT
Else if M = 1 and WETTYPE = 1 thenNUMTMPER, POROSITY, HYDCONP
End if
If M = 1 then
If PERCINF = 1 thenPERCINFA
End if
OUTLETIf OUTLET = 1 then
HOUT, OUTWIDTH, HOVERElse if OUTLET = 2 then
181
ANGVNOT, DISEFFC, KHCOEF, HOUT,HOVERElse if OUTLET = 3 then
HOUT, AREAPIPE, CONTCElse if OUTLET = 4 then
FLOWOUT, TOPPUMPElse if OUTLET = 5 then
ALPHA, BETA, COEFK, HOUTElse if OUTLET = 6 then
HOUTEnd if
End if
If M>1 thenNUMTMPER, CHECK1, CHECK2
If CHECK1 = 0 thenPORPEAT, HYDCONP, POROSITY
End if
If CHECK2 = 0 thenIf OUTLET = 1 then
HOUT, OUTWIDTH, HOVER, NCONTElse if OUTLET = 2 then
ANGVNOT, DISEFFC, KHCOEF, HOUTElse if OUTLET = 3 then
HOUT, AREAPIPE, CONTCElse if OUTLET = 4 then
FLOWOUT, TOPPUMPElse if OUTLET = 5 then
ALPHA, BETA, COEFK, HOUTElse if OUTLET = 6 then
HOUTEnd if
End ifEnd if
-HYDTM (hydrologic time period)
If HYDTYPE = 0 thenIf EVAP = 0 then
WATINPUT (J), PRECIPR, AIRTEMP, DAYLENElse if EVAP = 1 then
WATINPUT (J), PRECIPR, PANEVAPEnd if
If POINT = 1 then
182
APFLOW (J)End if
Else if HYDTYPE =1 then
If EVAP = 0 thenPRECIPR, AIRTEMP, DAYLEN
Else if EVAP = 1 thenPRECIPR, PANEVAP
End ifEnd if
If POINT = 1 thenAPFLOW (J)
End if
BIOMASS submodel:
-VEGST (biomass season period)
-This is only called if NITCYCLE = 1 or SEDCYCLE =1
If M = 1 and WETTYPE = 0 thenBIOINIT, STANDIN, PEATACRW, PEATACRB, PEATDENS,PRATEUP,BIODENS, STDDENS, PBIOUW, PSTDUW
Else if M = 1 and WETTYPE = 1 thenBIOINIT, STANDIN, PEATACRB, PEATDENS, PRATEUP
End if
If M =1 thenBIOMREML, DAYREML, ABIOREML, BIOMREMD, DAYREMD,ABIOREMD, DAYWINS, DAYWINE, DEGBIO, DAYSDEG
If M>1 thenCHECK 1, CHECK2
If CHECK1 = 0 thenIf WETTYPE = 0 then
PEATACRW, PEATACRB, PRATEUP, BIODENS,STDDENS, PBIOUW, PSTDUW
Else if WETTYPE = 1 thenPEATACRB, PRATEUP
End if
If CHECK2 = 0 then
183
BIOMREML, DAYREML, ABIOREML, BIOMREMD, DAYREMD,ABIOREMD, DAYWINS, DAYWINE, DEGBIO, DAYSDEG
-VEGTM (biomass time period)
BIOMGRR
NCOB cycles:-Encompasses the carbon, bacteria, nitrogen, and dissolved oxygen (DO) cycles-These are called if NITCYCLE = 1
BACTERIA submodel:
-BACSTR (bacteria season period)
If M = 1 and WETTYPE = 0 thenNITROSIW, NITROSIB, NDRATEW, NDRATEB, NDOHSATW, NDOHSATB,
NMAXGRW, NMAXGRB, NNH4HSCW, NNH4HSCBHETEROIW, HETEROIB, AEMAXGRW,AEMAXGRB, ANMAXGRW, ANMAXGRB,HTDRW, HTDRB, HTDOHSCW, HTDOHSCB, HNO3HSCW, HNO3HSCB,HORGHSCW, HORGHSCB
Else if M = 1 and WETTYPE = 1 thenNITROSIB, NDRATEB, NDOHSATB, NMAXGRB, NNH4HSCBHETEROIB, AEMAXGRB, ANMAXGRB, HTDRB, HTDOHSCB, HNO3HSCB,HORGHSCB
End if
If M>1 thenCHECK 1, CHECK2
If CHECK1 = 0 thenIf WETTYPE = 0 then
NDRATEW, NDRATEB, NDOHSATW, NDOHSATB,NMAXGRW, NMAXGRB, NNH4HSCW, NNH4HSCB
Else if WETTYPE = 1 thenNDRATEB, NDOHSATB, NMAXGRB, NNH4HSCB
End if
If CHECK2 = 0 thenIf WETTYPE = 0 then
AEMAXGRW, AEMAXGRB, ANMAXGRW, ANMAXGRB, HTDRW, HTDRB, HTDOHSCW, HTDOHSCB,
HNO3HSCW, HNO3HSCB, HORGHSCW, HORGHSCB
184
Else if WETTYPE = 1 thenAEMAXGRB, ANMAXGRB, HTDRB,HTDOHSCB, HNO3HSCB, HORGHSCB
Endif
-BACTIME (bacteria time period)
If (WETTYPE=0) thenWATTEMPW, WATTEMPB
Else if (WETTYPE=1) thenWATTEMPB
CARBON submodel:
-CARSTR (carbon season period)
If M = 1 and WETTYPE = 0 thenREFCINIT, DOCINITW, DOCINITB, POCINITW, POCINITB,BIOCCONT, BODCFRAC, BODPFRAC, LEACHR, MICROBEC, PEATCC,POCFALL, POCRES, MANNC, RESTHC, POCSIZE, MTCDOC, POCCOUT
Else if M = 1 and WETTYPE = 1 thenREFCINIT, DOCINITB, POCINITB, BIOCCONT,BODCFRAC, BODPFRAC, LEACHR, MICROBEC, PEATCC
End if
If POINT = 1 thenBODCONC
If PERCINF = 1 thenDOCININC
If HYDTYPE =1 thenBODRC
If M>1 thenCHECK 1, CHECK2
If CHECK1 = 0 thenIf WETTYPE = 0 then
BIOCCONT, BODCFRAC, BODPFRAC, LEACHR, MICROBEC,PEATCC, POCFALL, POCRES, MANNC, RESTHC, POCSIZE,MTCDOC, POCCOUT
Else if WETTYPE = 1 thenBIOCCONT, BODCFRAC, BODPFRAC,LEACHR, MICROBEC, PEATCC
End if
185
If CHECK2 = 0 thenIf POINT = 1 then
BODCONCIf PERCINF =1 then
DOCININCIf HYDTYPE = 1 then
BODRC
-CARTM (carbon time period)
If HYDTYPE = 0 thenBODINFCO
NITROGEN submodel:
-NITSTR (nitrogen season cycle)
If M = 1 and WETTYPE = 0 thenATDEP, NITFIX, VOLAT,DONINITW, DONINITB, IMMINITW, IMMINITB, NH4INITW, NH4INITB,NO3INITW, NO3INITB, PONINITW, PONINITB, REFNINIT, BIOMCN,
BIOMPN, HTNO3YW, HTNO3YB, MICRONC, NSYIELDW, NSYIELDB, ONPARTF, PEATNC, PONRES, PONFALL, MTCDON, MTCNH4, MTCNO3, PONSIZE, RESTHN, PONCOUT,Else if M = 1 and WETTYPE = 1 then
ATDEP, NITFIX, VOLAT,DONINITB, IMMINITB, NH4INITB, NO3INITB, PONINITB, REFNINIT,BIOMCN, BIOMPN, HTNO3YB, MICRONC, NSYIELDB, ONPARTF, PEATNC
End if
If NITFIX = 1 thenNITFRATE
If VOLAT =1 thenVOLATR
If ATDEP =1 thenDADDON, WADDON, DADPON, WADPON,DADNH4, WADNH4, DADNO3, WADNO3
If POINT = 1 thenDONCONC, PONCONC, NO3CONC, NH4CONC
If PERCINF = 1 thenDONININC, NH4ININC, NO3ININC
If HYDTYPE = 1 thenORGNRC, NH4RC, NO3RC
186
If M>1 thenCHECK1, CHECK2, CHECK3, CHECK4, CHECK5, CHECK6, CHECK7
If CHECK1 = 0 thenIf WETTYPE = 0 then
BIOMCN, BIOMPN, HTNO3YW,HTNO3YB, MICRONC, NSYIELDW, NSYIELDB, ONPARTF, PEATNC, PONRES, PONFALL, MTCDON, MTCNH4, MTCNO3, MANNC, PONSIZE, RESTHN, PONCOUT
Else if WETTYPE = 1 thenBIOMCN, BIOMPN, HTNO3YB, MICRONC,NSYIELDB, ONPARTF, PEATNC
End if
If CHECK2 = 0 thenNITFRATE
End if
If CHECK3 = 0 thenVOLATR
End if
If CHECK4 = 0 thenDADDON, WADDON, DADPON, WADPON,DADNH4, WADNH4, DADNO3, WADNO3
End if
If CHECK5 = 0 then DONCONC, PONCONC, NO3CONC, NH4CONC
End if
If CHECK6 = 0 thenDONININC, NH4ININC, NO3ININC
End if
IF CHECK7 = 0 thenORGNRC, NH4RC, NO3RC
End if
-NITTIME (nitrogen time period)
If VOLAT = 0 and HYDTYPE = 0 thenORGNINC, NH4INC, NO3INC
Else if VOLAT = 1 and HYDTYPE = 1 thenORGNINC, NH4INC, NO3INC, pH
Else if VOLAT = 1 and HYDTYPE =1 thenpH
187
End if
DISSOLVED OXYGEN submodel:
-OXYSTR (dissolved oxygen season period)
If M = 1 and WETTYPE = 0 thenDOINITW, DOINITB, HTDOYW, HTDOYB, NSDOYW, NSDOYB,DOCONCP, MTDOX, MTFWSDOC, DOXYCSAT
Else if M = 1 and WETTYPE = 1 thenDOINITB, HTDOYB, NSDOYB, DOCONCP
End if
If POINT = 1 thenOXYCONC
If PERCINF = 1 thenDOININC
If HYDTYPE = 1 thenOXDRC
If M>1 thenCHECK1, CHECK2
If CHECK1 = 0 thenIf WETTYPE = 0 then
HTDOYW, HTDOYB, NSDOYW, NSDOYB, DOCONCP,MTDOX, MTFWSDOC, DOXYCSAT
Else if WETTYPE = 1 thenHTDOYB, NSDOYB, DOCONCP
End if
If CHECK2 = 0 thenIf POINT = 1 then
OXYCONCIf PERCINF = 1 then
DOININCIf HYDTYPE =1 then
OXDRC
-OXYTIME (dissolved oxygen time period)
If HYDTYPE = 0 thenBIOOXRB, DOCONCIN
Else if HYDTYPE =1 thenBIOOXRB
188
End if
SEDIMENT submodel:
-SEDSTR (sediment season period)-Can only be simulated when HYDTYPE = 0
If M = 1 thenSEDCAT, SEDRES
Now depending on the number of sediment categories (SEDCAT)
DO SEDCLASS=1,SEDCATSEDSIZE (SEDCLASS), SEDFALL (SEDCLASS), SEDINITW (SEDCLASS),SEDINITB (SEDCLASS), SEDSPG (SEDCLASS), SEDPER (SEDCLASS)
CONTINUE
If M = 1 thenRESTHICK, MANNC, DECOMPR, PSEDDEP
If POINT =1 thenSEDCONC
If M>1 thenCHECK
IF CHECK = 0 thenRESTHICK, MANNC, DECOMPR
-SEDTIME (sediment time period)
SEDINT
PHOSPHOROUS submodel:
-PHOSSTR (phosphorous season period)-Can only be called if SEDCYCLE =1
M = 1 thenDTPHOSIW, DTPHOSIB, BTPHOSI, PPHOSI, PMINPPC,PRMINBPC, ADSORP, MTCPHOS, BIOMPP
If PERCINF = 1 thenPININC
If ADSORP = 0 thenFREUNDK, FREUNDN
189
Else if ADSORP = 1 or ADSORP = 2 thenLINPARTC
End if
IF POINT =1 or ADSORP =2 thenPHOSCON
If M>1 thenCHECK1, CHECK2
If CHECK1 = 0 thenPMINPPC, PRMINBPC, MTCPHOS, BIOMPP
If PERCINF = 1 thenPININC
End if
If CHECK2 = 0 thenIf ADSORP = 0 then
FREUNDK, FREUNDN, PHOSCONElse if ADSORP = 1 then
LINPARTC, PHOSCONElse if ADSORP =2 then
PHOSCONEnd if
-PHOSTIME (phosphorous time period)
DISPHOSC
If ADSORP = 2 thenPPHOSCAL
190
Appendix C: Model Fortran Code for the SET-WET model
MAIN PROGRAM
PROGRAM WETLANDC MAIN CODE FOR THE SET-WET MODEL (Ver. 1) DEVELOPED BY ERIK LEECC SET-WET IS A CSTR, DYNAMIC, LONG TERM SIMULATION MODEL DESIGNED TO HELPC OPTIMIZE THE CONTROL OF NPS POLLUTION WITH THE USE OF CONSTRUCTEDC WETLANDS. SET-WET MODELS THE HYDROLOGIC, NITROGEN, CARBON, DISSOLVEDC OXYGEN, BACTERIA, VEGTATIVE, SEDIMENT AND PHOSPHOROUS CYCLES OF AC WETLAND.
COMMON /DESCRIBE/ LENGTH,WIDTH,HO,HB,SOREAL WATERVOL(0:500),WATERB(0:500),WATINPUT(0:500),
/ PRECIP(0:500),EVAPT(0:500),HI(0:500),PHYSDEGC(0:500),HT(0:500),SEDSIZE(5), / SEDINITW(5),SEDINITB(5),SEDBV(5), / SEDBSA(5),RESUSP(5,0:500),SEDFALL(5),SEDQTYB(5,0:500), / SEDPER(5,0:500),SEDINW(5,0:500),PHOSPER(5,0:500), / SEDDEP(5,0:500),SEDQTYW(5,0:500),WATINT(0:500), / SEDDELTA(6),DTPHOSW(0:500),BTPHOST(0:500), / OUTFLOW(0:500),HTGROWW(0:500),HTGROWB(0:500),PPHOS(5,0:500), / IMMNB(0:500),NH4CW(0:500),PONCB(0:500),DTPHOSB(0:500), / DONW(0:500),DONB(0:500),PPHOST(0:500),BIOMASST(0:500), / IMMNW(0:500),NH4B(0:500),NO3W(0:500),NO3B(0:500),PONW(0:500), / PONB(0:500),TONCW(0:500),TONCB(0:500),DONOUT(0:500), / REFN(0:500),TONW(0:500),TONB(0:500),DONIN(0:500), / HNH4IMMW(0:500),HNH4IMMB(0:500),DONIMMW(0:500), / DONAMMW(0:500),DONAMMB(0:500),FIXNIT(0:500),DONAT(0:500), / DONDIF(0:500),NH4W(0:500),PONAMMB(0:500),DONIMMB(0:500), / NLEACH(0:500),NITUP(0:500),AMMUP(0:500),DONCW(0:500), / PONIMMW(0:500),PONIMMB(0:500),DEATHB(0:500),IMMNOUT(0:500), / IMMNOUTD(0:500),NH4IN(0:500),DONMINW(0:500),DONMINB(0:500), / PONMINW(0:500),PONMINB(0:500),NH4OUT(0:500),NITRIFW(0:500), / NITRIFB(0:500),NH4DIF(0:500),NH4AT(0:500),VOLATIZ(0:500), / NO3IN(0:500),NO3OUT(0:500),DENITW(0:500),DENITB(0:500), / NO3AT(0:500),NO3DIF(0:500),DEATHW(0:500),PONIN(0:500), / PEATNACW(0:500),PEATNACB(0:500),PONAMMW(0:500),APFLOW(0:500), / PONAT(0:500),PONSET(0:500),PONRE(0:500),DONCB(0:500), / NH4CB(0:500),NO3CW(0:500),NO3CB(0:500),PONCW(0:500), / TOCW(0:500),MICTCNW(0:500),DOCW(0:500),PHOSPERB(5,0:500), / TOCB(0:500),HTDEATHW(0:500),NO3(0:500),ANHTGB(0:500), / MICTCNB(0:500),DOCB(0:500),HTYIELDW(0:500),HTYIELDB(0:500), / BIOMGROW(0:500),NSGROWB(0:500),POCW(0:500),POCB(0:500), / NSDEATHW(0:500),HTDEATHB(0:500),DONIMM(0:500),PON(0:500), / NH4(0:500),NSGROWW(0:500),NH4ATDEP(0:500),ANHTGW(0:500), / PEATACRW(0:500),HETEROW(0:500),NITROSOW(0:500), / ANFRACW(0:500),ANFRACB(0:500),AEROHTGW(0:500),NITROSOB(0:500), / NDEATHW(0:500),NDEATHB(0:500),HETEROB(0:500),AEROHTGB(0:500), / DOCMINIB(0:500),TOCCB(0:500),DOCCB(0:500),DOCMINIW(0:500), / POCMINIB(0:500),PEATCACB(0:500),PEATCACW(0:500),PONOUT(0:500), / POCMINIW(0:500),POCOUT(0:500),DOCOUT(0:500),POCRE(0:500), / BIOMASS(0:500),DOCCW(0:500),POCCW(0:500),POCCB(0:500), / STANDDC(0:500),TOCCW(0:500),REFC(0:500),BIOMDTH(0:500), / BIOMOUT(0:500),SDEADOUT(0:500),STANDDT(0:500), / SOBODIN(0:500),PBODIN(0:500),PHYSDEG(0:500),MICRODW(0:500), / MICRODB(0:500),DOCDIF(0:500),POCSET(0:500),DOCLEACH(0:500), / DOXYW(0:500),DOXYB(0:500),DOINF(0:500),DOOUT(0:500), / DOXYCW(0:500),DOXYCB(0:500),NSRESPW(0:500),NSRESPB(0:500), / DIFOXY(0:500),DIFWS(0:500),HTRESPW(0:500),HTRESPB(0:500), / BTPHOS(5,0:500),SEDPART(5,0:500),PONEC(0:500),SEDSPG(5), / PHOSPERW(5,0:500),NH4EC(0:500),NO3EC(0:500),DONEC(0:500), / SEDOUT(5,0:500),PERCINFT(0:500),BIOMASSV(0:500), / STANDDV(0:500),DTPHOSCW (0:500),DTPHOSCB(0:500), / SEDOUTT(0:500),SEDOUTC(0:500),BOD5CW(0:500),BOD5CB(0:500), / DPHOSOUT(0:500),PPHOSOUT(0:500),DPHOSEC(0:500),TPHOSEC(0:500)
INTEGER NITCYCLE,SEDCYCLE,PCYCLE,NUMSTPER,NUMTMPER,POINT, / STRPER,TIMPER,SEDCAT,SEDCLASS,WETTYPE,EVAP
191
REAL NITROSIW,NITROSIB,DOINITW,DOINITB,POCINITW,POCINITB, / REFCINIT,STANDIN,IMMINITW,IMMINITB,NH4INITW,NH4INITB,HRT, / NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,HETEROIW, / HETEROIB,LENGTH,WIDTH,HO,HB,SO,BIOINIT,PEATACRB,DECOMPR
HYDTYPE=0POINT=0NITCYCLE=0PCYCLE=0SEDCYCLE=0WTRSHDAR=0SCCURVE=0ATDEP=0NITFIX=0VOLAT=0
C OPEN THE MANDATORY UNIT FILES
OPEN (UNIT=2, FILE='base.INP', STATUS='OLD')OPEN (UNIT=3, FILE='hydro.INP', STATUS='OLD')OPEN (UNIT=15, FILE='hydro.OUT', STATUS='OLD')CALL BASE (NUMSTPER,HYDTYPE,NITCYCLE,PCYCLE,SEDCYCLE,
/ WTRSHDAR,SCCURVE,POINT,HEATINDX,EVAPCOEF,HTI,HII,WETTYPE, / EVAP,POROSITY,PERCINF)
IF (NITCYCLE.EQ.1) THENOPEN (UNIT=6, FILE='bac.INP', STATUS='OLD')OPEN (UNIT=8, FILE='nit.INP', STATUS='OLD')OPEN (UNIT=10, FILE='oxy.INP', STATUS='OLD')OPEN (UNIT=12, FILE='car.INP', STATUS='OLD')OPEN (UNIT=16, FILE='bac.OUT', STATUS='OLD')OPEN (UNIT=17, FILE='nit1.OUT', STATUS='OLD')OPEN (UNIT=23, FILE='nit2.OUT',STATUS='OLD')OPEN (UNIT=19, FILE='oxy.OUT', STATUS='OLD')OPEN (UNIT=20, FILE='car1.OUT', STATUS='OLD')OPEN (UNIT=21, FILE='car2.OUT', STATUS='OLD')OPEN (UNIT=24, FILE='bio.INP', STATUS='OLD')OPEN (UNIT=32, FILE='nconc.OUT', STATUS='OLD')
ENDIF
IF (SEDCYCLE.EQ.1 .AND. PCYCLE.EQ.0) THENOPEN (UNIT=9, FILE='sed.INP', STATUS='OLD')OPEN (UNIT=18, FILE='sedw.OUT', STATUS='OLD')OPEN (UNIT=22, FILE='sedb.OUT', STATUS='OLD')
IF (NITCYCLE .EQ. 0) THENOPEN (UNIT=24,FILE='bio.INP',STATUS='OLD')OPEN (UNIT=32, FILE='’nconc.OUT', STATUS='OLD')
ENDIFENDIF
IF (PCYCLE.EQ.1 .AND. SEDCYCLE.EQ.1) THENOPEN (UNIT=9, FILE='sed.INP', STATUS='OLD')OPEN (UNIT=18, FILE='sedw.OUT', STATUS='OLD')OPEN (UNIT=22, FILE='sedb.OUT', STATUS='OLD')OPEN (UNIT=14, FILE='phos.INP', STATUS='OLD')OPEN (UNIT=25, FILE='phos.OUT', STATUS='OLD')
IF (NITCYCLE .EQ. 0) THENOPEN (UNIT=24,FILE='bio.INP',STATUS='OLD')OPEN (UNIT=32, FILE='nconc.OUT', STATUS='OLD')
ENDIF
ELSEIF (PCYCLE.EQ.1 .AND.SEDCYCLE.EQ.0) THENWRITE(*,6)
6 FORMAT ('To model the phosphorous cycle, the ' / 'sediment cycle must also be chosen.'/'Please enter ' / 'a 1 for input SEDCYCLE in the base file or the',/, / 'phosphorous model will not run.',//,' Simulation terminated.')
GO TO 1000ENDIF
192
IF (WETTYPE.EQ.1 .AND. (PCYCLE.EQ.1 .OR. SEDCYCLE.EQ.1)) THENWRITE (*,7)
7 FORMAT ('The sediment cycle and/or phosphorous cycle are ' / 'currently not'/' modeled for SSF wetlands.'//'Please change' / 'data input accordingly, and enter a zero in both '/ / 'SEDCYCLE and PCYCLE in the base file or the model will not ' / 'run properly.'//)
GO TO 1000ENDIF
C START OF SEASON PERIOD LOOP(S)
DO 400 STRPER=1,NUMSTPERM=STRPERCALL HYDSTR(NUMTMPER,POINT,PORPEAT,M,WETTYPE,HYDCONP,
/ POROSITY,PERCINF,PERCINFA)
IF (NITCYCLE.EQ.1. OR. SEDCYCLE.EQ.1) THENCALL VEGST(BIOINIT,BIOMREML,DAYREML,ABIOREML,BIOMREMD,
/ DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,STANDIN, / PEATACRW,PEATACRB,M,PEATDENS,WETTYPE,PRATEUP,PBIOUW, / BIODENS,STDDENS,PSTDUW,DAYWINE)
ENDIF
IF (NITCYCLE.EQ.1) THENCALL CARSTR (DOCINITW,DOCINITB,POCINITW,POCINITB,
/ POCFALL,POCRES,BIOCCONT,BODCFRAC,MTCDOC,POINT,PERCINF, / BODPFRAC,LEACHR,MICROBEC,PEATCC,BODCONC,HYDTYPE,BODRC, / MANNC,POCSIZE,RESTHC,M,REFCINIT,WETTYPE,POCCOUT,DOCININC)
CALL BACSTR(NITROSIW,NITROSIB,HETEROIW,HETEROIB,M,NDRATEW, / NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB, / AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW, / HTDOHSCB,HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,WETTYPE)
CALL NITSTR (POINT,HYDTYPE,MTCDON,NSYIELDB,M,WETTYPE, / NITFRATE,VOLATR,DONCONC,PONCONC,NO3CONC,NH4CONC,MTCNO3, / DONINITW,DONINITB,IMMINITW,IMMINITB,NH4INITW,MTCNH4,HTNO3YB, / NH4INITB,NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,PONCOUT, / BIOMCN,HTNO3YW,MICRONC,NSYIELDW,ONPARTF,PEATNC,BIOMPN,RESTHN, / DADDON,WADDON,DADPON,WADPON,DADNH4,WADNH4,DADNO3,WADNO3, / ORGNRC,NH4RC,NO3RC,PONSIZE,ATDEP,NITFIX,VOLAT,PONRES,PONFALL, / PERCINF,DONININC,NH4ININC,NO3ININC)
CALL OXYSTR (DOINITW,DOINITB,HTDOYB,NSDOYB,HTDOYW, / NSDOYW,DOCONCP,OXYCONC,OXDRC,MTFWSDOC,PERCINF, / MTDOX,POINT,HYDTYPE,M,DOXYCSAT,WETTYPE,DOININC)
ENDIF
IF (SEDCYCLE.EQ.1) THENCALL SEDSTR (SEDSIZE,SEDFALL,SEDINITW,SEDINITB,SEDPER,
/ RESTHICK,SEDBV,SEDBSA,SEDSPG,SEDCAT,MANNC,M,SEDRC, / POINT,SEDCONC,SEDRES,DECOMPR,PSEDDEP) ENDIF
IF (PCYCLE.EQ.1) THENCALL PHOSSTR(DTPHOSIW,PPHOSI,BTPHOSI,M,MTCPHOS,
/ PMINPPC,LINPARTC,FREUNDK,FREUNDN,ADSORP,BIOMPP, / PRMINBPC,POINT,PHOSCON,PHOSRC,DTPHOSIB,PININC,PERCINF)
END IF
C START OF TIME PERIOD LOOP(S)
DO 365 TIMPER=1,NUMTMPERJ=TIMPERK=NUMTMPER
IF (M.EQ.1 .AND. J.EQ.1) THENWRITE (32,*) 'OUTPUT DATA FOR EFFLUENT CONC. IN WETLAND'WRITE(32,275)
193
275 FORMAT (T12,'NH4EC',T23,'NO3EC',T33,'DONEC',T43,'PONEC', / T52,'DOXYCW', t61 'BOD5EC', t71, 'sedoutc', T81, 'DPHOSEC', / T91, 'TPHOSEC')
ENDIF
IF (NITCYCLE.EQ.1 .OR. SEDCYCLE.EQ.1) THENIF (M.EQ.1 .AND. J.EQ.1) THEN
C FROM VEGCYCLEBIOMASST(0)=BIOINITSTANDDT(0)=STANDINBIOMASSV(0)=BIOMASST(0)/1000*(1/BIODENS)STANDDV(0)=STANDDT(0)/1000*(1/STDDENS)
ENDIF
CALL VEGTM(BIOMASST,BIOMREML,DAYREML,ABIOREML,BIOMREMD, / DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,PHYSDEG, / BIOMGROW,STANDDT,BIOMOUT,BIOMDTH,M,J,K,WETTYPE, / BIODENS,STDDENS,BIOMASSV,STANDDV,DAYWINE)
ENDIF
CALL HYDROT(HYDTYPE,APFLOW,WATERVOL,WATERB,POROSITY, / HEATINDX,EVAPCOEF,HI,HT,WATINPUT,PRECIP,PRECIPR,EVAPT, / OUTFLOW,WATERVEL,HTI,HII,WTRSHDAR,SCCURVE,J,K,M, / PORPEAT,NITCYCLE,SEDCYCLE,WATINT,WETTYPE,EVAP,HYDCONP, / PERCINF,PERCINFA,PERCINFT,BIODENS,STDDENS,PBIOUW, / BIOMASSV,STANDDV,PSTDUW,HRT)
IF(NITCYCLE.EQ.1) THENIF (M.EQ.1 .AND.J.EQ.1) THEN
C OXYGENIF (WETTYPE.EQ.0) THEN
DOXYW(0)=DOINITWDOXYCW(0)=DOXYW(0)/WATERVOL(0)
ENDIFDOXYB(0)=DOINITBDOXYCB(0)=DOXYB(0)/WATERB(0)
C CARBONIF (WETTYPE.EQ.0) THEN
DOCW(0)=DOCINITWDOCCW(0)=DOCW(0)/WATERVOL(0)POCW(0)=POCINITWPOCCW(0)=POCW(0)/WATERVOL(0)TOCW(0)=DOCW(0)+POCW(0)TOCCW(0)=TOCW(0)/WATERVOL(0)
ENDIFDOCB(0)=DOCINITBDOCCB(0)=DOCB(0)/WATERB(0)POCB(0)=POCINITBPOCCB(0)=POCCB(0)/WATERB(0)REFC(0)=REFCINITTOCB(0)=DOCB(0)+POCB(0)TOCCB(0)=TOCB(0)/WATERB(0)BIOMASS(0)=BIOMASST(0)*BIOCCONTSTANDDC(0)=STANDDT(0)*BIOCCONT
C NITROGENIF (WETTYPE.EQ.0)THEN
DONW(0)=DONINITWDONCW(0)=DONW(0)/WATERVOL(0)IMMNW(0)=IMMINITWNH4W(0)=NH4INITWNH4CW(0)=NH4W(0)/WATERVOL(0)NO3W(0)=NO3INITWNO3CW(0)=NO3W(0)/WATERVOL(0)PONW(0)=PONINITWPONCW(0)=PONW(0)/WATERVOL(0)TONW(0)=DONW(0)+PONW(0)
194
TONCW(0)=TONW(0)/WATERVOL(0)ENDIF
DONB(0)=DONINITBDONCB(0)=DONB(0)/WATERB(0)IMMNB(0)=IMMINITBNH4B(0)=NH4INITBNH4CB(0)=NH4B(0)/WATERB(0)NO3B(0)=NO3INITBNO3CB(0)=NO3B(0)/WATERB(0)PONB(0)=PONINITBPONCB(0)=PONB(0)/WATERB(0)REFN(0)=REFNINITTONB(0)=DONB(0)+PONB(0)TONCB(0)=TONB(0)/WATERB(0)
C BACTERIAIF (WETTYPE.EQ.0) THEN
NITROSOW(0)=NITROSIWHETEROW(0)=HETEROIW
ENDIFNITROSOB(0)=NITROSIBHETEROB(0)=HETEROIB
ENDIF
CALL BACTIME (DOXYCW,NSGROWW,HTGROWW,NSGROWB,HTGROWB, / TOCCB,TOCCW,DOXYCB,NH4CW,NH4CB,DOXYW,DOXYB,NO3W,NO3B,NO3CW, / NO3CB,NITROSOW,NITROSOB,HETEROW,HETEROB,NITROSIW,NITROSIB, / HETEROIW,HETEROIB,J,K,M,NDRATEW,NDRATEB,NDOHSATW,ANMAXGRW, / NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,AEMAXGRW, / AEMAXGRB,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW, / HNO3HSCB,HORGHSCW,HORGHSCB,AEROHTGB,AEROHTGW, / NDEATHW,NDEATHB,HTDEATHW,HTDEATHB,WETTYPE,ANHTGW,ANHTGB)
CALL CARTIME(HI,HT,HYDTYPE,BIOMASST,BIOCCONT,BODCFRAC, / BODPFRAC,LEACHR,MICROBEC,PEATCC,REFCINIT,PHYSDEGC,PERCINFT, / POCRES,POCFALL,WATINPUT,HTGROWW,HTGROWB,HTYIELDW,WETTYPE, / HTYIELDB,DONW,DONB,TONW,TONB,DOXYCW,DOXYCB,BIOMASS,DOCININC, / BODRC,DAYSDEG,OUTFLOW,BODCONC,STANDDT,DOCLEACH,MICTCNB, / DOCW,DOCB,POCW,POCB,DOCCW,DOCCB,WATERVOL,WATERB,J,K,M, / POCCW,POCCB,TOCW,TOCB,TOCCW,TOCCB,REFC,STANDDC,MICTCNW, / PEATCACW,PEATCACB,MANNC,RESTHC,POCSIZE,APFLOW,NDEATHW,POCCOUT, / NDEATHB,HTDEATHW,HTDEATHB,PEATACRW,PEATACRB,PONW,PONB,MICDTHW, / MICDTHB,MTCDOC,WATERVEL,DOCINITW,DOCINITB,POCINITW,POCINITB, / PHYSDEG,BIOMDTH,SDEADOUT,POCOUT,DOCOUT)
CALL NITTIME (WATINPUT,PRECIP,OUTFLOW,BIOMGROW,DONW,DONB, / DONCW,DONCB,IMMNW,IMMNB,NH4W,NH4B,NH4CW,NH4CB,NO3W,NO3B, / NO3CW,NO3CB,PONW,PONB,PONCW,PONCB,NO3AT,NO3,MTCNH4, / REFN,TONW,TONB,TONCW,TONCB,BIOMCN,HTNO3YW,MICRONC,BIOMPN, / ONPARTF,PEATNC,DOCLEACH,PHYSDEGC,PONINITB,REFNINIT, / ATDEP,NITFIX,NITFRATE,VOLAT,VOLATR,WATERVOL,WATERB, / PONRES,PONFALL,DONCONC,PONCONC,NO3CONC,J,K,M,HTGROWW,HTGROWB, / NH4CONC,BIOMREML,BIOMREMD,DAYREML,DAYREMD,ABIOREMD, / ABIOREML,MTCDON,HI,HT,HYDTYPE,MANNC,PONSIZE,MTCNO3, / RESTHN,APFLOW,TOCW,MICTCNW,DOCW,TOCB,MICTCNB,DOCB,HTNO3YB, / HTYIELDW,HTYIELDB,NSGROWB,POCW,POCB,HTDEATHW,PEATACRW, / NSDEATHW,HTDEATHB,NSDEATHB,DONIMM,PONIMMW,DONIMMB,PEATACRB, / PON,NSGROWW,NSYIELDW,NSYIELDB,NH4ATDEP,ANHTGW,ANHTGB, / WATERVEL,DADDON,WADDON,DADPON,WADPON,DADNH4,NO3OUT,NH4OUT, / WADNH4,DADNO3,WADNO3,NO3RC,NH4RC,ORGNRC,WETTYPE,DONOUT,PONOUT, / NH4EC,NO3EC,DONEC,PONEC,PONCOUT,PRATEUP,DONININC,NH4ININC, / NO3ININC,PERCINFT)
CALL OXYTIME (DOINITW,DOINITB,HTDOYW,NSDOYW,NSDOYB,OUTFLOW, / NSGROWW,AEROHTGB,J,K,M,NSGROWB,AEROHTGW,WATINPUT,PRECIP, / HTDOYB,DOXYW,DOXYB,OXYCONC,OXDRC,APFLOW,MTFWSDOC, / MTDOX,DOXYCW,DOXYCB,HYDTYPE,HI,HT,WATERVEL,DOCONCP,WATERVOL, / WATERB,DOXYCSAT,WETTYPE,PERCINFT,DOININC,BIOMASST)
ENDIF
195
IF (SEDCYCLE.EQ.1) THENCALL SEDTIME (OUTFLOW,WATERVEL,HT,HI,PHOSPERB,SEDOUTT,
/ RESTHICK,SEDSIZE,SEDFALL,SEDPER,SEDQTYW,SEDQTYB,SEDRES, / MANNC,PHOSPER,J,K,M,SEDRC,WATINPUT,SEDCAT,APFLOW,SEDCONC, / SEDINITW,SEDINITB,SEDBSA,SEDBV,SEDSPG,WATERVOL,DECOMPR, / SEDPART,PHOSPERW,SEDDEP,RESUSP,SEDOUT,PCYCLE,PHYSDEG,PSEDDEP) ENDIF
C INITIAL DETERMINATION OF WATER VOLUME IF NOT AFFECTED BY NUTRIENTC OR SEDIMENT PROCESSESC
IF (NITCYCLE.EQ.0 .AND. SEDCYCLE.EQ.0) THENWATERVOL(J)=WATERVOL(J-1)+WATINT(J)-EVAPT(J)-OUTFLOW(J)-PERCINFT(J)ENDIF
IF (WETTYPE.EQ.0) THENIF (NITCYCLE.EQ.1. .OR. SEDCYCLE.EQ.1) THENCALL DELTAH(DELTAHT,NITCYCLE,SEDCYCLE,SEDTOTAL,SEDDELTA,
/ SEDCAT,SEDSPG,PEATACRB,PEATDENS,HT,NUMTMPER,J,SEDQTYB,WATERB, / PORPEAT)
WATERVOL(J)=WATERVOL(J-1)+WATINT(J)-EVAPT(J) / -OUTFLOW(J)-PERCINFT(J)+(BIOMASSV(J-1)*PBIOUW) / +(STANDDV(J-1)*PSTDUW)-(BIOMASSV(J)*PBIOUW) / -(STANDDV(J)*PSTDUW)-(LENGTH*WIDTH*(HT(J)-HT(J-1)))
ENDIFELSEIF (WETTYPE.EQ.1) THENWATERB(J)=WATERB(J-1)+WATINT(J)-EVAPT(J)-OUTFLOW(J)-PERCINFT(J)ENDIF
C DETERMINATION OF HYDRAULIC RETENTION TIMEC
IF (WETTYPE.EQ.0) THENHRT=WATERVOL(J)/OUTFLOW(J)ELSEIF (WETTYPE.EQ.1) THENHRT=WATERB(J)/OUTFLOW(J)ENDIF
IF (PCYCLE.EQ.1) THENIF (M.EQ.1 .AND. J.EQ.1) THENDTPHOSCW(0)=DTPHOSIW/WATERVOL(0)DTPHOSCB(0)=DTPHOSIB/WATERB(0)ENDIF
CALL PHOSTIME (DTPHOSW,PPHOS,BTPHOS,OUTFLOW,ADSORP,PMINPPC, / FREUNDK,FREUNDN,SEDINITB,SEDINITW,SEDQTYB,SEDQTYW,SEDCAT, / SEDBSA,HI,HT,PPHOST,BTPHOST,SEDINW,APFLOW,PHOSCON, / PHOSPER,BIOMPP,BIOMASST,PHYSDEG,DTPHOSIW,PPHOSI, / WATINPUT,SEDDEP,RESUSP,PRMINBPC,MTCPHOS,BTPHOSI,WATERVOL, / J,K,M,LINPARTC,POINT,PHOSRC,PHOSPERB,DTPHOSIB,DTPHOSB, / WATERB,SEDSPG,SEDBV,PHOSPERW,BIOMGROW,SEDOUT,PRATEUP, / PERCINFT,PININC,DTPHOSCW,DTPHOSCB,DPHOSOUT,PPHOSOUT)
ENDIF
C WRITE VALUES TO OUTPUT FILESC
IF (WETTYPE.EQ.0) THENWRITE(15,75) M,J,WATERVOL(J),WATERB(J),WATINPUT(J),PRECIP(J),
/ WATINT(J),OUTFLOW(J),EVAPT(J),HRT, hi(j), ht(j) 75 FORMAT (T1, (I2), T4, (I3), T7, 6(F9.2, 2X),T73,(F5.1,1X),T81,3(F5.2,2x))
ELSEIF(WETTYPE.EQ.1) THENWRITE(15,100)M,J,WATERB(J),WATINPUT(J),PRECIP(J),WATINT(J),
/ OUTFLOW(J),EVAPT(J),hrt,hi(j),ht(j) 100 FORMAT (T1, (I1), T3, (I3), T7, 6(F9.2, 2X),T73,3(F5.2,1X))
ENDIFIF (HI(J) .GT. HO) THENWRITE(*,101) M,J
101 FORMAT ('THE WETLAND SYSTEM HAS OVERFLOWN THE BANKS AT SEASON' / ' PERIOD ', I2,', TIME PERIOD ', I3)
ENDIFC DETERMINATION OF NUTIENT CONCENTRATIONS
196
CIF (NITCYCLE.EQ.1) THEN
C CARBONIF(WETTYPE.EQ.0) THENDOCCW(J)=DOCW(J)/WATERVOL(J)POCCW(J)=POCW(J)/WATERVOL(J)TOCW(J)=DOCW(J)+POCW(J)TOCCW(J)=TOCW(J)/WATERVOL(J)BOD5CW(J)=((DOCOUT(J)+POCOUT(J))/(1.4*BODCFRAC*OUTFLOW(J)))ENDIFDOCCB(J)=DOCB(J)/WATERB(J)POCCB(J)=POCB(J)/WATERB(J)TOCB(J)=DOCB(J)+POCB(J)TOCCB(J)=TOCB(J)/WATERB(J)BOD5CB(J)=DOCCB(J)/(1.4*BODCFRAC*OUTFLOW(J))
C NITROGENIF (WETTYPE.EQ.0) THENDONCW(J)=DONW(J)/WATERVOL(J)NH4CW(J)=NH4W(J)/WATERVOL(J)NO3CW(J)=NO3W(J)/WATERVOL(J)PONCW(J)=PONW(J)/WATERVOL(J)TONCW(J)=TONW(J)/WATERVOL(J)ENDIFDONCB(J)=DONB(J)/WATERB(J)NH4CB(J)=NH4B(J)/WATERB(J)NO3CB(J)=NO3B(J)/WATERB(J)PONCB(J)=PONB(J)/WATERB(J)TONCB(J)=TONB(J)/WATERB(J)
C OXYGEN/NO BACTERIA NECESSARYIF (WETTYPE.EQ.0) THENDOXYCW(J)=DOXYW(J)/WATERVOL(J)ENDIFDOXYCB(J)=DOXYB(J)/WATERB(J)
ENDIFC SEDIMENT
IF (SEDCYCLE.EQ.1) THENSEDOUTC(J)=SEDOUTT(J)/OUTFLOW(J)
ENDIFC PHOSPHOROUS
IF (PCYCLE.EQ.1) THENDTPHOSCW(J)=DTPHOSW(J)/WATERVOL(J)DTPHOSCB(J)=DTPHOSB(J)/WATERB(J)DPHOSEC(J)=DPHOSOUT(J)/OUTFLOW(J)TPHOSEC(J)=(DPHOSOUT(J)+PPHOSOUT(J))/OUTFLOW(J)
ENDIF
C WRITE TO OUTPUT FILESC BACTERIA
WRITE (16,300) M,J,NITROSOW(J),NITROSOB(J),HETEROW(J),HETEROB(J) 300 FORMAT(T1, (I2),T4,(I3),T8,4(F12.2,3X))C CARBON
WRITE (20,305) M,J,BIOMASS(J),DOCW(J), DOCB(J), POCW(J),POCB(J) 305 FORMAT(T1, (I2),T4,(I3),T8,F11.2,T21,2(F12.2,2X),T50,2(E11.5,2X))
WRITE (21,310) M,J,REFC(J),STANDDC(J),TOCW(J),TOCB(J) 310 FORMAT (T1, (I2),T4,(I3),T8,F11.2,T21,3(F11.2,2X))C NITROGEN
WRITE(17,315) M,J, NO3W(J),NO3B(J),NH4W(J),NH4B(J),IMMNW(J), / IMMNB(J) 315 FORMAT (T1, (I2),T4,(I3),T8,10(F9.1,2X))
WRITE(23 ,320) M,J,DONW(J),DONB(J),PONW(J),PONB(J),REFN(J),TONW(J), / TONB(J),NO3OUT(J) 320 FORMAT (T1, (I2),T4,(I3),T8,8(F9.1,2X))C OXYGEN
WRITE (19,325) M,J,DOXYW(J),DOXYB(J),DOXYCW(J),DOXYCB(J), / DOINF(J),BIOFLUXB,DOOUT(J),HTRESPW(J),HTRESPB(J),NSRESPW(J), / NSRESPB(J),DIFOXY(J),DIFWS(J) 325 FORMAT(T1, (I2),T5,(I3),T9,13(F9.2,2X))C EFFLUENT CONCENTRATIONS
WRITE (32,327) M,J,NH4EC(J),NO3EC(J),DONEC(J),PONEC(J), / doxycw(j),BOD5CW(j),SEDOUTC(J),DPHOSEC(J),TPHOSEC(J)
197
327 FORMAT(T1, (I2),T5,(I3),T9,9(F9.4,1X))C SEDIMENT
IF (SEDCYCLE.EQ.1)THENWRITE(18,330) M,J,SEDQTYW(1,J),SEDQTYW(2,J),SEDQTYW(3,J),
/ SEDQTYW(4,J),SEDQTYW(5,J) 330 FORMAT (T1, (I2), T3, (I3), T8, 5(f11.1, 2X))
WRITE(22,335) M,J,SEDQTYB(1,J),SEDQTYB(2,J),SEDQTYB(3,J), / SEDQTYB(4,J),SEDQTYB(5,J) 335 FORMAT (T1, (I2), T3, (I3), T8, 5(E11.5, 2X))
ENDIFC PHOSPHOROUSC IF (PCYCLE.EQ.1) THEN
WRITE(25,340) M,J,DTPHOSW(J),DTPHOSB(J),BTPHOST(J),PPHOST(J), / DTPHOSCW(J),(PPHOST(J)/WATERVOL(J)) 340 FORMAT(T1, (I2),T5,(I3),T9,4(F10.2,2X),T59,2(F5.2,2X))C ENDIFC 365 CONTINUE
C SET NEW SEASON PERIOD VALUES EQUAL TO END OF PREVIOUS SEASON PERIODC
IF (WETTYPE.EQ.0) THENWATERVOL(0)=WATERVOL(J)
IF (NITCYCLE.EQ.0 .AND. SEDCYCLE.EQ.0) THENWATERB(0)=WATERB(J)
ENDIFHI(0)=HI(J)
ELSEIF (WETTYPE.EQ.1) THENWATERB(0)=WATERB(J)HI(0)=HI(J)
ENDIFC
IF (NITCYCLE.EQ.1 .OR. PCYCLE.EQ.1) THENBIOMASST(0)=BIOMASST(J)STANDDT(0)=STANDDT(J)
ENDIFIF (NITCYCLE.EQ.1) THEN
C BACIF (WETTYPE.EQ.0) THEN
NITROSOW(0)=NITROSOW(J)HETEROW(0)=HETEROW(J)
ENDIFNITROSOB(0)=NITROSOB(J)HETEROB(0)=HETEROB(J)
C CARBONIF (WETTYPE.EQ.0) THEN
DOCW(0)=DOCW(J)POCW(0)=POCW(J)DOCCW(0)=DOCCW(J)POCCW(0)=POCCW(J)TOCW(0)=TOCW(J)TOCCW(0)=TOCCW(J)BOD5CW(0)=BOD5CW(J)
ENDIFDOCB(0)=DOCB(J)POCB(0)=POCB(J)DOCCB(0)=DOCCB(J)POCCB(0)=POCCB(J)TOCB(0)=TOCB(J)TOCCB(0)=TOCCB(J)BOD5CB(0)=BOD5CB(J)REFC(0)=REFC(J)BIOMASS(0)=BIOMASS(J)STANDDC(0)=STANDDC(J)
C OXYGENIF (WETTYPE.EQ.0) THEN
DOXYW(0)=DOXYW(J)DOXYCW(0)=DOXYCW(J)
ENDIFDOXYB(0)=DOXYB(J)
198
DOXYCB(0)=DOXYCB(J)C NITROGEN
IF (WETTYPE.EQ.0) THENDONW(0)=DONW(J)DONCW(0)=DONCW(J)IMMNW(0)=IMMNW(J)NH4W(0)=NH4W(J)NH4CW(0)=NH4CW(J)NO3W(0)=NO3W(J)NO3CW(0)=NO3CW(J)PONW(0)=PONW(J)PONCW(0)=PONCW(J)TONW(0)=TONW(J)TONCW(0)=TONCW(J)
ENDIFDONB(0)=DONB(J)DONCB(0)=DONCB(J)IMMNB(0)=IMMNB(J)NH4B(0)=NH4B(J)NH4CB(0)=NH4CB(J)NO3B(0)=NO3B(J)NO3CB(0)=NO3CB(J)PONB(0)=PONB(J)PONCB(0)=PONCB(J)REFN(0)=REFN(J)TONB(0)=TONB(J)TONCB(0)=TONCB(J)
ENDIFC PHOSPHORUS
IF (PCYCLE.EQ.1) THENDTPHOSCW(0)=DTPHOSCW(J)DTPHOSCB(0)=DTPHOSCB(J)
ENDIF 400 CONTINUE
C SCREEN HELPERSC
IF (NITCYCLE.EQ.1) THENWRITE(*,425)
425 FORMAT ('Nitrogen cycle is included in the simulation.')ELSEIF (NITCYCLE.EQ.0) THENWRITE(*,450)
450 FORMAT ('Nitrogen cycle is not included in the simulation.')ENDIFIF (SEDCYCLE.EQ.1) THENWRITE(*,500)
500 FORMAT (/,'Sediment cycle is included in the simulation.')ELSEIF (SEDCYCLE.EQ.0) THENWRITE(*,600)
600 FORMAT (/,'Sediment cycle is not included in the simulation.') ENDIF
IF (PCYCLE.EQ.1) THENWRITE(*,700)
700 FORMAT (/,'Phosphorous cycle is included in the simulation.',/)ELSEIF (PCYCLE.EQ.0) THENWRITE(*,800)
800 FORMAT (/,'Phosphorous cycle is not included in the ' / 'simulation.',/)
ENDIFC END OF SIMULATION
WRITE (*,900) 900 FORMAT ('Normal termination of simulation.',/)C 1000 END PROGRAM WETLAND
199
BASE SUBMODEL
SUBROUTINE BASE (NUMSTPER,HYDTYPE,NITCYCLE,PCYCLE,SEDCYCLE, / WTRSHDAR,SCCURVE,POINT,HEATINDX,EVAPCOEF,HTI,HII,WETTYPE, / EVAP,POROSITY,PERCINF)
COMMON /DESCRIBE/ LENGTH,WIDTH,HO,HB,SOINTEGER HYDTYPE,POINT,NUMSTPER,NITCYCLE,PCYCLE,SEDCYCLE,WETTYPE,
/ EVAP,PERCINF
REAL LENGTH,WIDTH,HO,HB,HTI,HII,SO,WTRSHDAR, / SCCURVE,TJ,HEATIJ,HEATINDX,EVAPCOEF
HEATINDX=0
C READ IN DESCRIPTION OF WHICH CYCLES AND PROCESSES ARE IN SIMULATIONC
READ(2,*) WETTYPE,HYDTYPE,POINT,NITCYCLE,SEDCYCLE,PCYCLE, / EVAP,PERCINF
C DESCRIPTION OF WETLAND SIZE AND INITIAL WATER LEVELSC
IF (HYDTYPE.EQ.0) THENREAD (2,*) NUMSTPER,LENGTH,WIDTH,HO,HB,HTI,HII,SOENDIFIF (HYDTYPE.EQ.1) THENREAD (2,*) NUMSTPER,LENGTH,WIDTH,HO,HB,HTI,HII,SO,WTRSHDAR,SCCURVEENDIF
IF (EVAP.EQ.0) THENHEATINDX=0
DO 75 KEEP=1,12HEATIJ=0READ(2,*) TJHEATIJ=(TJ/5)**1.514HEATINDX=HEATINDX+HEATIJ
75 CONTINUEEVAPCOEF=6.75E-7*(HEATINDX**3)-7.71E-5*(HEATINDX**2)
/ +1.792E-2*HEATINDX+0.49239ELSE IF (EVAP.EQ.1) THENHEATINDX=0EVAPCOEF=0ENDIF
CEND
200
HYDROLOGIC SUBMODELS
SUBROUTINE HYDSTR(NUMTMPER,POINT,PORPEAT,M,WETTYPE,HYDCONP, / POROSITY,PERCINF,PERCINFA)
COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT, / TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK, / KHCOEF
INTEGER POINT,NUMTMPER,CHECK1,CHECK2,OUTLET,WETTYPE,PERCINFREAL HOUT,OUTWIDTH,HOVER,ANGVNOT,HYDCONP,
/ FLOWOUT,TOPPUMP,AREAPIPE,CONTC,DISEFFC,PERCINFA / ALPHA,BETA,COEFK,PORPEAT,KHCOEF,POROSITY
C NUMBER OF TIME PERIODS IN SEASON PERIODC
IF (M.EQ.1) THENIF (WETTYPE.EQ.0) THEN
READ(3,*) NUMTMPER,PORPEATHYDCONP=0.0POROSITY=0
ELSEIF (WETTYPE.EQ.1)THENREAD (3,*) NUMTMPER,POROSITY,HYDCONPPORPEAT=0
ENDIF
IF (PERCINF.EQ.1) THENREAD (3,*) PERCINFA
ELSEIF (PERCINF.EQ.0) THENPERCINFA=0.
ENDIF
HOUT=0.OUTWIDTH=0.HOVER=0.ANGVNOT=0.FLOWOUT=0.TOPPUMP=0.AREAPIPE=0.CONTC=0.DISEFFC=0.ALPHA=0.BEATA=0.COEFK=0.KHCOEF=0.
READ (3,*) OUTLETIF (OUTLET.EQ.1) THEN
READ (3,*) HOUT,OUTWIDTH,HOVERELSEIF (OUTLET.EQ.2)THEN
READ(3,*) ANGVNOT,DISEFFC,KHCOEF,HOUT,HOVERELSEIF (OUTLET.EQ.3) THEN
READ (3,*) HOUT, AREAPIPE, CONTCELSEIF (OUTLET.EQ.4) THEN
READ (3,*) FLOWOUT,TOPPUMPELSEIF (OUTLET.EQ.5) THEN
READ (3,*) ALPHA,BETA,COEFK,HOUTELSEIF (OUTLET.EQ.6)THEN
READ(3,*) HOUTEND IF
ENDIF
C CHECK VALUES TO SEE IF CHANGING BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (3,*) c, NUMTMPER,CHECK1,CHECK2IF (CHECK1.EQ.0) THEN
READ(3,*) PORPEAT,HYDCONP,POROSITYIF (PERCINF.EQ.1) THEN
READ(3,*) PERCINFAENDIF
201
ENDIFIF (CHECK2.EQ.0) THEN
IF (OUTLET.EQ.1) THENREAD(3,*) HOUT,OUTWIDTH,HOVER
ELSEIF (OUTLET.EQ.2) THENREAD(3,*) ANGVNOT,DISEFFC,KHCOEF,HOUT,HOVER
ELSEIF (OUTLET.EQ.3) THENREAD(3,*) HOUT,AREAPIPE,CONTC
ELSEIF (OUTLET.EQ.4) THENREAD(3,*)FLOWOUT,TOPPUMP
ELSEIF (OUTLET.EQ.5) THENREAD(3,*)ALPHA,BETA,COEFK,HOUT
ELSEIF (OUTLET.EQ.6) THENREAD(3,*)HOUT
ENDIFENDIF
ENDIF
ENDCC
SUBROUTINE HYDROT(HYDTYPE,APFLOW,WATERVOL,WATERB,POROSITY, / HEATINDX,EVAPCOEF,HI,HT,WATINPUT,PRECIP,PRECIPR,EVAPT, / OUTFLOW,WATERVEL,HTI,HII,WTRSHDAR,SCCURVE,J,K,M, / PORPEAT,NITCYCLE,SEDCYCLE,WATINT,WETTYPE,EVAP,HYDCONP, / PERCINF,PERCINFA,PERCINFT,BIODENS,STDDENS,PBIOUW, / BIOMASSV,STANDDV,PSTDUW,HRT)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SOCOMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
/ TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK, / KHCOEF
REAL WATERVOL(0:500),WATERB(0:500),WATINPUT(0:500), / WATINT(0:500),PRECIP(0:500),EVAPT(0:500),OUTFLOW(0:500), / HI(0:500),HT(0:500),HYDCONP,PANEVAP,POROSITY,PERCINFA, / EVAPTT(24),WATINTT(24),OUTFLOWT(24),HIT(0:24),APFLOW(0:500), / PERCINFT(0:500), PERCITT(24),BIOMASSV(0:500),STANDDV(0:500)
INTEGER OUTLET,HYDTYPE,NITCYCLE,SEDCYCLE,WETTYPE,EVAP,TIME, / PERCINF
REAL FT3TOM3,ALPHA,BETA,COEFK,AIRTEMP,DISEFFC / PORPEAT,WTRSHDAR,SCCURVE,HOUT,OUTWIDTH,HOVER,ANGVNOT, / LENGTH,WIDTH,HO,SO,HB,FLOWOUT,TOPPUMP,AREAPIPE,CONTC, / PRECIPR,KHCOEF,M3TOFT3,MTOFT,FTTOM,DAYLEN,HTI,HII, / BIODENS,STDDENS,PBIOUW,PSTDUW,HRT
C INITIAL CONDITIONS OF WATER IN SYSTEMC
IF (J.EQ.1 .AND. M.EQ.1) THENHI(0)=HIIHT(0)=HTIIF (WETTYPE.EQ.0) THENIF (NITCYCLE.EQ.1 .OR. SEDCYCLE.EQ.1) THENWATERVOL(0)=(LENGTH*WIDTH*(HI(0)-HT(0)))
/ -((BIOMASSV(0)*PBIOUW)+(STANDDV(0)*PSTDUW))WATERB(0)=LENGTH*WIDTH*(HT(0)-HB)*PORPEATELSEIF (NITCYCLE.EQ.0 .AND. SEDCYCLE.EQ.0) THENWATERVOL(0)=(LENGTH*WIDTH*(HI(0)-HT(0)))WATERB(0)=LENGTH*WIDTH*(HT(0)-HB)*PORPEATENDIF
C WRITE TO OUTPUT FILESC
WRITE (15,*) 'OUTPUT DATA FOR HYDROLOGIC COUNTS IN WETLAND'WRITE(15,75)
75 FORMAT (T8, 'WATERVOL',T21,'WATERB',T30'WATINPUT',T43,'PRECIP', / T54,'WATINT',T64, 'OUTFLOW',T73, 'EVAPT',T81, 'HRT')
WRITE(15,100) M,WATERVOL(0),WATERB(0) 100 FORMAT (T2, (I1), T4,(' 0 '),T7,2(F9.2,2X))
ELSE IF (WETTYPE.EQ.1) THEN
202
WATERB(0)=LENGTH*WIDTH*(HI(0)-HB)*POROSITYWRITE (15,*) 'OUTPUT DATA FOR HYDROLOGIC COUNTS IN WETLAND'WRITE(15,125)
125 FORMAT (T10, 'WATERB',T19,'WATINPUT',T32'PRECIP',T43,'WATINT', / T53,'OUTFLOW',T66, 'EVAPT',T76, 'HRT')
WRITE(15,150) M,WATERB(0) 150 FORMAT (T1, (I1), T4,(' 0 '),T7,(F9.2,2X))
ENDIFENDIF
IF (NITCYCLE.EQ.0 .AND.SEDCYCLE.EQ.0) THENIF (WETTYPE.EQ.0)THEN
HT(J)=HTI WATERB(J)=LENGTH*WIDTH*(HT(J)-HB)*PORPEATENDIFENDIF
IF (WETTYPE.EQ.1) THENHT(J)=HTI
ENDIF
C READ IN INPUTC
IF (HYDTYPE.EQ.0) THENIF (EVAP.EQ.0) THENREAD (3,*) WATINPUT(J),PRECIPR,AIRTEMP,DAYLENELSEIF (EVAP.EQ.1) THENREAD(3,*)WATINPUT(J),PRECIPR,PANEVAPENDIFIF (POINT.EQ.1) THEN
READ(3,*) APFLOW(J)ELSEIF (POINT.EQ.0) THEN
APFLOW(J)=0ENDIF
PRECIP(J)=PRECIPR*WIDTH*LENGTHWATINT(J)=WATINPUT(J)+PRECIP(J)+APFLOW(J)ENDIF
IF (HYDTYPE.EQ.1) THENIF (EVAP.EQ.0) THENREAD (3,*)PRECIPR,AIRTEMP,DAYLENELSEIF (EVAP.EQ.1) THENREAD(3,*)PRECIPR,PANEVAPENDIF
IF (POINT.EQ.1) THENREAD(3,*) APFLOW(J)
ELSEIF (POINT.EQ.0) THENAPFLOW(J)=0
ENDIF
C DETERMINATION OF HYDROLOGIC INPUT WITH SCS CURVE NUMBER APPROACHC
PRECIP(J)=PRECIPR*WIDTH*LENGTHSTORPARM=25400./SCCURVE-254.FLOWRATE=(((PRECIPR*1000.)-0.2*STORPARM)**2.)/((PRECIPR*1000.)
/ +0.8*STORPARM)WATINPUT(J)=(FLOWRATE/1000.)*(WTRSHDAR-LENGTH*WIDTH)WATINT(J)=WATINPUT(J)+PRECIP(J)+APFLOW(J)ENDIF
C EVAPORATION CALCULATIONSC
IF (EVAP.EQ.0) THENIF(AIRTEMP.LE.0)THEN
EVAPT(J)=0.ELSEIF(AIRTEMP.GT.0) THEN
EVAPT(J)=(1.6*DAYLEN)*((10.*AIRTEMP/HEATINDX)**EVAPCOEF)* / WIDTH*LENGTH/(30.*100.)
203
ENDIFELSEIF (EVAP.EQ.1) THEN
EVAPT(J)=PANEVAP*LENGTH*WIDTHENDIF
C PERCOLATION/INFILTRATION CALCULATIONSC
IF (WETTYPE.EQ.0) THENIF (PERCINF.EQ.0) THEN
PERCINFT(J)=0ELSEIF (PERCINF.EQ.1) THEN
PERCINFT(J)=PERCINFA*LENGTH*WIDTHENDIFELSEIF (WETTYPE.EQ.1) THENIF (PERCINF.EQ.0) THEN
PERCINFT(J)=0ELSEIF (PERCINF.EQ.1) THEN
PERCINFT(J)=PERCINFA*LENGTH*WIDTHENDIFENDIF
C DETERMINATION OF OUTFLOW OVER 24 HOUR TIME FRAMEC
DO 200 TIME=1,24IF (TIME.EQ.1) THEN
HIT(0)=HI(J-1)OUTFLOW(J)=0
ENDIFEVAPTT(TIME)=EVAPT(J)/24WATINTT(TIME)=WATINT(J)/24PERCITT(TIME)=PERCINFT(J)/24
IF (OUTLET.EQ.1) THENIF (HIT(TIME-1).LE.HOUT) THEN
OUTFLOWT(TIME)=0ELSEIF (HIT(TIME-1).GT.HOUT .AND. HIT(TIME-1).LT.HOVER) THEN
OUTFLOWT(TIME)=1.84*(OUTWIDTH-(0.2*(HIT(TIME-1)-HOUT)))* / ((HIT(TIME-1)-HOUT)**1.5)*3600
ELSEIF (HIT(TIME-1).GT.HOVER) THENOUTFLOWT(TIME)=((1.843*(OUTWIDTH-(0.2*(HOVER-HOUT))*
/ ((HOVER-HOUT)**1.5)))+((HIT(TIME-1)-HOVER)*WIDTH))*3600END IF
ELSE IF (OUTLET.EQ.2) THENIF (HIT(TIME-1).LE.HOUT) THEN
OUTFLOWT(TIME)=0ELSEIF (HIT(TIME-1).GT.HOUT) THEN
OUTFLOWT(TIME)=2.363*DISEFFC*(TAN(ANGVNOT/360*22/7))* / (((HIT(TIME-1)-HOUT)+KHCOEF)**(5./2.))*3600
ELSEIF (HIT(TIME-1).GT.HOVER) THENOUTFLOWT(TIME)=(2.363*DISEFFC*(TAN(ANGVNOT/360*22/7))*
/ (((HOVER-HOUT)+KHCOEF)**(5./2.))+(HIT(TIME-1)-HOVER*WIDTH)) / *3600
ENDIFELSE IF (OUTLET.EQ.3) THEN
IF (HIT(TIME-1).LE.HOUT) THENOUTFLOWT(TIME)=0
ELSE IF (HIT(TIME-1).GT.HOUT) THENOUTFLOWT(TIME)=3600*CONTC*AREAPIPE*(SQRT(2.*9.81*
/ (HIT(TIME-1)-HOUT)))ENDIF
ELSE IF (OUTLET.EQ.4) THENIF (HIT(TIME-1).LE.TOPPUMP) THEN
OUTFLOWT(TIME)=0ELSEIF (HIT(TIME-1).GT.TOPPUMP) THEN
OUTFLOWT(TIME)=FLOWOUT/24ENDIF
ELSEIF (OUTLET.EQ.5) THENIF (HIT(TIME-1).LE.HOUT) THEN
WATERVEL=0OUTFLOWT(TIME)=0
204
ELSEIF(HIT(TIME-1).GT.HOUT) THENWATERVEL=COEFK*(((HIT(TIME-1)-HT(J-1))/2)**BETA)*
/ (SO**ALPHA)/((HIT(TIME-1)-HT(J-1))/2)OUTFLOWT(TIME)=WATERVEL*WIDTH*(HIT(TIME-1)-HT(J-1))/24
ENDIFELSE IF (OUTLET.EQ.6) THEN
IF (HIT(TIME-1).GE.HOUT)THENOUTFLOWT(TIME)=MAX(WIDTH*(HIT(TIME-1)-HOUT)*HYDCONP*SO/24,
/ (WATERB(J-1)*HYDCONP/(LENGTH**2*POROSITY)*(WATERB(J-1)/ / (LENGTH*WIDTH*POROSITY)-HOUT)))/24
ELSEIF(HIT(TIME-1).LT.HOUT) THENOUTFLOWT(TIME)=0.0
ENDIFENDIF
IF(WETTYPE.EQ.0) THENHIT(TIME)=HIT(TIME-1)+((WATINTT(TIME)-EVAPTT(TIME)-OUTFLOWT(TIME)
/ -PERCITT(TIME))/(LENGTH*WIDTH))ELSEIF (WETTYPE.EQ.1) THENHIT(TIME)=HIT(TIME-1)+((WATINTT(TIME)-EVAPTT(TIME)-OUTFLOWT(TIME)
/ -PERCITT(TIME))/(LENGTH*WIDTH*POROSITY))ENDIFOUTFLOW(J)=OUTFLOW(J)+OUTFLOWT(TIME)
200 CONTINUE
IF (OUTLET.EQ.1 .OR. OUTLET.EQ.2 .OR. OUTLET.EQ.3 .OR. / OUTLET.EQ.4) THEN
WATERVEL=OUTFLOW(J)/(WIDTH*(HI(J-1)-HT(J-1)))ENDIF
HI(J)=HIT(24)
END
205
VEGETATION SUBMODELS
SUBROUTINE VEGST(BIOINIT,BIOMREML,DAYREML,ABIOREML,BIOMREMD, / DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,STANDIN, / PEATACRW,PEATACRB,M,PEATDENS,WETTYPE,PRATEUP,PBIOUW, / BIODENS,STDDENS,PSTDUW,DAYWINE)
REAL BIOINIT,ABIOREML,DEGBIO,DAYSDEG,ABIOREMD,REFCINIT,PSTDUW, / STANDIN,PEATACRW,PEATACRB,PEATDENS,STDDENS,BIODENS,PBIOUW
INTEGER BIOMREML,DAYREML,BIOMREMD,DAYREMD,CHECK1,CHECK2, / WETTYPE,DAYWINS,DAYWINE
C READ INITIAL INPUT VALUESC
IF (M.EQ.1) THENIF (WETTYPE.EQ.0) THENREAD (24,*) BIOINIT,STANDIN,PEATACRW,PEATACRB,PEATDENS,
/ PRATEUP,BIODENS,STDDENS,PBIOUW,PSTDUWELSEIF (WETTYPE.EQ.1) THENREAD (24,*) BIOINIT,STANDIN,PEATACRB,PEATDENS,PRATEUPENDIFREAD (24,*) BIOMREML,DAYREML,ABIOREML,BIOMREMD,
/ DAYREMD,ABIOREMD,DAYWINS,DAYWINE,DEGBIO,DAYSDEG ENDIFCC CHECK FOR CHANGES IN PARAMETERS BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD(24,*) c, CHECK 1,CHECK2IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THENREAD (24,*) PEATACRW,PEATACRB,PRATEUP,BIODENS,
/ STDDENS,PBIOUW,PSTDUWELSEIF(WETTYPE.EQ.1)THEN
READ(24,*) PEATACRB,PRATEUPENDIF
ENDIFIF (CHECK2.EQ.0) THEN
READ(24,*) BIOMREML,DAYREML,ABIOREML,BIOMREMD,DAYREMD, / ABIOREMD,DAYWINS,DAYWINE,DEGBIO,DAYSDEG
ENDIFENDIFEND
CC
SUBROUTINE VEGTM(BIOMASST,BIOMREML,DAYREML,ABIOREML,BIOMREMD, / DAYREMD,ABIOREMD,DAYWINS,DEGBIO,DAYSDEG,PHYSDEG, / BIOMGROW,STANDDT,BIOMOUT,BIOMDTH,M,J,K,WETTYPE, / BIODENS,STDDENS,BIOMASSV,STANDDV,DAYWINE)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SOREAL BIOMASST(0:500),BIOMGROW(0:500),BIOMDTH(0:500),
/ SDEADOUT(0:500),PHYSDEG(0:500),STANDDT(0:500),BIOMOUT(0:500), / BIOMASSV(0:500),STANDDV(0:500)
REAL ABIOREML,DEGBIO,DAYSDEG,ABIOREMD,WINKCO, / LENGTH,WIDTH,HO,HB,SO,PRATEUP,BIODENS,STDDENS
INTEGER BIOMREML,DAYREML,BIOMREMD,DAYREMD,WINTER,WETTYPE, / DAYWINS,DAYWINE
PARAMETER (BIODEGR=.012616904)
READ(24,*) BIOMGRR
IF (DAYWINS.EQ.0) THENWINTER=0GO TO 10
ENDIF
IF (J .LT. DAYWINS) THENWINTER=0
ELSE IF ((J .GE. DAYWINS) .AND. (J .LE. DAYWINE)) THEN
206
WINTER=1ELSE IF (J .GT. DAYWINE) THEN
WINTER=0ENDIF
10 IF (WINTER.EQ.0) THENBIOMGROW(J)=LENGTH*WIDTH*BIOMGRRBIOMDTH(J)=0.0
ENDIF
IF (WINTER.EQ.1) THENCC CALCULATION OF EXPONENTIAL DECAYC
WINKCO=-LOG((DEGBIO)/DAYSDEG)BIOMDTH(J)=BIOMASST(J-1)*(EXP(-WINKCO))BIOMGROW(J)=0.0ENDIF
CIF (BIOMREML.EQ.0) THEN
BIOMOUT(J)=0.0ELSE IF (BIOMREML.EQ.1) THEN
IF (J.EQ.DAYREML) THENBIOMOUT(J)=ABIOREML
ELSEBIOMOUT(J)=0.0
ENDIFENDIF
IF (BIOMREMD.EQ.0) THENSDEADOUT(J)=0.0
ELSE IF (BIOMREMD.EQ.1) THENIF (J.EQ.DAYREMD) THEN
SDEADOUT(J)=ABIOREMDELSE
SDEADOUT(J)=0.0ENDIF
ENDIFPHYSDEG(J)=BIODEGR*STANDDT(J-1)
CC TOTAL BIOMASS AND STANDING DEAD MASS BALANCESC
BIOMASST(J)=BIOMASST(J-1)+BIOMGROW(J)-BIOMDTH(J)-BIOMOUT(J)STANDDT(J)=STANDDT(J-1)+BIOMDTH(J)-PHYSDEG(J)-SDEADOUT(J)BIOMASSV(J)=BIOMASST(J)/1000*(1/BIODENS)STANDDV(J)=STANDDT(J)/1000*(1/STDDENS)END
207
CARBON SUBMODELS
SUBROUTINE CARSTR(DOCINITW,DOCINITB,POCINITW,POCINITB, / POCFALL,POCRES,BIOCCONT,BODCFRAC,MTCDOC,POINT,PERCINF, / BODPFRAC,LEACHR,MICROBEC,PEATCC,BODCONC,HYDTYPE,BODRC, / MANNC,POCSIZE,RESTHC,M,REFCINIT,WETTYPE,POCCOUT,DOCININC)
REAL DOCINITW,DOCINITB,POCINITW,POCINITB,REFCINIT, / BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,MTCDOC, / PEATCC,POCFALL,POCRES,POCCOUT,BODRC / BODCONC,MANNC,RESTHC,POCSIZE,DOCININC
INTEGER CHECK1,CHECK2,HYDTYPE,CHECK3,WETTYPE,POINT,PERCINF
C READ IN INITIAL DATAC
IF (M.EQ.1) THENIF (WETTYPE.EQ.0) THENREAD (12,*) REFCINIT,DOCINITW,DOCINITB,POCINITW,POCINITB,
/ BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,PEATCC, / POCFALL,POCRES,MANNC,RESTHC,POCSIZE,MTCDOC,POCCOUT
ELSEIF (WETTYPE.EQ.1) THENREAD (12,*) REFCINIT,DOCINITB,POCINITB,BIOCCONT,
/ BODCFRAC,BODPFRAC,LEACHR,MICROBEC,PEATCCENDIF
IF (POINT.EQ.0) THENBODCONC=0.
ELSEIF (POINT.EQ.1) THENREAD (12,*) BODCONC
ENDIF
IF (PERCINF.EQ.0) THENDOCININC=0.0
ELSEIF (PERCINF.EQ.1) THENREAD (12,*) DOCININC
ENDIF
IF (HYDTYPE.EQ.0) THENBODRC=0.0
ELSEIF (HYDTYPE.EQ.1) THENREAD(12,*) BODRC
ENDIFENDIF
C CHECK VALUES TO SEE IF CHANGING BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (12,*) C, CHECK1,CHECK2IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0)THENREAD (12,*) BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,
/ PEATCC,POCFALL,POCRES,MANNC,RESTHC,POCSIZE, / MTCDOC,POCCOUT
ELSEIF(WETTYPE.EQ.1) THENREAD (12,*) BIOCCONT,BODCFRAC,BODPFRAC,
/ LEACHR,MICROBEC,PEATCCENDIF
ENDIFIF (CHECK2.EQ.0) THEN
IF (POINT.EQ.1) THENREAD(12,*) BODCONCENDIFIF (PERCINF.EQ.1) THENREAD(12,*) DOCININCENDIFIF(HYDTYPE.EQ.1) THENREAD(12,*) BODRCENDIF
ENDIF
208
ENDIFEND
CSUBROUTINE CARTIME(HI,HT,HYDTYPE,BIOMASST,BIOCCONT,BODCFRAC,
/ BODPFRAC,LEACHR,MICROBEC,PEATCC,REFCINIT,PHYSDEGC,PERCINFT, / POCRES,POCFALL,WATINPUT,HTGROWW,HTGROWB,HTYIELDW,WETTYPE, / HTYIELDB,DONW,DONB,TONW,TONB,DOXYCW,DOXYCB,BIOMASS,DOCININC, / BODRC,DAYSDEG,OUTFLOW,BODCONC,STANDDT,DOCLEACH,MICTCNB, / DOCW,DOCB,POCW,POCB,DOCCW,DOCCB,WATERVOL,WATERB,J,K,M, / POCCW,POCCB,TOCW,TOCB,TOCCW,TOCCB,REFC,STANDDC,MICTCNW, / PEATCACW,PEATCACB,MANNC,RESTHC,POCSIZE,APFLOW,NDEATHW,POCCOUT, / NDEATHB,HTDEATHW,HTDEATHB,PEATACRW,PEATACRB,PONW,PONB,MICDTHW, / MICDTHB,MTCDOC,WATERVEL,DOCINITW,DOCINITB,POCINITW,POCINITB, / PHYSDEG,BIOMDTH,SDEADOUT,POCOUT,DOCOUT)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SOCOMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
/ TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK, / KHCOEF
REAL HTYIELDB(0:500),HTYIELDW(0:500),WATERVOL(0:500), / WATERB(0:500),DOCMINIB(0:500),TOCCB(0:500),DOCCB(0:500), / POCMINIB(0:500),PEATCACB(0:500),PEATCACW(0:500), / DOCMINIW(0:500),BIOMASST(0:500),BIOMASS(0:500),STANDDT(0:500), / POCMINIW(0:500),MICTCNW(0:500),MICTCNB(0:500),DOXYCB(0:500), / POCOUT(0:500),DOCOUT(0:500),POCRE(0:500),DOXYCW(0:500), / DOCW(0:500),DOCCW(0:500),DOCB(0:500),PHYSDEGC(0:500), / POCW(0:500),POCCW(0:500),POCB(0:500),POCCB(0:500),REFC(0:500), / STANDDC(0:500),TOCW(0:500),TOCCW(0:500),TOCB(0:500), / BIOMGROW(0:500),BIOMDTH(0:500), BIOMOUT(0:500),PONB(0:500), / SOBODIN(0:500),PBODIN(0:500),MICRODW(0:500),PERCINFT(0:500), / MICRODB(0:500),DOCMT(0:500),POCSET(0:500),DOCLEACH(0:500), / HI(0:500),HT(0:500),SDEADOUT(0:500),PONW(0:500), / WATINPUT(0:500),OUTFLOW(0:500),TONW(0:500),TONB(0:500), / HTGROWB(0:500),HTGROWW(0:500),DONW(0:500),DONB(0:500), / NDEATHW(0:500),NDEATHB(0:500),HTDEATHW(0:500),HTDEATHB(0:500), / DOCPERC(0:500),APFLOW(0:500), PHYSDEG(0:500)
INTEGER HYDTYPE,WETTYPE,OUTLETC / BIOMREML,DAYREML,WINTER,BIOMREMD,DAYREMD
REAL DOCINITW,DOCINITB,POCINITW,POCINITB,REFCINIT,POCRES, / BIOCCONT,BODCFRAC,BODPFRAC,LEACHR,MICROBEC,WATERVEL, / PEATCC,RESTHC,POCSIZE,LENGTH,WIDTH,HO,HB,SO,MANNC, / ABIOREML,DAYWIN,DEGBIO,POCFALL,BODCONC,ABIOREMD,RESUSPV, / DAYSDEG,MTCDOC,BODRC,PEATACRB,PEATACRW,POCCOUT
C BIODEGREDATION SET AT 95% OVER THE COURSE OF ONE YEARPARAMETER (BIODEGR=.008207485)
C READ IN INPUT VALUESC
IF (HYDTYPE.EQ.0) THENREAD(12,*) BODINFCOENDIF
C DETERMINE HETEROTROPHIC YIELDC
IF(WETTYPE.EQ.0) THENIF (DOXYCW(J-1).EQ.0) THEN
HTYIELDW=.01ELSEIF (DOXYCW(J-1).GE.0 .AND. DOXYCW(J-1).LE.2) THEN
HTYIELDW(J)=DOXYCW(J-1)*.05ELSEIF (DOXYCW(J-1).GT.2. .AND. DOXYCW(J-1).LE.10) THEN
HTYIELDW(J)=.1+((DOXYCW(J-1)-2.)*.1)ELSEIF (DOXYCW(J-1).GT.10.) THEN
HTYIELDW(J)=0.9ENDIFENDIF
IF (DOXYCB(J-1).EQ.0.0) THENHTYIELDB(J)=.01
209
ELSEIF (DOXYCB(J-1).GE.0. .AND. DOXYCB(J-1).LE.2.) THENHTYIELDB(J)=DOXYCB(J-1)*.05
ELSEIF (DOXYCB(J-1).GT.2. .AND. DOXYCB(J-1).LE.10.) THENHTYIELDB(J)=.1+((DOXYCB(J-1)-2.)*.1)
ELSEIF (DOXYCB(J-1).GT.10.) THENHTYIELDB(J)=0.9
ENDIF
C DETERMINE THE MICROBIAL CARBON:NITROGEN RATIOC
IF (WETTYPE.EQ.0) THENIF (DOXYCW(J-1).GE.0. .AND. DOXYCW(J-1).LE.5.) THEN
MICTCNW(J)=80.-10.*DOXYCW(J-1)ELSEIF (DOXYCW(J-1).GE.5.) THEN
MICTCNW(J)=30.ENDIFIF (M.EQ.1 .AND. J. EQ.1) THENMICTCNW(0)=MICTCNW(J)ENDIFENDIF
IF (DOXYCB(J-1).GE.0. .AND. DOXYCB(J-1).LE.5.) THENMICTCNB(J)=80.-10.*DOXYCB(J-1)
ELSEIF (DOXYCB(J-1).GE.5.) THENMICTCNB(J)=30.
ENDIFIF (M.EQ.1 .AND. J.EQ.1) THENMICTCNB(0)=MICTCNB(J)ENDIF
C DOC RELATIONSC
IF (HYDTYPE.EQ.0) THENSOBODIN(J)=1.4*(1-BODPFRAC)*BODCFRAC*(BODINFCO*(WATINPUT(J)+
/ APFLOW(J)*(BODCONC)))ELSEIF (HYDTYPE.EQ.1) THENSOBODIN(J)=1.4*(1-BODPFRAC)*BODCFRAC*(BODRC*WATINPUT(J)+
/ APFLOW(J)*BODCONC)ENDIFDOCLEACH(J)=LENGTH*WIDTH*LEACHRIF(WETTYPE.EQ.0) THENIF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENDOCOUT(J)=DOCW(J-1)*OUTFLOW(J)/(WATERVOL(J-1))ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THENDOCOUT(J)=.75*DOCW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ENDIFELSEIF (WETTYPE.EQ.1) THENDOCOUT(J)=DOCB(J-1)*OUTFLOW(J)/(WATERB(J-1))ENDIF
IF (WETTYPE.EQ.0) THENIF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
DOCMINIW(J)=DONW(J-1)/TONW(J-1)*HTGROWW(J)/HTYIELDW(J)ELSE
DOCMINIW(J)=DOCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)ENDIFENDIFIF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
DOCMINIB(J)=DONB(J-1)/TONB(J-1)*HTGROWB(J)/HTYIELDB(J)ELSE
DOCMINIB(J)=DOCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)ENDIF
IF (WETTYPE.EQ.0) THENDOCMT(J)=864*LENGTH*WIDTH*MTCDOC*(DOCCB(J-1)-DOCCW(J-1))ENDIF
IF (PERCINFT(J).GT.0.) THENDOCPERC(J)=DOCB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THEN
210
DOCPERC(J)=0.0ELSEIF (PERCINFT(J).LT.0.) THEN
DOCPERC(J)=PERCINFT(J)*DOCININCENDIF
C MASS BALANCE FOR DOCC
IF (WETTYPE.EQ.0) THENDOCW(J)=DOCW(J-1)+SOBODIN(J)-DOCOUT(J)-DOCMINIW(J)+DOCMT(J)DOCB(J)=DOCB(J-1)-DOCMINIB(J)-DOCMT(J)+DOCLEACH(J)-DOCPERC(J)ELSEIF (WETTYPE.EQ.1) THENDOCB(J)=DOCB(J-1)+SOBODIN(J)+DOCLEACH(J)-DOCOUT(J)
/ -DOCMINIB(J)-DOCPERC(J)ENDIF
C POC RELATIONSC
IF (HYDTYPE.EQ.0) THENPBODIN(J)=1.4*(BODPFRAC)*BODCFRAC*(BODINFCO*WATINPUT(J)+
/ APFLOW(J)*BODCONC)ELSEIF (HYDTYPE.EQ.1) THENPBODIN(J)=1.4*(BODPFRAC)*BODCFRAC*(BODRC*WATINPUT(J)+
/ APFLOW(J)*BODCONC)ENDIF
PHYSDEGC(J)=BIODEGR*STANDDC(J-1)
IF(WETTYPE.EQ.0) THENMICRODW(J)=(NDEATHW(J)+HTDEATHW(J))*MICROBECPEATCACW(J)=PEATACRW*PEATCC
ENDIFMICRODB(J)=(NDEATHB(J)+HTDEATHB(J))*MICROBECPEATCACB(J)=PEATACRB*PEATCC
IF (WETTYPE.EQ.0) THENIF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
POCMINIW(J)=PONW(J-1)/TONW(J-1)*HTGROWW(J)/HTYIELDW(J)ELSE
POCMINIW(J)=POCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)ENDIFENDIFIF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
POCMINIB(J)=PONB(J-1)/TONB(J-1)*HTGROWB(J)/HTYIELDB(J)ELSE
POCMINIB(J)=POCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)ENDIF
IF (WETTYPE.EQ.0) THENIF (POCFALL.LT.(HI(J-1)-HT(J-1))) THEN
POCSET(J)=POCW(J-1)*(POCFALL)/(HI(J-1)-HT(J-1))ELSE IF (POCFALL.GE.(HI(J-1)-HT(J-1)))THEN
POCSET(J)=POCW(J-1)ENDIF
ENDIF
IF(WETTYPE.EQ.0) THENRESUSPV=7.2*(((POCFALL/86400)**1/3)*
/ ((HI(J-1)-HT(J-1))**(1./6.))/(MANNC*((POCSIZE**(2./3.)))))IF (WATERVEL.GE.RESUSPV) THEN
POCRE(J)=POCB(J-1)*POCRES*RESTHC/(HT(J-1)-HB)ELSE
POCRE(J)=0ENDIFENDIF
IF (WETTYPE.EQ.0) THENIF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
/ OUTLET.EQ.5) THENIF (POCFALL .GT. (HI(J-1)-HT(J-1))) THEN
POCOUT(J)=0.0
211
ELSEIF (POCFALL .LE. (HI(J-1)-HT(J-1)))THENIF (OUTFLOW(J) .LE. WATERVOL(J-1)) THEN
POCOUT(J)=POCCOUT*POCW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ELSEIF (OUTFLOW(J) .GT. WATERVOL(J-1)) THEN
POCOUT(J)=POCCOUT*.5*POCW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ENDIF
ENDIFELSEIF (OUTLET.EQ.4) THENIF (POCFALL .GT. (HI(J-1)-HT(J-1))) THEN
POCOUT(J)=0.ELSEIF (POCFALL .LE. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J) .LE. WATERVOL(J-1)) THENPOCOUT(J)=((POCW(J-1))*OUTFLOW(J)/WATERVOL(J-1)*POCCOUT)
ELSEIF (OUTFLOW(J) .GT. WATERVOL(J-1)) THENPOCOUT(J)= ((POCW(J-1))*.5*OUTFLOW(J)/WATERVOL(J-1)*POCCOUT)
ENDIFENDIFENDIF
ENDIF
C POC MASS BALANCESC
IF (WETTYPE.EQ.0) THENPOCW(J)=POCW(J-1)+PBODIN(J)+MICRODW(J)-PEATCACW(J)
/ -POCMINIW(J)-POCSET(J)+POCRE(J)-POCOUT(J)+PHYSDEGC(J)POCB(J)=POCB(J-1)+MICRODB(J)-PEATCACB(J)-POCMINIB(J)
/ +POCSET(J)-POCRE(J)ELSEIF (WETTYPE.EQ.1) THENPOCB(J)=POCB(J-1)+PHYSDEGC(J)+PBODIN(J)+MICRODB(J)-PEATCACB(J)
/ -POCMINIB(J)ENDIF
C CARBON BIOMASS/REFRACTORY CARBON/STANDING DEAD CARBONC MASS BALANCEC
BIOMASS(J)=BIOMASS(J-1)+((BIOMASST(J)-BIOMASST(J-1))*BIOCCONT)
IF (WETTYPE.EQ.0) THENREFC(J)=REFC(J-1)+PEATCACW(J)+PEATCACB(J)ELSEIF (WETTYPE.EQ.1) THENREFC(J)=REFC(J-1)+PEATCACB(J)ENDIF
STANDDC(J)=STANDDC(J-1)+((BIOMDTH(J)-PHYSDEG(J)-SDEADOUT(J)) / *BIOCCONT)-DOCLEACH(J)
IF (M.EQ.1 .AND. J.EQ.1) THENWRITE (20,*) 'OUTPUT DATA FOR CARBON COUNTS IN WETLAND'WRITE(20,150)
150 FORMAT(T10,'BIOMASS',T24,'DOCW',T35,'DOCB',T46,'POCW',T57,'POCB')WRITE(20,175) M,BIOMASS(0),DOCW(0),DOCB(0),POCW(0),POCB(0)
175 FORMAT(T1,(I1),T5,(' 0'),T8,F11.2,T21,2(F11.2,2X),t46,2(E11.5,2X)) WRITE (21,*) 'OUTPUT DATA FOR CARBON COUNTS IN WETLAND' WRITE(21,200) 200 FORMAT (T13,'REFC',T23,'STANDDC',T35,'TOCW',T46,'TOCB')
WRITE(21,225) M,REFC(0),STANDDC(0),TOCW(0),TOCB(0) 225 FORMAT (T1, (I2),T5,(' 0'),T8,F11.2,T21,3(F9.2,2X))
ENDIFC
END
212
BACTERIA SUBMODELS
SUBROUTINE BACSTR(NITROSIW,NITROSIB,HETEROIW,HETEROIB,M,NDRATEW, / NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB, / AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW, / HTDOHSCB,HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,WETTYPE)
REAL NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB, / NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,HETEROIW,HETEROIB, / AEMAXGRB,ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB, / HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB,AEMAXGRW
INTEGER CHECK1,CHECK2,WETTYPE,M
C READ IN INITIAL VALUESC
IF (M.EQ.1) THENIF (WETTYPE.EQ.0) THENREAD (6,*) NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,
/ NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCBREAD(6,*)HETEROIW,HETEROIB,AEMAXGRW,AEMAXGRB,ANMAXGRW,
/ ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW,HNO3HSCB, / HORGHSCW,HORGHSCB
ELSEIF (WETTYPE.EQ.1) THENREAD(6,*) NITROSIB,NDRATEB,NDOHSATB,NMAXGRB,NNH4HSCBREAD(6,*) HETEROIB,AEMAXGRB,ANMAXGRB,HTDRB,
/ HTDOHSCB,HNO3HSCB,HORGHSCBENDIFENDIF
C CHECK TO SEE VALUE CHANGES BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (6,*) C, CHECK1,CHECK2IF (CHECK1.EQ.0) THEN
IF (WETTYPE.EQ.0) THENREAD(6,*)NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,
/ NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCBELSEIF(WETTYPE.EQ.1) THENREAD(6,*)NDRATEB,NDOHSATB,NMAXGRB,NNH4HSCBENDIF
ENDIF
IF (CHECK2.EQ.0) THENIF (WETTYPE.EQ.0) THENREAD (6,*) AEMAXGRW,AEMAXGRB,ANMAXGRW,ANMAXGRB,
/ HTDRW,HTDRB,HTDOHSCW,HTDOHSCB, / HNO3HSCW,HNO3HSCB,HORGHSCW,HORGHSCB
ELSEIF (WETTYPE.EQ.1) THENREAD (6,*) AEMAXGRB,ANMAXGRB,HTDRB,
/ HTDOHSCB,HNO3HSCB,HORGHSCBENDIF
ENDIFENDIFEND
CC
SUBROUTINE BACTIME (DOXYCW,NSGROWW,HTGROWW,NSGROWB,HTGROWB, / TOCCB,TOCCW,DOXYCB,NH4CW,NH4CB,DOXYW,DOXYB,NO3W,NO3B,NO3CW, / NO3CB,NITROSOW,NITROSOB,HETEROW,HETEROB,NITROSIW,NITROSIB, / HETEROIW,HETEROIB,J,K,M,NDRATEW,NDRATEB,NDOHSATW,ANMAXGRW, / NDOHSATB,NMAXGRW,NMAXGRB,NNH4HSCW,NNH4HSCB,AEMAXGRW,AEMAXGRB, / ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW,HNO3HSCB, / HORGHSCW,HORGHSCB,AEROHTGB,AEROHTGW,NDEATHW,NDEATHB,HTDEATHW, / HTDEATHB,WETTYPE,ANHTGW,ANHTGB)
REAL HETEROW(0:500),NITROSOW(0:500),NITROSOB(0:500), / ANFRACW(0:500),ANFRACB(0:500),HTDEATHB(0:500),DOXYCW(0:500), / NSGROWW(0:500),NSGROWB(0:500),AEROHTGW(0:500),ANHTGW(0:500), / HTGROWW(0:500),HTDEATHW(0:500),NDEATHW(0:500),NDEATHB(0:500),
213
/ HETEROB(0:500),AEROHTGB(0:500),ANHTGB(0:500),HTGROWB(0:500), / DOXYCB(0:500),DOXYW(0:500),DOXYB(0:500),NH4CW(0:500), / NH4CB(0:500),TOCCW(0:500),TOCCB(0:500),NO3CW(0:500), / NO3W(0:500),NO3B(0:500),NO3CB(0:500)
REAL NITROSIW,NITROSIB,NDRATEW,NDRATEB,NDOHSATW,NDOHSATB,NMAXGRW, / NMAXGRB,NNH4HSCW,NNH4HSCB,HETEROIW,HETEROIB,AEMAXGRW,AEMAXGRB, / ANMAXGRW,ANMAXGRB,HTDRW,HTDRB,HTDOHSCW,HTDOHSCB,HNO3HSCW, / HNO3HSCB,HORGHSCW,HORGHSCB,WATTEMPW,WATTEMPB,NSTEMPFW, / HTTEMPFW,NSTEMPFB,HTTEMPFB
INTEGER WETTYPE,J,K,M
C READ IN DAILY INPUT VALUESC
IF (WETTYPE.EQ.0) THENREAD(6,*) WATTEMPW,WATTEMPB
ELSEIF (WETTYPE.EQ.1) THENREAD(6,*) WATTEMPB
ENDIF
C BACTERIA TEMPERATURE FACTORC
IF (WETTYPE.EQ.0) THENIF (WATTEMPW.LT.0) THEN
NSTEMPFW=0.HTTEMPFW=0.
ELSEIF (WATTEMPW.GE.0. .AND. WATTEMPW.LE.15.) THENNSTEMPFW=WATTEMPW/15.HTTEMPFW=WATTEMPW/15.
ELSEIF (WATTEMPW.GT.15. .AND. WATTEMPW.LE.35.) THENNSTEMPFW=1.HTTEMPFW=1.
ELSEIF (WATTEMPW.GT.35. .AND. WATTEMPW.LE.40.) THENNSTEMPFW=((40.-WATTEMPW)/5.)HTTEMPFW=((40.-WATTEMPW)/5.)
ELSEIF (WATTEMPW.GT.40.) THENNSTEMPFW=0.0HTTEMPFW=0.0
ENDIFENDIF
IF (WATTEMPB.LT.0) THENNSTEMPFB=0.HTTEMPFB=0.
ELSEIF (WATTEMPB.GE.0. .AND. WATTEMPB.LE.15.) THENNSTEMPFB=WATTEMPB/15.HTTEMPFB=WATTEMPB/15.
ELSEIF (WATTEMPB.GT.15. .AND. WATTEMPB.LE.35.) THENNSTEMPFB=1.HTTEMPFB=1.
ELSEIF (WATTEMPB.GT.35. .AND. WATTEMPB.LE.40.) THENNSTEMPFB=((40.-WATTEMPB)/5.)HTTEMPFB=((40.-WATTEMPB)/5.)
ELSEIF (WATTEMPW.GT.40.) THENNSTEMPFB=0.0HTTEMPFB=0.0
ENDIF
C ANAEROBE FRACTION DETERMINATIONC
IF (WETTYPE.EQ.0) THENIF (DOXYCW(J-1).GE.0. .AND. DOXYCW(J-1).LE.8.) THEN
ANFRACW(J)=0.8-(DOXYCW(J-1)*0.075)ELSEIF (DOXYCW(J-1).GT.8.) THEN
ANFRACW(J)=0.2ENDIFENDIF
IF (DOXYCB(J-1).GE.0. .AND. DOXYCB(J-1).LE.8.) THENANFRACB(J)=0.8-(DOXYCB(J-1)*0.075)
214
ELSEIF (DOXYCB(J-1).GT.8.) THENANFRACB(J)=0.2
ENDIF
C NITROSOMONAS GROWTH AND DEATHC
IF (WETTYPE.EQ.0) THEN NSGROWW(J)=(NMAXGRW*((NH4CW(J-1))/(NNH4HSCW+ / (NH4CW(J-1))))*((DOXYCW(J-1))/(NDOHSATW+(DOXYCW(J-1)))) / *NITROSOW(J-1)*NSTEMPFW)
NDEATHW(J)=NDRATEW*NITROSOW(J-1)ENDIF NSGROWB(J)=(NMAXGRB*((NH4CB(J-1))/
/ (NNH4HSCB+(NH4CB(J-1))))*((DOXYCB(J-1))/(NDOHSATB+ / (DOXYCB(J-1))))*NITROSOB(J-1)*NSTEMPFB)
NDEATHB(J)=NDRATEB*NITROSOB(J-1)
C NITROSOMONAS MASS BALANCEC
IF (WETTYPE.EQ.0) THENNITROSOW(J)=NITROSOW(J-1)+NSGROWW(J)-NDEATHW(J)
ENDIFNITROSOB(J)=NITROSOB(J-1)+NSGROWB(J)-NDEATHB(J)
C HETEROTROPHS GROWTH AND DEATHC
IF (WETTYPE.EQ.0) THENAEROHTGW(J)=(AEMAXGRW*(TOCCW(J-1))/
/ ((TOCCW(J-1))+HORGHSCW)*HTTEMPFW*(DOXYCW(J-1)) / /((DOXYCW(J-1))+HTDOHSCW)*(1-ANFRACW(J))*HETEROW(J-1))
ANHTGW(J)=((ANMAXGRW*TOCCW(J-1))/ / ((TOCCW(J-1))+HORGHSCW)*HTTEMPFW*HTDOHSCW/(HTDOHSCW+ / (DOXYCW(J-1)))*(NO3CW(J-1))/((NO3CW(J-1)) / +HNO3HSCW)*ANFRACW(J)*HETEROW(J-1))
HTGROWW(J)=AEROHTGW(J)+ANHTGW(J)HTDEATHW(J)=HTDRW*HETEROW(J-1)ENDIF
AEROHTGB(J)=(AEMAXGRB*(TOCCB(J-1))/ / ((TOCCB(J-1))+HORGHSCB)*HTTEMPFB*(DOXYCB(J-1))/ / ((DOXYCB(J-1))+HTDOHSCB)*(1-ANFRACB(J))*HETEROB(J-1))
ANHTGB(J)=((ANMAXGRB*TOCCB(J-1))/ / ((TOCCB(J-1))+HORGHSCB)*HTTEMPFB*HTDOHSCB/(HTDOHSCB+ / (DOXYCB(J-1)))*(NO3CB(J-1))/((NO3CB(J-1)) / +HNO3HSCB)*ANFRACB(J)*HETEROB(J-1))
HTGROWB(J)=AEROHTGB(J)+ANHTGB(J)HTDEATHB(J)=HTDRB*HETEROB(J-1)
C MASS BALANCE FOR HETEROTROPHSC
IF (WETTYPE.EQ.0)THENHETEROW(J)=HETEROW(J-1)+HTGROWW(J)-HTDEATHW(J)ENDIFHETEROB(J)=HETEROB(J-1)+HTGROWB(J)-HTDEATHB(J)
C OUTPUT DATAC
IF (M.EQ.1 .AND. J.EQ.1) THENWRITE (16,*) 'OUTPUT DATA FOR BACTERIA COUNTS IN WETLAND'WRITE(16,100)
100 FORMAT (T12,'NITROSOW',T27,'NITROSOB',T43,'HETEROW', / T57,'HETEROB')
WRITE(16,125)M,NITROSOW(0),NITROSOB(0),HETEROW(0),HETEROB(0) 125 FORMAT (T1, (I2),T5,(' 0'),T8,4(F12.2,3X))
ENDIFEND
215
NITROGEN SUBMODELS
SUBROUTINE NITSTR (POINT,HYDTYPE,MTCDON,NSYIELDB,M,WETTYPE, / NITFRATE,VOLATR,DONCONC,PONCONC,NO3CONC,NH4CONC,MTCNO3, / DONINITW,DONINITB,IMMINITW,IMMINITB,NH4INITW,MTCNH4,HTNO3YB, / NH4INITB,NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,PONCOUT, / BIOMCN,HTNO3YW,MICRONC,NSYIELDW,ONPARTF,PEATNC,BIOMPN,RESTHN, / DADDON,WADDON,DADPON,WADPON,DADNH4,WADNH4,DADNO3,WADNO3, / ORGNRC,NH4RC,NO3RC,PONSIZE,ATDEP,NITFIX,VOLAT,PONRES,PONFALL, / PERCINF,DONININC,NH4ININC,NO3ININC)
INTEGER ATDEP,NITFIX,VOLAT,CHECK1,CHECK2,CHECK7 / CHECK3,CHECK4,CHECK5,CHECK6,POINT,WETTYPE,PERCINF
REAL NITFRATE,VOLATR,DONCONC,PONCONC,NO3CONC,NH4CONC, / DONINITW,DONINITB,IMMINITW,IMMINITB,NH4INITW,MTCNH4, / NH4INITB,NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT, / BIOMCN,HTNO3YW,MICRONC,NSYIELDW,NSYIELDB,ONPARTF,PEATNC, / DADDON,WADDON,DADPON,WADPON,DADNH4,WADNH4,DADNO3, / WADNO3,ORGNRC,NH4RC,NO3RC,MTCDON,PONSIZE,RESTHN,PONCOUT, / MTCNO3,BIOMPN,PONFALL,HTNO3YB,DONININC,NH4ININC,NO3ININC
IF(M.EQ.1) THEN
C READ IN INITIAL NITROGEN DATAC
IF (WETTYPE.EQ.0) THENREAD (8,*) ATDEP,NITFIX,VOLAT,
/ DONINITW,DONINITB,IMMINITW,IMMINITB,NH4INITW,NH4INITB, / NO3INITW,NO3INITB,PONINITW,PONINITB,REFNINIT,BIOMCN, / BIOMPN,HTNO3YW,HTNO3YB,MICRONC,NSYIELDW,NSYIELDB, / ONPARTF,PEATNC,PONRES,PONFALL,MTCDON,MTCNH4, / MTCNO3,PONSIZE,RESTHN,PONCOUT
ELSEIF(WETTYPE.EQ.1) THENREAD(8,*) ATDEP,NITFIX,VOLAT,
/ DONINITB,IMMINITB,NH4INITB,NO3INITB,PONINITB,REFNINIT, / BIOMCN,BIOMPN,HTNO3YB,MICRONC,NSYIELDB,ONPARTF,PEATNC
ENDIF
C NITROGEN FIXATIONC
IF (NITFIX.EQ.0) THENNITFRATE=0.0
ELSEIF (NITFIX.EQ.1) THENREAD(8,*) NITFRATE
ENDIF
C VOLATILIZATIONC
IF (VOLAT.EQ.0) THENVOLATR=0.0
ELSEREAD(8,*) VOLATR
ENDIF
C ATMOSPHERIC DEPOSITIONC
IF (ATDEP.EQ.0) THENDADDON=0.0WADDON=0.0DADPON=0.0WADPON=0.0DADNH4=0.0WADNH4=0.0DADNO3=0.0WADN03=0.0
ELSEREAD(8,*)DADDON,WADDON,DADPON,WADPON,
/ DADNH4,WADNH4,DADNO3,WADNO3ENDIF
216
C POINT FLOWC
IF (POINT.EQ.1) THENREAD (8,*) DONCONC,PONCONC,NO3CONC,NH4CONC
ELSEDONCONC=0.0PONCONC=0.0NO3CONC=0.0NH4CONC=0.0
ENDIF
C PERCOLATION/INFILITRATIONC
IF (PERCINF.EQ.0) THENDONININC=0.0NH4ININC=0.0NO3ININC=0.0
ELSEIF (PERCINF.EQ.1) THENREAD (8,*) DONININC,NH4ININC,NO3ININC
ENDIF
C INPUT TYPEC
IF (HYDTYPE.EQ.0) THENORGNRC=0.0NH4RC=0.0NO3RC=0.0
ELSEIF (HYDTYPE.EQ.1) THENREAD (8,*) ORGNRC,NH4RC,NO3RC
ENDIFENDIF
C CHECK FOR PARAMETER VALUE CHANGES BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (8,*)c,CHECK1,CHECK2,CHECK3,CHECK4,CHECK5,CHECK6,CHECK7
IF (CHECK1.EQ.0) THENIF (WETTYPE.EQ.0) THENREAD(8,*)BIOMCN,BIOMPN,HTNO3YW,HTNO3YB,MICRONC,NSYIELDW,
/ NSYIELDB,ONPARTF,PEATNC,PONRES,PONFALL,MTCDON, / MTCNH4,MTCNO3,PONSIZE,RESTHN,PONCOUT
ELSEIF(WETTYPE.EQ.1)THENREAD(8,*)BIOMCN,BIOMPN,HTNO3YB,MICRONC,
/ NSYIELDB,ONPARTF,PEATNCENDIF
ENDIF
IF (CHECK2.EQ.0) THENREAD(8,*) NITFRATE
ENDIFIF (CHECK3.EQ.0) THEN
READ(8,*) VOLATRENDIFIF (CHECK4.EQ.0) THEN
READ (8,*) DADDON,WADDON,DADPON,WADPON, / DADNH4,WADNH4,DADNO3,WADNO3
ENDIFIF (CHECK5.EQ.0) THEN
READ (8,*) DONCONC,PONCONC,NO3CONC,NH4CONCENDIFIF (CHECK6.EQ.0) THEN
READ(8,*) DONININC,NH4ININC,NO3ININCENDIFIF (CHECK7.EQ.0) THEN
READ(8,*) ORGNRC,NH4RC,NO3RCENDIFENDIFEND
C
217
CSUBROUTINE NITTIME (WATINPUT,PRECIP,OUTFLOW,BIOMGROW,DONW,DONB,
/ DONCW,DONCB,IMMNW,IMMNB,NH4W,NH4B,NH4CW,NH4CB,NO3W,NO3B, / NO3CW,NO3CB,PONW,PONB,PONCW,PONCB,NO3AT,NO3,MTCNH4, / REFN,TONW,TONB,TONCW,TONCB,BIOMCN,HTNO3YW,MICRONC,BIOMPN, / ONPARTF,PEATNC,DOCLEACH,PHYSDEGC,PONINITB,REFNINIT, / ATDEP,NITFIX,NITFRATE,VOLAT,VOLATR,WATERVOL,WATERB, / PONRES,PONFALL,DONCONC,PONCONC,NO3CONC,J,K,M,HTGROWW,HTGROWB, / NH4CONC,BIOMREML,BIOMREMD,DAYREML,DAYREMD,ABIOREMD, / ABIOREML,MTCDON,HI,HT,HYDTYPE,MANNC,PONSIZE,MTCNO3, / RESTHN,APFLOW,TOCW,MICTCNW,DOCW,TOCB,MICTCNB,DOCB,HTNO3YB, / HTYIELDW,HTYIELDB,NSGROWB,POCW,POCB,HTDEATHW,PEATACRW, / NSDEATHW,HTDEATHB,NSDEATHB,DONIMM,PONIMMW,DONIMMB,PEATACRB, / PON,NSGROWW,NSYIELDW,NSYIELDB,NH4ATDEP,ANHTGW,ANHTGB, / WATERVEL,DADDON,WADDON,DADPON,WADPON,DADNH4,NO3OUT,NH4OUT, / WADNH4,DADNO3,WADNO3,NO3RC,NH4RC,ORGNRC,WETTYPE,DONOUT,PONOUT, / NH4EC,NO3EC,DONEC,PONEC,PONCOUT,PRATEUP,DONININC,NH4ININC, / NO3ININC,PERCINFT)C
REAL OUTFLOW(0:500),HTGROWW(0:500),HTGROWB(0:500), / IMMNB(0:500),NH4CW(0:500),PONCB(0:500),WATERVOL(0:500), / WATERB(0:500),HI(0:500),HT(0:500),DONW(0:500),DONB(0:500), / IMMNW(0:500),NH4B(0:500),NO3W(0:500),NO3B(0:500),PONW(0:500), / PONB(0:500),TONCW(0:500),TONCB(0:500),DONOUT(0:500), / REFN(0:500),TONW(0:500),TONB(0:500),DONIN(0:500), / HNH4IMMW(0:500),HNH4IMMB(0:500),DONIMMW(0:500),DONIMMB(0:500), / DONMINW(0:500),DONMINB(0:500),FIXNIT(0:500),DONAT(0:500), / PRECIP(0:500),WATINPUT(0:500),DONMT(0:500),NH4W(0:500), / NLEACH(0:500),NITUPB(0:500),AMMUPB(0:500),DONCW(0:500), / PONIMMW(0:500),PONIMMB(0:500),DEATHB(0:500),IMMNOUT(0:500), / IMMNOUTD(0:500),NH4IN(0:500),NH4OUT(0:500),NITRIFW(0:500), / NITRIFB(0:500),NH4MT(0:500),NH4AT(0:500),VOLATIZ(0:500), / NO3IN(0:500),NO3OUT(0:500),DENITW(0:500),DENITB(0:500), / NO3AT(0:500),NO3MT(0:500),DEATHW(0:500),PONIN(0:500), / PEATNACW(0:500),PEATNACB(0:500),PONMINW(0:500),PONMINB(0:500), / PONAT(0:500),PONSET(0:500),PONRE(0:500),DONCB(0:500), / NH4CB(0:500),NO3CW(0:500),NO3CB(0:500),PONCW(0:500), / TOCW(0:500),MICTCNW(0:500),DOCW(0:500),NITUPW(0:500), / TOCB(0:500),HTDEATHW(0:500),NO3(0:500),ANHTGB(0:500), / MICTCNB(0:500),DOCB(0:500),HTYIELDW(0:500),HTYIELDB(0:500), / BIOMGROW(0:500),NSGROWB(0:500),POCW(0:500),POCB(0:500), / NSDEATHW(0:500),HTDEATHB(0:500),DONIMM(0:500),PON(0:500), / NSGROWW(0:500),NH4ATDEP(0:500),ANHTGW(0:500),DOCLEACH(0:500), / PEATACW(0:500),PEATACB(0:500),PHYSDEGC(0:500), / NSDEATHB(0:500),PONOUT(0:500),NH4EC(0:500),NO3EC(0:500), / DONEC(0:500),PONEC(0:500),AMMUPW(0:500),APFLOW(0:500), / PERCINFT(0:500),DONPERC(0:500),NH4PERC(0:500),NO3PERC(0:500)
INTEGER NITFIX,ATDEP,VOLAT,DAYREMD,DAYREML,BIOMREML,BIOMREMD, / HYDTYPE,WETTYPE,OUTLET
REAL MTCNH4,NSYIELDW,MTCNO3,NSYIELDB,WATERVEL,NH4RC, / BIOMCN,HTNO3YW,HTNO3YB,MICRONC,ONPARTF,PEATNC,BIOMPN / NITFRATE,VOLATR,PONRES,NO3RC,ORGNRC,DONINITW,DONINITB, / PONFALL,DONCONC,PONCONC,NO3CONC,NH4CONC,REFNINIT, / ABIOREMD,ABIOREML,MTCDON,NO3INITW,NO3INITB,PONINITW,PONINITB, / RESUSPV,MANNC,PONSIZE,RESTHN,DADNH4,WADNH4,DADNO3,WADNO3, / DADDON,WADDON,DADPON,WADPON,NH4INITB,NH4INITW, / LENGTH,WIDTH,HO,HB,SO,ORGNINC,NH4INC,NO3INC, / HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,KHCOEF,PONCOUT, / TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,PRATEUP
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,S0COMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
/ TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK, / KHCOEF
C READ INPUT VALUES FOR DAILY INFLOWC
IF (VOLAT.EQ.0. .AND. HYDTYPE.EQ.0) THENREAD(8,*)ORGNINC,NH4INC,NO3INC
ELSE IF (VOLAT.EQ.1 .AND. HYDTYPE.EQ.0) THEN
218
READ(8,*)ORGNINC,NH4INC,NO3INC,PHELSEIF (VOLAT.EQ.1 .AND. HYDTYPE.EQ.1) THEN
READ (8,*) PHENDIFNLEACH(J)=DOCLEACH(J)/BIOMCN
C DON PROCESSESC
IF (WETTYPE.EQ.0) THENIF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENDONOUT(J)=DONW(J-1)*OUTFLOW(J)/(WATERVOL(J-1))ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THENDONOUT(J)=DONW(J-1)*.75ENDIFELSEIF (WETTYPE.EQ.1) THENDONOUT(J)=DONB(J-1)*OUTFLOW(J)/WATERB(J-1)ENDIFDONEC(J)=DONOUT(J)/OUTFLOW(J)
IF (HYDTYPE.EQ.0) THENDONIN(J)=(1-ONPARTF)*(WATINPUT(J)*ORGNINC+APFLOW(J)*DONCONC)
ELSEIF (HYDTYPE.EQ.1) THENDONIN(J)=(1-ONPARTF)*(WATINPUT(J)*ORGNRC+APFLOW(J)*DONCONC)
ENDIF
IF (WETTYPE.EQ.0)THENHNH4IMMW(J)=MICRONC*HTGROWW(J)*NH4W(J-1)/(TONW(J-1)+NH4W(J-1))ENDIFHNH4IMMB(J)=MICRONC*HTGROWB(J)*NH4B(J-1)/(TONB(J-1)+NH4B(J-1))
IF (WETTYPE.EQ.0) THENIF ((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
DONIMMW(J)=(DONW(J-1)/(TONW(J-1)+NH4W(J-1))*MICRONC*HTGROWW(J))ELSE
DONIMMW(J)=(DOCW(J-1)/TOCW(J-1))*(MICRONC*HTGROWW(J)-HNH4IMMW(J))ENDIF
ENDIFIF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THEN
DONIMMB(J)=(DONB(J-1)/(TONB(J-1)+NH4B(J-1))*MICRONC*HTGROWB(J))ELSE
DONIMMB(J)=(DOCB(J-1)/TOCB(J-1))*(MICRONC*HTGROWB(J)-HNH4IMMB(J))ENDIF
IF (WETTYPE.EQ.0) THENIF (((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) .OR. (DOCW(J-1).LT.0.1)) THEN
DONMINW(J)=0.0ELSE
DONMINW(J)=(DOCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J) / *DONW(J-1)/DOCW(J-1)-DONIMMW(J))
IF (DONMINW(J).GE.0) THENDONMINW(J)=DONMINW(J)
ELSEDONMINW(J)=0
ENDIFENDIFENDIF
IF (((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) .OR. (DOCB(J-1).LT.0.1)) THENDONMINB(J)=0.0
ELSEDONMINB(J)=(DOCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
/ *DONB(J-1)/DOCB(J-1)-DONIMMB(J))IF(DONMINB(J).LE. 0) THEN
DONMINB(J)=0.0ENDIF
ENDIF
IF (NITFIX.EQ.1) THENFIXNIT(J)=NITFRATE*LENGTH*WIDTH
219
ENDIF
IF (ATDEP.EQ.1) THENIF (PRECIP(J).EQ.0.) THEN
DONAT(J)=DADDON*LENGTH*WIDTHELSE
DONAT(J)=PRECIP(J)*WADDONENDIF
ENDIF
IF (WETTYPE.EQ.0) THENDONMT(J)=864*LENGTH*WIDTH*MTCDON*(DONCB(J-1)-DONCW(J-1))ENDIF
IF (PERCINFT(J).GT.0.) THENDONPERC(J)=DONB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THENDONPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THENDONPERC(J)=PERCINFT(J)*DONININC
ENDIF
C DON MASS BALANCESC
IF (WETTYPE.EQ.0) THENDONW(J)=DONW(J-1)+DONIN(J)-DONOUT(J)-DONIMMW(J)
/ -DONMINW(J)+FIXNIT(J)+DONAT(J)+DONMT(J)DONB(J)=DONB(J-1)-DONIMMB(J)-DONMINB(J)-DONMT(J)-DONPERC(J)
/ +NLEACH(J)ELSEIF (WETTYPE.EQ.1) THEN
DONB(J)=DONB(J-1)+DONIN(J)+NLEACH(J)-DONOUT(J)-DONIMMB(J)- / DONMINB(J)+FIXNIT(J)+DONAT(J)-DONPERC(J)
ENDIF
C NO3 PROCESSESC
IF (WETTYPE.EQ.0) THENIF (NO3B(J-1).GT.(PRATEUP*BIOMGROW(J)/BIOMPN)) THEN
NITUPB(J)=PRATEUP*BIOMGROW(J)/BIOMPNELSE
NITUPB(J)=NO3B(J-1)ENDIF
IF (NO3W(J-1) .GT. ((1-PRATEUP)*BIOMGROW(J)/BIOMPN)) THENNITUPW(J)=(BIOMGROW(J)*(1-PRATEUP))/BIOMPN
ELSENITUPW(J)=NO3W(J-1)
ENDIFENDIFIF (WETTYPE.EQ.1) THENIF (NO3B(J-1).GT.(BIOMGROW(J)/BIOMPN)) THEN
NITUPB(J)=BIOMGROW(J)/BIOMPNELSE
NITUPB(J)=NO3B(J-1)ENDIFENDIF
C AMMONIUM PROCESSESC
IF (WETTYPE.EQ.0) THENAMMUPW(J)=NSGROWW(J)*MICRONC+(BIOMGROW(J)/BIOMPN*(1-PRATEUP)
/ -(NITUPW(J)))+HNH4IMMW(J)AMMUPB(J)=NSGROWB(J)*MICRONC+((BIOMGROW(J)*PRATEUP/BIOMPN)
/ -(NITUPB(J)))+HNH4IMMB(J)ELSEIF (WETTYPE.EQ.1) THENAMMUPB(J)=NSGROWB(J)*MICRONC+(BIOMGROW(J)/BIOMPN
/ -(NITUPB(J)))+HNH4IMMB(J)ENDIF
220
IF (WETTYPE.EQ.0) THENIF ((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1)) THEN
PONIMMW(J)=PONW(J-1)/(TONW(J-1)+NH4W(J-1))*MICRONC*HTGROWW(J)ELSE
PONIMMW(J)=POCW(J-1)/TOCW(J-1)*(MICRONC*HTGROWW(J)- / (HNH4IMMW(J)))
ENDIF
DEATHW(J)=.2*PHYSDEGC(J)/BIOMCN+MICRONC*(HTDEATHW(J)+NSDEATHW(J))ENDIF
IF ((TOCB(J-1)/TONB(J-1)).GT.MICTCNB(J-1)) THENPONIMMB(J)=PONB(J-1)/(TONB(J-1)+NH4B(J-1))*MICRONC*HTGROWB(J)
ELSEPONIMMB(J)=POCB(J-1)/(TOCB(J-1))*(MICRONC*HTGROWB(J)-
/ (HNH4IMMB(J)))ENDIFDEATHB(J)=.8*PHYSDEGC(J)/BIOMCN+MICRONC*(HTDEATHB(J)+NSDEATHB(J))
CC HERE I AM GOING TO PUT INFORMATION ON THE REMOVAL OF THEC BIOMASS FROM THE SYSTEM WHICH WILL THEN REMOVE A CERTAINC PERCENTAGE OF IMMOBILIZED N THE SYSTEM. WHAT I NEEDC TO FIGURE OUT IS THE AMOUNT THAT NEEDS TO BE REMOVED BASEDC ON THE SEPARATION RATIO BETWEEN THE PLANTS AND MICROBESC BUT WHAT IS THIS SEPARATION
IF (BIOMREML.EQ.0) THENIMMNOUT(J)=0.0
ELSE IF (BIOMREML.EQ.1) THENIF (J.EQ.DAYREML) THEN
IMMNOUT(J)=ABIOREML/BIOMPNELSE
IMMNOUT(J)=0.0ENDIF
ENDIFC
IF (BIOMREMD.EQ.0) THENIMMNOUTD(J)=0.0
ELSE IF (BIOMREMD.EQ.1) THENIF (J.EQ.DAYREMD) THEN
IMMNOUTD(J)=ABIOREMD/BIOMPNELSE
IMMNOUTD(J)=0.0ENDIF
ENDIFC PUT THE RK FOR IMMOBILIZED N HERE
IF (WETTYPE.EQ.0) THENIMMNW(J)=IMMNW(J-1)+DONIMMW(J)+PONIMMW(J)-DEATHW(J)+NITUPW(J)+
/ AMMUPW(J)IMMNB(J)=IMMNB(J-1)+NITUPB(J)+DONIMMB(J)+AMMUPB(J)
/ +PONIMMB(J)-DEATHB(J)-IMMNOUT(J)-IMMNOUTD(J)ELSEIF (WETTYPE.EQ.1) THENIMMNB(J)=IMMNB(J-1)+NITUPB(J)+DONIMMB(J)+AMMUPB(J)+PONIMMB(J)
/ -DEATHB(J)-IMMNOUT(J)-IMMNOUTD(J)ENDIF
C NH4IN=NH4 INFLUENTC PONMINW=PON MINERALIZATION //PONMINBC NH4OUT=NH4 OUTFLOWC AMMUP=ALREADY CALCULATEDC NITRIFW=NITRFICATION RATE //NITRIFB
IF (HYDTYPE.EQ.0) THENNH4IN(J)=(NH4INC*WATINPUT(J))+NH4CONC*APFLOW(J)
ELSEIF (HYDTYPE.EQ.1) THENNH4IN(J)=NH4RC*WATINPUT(J)+NH4CONC*APFLOW(J)
ENDIFC
IF (WETTYPE.EQ.0) THENIF (OUTFLOW(J).LT.WATERVOL(J-1)) THENNH4OUT(J)=NH4W(J-1)/(WATERVOL(J-1))*OUTFLOW(J)ELSEIF (OUTFLOW(J).GE.WATERVOL(J-1)) THENNH4OUT(J)=NH4W(J-1)*.75
221
ENDIFELSEIF (WETTYPE.EQ.1) THENNH4OUT(J)=NH4B(J-1)/WATERB(J-1)*OUTFLOW(J)ENDIFNH4EC(J)=NH4OUT(J)/OUTFLOW(J)IF (WETTYPE.EQ.0) THENNITRIFW(J)=NSGROWW(J)/NSYIELDWENDIFNITRIFB(J)=NSGROWB(J)/NSYIELDB
CC NEED A DIFFUSION EQUATION HERE THE EQUATION SHOULD BASICALLYC GO WITH A DIFFUSION COEFFICIENT AND THEN ID ONE SIDE ISC HIGHER THAN THE OTHER THE GRADIENT GOES IN AC CERTAIN DIRECTION
IF (WETTYPE.EQ.0) THENNH4MT(J)=864*LENGTH*WIDTH*MTCNH4*(NH4CB(J-1)-NH4CW(J-1))
c nh4mt(j)=-.005*nh4w(j-1)ENDIF
CC VOLATILIZATION AND ATDEP WILL BE ADDED HEREC
IF (ATDEP.EQ.1) THENIF (PRECIP(J).EQ.0.) THEN
NH4AT(J)=DADNH4*LENGTH*WIDTHELSE
NH4AT(J)=PRECIP(J)*WADNH4ENDIF
ENDIFIF(VOLAT.EQ.1) THEN
IF (PH.GT.8.) THENC this is form the paper by rao and jessup on the Nflood model
PHDEP=5.8*(10**(PH-10))C VOLATIZ = THE AMOUNT OF VOLATIZATION, PHDEP IS THE PARTITIONINGC BETWEEN NH3 AND NH4+, AND VOLAT C IS A RATE IN
IF (WETTYPE.EQ.0) THENVOLATIZ(J)=VOLATR*PHDEP*NH4W(J-1)ELSEIF (WETTYPE.EQ.1) THENVOLATIZ(J)=VOLATR*PHDEP*NH4B(J-1)ENDIFENDIF
ENDIFC NEED TO HAVE A PH CHECK WHERE IT ONLY COUNTS IF THE PH ISC ABOVE A CERTAIN NUMBER AND PERHAPS A MINIMUM NH3 CONC,C THAT IS NECESSARY FOR THE VOLATILIZATION TO OCCURC
IF (PERCINFT(J).GT.0.) THENNH4PERC(J)=NH4B(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THENNH4PERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THENNH4PERC(J)=PERCINFT(J)*NH4ININC
ENDIFC
IF (WETTYPE.EQ.0) THENIF((TOCW(J-1)/TONW(J-1)).GT.MICTCNW(J-1) .OR.
/ (POCW(J-1).LT.0.1)) THENPONMINW(J)=0.0
ELSEPONMINW(J)=POCW(J-1)/TOCW(J-1)*HTGROWW(J)/HTYIELDW(J)
/ *PONW(J-1)/POCW(J-1)- / (POCW(J-1)/TOCW(J-1)*(MICRONC*HTGROWW(J)-HNH4IMMW(J)))
ENDIFENDIFIF((TOCB(J-1)/TONB(J-1)) .GT. MICTCNB(J-1) .OR.
/ (POCB(J-1) .LT. 0.1)) THENPONMINB(J)=0.0
ELSEPONMINB(J)=POCB(J-1)/TOCB(J-1)*HTGROWB(J)/HTYIELDB(J)
/ *PONB(J-1)/POCB(J-1)-
222
/ (POCB(J-1)/TOCB(J-1)*(MICRONC*HTGROWB(J)-HNH4IMMB(J)))IF (PONMINB(J).LE.0) THENPONMINB(J)=0.0ENDIF
ENDIF
C AMMONIUM MASS BALANCEC
IF (WETTYPE.EQ.0) THENNH4W(J)=NH4W(J-1)+NH4IN(J)+DONMINW(J)+PONMINW(J)-NH4OUT(J)
/ -NITRIFW(J)+NH4ATDEP(J)-VOLATIZ(J)+NH4MT(J)-AMMUPW(J)
NH4B(J)=NH4B(J-1)+DONMINB(J)+PONMINB(J)-AMMUPB(J)-NITRIFB(J)-NH4MT(J)
ELSEIF (WETYPE.EQ.1) THENNH4B(J)=NH4B(J-1)+NH4IN(J)+DONMINB(J)+PONMINB(J)-NH4OUT(J)-
/ AMMUPB(J)-NITRIFB(J)+NH4ATDEP(J)-VOLATIZ(J)ENDIF
C NITRATE PROCESSESC
IF (HYDTYPE.EQ.0) THENNO3IN(J)=NO3INC*WATINPUT(J)+NO3CONC*APFLOW(J)
ELSEIF (HYDTYPE.EQ.1)THENNO3IN(J)=NO3RC*WATINPUT(J)+NO3CONC*APFLOW(J)
ENDIF
IF (WETTYPE.EQ.0) THENIF (OUTFLOW(J) .LT. WATERVOL (J-1)) THEN
NO3OUT(J)=NO3W(J-1)/(WATERVOL(J-1))*OUTFLOW(J)ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THEN
NO3OUT(J)=NO3W(J-1)*.75ENDIFELSEIF (WETTYPE.EQ.1) THEN
NO3OUT(J)=NO3B(J-1)/WATERB(J-1)*OUTFLOW(J)ENDIFNO3EC(J)=NO3OUT(J)/OUTFLOW(J)
IF (WETTYPE.EQ.0) THENDENITW(J)=ANHTGW(J)/HTNO3YW
ENDIFDENITB(J)=ANHTGB(J)/HTNO3YB
IF (ATDEP.EQ.1) THENIF (PRECIP(J).EQ.0.) THEN
NO3AT(J)=DADNO3*LENGTH*WIDTHELSE
NO3AT(J)=PRECIP(J)*WADNO3ENDIF
ENDIF
IF (WETTYPE.EQ.0) THENNO3MT(J)=864*LENGTH*WIDTH*MTCNO3*(NO3CB(J-1)-NO3CW(J-1))ENDIF
IF (PERCINFT(J).GT.0.) THENNO3PERC(J)=NO3B(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THENNO3PERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THENNO3PERC(J)=PERCINFT(J)*NO3ININC
ENDIF
C MASS BALANCE FOR NITRATEC
IF(WETTYPE.EQ.0) THENNO3W(J)=NO3W(J-1)+NO3IN(J)+NITRIFW(J)-NO3OUT(J)-DENITW(J)
/ +NO3AT(J)+NO3MT(J)-NITUPW(J)NO3B(J)=NO3B(J-1)+NITRIFB(J)-NITUPB(J)-DENITB(J)-NO3MT(J)--NO3PERC(J)
223
ELSEIF (WETTYPE.EQ.1) THENNO3B(J)=NO3B(J-1)+NO3IN(J)+NITRIFB(J)-NITUPB(J)-NO3OUT(J)-
/ DENITB(J)+NO3AT(J)-NO3PERC(J)ENDIF
C PON PROCESSESC
IF (HYDTYPE.EQ.0) THENPONIN(J)=ONPARTF*(WATINPUT(J)*ORGNINC+APFLOW(J)*PONCONC)
ELSEIF (HYDTYPE.EQ.1) THENPONIN(J)=(ONPARTF)*(WATINPUT(J)*ORGNRC+APFLOW(J)*PONCONC)
ENDIF
IF (WETTYPE.EQ.0) THENPEATNACW(J)=PEATACRW*PEATNCENDIFPEATNACB(J)=PEATACRB*PEATNC
IF (ATDEP.EQ.1) THENIF (PRECIP(J).EQ.0.) THEN
PONAT(J)=DADPON*LENGTH*WIDTHELSE
PONAT(J)=PRECIP(J)*WADPONENDIF
ENDIF
IF (WETTYPE.EQ.0) THENIF(PONFALL.LT.(HI(J-1)-HT(J-1)))THEN
PONSET(J)=PONW(J-1)*(PONFALL)/(HI(J-1)-HT(J-1))ELSEIF (PONFALL.GE.(HI(J-1)-HT(J-1)))THEN
PONSET(J)=PONW(J-1)ENDIF
RESUSPV=7.2*(((PONFALL/86400)**1/3)* / ((HI(J-1)-HT(J-1))**(1./6.))/(MANNC* / ((PONSIZE**(2./3.)))))
IF (WATERVEL.GE.RESUSPV) THENPONRE(J)=PONB(J-1)*PONRES*RESTHN/(HT(J-1)-HB)ELSEPONRE(J)=0ENDIFENDIF
IF (WETTYPE.EQ.0) THENIF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
/ OUTLET.EQ.5) THENIF (PONFALL .GT. (HI(J-1)-HT(J-1))) THENPONOUT(J)=0.0ELSEIF (PONFALL .LT. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J).LT.WATERVOL(J-1)) THENPONOUT(J)=PONCOUT*((PONW(J-1))*OUTFLOW(J)/WATERVOL(J-1))
ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THENPONOUT(J)=.5*PONCOUT*((PONW(j-1))*OUTFLOW(J)/WATERVOL(J-1))
ENDIFENDIFELSEIF (OUTLET.EQ.4)THENIF (PONFALL .GT. (HI(J-1)-HT(J-1))) THENPONOUT(J)=0.ELSEIF (PONFALL .LT. (HI(J-1)-HT(J-1)))THEN
IF (OUTFLOW(J).LE.WATERVOL(J-1)) THENPONOUT(J)=((PONW(J-1))*OUTFLOW(J)/WATERVOL(J-1)*PONCOUT)
ELSEIF (OUTFLOW(J).GE.WATERVOL(J-1)) THENPONOUT(J)=((PONW(J-1))*.5*OUTFLOW(J-1)/WATERVOL(J-1)*PONCOUT)
ENDIFENDIFENDIF
ENDIFPONEC(J)=PONOUT(J)/OUTFLOW(J)
C MASS BALANCE FOR PON
224
CIF (WETTYPE.EQ.0) THENPONW(J)=PONW(J-1)+DEATHW(J)+PONIN(J)-PEATNACW(J)-PONIMMW(J)
/ -PONMINW(J)-PONSET(J)+PONRE(J)-PONOUT(J)+PONAT(J)PONB(J)=PONB(J-1)+DEATHB(J)-PEATNACB(J)-PONIMMB(J)
/ -PONMINB(J)+PONSET(J)-PONRE(J)ELSEIF (WETTYPE.EQ.1) THENPONB(J)=PONB(J-1)+DEATHB(J)+PONIN(J)-PEATNACB(J)-PONIMMB(J)-
/ PONMINB(J)+PONAT(J)ENDIF
C MASS BALANCE FOR REFRACTORY NC
IF (WETTYPE.EQ.0) THENREFN(J)=REFN(J-1)+PEATNACW(J)+PEATNACB(J)TONW(J)=PONW(J)+DONW(J)TONB(J)=PONB(J)+DONB(J)ELSEIF (WETTYPE.EQ.1) THENREFN(J)=REFN(J-1)+PEATNACB(J)TONB(J)=PONB(J)+DONB(J)ENDIF
C WRITE TO OUTPUT FILESC
IF (J.EQ.1 .AND.M.EQ.1) THENWRITE (17,*) 'OUTPUT DATA FOR NITROGEN COUNTS IN WETLAND'WRITE(17,235)
235 FORMAT (T13,'NO3W',T24,'NO3B',T34,'NH4W',T45,'NH4B',T57 / ,'IMMNW',T68,'IMMNB')
WRITE(17,245) M,NO3W(0),NO3B(0),NH4W(0),NH4B(0),IMMNW(0),IMMNB(0) 245 FORMAT (T1, (I2),T5,(' 0'),T8,6(F9.2,2X))
WRITE(23,*) 'OUTPUT DATA FOR NITROGEN COUNTS IN WETLAND'WRITE(23,255)
255 FORMAT (T13,'DONW',T24,'DONB',T35,'PONW',T46,'PONB',T57 / ,'REFN',T68,'TONW',T79,'TONB')
WRITE(23,265) M,DONW(0),DONB(0),PONW(0),PONB(0),REFN(0),TONW(0), / TONB(0) 265 FORMAT (T1, (I2),T5,(' 0'),T8,7(F9.2,2X))
ENDIFC
END
225
OXYGEN SUBMODELS
SUBROUTINE OXYSTR (DOINITW,DOINITB,HTDOYB,NSDOYB,HTDOYW, / NSDOYW,DOCONCP,OXYCONC,OXDRC,MTFWSDOC,PERCINF, / MTDOX,POINT,HYDTYPE,M,DOXYCSAT,WETTYPE,DOININC)
REAL DOINITW,DOINITB,HTDOYW,NSDOYW,HTDOYB,NSDOYB,DOCONCP, / OXYCONC,OXDRC,MTFWSDOC,MTDOX,DOXYCSAT,DOININC
INTEGER CHECK1,CHECK2,POINT,HYDTYPE,WETTYPE,PERCINF
C READ IN INITIAL INPUT VALUESC
IF (M.EQ.1) THENIF (WETTYPE.EQ.0) THEN
READ (10,*) DOINITW,DOINITB,HTDOYW,HTDOYB,NSDOYW,NSDOYB, / DOCONCP,MTDOX,MTFWSDOC,DOXYCSAT
ELSEIF(WETTYPE.EQ.1) THENREAD(10,*) DOINITB,HTDOYB,NSDOYB,DOCONCPENDIF
IF (POINT.EQ.0) THENOXYCONC=0.
ELSEIF (POINT.EQ.1) THENREAD (10,*) OXYCONC
ENDIF
IF (PERCINF.EQ.0) THENDOININC=0.0
ELSEIF (PERCINF.EQ.1) THENREAD (12,*) DOININC
ENDIF
IF (HYDTYPE.EQ.0 ) THENOXDRC=0.0
ELSEIF (HYDTTYPE.EQ.1) THENREAD(10,*) OXDRC
ENDIFENDIF
CC CHECK FOR VALUE CHANGES BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (10,*)c, CHECK1,CHECK2
IF (CHECK1.EQ.0) THENIF (WETTYPE.EQ.0) THENREAD(10,*)HTDOYW,HTDOYB,NSDOYW,NSDOYB,DOCONCP,
/ MTDOX,MTFWSDOC,DOXYCSATELSEIF (WETTYPE.EQ.1) THENREAD(10,*) HTDOYB,NSDOYB,DOCONCPENDIF
ENDIF
IF (CHECK2.EQ.0) THENIF(POINT.EQ.1) THEN
READ(10,*) OXYCONCENDIFIF (PERCINF.EQ.1) THEN
READ (10,*) DOININCENDIFIF (HYDTYPE.EQ.1) THEN
READ(10,*) OXDRCENDIF
ENDIFENDIF
CEND
CC
226
SUBROUTINE OXYTIME (DOINITW,DOINITB,HTDOYW,NSDOYW,NSDOYB,OUTFLOW, / NSGROWW,AEROHTGB,J,K,M,NSGROWB,AEROHTGW,WATINPUT,PRECIP, / HTDOYB,DOXYW,DOXYB,OXYCONC,OXDRC,APFLOW,MTFWSDOC, / MTDOX,DOXYCW,DOXYCB,HYDTYPE,HI,HT,WATERVEL,DOCONCP,WATERVOL, / WATERB,DOXYCSAT,WETTYPE,PERCINFT,DOININC,BIOMASST)
REAL DOXYW(0:500),DOXYB(0:500),WATERVOL(0:500),WATERB(0:500), / OUTFLOW(0:500),DOINF(0:500),DOOUT(0:500),HI(0:500),HT(0:500), / NSGROWW(0:500),AEROHTGB(0:500),DOXYCW(0:500),DOXYCB(0:500), / NSGROWB(0:500),AEROHTGW(0:500),WATINPUT(0:500),PRECIP(0:500), / NSRESPW(0:500),NSRESPB(0:500),MTDOXY(0:500),MTFWS(0:500), / HTRESPW(0:500),HTRESPB(0:500),DOPERC(0:500),PERCINFT(0:500), / APFLOW(0:500),BIOMASST(0:500)
REAL DOINITW,DOINITB,HTDOYW,NSDOYW,NSDOYB,HTDOYB,OXYCONC,OXDRC, / DOCONCP,MTFWSDOC,MTDOX,LENGTH,WIDTH,HO,SO,HB / WATERVEL,BIOOXRB,BIOFLUXB,DOXYCSAT,DOININC
INTEGER HYDTYPE,WETTYPE,J,K,MCOMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SO
C READ IN OXYGEN INPUT VALUESC
IF (HYDTYPE.EQ.0) THENREAD (10,*) BIOOXRB,DOCONCINELSEIF (HYDTYPE.EQ.1) THENREAD (10,*) BIOOXRBENDIF
IF (HYDTYPE.EQ.0) THENDOINF(J)=DOCONCIN*WATINPUT(J)+PRECIP(J)*DOCONCP+APFLOW(J)*OXYCONCELSEIF (HYDTYPE.EQ.1) THENDOINF(J)=OXDRC*WATINPUT(J)+APFLOW(J)*OXYCONC+PRECIP(J)*DOCONCPENDIF
BIOFLUXB=BIOOXRB*LENGTH*WIDTH
IF (WETTYPE.EQ.0) THENIF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENDOOUT(J)=DOXYW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THENDOOUT(J)=.75*DOXYW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ENDIFELSEIF (WETTYPE.EQ.1) THENDOOUT(J)=DOXYB(J-1)*OUTFLOW(J)/WATERB(J-1)ENDIF
C MICROBIAL RESPIRATIONC
IF (WETTYPE.EQ.0) THENHTRESPW(J)=AEROHTGW(J)/HTDOYWNSRESPW(J)=NSGROWW(J)/NSDOYWENDIFHTRESPB(J)=AEROHTGB(J)/HTDOYBNSRESPB(J)=NSGROWB(J)/NSDOYB
IF (WETTYPE.EQ.0) THENMTDOXY(J)=864*LENGTH*WIDTH*MTDOX*(DOXYCB(J-1)-DOXYCW(J-1))MTFWS(J)=864*LENGTH*WIDTH*MTFWSDOC*(DOXYCSAT-DOXYCW(J-1))ENDIF
IF (PERCINFT(J).GT.0.) THENDOPERC(J)=DOXYB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THENDOPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THENDOPERC(J)=PERCINFT(J)*DOININC
ENDIFCC MASS BALANCE FOR DISSOLVED OXYGEN
227
CIF (WETTYPE.EQ.0) THENDOXYW(J)=DOXYW(J-1)+DOINF(J)+MTDOXY(J)+MTFWS(J)-DOOUT(J)
/ -HTRESPW(J)-NSRESPW(J)DOXYB(J)=DOXYB(J-1)+BIOFLUXB-HTRESPB(J)-NSRESPB(J)-MTDOXY(J)
/ -DOPERC(J)ELSEIF (WETTYPE.EQ.1) THENDOXYB(J)=DOXYB(J-1)+DOINF(J)-DOOUT(J)+BIOFLUXB-HTRESPB(J)
/ -NSRESPB(J)-DOPERC(J)ENDIF
CC WRITE TO OUTPUT FILESC
IF (M.EQ.1 .AND.J.EQ.1) THENWRITE (19,*) 'OUTPUT DATA FOR DISSOLVED OXYGEN COUNTS IN WETLAND'WRITE(19,290)
290 FORMAT (T12,'DOXYW',T23,'DOXYB',T33,'DOXYCW',T44,'DOXYCB')WRITE(19,295) M,DOXYW(0),DOXYB(0),DOXYCW(0),DOXYCB(0)
295 FORMAT (T1, (I2),T6,(' 0'),T9,4(F9.2,2X)) ENDIFC
END
228
PHOSPHOROUS SUBMODELS
SUBROUTINE PHOSSTR(DTPHOSIW,PPHOSI,BTPHOSI,M,MTCPHOS, / PMINPPC,LINPARTC,FREUNDK,FREUNDN,ADSORP,BIOMPP, / PRMINBPC,POINT,PHOSCON,PHOSRC,DTPHOSIB,PININC,PERCINF)
INTEGER ADSORP,POINT,CHECK1,CHECK2,PERCINF
REAL DTPHOSI,PPHOSI,BTPHOSI,PMINPPC,PRMINBPC,PHOSCON, / LINPARTC,FREUNDK,FREUNDN,BIOMPP,MTCPHOS,PININC / PHOSRCCC INITIAL INPUT VALUESC
IF (M.EQ.1) THENREAD(14,*) DTPHOSIW,DTPHOSIB,BTPHOSI,PPHOSI,PMINPPC,
/ PRMINBPC,ADSORP,MTCPHOS,BIOMPP
IF (PERCINF.EQ.0) THENPININC=0.0
ELSEIF (PERCINF.EQ.1) THENREAD (10,*) PININC
ENDIF
IF (ADSORP.EQ.0) THENREAD (14,*) FREUNDK,FREUNDNLINPARTC=0.
ELSEIF (ADSORP.EQ.1) THENREAD(14,*) LINPARTCFREUNDK=0.FREUNDN=0.
ELSEIF (ADSORP.EQ.2) THENFREUNDK=0FREUNDN=0READ(14,*) LINPARTC
ENDIF
IF (POINT.EQ.1 .OR. ADSORP.EQ.2) THENREAD(14,*) PHOSCON
ELSEPHOSCON=0.0
ENDIFENDIF
CC CHECK FOR CHANGES BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (14,*) c, CHECK1,CHECK2IF (CHECK1.EQ.0) THEN
READ (14,*) PMINPPC,PRMINBPC,MTCPHOS,BIOMPPIF (PERCINF.EQ.1) THEN
READ(14,*) PININCENDIF
ENDIFIF (CHECK2.EQ.0) THEN
IF (ADSORP.EQ.0) THENREAD (14,*)FREUNDK,FREUNDN,PHOSCON
ELSEIF (ADSORP.EQ.1) THENREAD(14,*) LINPARTC,PHOSCON
ELSEIF (ADSORP.EQ.2) THENREAD(14,*) PHOSCON
ENDIFENDIF
ENDIFEND
CC
SUBROUTINE PHOSTIME (DTPHOSW,PPHOS,BTPHOS,OUTFLOW,ADSORP,PMINPPC, / FREUNDK,FREUNDN,SEDINITB,SEDINITW,SEDQTYB,SEDQTYW,SEDCAT,
229
/ SEDBSA,HI,HT,PPHOST,BTPHOST,SEDINW,APFLOW,PHOSCON, / PHOSPER,BIOMPP,BIOMASST,PHYSDEG,DTPHOSIW,PPHOSI, / WATINPUT,SEDDEP,RESUSP,PRMINBPC,MTCPHOS,BTPHOSI,WATERVOL, / J,K,M,LINPARTC,POINT,PHOSRC,PHOSPERB,DTPHOSIB,DTPHOSB, / WATERB,SEDSPG,SEDBV,PHOSPERW,BIOMGROW,SEDOUT,PRATEUP, / PERCINFT,PININC,DTPHOSCW,DTPHOSCB,DPHOSOUT,PPHOSOUT)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SOREAL DTPHOSW(0:500),SEDINITB(5),PPHOSIN(5,0:500),SEDINITW(5),
/ SEDQTYB(5,0:500),SEDQTYW(5,0:500),SEDBSA(5),PPHOS(5,0:500), / BTPHOS(5,0:500),PPHOSOUT(5,0:500),PPHOSSET(5,0:500), / PPHOSRES(5,0:500),HI(0:500),HT(0:500),DISPHOSI(0:500), / SEDINW(5,0:500),PHOSPER(5,0:500),WATERB(0:500), / PRMINBPT(0:500),PHYSDEG(0:500),BIOMASST(0:500),OUTFLOW(0:500), / PPHOST(0:500),BTPHOST(0:500),PPHOSINT(0:500),PMINPP(5,0:500), / PRMINBP(5,0:500),DPHOSOUT(0:500),WATERVOL(0:500), / PHOSMT(0:500),DEADPHOT(0:500),PHOSUPB(0:500),DTPHOSB(0:500), / DEADPHOS(5,0:500),PPHOSUP(5,0:500),WATINPUT(0:500), / PMINPPT(0:500) ,PPHOSCW(0:500),SEDDEP(5,0:500), / SEDINCV(5,0:500),SEDPART(5,0:500),SEDBV(5),BIOMGROW(0:500), / RESUSP(5,0:500),PHOSPERB(5,0:500),SEDSPG(5), / SEDTSAW(5,0:500),SEDTSAB(5,0:500),PHOSPERW(5,0:500), / SEDOUT(5,0:500),PHOSUPW(0:500),APFLOW(0:500),PERCINFT(0:500), / PHOSPERC(0:500),DTPHOSCW(0:500),DTPHOSCB(0:500)
INTEGER ADSORP,POINT,J,K,M,SEDCAT,SEDCLASSREAL DISPHOSC,DTPHOSIB,PPHOSI,FREUNDN,FREUNDK,LINPARTC,MTCPHOS,
/ LENGTH,WIDTH,HO,HB,SO,PPHOSINC,BTPHOSI,PRMINBPC, / BIOMPP,PHOSCON,PPHOSINP,PHOSRC,PPHOSCAL,DTPHOSIW,PININC / PPHOSTOT,SEDTSATP,SEDTSATU,SEDTSATC,PRATEUP
IF (M.EQ.1 .AND. J.EQ.1) THENDTPHOSW(0)=DTPHOSIWDTPHOSB(0)=DTPHOSIBPPHOST(0)=PPHOSIBTPHOST(0)=BTPHOSIPPHOSCW(0)=PPHOST(0)/WATERVOL(0)
DO 25 SEDCLASS=1,SEDCATPPHOS(SEDCLASS,0)=PPHOSI*PHOSPERW(SEDCLASS,0)
25 CONTINUE
DO 50 SEDCLASS=1,SEDCATBTPHOS(SEDCLASS,0)=BTPHOSI*PHOSPERB(SEDCLASS,0)
50 CONTINUEENDIF
IF (J.EQ.1) THENDO 60 CLEAR=1,100PMINPPT(CLEAR)=0.0PRMINBPT(CLEAR)=0.0BTPHOST(CLEAR)=0.0PPHOST(CLEAR)=0.0
60 CONTINUEENDIF
IF (HYDTYPE.EQ.0 .AND. (ADSORP.EQ.0 .OR. ADSORP.EQ.1)) THENREAD (14,*) DISPHOSCELSEIF (HYDTYPE.EQ.0 .AND. ADSORP.EQ.2) THENREAD (14,*) DISPHOSC, PPHOSCALENDIFIF (HYDTYPE.EQ.1) THENDISPHOSC=PHOSRCENDIF
IF (ADSORP.EQ.0) THENPPHOSINC=FREUNDK*(DISPHOSC**(1./FREUNDN))/1000IF (POINT.EQ.1) THEN
PPHOSINP=FREUNDK*((PHOSCON)**(1./FREUNDN))/1000ELSE IF (POINT.EQ.0) THEN
PPHOSINP=0.0
230
ENDIFELSEIF(ADSORP.EQ.1) THEN
PPHOSINC=DISPHOSC*LINPARTC/1000IF (POINT.EQ.1) THEN
PPHOSINP=(PHOSCON)*LINPARTC/1000ELSEIF (POINT.EQ.0) THEN
PPHOSINP=0.0ENDIF
ELSEIF (ADSORP.EQ.2) THENIF (POINT.EQ.1) THEN
PPHOSINP=(PHOSCON)*LINPARTC/1000ELSEIF (POINT.EQ.0) THEN
PPHOSINP=0.0ENDIF
ENDIF
IF (ADSORP.EQ.0 .OR. ADSORP.EQ.1) THENPPHOSINT(J)=WATINPUT(J)*PPHOSINC+APFLOW(J)*PPHOSINP
ELSEIF (ADSORP.EQ.2) THENPPHOSINT(J)=PPHOSCAL+APFLOW(J)*PPHOSINP
ENDIF
IF (HYDTYPE.EQ.0) THENDISPHOSI(J)=WATINPUT(J)*DISPHOSC+APFLOW(J)*PHOSCON
ELSEIF (HYDTYPE.EQ.1) THENDISPHOSI(J)=WATINPUT(J)*PHOSRC+APFLOW(J)*PHOSCON
ENDIF
DO 75 SEDCLASS=1,SEDCATPMINPP(SEDCLASS,J)=PMINPPC*PPHOS(SEDCLASS,J-1)PRMINBP(SEDCLASS,J)=PRMINBPC*BTPHOS(SEDCLASS,J-1)PMINPPT(J)=PMINPPT(J)+PMINPP(SEDCLASS,J)PRMINBPT(J)=PRMINBPT(J)+PRMINBP(SEDCLASS,J)
75 CONTINUE
IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENDPHOSOUT(J)=DTPHOSW(J-1)*OUTFLOW(J)/WATERVOL(J-1)ELSEIF (OUTFLOW(J) .GE. WATERVOL (J-1)) THENDPHOSOUT(J)=.75*DTPHOSW(J-1)* OUTFLOW(J)/WATERVOL(J-1)ENDIF
PHOSMT(J)=864*LENGTH*WIDTH*MTCPHOS*(DTPHOSCB(J-1)-DTPHOSCW(J-1))
IF (DTPHOSB(J-1).GT.(PRATEUP*BIOMGROW(J)/BIOMPP)) THENPHOSUPB(J)=BIOMGROW(J)/BIOMPP*PRATEUP
ELSEPHOSUPB(J)=DTPHOSB(J-1)
ENDIF
IF (DTPHOSW(J-1).GT.((1-PRATEUP)*BIOMGROW(J)/BIOMPP)) THENPHOSUPW(J)=BIOMGROW(J)/BIOMPP*(1-PRATEUP)
ELSEPHOSUPW(J)=DTPHOSW(J-1)
ENDIF
IF (PERCINFT(J).GT.0.) THENPHOSPERC(J)=DTPHOSB(J-1)*PERCINFT(J)/WATERB(J-1)
ELSEIF (PERCINFT(J).EQ.0.) THENPHOSPERC(J)=0.0
ELSEIF (PERCINFT(J).LT.0.) THENPHOSPERC(J)=PERCINFT(J)*PININC
ENDIFCC MASS BALANCE FOR DISSOLVED PHOSPHOROUSC
DTPHOSW(J)=DTPHOSW(J-1)+DISPHOSI(J)+PMINPPT(J)- / DPHOSOUT(J)+PHOSMT(J)-PHOSUPW(J)
DTPHOSB(J)=DTPHOSB(J-1)-PHOSMT(J)+PRMINBPT(J)-PHOSUPB(J)+PHOSPERC(J)
231
DO 125 SEDCLASS=1,SEDCATPPHOSIN(SEDCLASS,J)=PPHOSINT(J)*PHOSPER(SEDCLASS,J)IF (OUTFLOW(J) .LT. WATERVOL(J-1)) THEN
PPHOSOUT(SEDCLASS,J)=SEDOUT(SEDCLASS,J)/ / SEDQTYW(SEDCLASS,J)*PPHOS(SEDCLASS,J-1)
ELSEIF (OUTFLOW(J).GE. WATERVOL(J-1)) THENPPHOSOUT(SEDCLASS,J)=.5*SEDOUT(SEDCLASS,J)/
/ SEDQTYW(SEDCLASS,J)*PPHOS(SEDCLASS,J-1)ENDIF
125 CONTINUE
DO 150 SEDCLASS=1,SEDCATPPHOSSET(SEDCLASS,J)=SEDDEP(SEDCLASS,J)/
/ SEDQTYW(SEDCLASS,J-1)*PPHOS(SEDCLASS,J-1) 150 CONTINUE
DO 175 SEDCLASS=1,SEDCATPPHOSRES(SEDCLASS,J)=RESUSP(SEDCLASS,J)
/ /SEDQTYB(SEDCLASS,J-1)*BTPHOS(SEDCLASS,J-1) 175 CONTINUE
DO 200 SEDCLASS=1,SEDCATPPHOS(SEDCLASS,J)=PPHOS(SEDCLASS,J-1)+PPHOSIN(SEDCLASS,J)-
/ PMINPP(SEDCLASS,J)-PPHOSSET(SEDCLASS,J)+PPHOSRES(SEDCLASS,J)- / PPHOSOUT(SEDCLASS,J) 200 CONTINUE
DO 225 SEDCLASS=1,SEDCATPPHOST(J)=PPHOST(J)+PPHOS(SEDCLASS,J)
225 CONTINUE
DEADPHOT(J)=PHYSDEG(J)/BIOMPPDO 300 SEDCLASS=1,SEDCAT
DEADPHOS(SEDCLASS,J)=DEADPHOT(J)*PHOSPERB(SEDCLASS,J-1) 300 CONTINUE
DO 350 SEDCLASS=1,SEDCATBTPHOS(SEDCLASS,J)=BTPHOS(SEDCLASS,J-1)+PPHOSSET(SEDCLASS,J)
/ +DEADPHOS(SEDCLASS,J)-PRMINBP(SEDCLASS,J)-PPHOSRES(SEDCLASS,J)BTPHOST(J)=BTPHOST(J)+BTPHOS(SEDCLASS,J)
350 CONTINUE
SEDTSATC=0DO 360 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,J)=SEDQTYB(SEDCLASS,J)*(1./ / (SEDSPG(SEDCLASS)/1000.))
SEDPART(SEDCLASS,J)=SEDINCV(SEDCLASS,J)/SEDBV(SEDCLASS)SEDTSAB(SEDCLASS,J)=SEDPART(SEDCLASS,J)*SEDBSA(SEDCLASS)SEDTSATC=SEDTSATC+SEDTSAB(SEDCLASS,J)
360 CONTINUE
DO 375 SEDCLASS=1,SEDCATPHOSPERB(SEDCLASS,J)=SEDTSAB(SEDCLASS,J)/SEDTSATC
375 CONTINUE
SEDTSATU=0DO 380 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,J)=SEDQTYW(SEDCLASS,J)*(1./ / (SEDSPG(SEDCLASS)/1000.))
SEDPART(SEDCLASS,J)=SEDINCV(SEDCLASS,J)/SEDBV(SEDCLASS)SEDTSAW(SEDCLASS,J)=SEDPART(SEDCLASS,J)*SEDBSA(SEDCLASS)SEDTSATU=SEDTSATU+SEDTSAW(SEDCLASS,J)
380 CONTINUE
DO 390 SEDCLASS=1,SEDCATPHOSPERW(SEDCLASS,J)=SEDTSAW(SEDCLASS,J)/SEDTSATU
390 CONTINUECC WRITE TO OUTPUT FILESC
232
IF (M.EQ.1. .AND. J. EQ.1) THENWRITE (25,*) 'OUTPUT DATA FOR PHOSPHOROUS COUNTS IN WETLAND'WRITE(25,400)
400 FORMAT (T12,'DTPHOSW',T24,'DTPHOSB',T36,'BTPHOST',T49,'PPHOST', / T56,'DTPHOSCW', T65,'PPHOSCW')
WRITE(25,425) M,DTPHOSIW,DTPHOSIB,BTPHOSI,PPHOSI, / (DTPHOSIW/WATERVOL(0)),PPHOSI/WATERVOL(0) 425 FORMAT(T1, (I2),T7,'0',T9,4(F10.2,2X),T59,2(F5.2,2X))
ENDIF
IF(J.EQ.K)THENDTPHOSW(0)=DTPHOSW(J)DTPHOSB(0)=DTPHOSB(J)DO 500 SEDCLASS=1,SEDCAT
BTPHOS(SEDCLASS,0)=BTPHOS(SEDCLASS,J)PPHOS(SEDCLASS,0)=PPHOS(SEDCLASS,J)
500 CONTINUEENDIFEND
233
SEDIMENT SUBMODELS
SUBROUTINE SEDSTR (SEDSIZE,SEDFALL,SEDINITW,SEDINITB,SEDPER, / RESTHICK,SEDBV,SEDBSA,SEDSPG,SEDCAT,MANNC,M,SEDRC, / POINT,SEDCONC,SEDRES,DECOMPR,PSEDDEP)
REAL SEDSIZE(5), SEDINITW(5), SEDFALL(5), SEDSPG(5), / SEDINITB(5),SEDPER(5),SEDBV(5),SEDBSA(5)
REAL MANNC,SEDRC,RESTHICK,SEDCONC,SEDRES,DECOMPR,PSEDDEPINTEGER M,SEDCAT,SEDCLASS,CHECK
IF (M.EQ.1) THENREAD (9,*) SEDCAT,SEDRES
DO 100 SEDCLASS=1,SEDCATREAD(9,*) SEDSIZE(SEDCLASS),SEDFALL(SEDCLASS),
/ SEDINITW(SEDCLASS),SEDINITB(SEDCLASS), / SEDSPG(SEDCLASS),SEDPER(SEDCLASS)100 CONTINUE
READ (9,*) RESTHICK,MANNC,DECOMPR, PSEDDEP
DO 200 SEDCLASS=1,SEDCATSEDBV(SEDCLASS)=(88./21.*(((SEDSIZE(SEDCLASS))/2)**3))SEDBSA(SEDCLASS)=(88./7.*((SEDSIZE(SEDCLASS)/2)**2))
200 CONTINUE
IF (POINT.EQ.1) THENREAD (9,*) SEDCONC
ELSESEDCONC=1.0
ENDIFENDIF
CC CHECK FOR VALUE CHANGES BETWEEN SEASON PERIODSC
IF (M.GT.1) THENREAD (9,*) c, CHECK
IF (CHECK.EQ.0) THENREAD (9,*) RESTHICK,MANNC,SEDCONC,
/ DECOMPR,SEDRES,PSEDDEPENDIF
ENDIFEND
CC
SUBROUTINE SEDTIME (OUTFLOW,WATERVEL,HT,HI,PHOSPERB,SEDOUTT, / RESTHICK,SEDSIZE,SEDFALL,SEDPER,SEDQTYW,SEDQTYB,SEDRES, / MANNC,PHOSPER,J,K,M,SEDRC,WATINPUT,SEDCAT,APFLOW,SEDCONC, / SEDINITW,SEDINITB,SEDBSA,SEDBV,SEDSPG,WATERVOL,DECOMPR, / SEDPART,PHOSPERW,SEDDEP,RESUSP,SEDOUT,PCYCLE,PHYSDEG,PSEDDEP)
COMMON /DESCRIBE/LENGTH,WIDTH,HO,HB,SOCOMMON/OUTFLOW/OUTLET,HOUT,OUTWIDTH,HOVER,ANGVNOT,FLOWOUT,
/ TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK, / KHCOEF
REAL RESUSP(5,0:500),RE1(5,0:500),RE2(5),SEDFALL(5), / SEDQTYB(5,0:500),SEDPER(5),WATINPUT(0:500),SEDTSAW(5,0:500), / SEDINW(5,0:500),SEDINCV(5,0:500),SEDPART(5,0:500), / SEDTSAP(5,0:500),PHOSPER(5,0:500), PHOSPERW(5,0:500), / RESUSPV(5,0:500),SEDOUT(5,0:500),SEDDEP(5,0:500), / HI(0:500),HT(0:500),SEDQTYW(5,0:500),SEDINITW(5),SEDINITB(5), / WATERVOL(0:500),SEDBV(5),SEDBSA(5),SEDSIZE(5),APFLOW(0:500), / OUTFLOW(0:500),PHOSPERB(5,0:500),SEDTSAB(5,0:500),SEDSPG(5), / PHYSDEG(0:500),SDECOMP(5,0:500),SEDOUTT(0:500)
REAL SEDRC,RESTHICK,SEDINT,LENGTH,WIDTH,HO,HB,SO, / SEDCONC,SEDTSATW,SEDTSATB,SEDTSATP,HOUT,OUTWIDTH,HOVER, / TOPPUMP,AREAPIPE,CONTC,DISEFFC,ALPHA,BETA,COEFK,SEDRES / KHCOEF,ANGVNOT,FLOWOUT,DECOMPR,WATERVEL,MANNC,PSEDDEP
INTEGER J,K,M,SEDCAT,SEDCLASS,OUTLET,PCYCLE
234
IF (J.EQ.1 .AND. M.EQ.1) THENDO 25 SEDCLASS=1,SEDCAT
SEDQTYW(SEDCLASS,0)=SEDINITW(SEDCLASS)SEDQTYB(SEDCLASS,0)=SEDINITB(SEDCLASS)
25 CONTINUEWRITE (18,*) 'OUTPUT DATA FOR SEDIMENT COUNTS IN WETLAND WATER'WRITE(18,50)
50 FORMAT (//,T8, 'SEDQTYW(1)',T21, 'SEDQTYW(2)',T34, 'SEDQTYW(3)', / T47, 'SEDQTYW(4)',T60, 'SEDQTYW(5)')
WRITE(18,75) M,SEDINITW(1),SEDINITW(2),SEDINITW(3), / SEDINITW(4),SEDINITW(5) 75 FORMAT (T1, (I1), T4,(' 0 '),T8,5(F11.1,2X))
WRITE (22,*) 'OUTPUT DATA FOR SEDIMENT COUNTS IN WETLAND BOTTOM'WRITE(22,80)
80 FORMAT (//,T8, 'SEDQTYB(1)',T21, 'SEDQTYB(2)',T34, 'SEDQTYB(3)', / T47,'SEDQTYB(4)',T60, 'SEDQTYB(5)')
WRITE(22,85) M,SEDINITB(1),SEDINITB(2),SEDINITB(3), / SEDINITB(4),SEDINITB(5) 85 FORMAT (T1, (I1), T4,(' 0 '),T8,5(F11.1,2X)) ENDIF
SEDOUTT(J)=0
IF (HYDTYPE.EQ.0) THENREAD (9,*) SEDINTELSE IF (HYDTYPE.EQ.1) THENSEDINT=WATINPUT(J)*SEDRCENDIF
DO 100 SEDCLASS=1,SEDCATSEDINW(SEDCLASS,J)=(SEDINT+APFLOW(J)*SEDCONC)*SEDPER(SEDCLASS)
100 CONTINUE
SEDINW(SEDCAT,J)=SEDINW(SEDCAT,J)+PHYSDEG(J)
IF (PCYCLE.EQ.1) THENSEDTSATP=0DO 150 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,J)=SEDINW(SEDCLASS,J)*(1./(SEDSPG(SEDCLASS)/1000.))SEDPART(SEDCLASS,J)=SEDINCV(SEDCLASS,J)/SEDBV(SEDCLASS)SEDTSAP(SEDCLASS,J)=SEDPART(SEDCLASS,J)*SEDBSA(SEDCLASS)SEDTSATP=SEDTSATP+SEDTSAP(SEDCLASS,J)
150 CONTINUE
DO 175 SEDCLASS=1,SEDCATPHOSPER(SEDCLASS,J)=SEDTSAP(SEDCLASS,J)/SEDTSATP
175 CONTINUE
IF (M.EQ.1 .AND. J.EQ.1) THENSEDTSATW=0DO 180 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,0)=SEDQTYW(SEDCLASS,0)*(1./(SEDSPG(SEDCLASS)/1000.))SEDPART(SEDCLASS,0)=SEDINCV(SEDCLASS,0)/SEDBV(SEDCLASS)SEDTSAW(SEDCLASS,0)=SEDPART(SEDCLASS,0)*SEDBSA(SEDCLASS)SEDTSATW=SEDTSATW+SEDTSAW(SEDCLASS,0)
180 CONTINUE
DO 185 SEDCLASS=1,SEDCATPHOSPERW(SEDCLASS,0)=SEDTSAW(SEDCLASS,0)/SEDTSATW
185 CONTINUE
SEDTSATB=0DO 190 SEDCLASS=1,SEDCAT
SEDINCV(SEDCLASS,0)=SEDQTYB(SEDCLASS,0)*(1./ / (SEDSPG(SEDCLASS)/1000.))
SEDPART(SEDCLASS,0)=SEDINCV(SEDCLASS,0)/SEDBV(SEDCLASS)SEDTSAB(SEDCLASS,0)=SEDPART(SEDCLASS,0)*SEDBSA(SEDCLASS)SEDTSATB=SEDTSATB+SEDTSAB(SEDCLASS,0)
190 CONTINUE
235
DO 195 SEDCLASS=1,SEDCATPHOSPERB(SEDCLASS,0)=SEDTSAB(SEDCLASS,0)/SEDTSATB
195 CONTINUEENDIFENDIF
IF (M.EQ.1 .AND. J.EQ.1) THENDO 300 SEDCLASS=1,SEDCAT
RE2(SEDCLASS)=(SEDSIZE(SEDCLASS)/1000.)*(SEDFALL(SEDCLASS)/86400)*1000/.001IF (RE2(SEDCLASS).GT.1.) THENWRITE(*,250) SEDCLASS
250 FORMAT(1X,'THE CRITERION FOR RESUSPENSION IS NOT MET, WATER ' / 'IS OUT OF THE',/,' LAMINAR FLOW RANGE FOR SEDIMENT' / ' PARTICLE ',I1,/)
ENDIF 300 CONTINUE
C1=0C2=0C3=0C4=0C5=0ENDIF
DO 400 SEDCLASS=1,SEDCATRESUSPV(SEDCLASS,J)=7.2*(((SEDFALL(SEDCLASS)/86400.)**(1./3.))
/ *((HI(J-1)-HT(J-1))**(1./6.))/(MANNC*(((SEDSIZE(SEDCLASS)/1000.)**(2./3.))))) 400 CONTINUECC CHECK IF RESUSPENSION THEORY IS METC
RE1B=(2.*((HI(J-1)-HT(J-1))**(1./6.)))/((MANNC*(9.81**(0.5))))DO 500 SEDCLASS=1,SEDCAT
RE1(SEDCLASS,J)=(SEDSIZE(SEDCLASS)/1000.)*1000* / (RESUSPV(SEDCLASS,J)/86400.)/0.001
IF (RE1(SEDCLASS,J).GT.RE1B) THENIF (SEDCLASS.EQ.1) THEN
IF (C1.EQ.0) THENWRITE (*,451)
451 FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR' / ' PARTICLE CATEGORY 1')
ENDIFC1=1
ENDIFIF (SEDCLASS.EQ.2) THEN
IF (C2.EQ.0) THENWRITE (*,452)
452 FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR' / ' PARTICLE CATEGORY 2')
ENDIFC2=1
ENDIFIF (SEDCLASS.EQ.3) THEN
IF (C3.EQ.0) THENWRITE (*,453)
453 FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR' / ' PARTICLE CATEGORY 3')
ENDIFC3=1ENDIFIF (SEDCLASS.EQ.4) THEN
IF (C4.EQ.0) THENWRITE (*,454)
454 FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR' / ' PARTICLE CATEGORY 4')
ENDIFC4=1
ENDIFIF (SEDCLASS.EQ.5) THEN
236
IF (C5.EQ.0) THENWRITE (*,455)
455 FORMAT ('OUT OF LAMINAR RANGE FOR RESUSPENSION THEORY FOR' / ' PARTICLE CATEGORY 5')
ENDIFC5=1
ENDIFENDIF
500 CONTINUE
DO 550 SEDCLASS=1,SEDCATIF (RESUSPV(SEDCLASS,J).LE.WATERVEL) THEN
RESUSP(SEDCLASS,J)=SEDQTYB(SEDCLASS,J-1)*SEDRES* / RESTHICK/(HT(J-1)-HB)
ELSEIF (RESUSPV(SEDCLASS,J).GT.WATERVEL) THENRESUSP(SEDCLASS,J)=0.0
ENDIF 550 CONTINUE
DO 600 SEDCLASS=1,SEDCATIF (SEDFALL(SEDCLASS) .LT. (HI(J-1)-HT(J-1))) THENSEDDEP(SEDCLASS,J)=(SEDQTYW(SEDCLASS,J-1))*
/ (SEDFALL(SEDCLASS))/(HI(J-1)-HT(J-1))*PSEDDEPELSE IF (SEDFALL(SEDCLASS) .GE. (HI(J-1)-HT(J-1)))THENSEDDEP(SEDCLASS,J)=SEDQTYW(SEDCLASS,J-1)*PSEDDEPENDIF
600 CONTINUE
DO 700 SEDCLASS=1,SEDCATIF (OUTLET.EQ.1 .OR. OUTLET.EQ.2. .OR. OUTLET.EQ.3. .OR.
/ OUTLET.EQ.5) THENIF (SEDFALL(SEDCLASS) .GT.
/ (HI(J-1)-HT(J-1))) THENSEDOUT(SEDCLASS,J)=0.ELSEIF (SEDFALL(SEDCLASS) .LE. (HI(J-1)-HT(J-1)))THENIF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENSEDOUT(SEDCLASS,J)=((SEDQTYW(SEDCLASS,J-1)))
/ *OUTFLOW(J)/WATERVOL(J-1)ELSEIF (OUTFLOW (J) .GE. WATERVOL(J-1)) THENSEDOUT(SEDCLASS,J)=(.5*(SEDQTYW(SEDCLASS,J-1)))
/ *OUTFLOW(J)/WATERVOL(J-1)ENDIFENDIFELSEIF (OUTLET.EQ.4) THENIF (SEDFALL(SEDCLASS) .GT.(HI(J-1)-HT(J-1))) THENSEDOUT(SEDCLASS,J)=0.ELSEIF (SEDFALL(SEDCLASS) .LE. (HI(J-1)-TOPPUMP))THENIF (OUTFLOW(J) .LT. WATERVOL(J-1)) THENSEDOUT(SEDCLASS,J)=(SEDQTYW(SEDCLASS,J-1))
/ *OUTFLOW(J)/WATERVOL(J-1)ELSEIF (OUTFLOW (J) .GE. WATERVOL(J-1)) THENSEDOUT(SEDCLASS,J)=(.5*SEDQTYW(SEDCLASS,J-1))
/ *OUTFLOW(J)/WATERVOL(J-1)ENDIFENDIFENDIFSEDOUTT(J)=SEDOUTT(J)+SEDOUT(SEDCLASS,J)
700 CONTINUECC AMOUNT OF DECOMPOSITION FROM PLANT LIFEC
SDECOMP(SEDCAT,J)=DECOMPR*SEDQTYB(SEDCAT,J-1)CC MASS BALANCE FOR SEDIMENTC
DO 800 SEDCLASS=1,SEDCATSEDQTYW(SEDCLASS,J)=SEDQTYW(SEDCLASS,J-1)+RESUSP(SEDCLASS,J)-
/ -SEDOUT(SEDCLASS,J)-SEDDEP(SEDCLASS,J)+SEDINW(SEDCLASS,J) 800 CONTINUE
237
DO 900 SEDCLASS=1,SEDCATSEDQTYB(SEDCLASS,J)=SEDQTYB(SEDCLASS,J-1)-RESUSP(SEDCLASS,J)
/ +SEDDEP(SEDCLASS,J)-SDECOMP(SEDCLASS,J) 900 CONTINUE
IF (J.EQ.K) THENDO 1000 SEDCLASS=1,SEDCAT
SEDQTYW(SEDCLASS,0)=SEDQTYW(SEDCLASS,J)SEDQTYB(SEDCLASS,0)=SEDQTYB(SEDCLASS,J)
1000 CONTINUEENDIF
CEND
238
DELTAHT SUBMODEL
SUBROUTINE DELTAH(DELTAHT,NITCYCLE,SEDCYCLE,SEDTOTAL,SEDDELTA, / SEDCAT,SEDSPG,PEATACRB,PEATDENS,HT,NUMTMPER,J,SEDQTYB,WATERB, / PORPEAT)
COMMON/DESCRIBE/LENGTH,WIDTH,HO,HB,SOREAL SEDDELTA(5),SEDQTYB(5,0:500),HT(0:500),PEATACRB,WATERB(0:500)REAL DELTAHT,PEATDENS,SEDSPG,PORPEAT,LENGTH,WIDTH,HO,HB,SOINTEGER NITCYCLE,SEDCYCLE,SEDCAT,J
IF (NITCYCLE.EQ.1 .AND. SEDCYCLE.EQ.0) THENDELTAHT=(PEATACRB/(PEATDENS*1000.))/(LENGTH*WIDTH)ENDIF
IF (NITCYCLE.EQ.0 .AND.SEDCYCLE.EQ.1) THENSEDTOTAL=0DO 75 SEDCLASS=1,SEDCATSEDDELTA(SEDCLASS)=SEDQTYB(SEDCLASS,J)-SEDQTYB(SEDCLASS,J-1)SEDTOTAL=SEDTOTAL+SEDDELTA(SEDCLASS)
75 CONTINUE
DELTAHT=(SEDTOTAL/(SEDSPG*1000000.))/(LENGTH*WIDTH)ENDIF
IF (NITCYCLE.EQ.1 .AND. SEDCYCLE.EQ.1) THENSEDTOTAL=0DO 85 SEDCLASS=1,SEDCAT SEDDELTA(SEDCLASS)=SEDQTYB(SEDCLASS,J)-SEDQTYB(SEDCLASS,J-1)SEDTOTAL=SEDTOTAL+SEDDELTA(SEDCLASS)
85 CONTINUE
DELTAHT=((SEDTOTAL/(SEDSPG*1000000.))+(PEATACRB/(PEATDENS* / 1000.)))/(LENGTH*WIDTH)
ENDIFHT(J)=HT(J-1)+DELTAHT
WATERB(J)=LENGTH*WIDTH*(HT(J)-HB)*PORPEAT
IF (J.EQ.NUMTMPER) THENHT(0)=HT(J)WATERB(0)=WATERB(J)
ENDIF
END
239
Appendix D: Symbol description for Figures 8 through 22
Figures 8 through 22 represent the relationships between the various pools and parametersthat are used to model the processes within each respective cycle of the SET-WET model. Thefigures detail what parameters are needed to determine the rate for each respective process andthe mass balances for each cycle. The following are descriptions for the symbols used.
240
Appendix E: Regression Graphs
FIGURE E.1.: SIMULATED AND OBSERVED VALUES FOR OUTFLOW, PLOTTED BESIDE THE DETERMINED LINEAR
REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.2.: SIMULATED AND OBSERVED VALUES FOR AMMONIUM CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
241
FIGURE E.3.: SIMULATED AND OBSERVED VALUES FOR NITRATE CONCENTRATION, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.4.: SIMULATED AND OBSERVED VALUES FOR ORGANIC NITROGEN CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
242
FIGURE E.5.: SIMULATED AND OBSERVED VALUES FOR DISSOLVED OXYGEN CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.6.: SIMULATED AND OBSERVED VALUES FOR BOD5 CONCENTRATIONS, PLOTTED BESIDE THE DETERMINED
LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
243
FIGURE E.7.: SIMULATED AND OBSERVED VALUES FOR TOTAL SUSPENDED SOLID CONCENTRATIONS, PLOTTED BESIDE
THE DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
FIGURE E.8.: SIMULATED AND OBSERVED VALUES FOR TOTAL PHOSPHOROUS CONCENTRATIONS, PLOTTED BESIDE THE
DETERMINED LINEAR REGRESSION WITH PREDICTION INTERVAL, AND IDEAL 1:1 LINE.
244
Appendix F: Sensitivity Analysis Tables
TABLE F.1: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 10% CHANGE IN BASE VALUESParameter Base Value NH4 NO3 DON PON DO BOD5 T SS DP T PBACTERIA
AEMAXGRB 0.01 (NC)/(+E) (NC)/(-E) (NC)/(-D) (NC)/(-C) (NC)/(-E) (NC)/(-C) A/A A/A A/AAEMAXGRW 0.05 (+D)/(+D) (-C)/(-C) (-D)/(-C) (-C)/(-C) (-D)/(-D) (-D)/(-C) A/A A/A A/AANMAXGRB 0.01 (+D)/(+D) (-D)/(-D) (-B)/(-B) (-B)/(-B) (-C)/(-C) (-B)/(-B) A/A A/A A/AANMAXGRW 0.05 (+B)/(+B) (-C)/(-C) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) A/A A/A A/A
HETROIB 240000 (NC)/(+E) (NC)/(-E) (NC)/(-C) (NC)/(-C) (NC)/(-E) (NC)/(-C) A/A A/A A/AHETEROIW 25000 (+D)/(+D) (-C)/(-C) (-C)/(-C) (-B)/(-B) (-D)/(-D) (-C)/(-C) A/A A/A A/AHNO3HSCB 0.15 (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B) A/A A/A A/AHNO3HSCW 0.15 (-B)/(-B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) A/A A/A A/AHORGHSCB 50 (-D)/(-D) (+C)/(+C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/AHORGHSCW 50 (-D)/(-D) (+C)/(+C) (+C)/(+C) (+C)/(+C) (+D)/(+D) (+C)/(+C) A/A A/A A/AHTDOHSCB 0.15 (-D)/(-E) (-C)/(-C) (+C)/(+C) (+C)/(+C) (+D)/(+D) (+C)/(+C) A/A A/A A/AHTDOHSCW 0.5 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/A
HTDRB 0.00125 (-D)/(-D) (+D)/(+D) (+B)/(+C) (+B)/(+B) (+D)/(+D) (+B)/(+B) A/A A/A A/AHTDRW 0.001 (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+C)/(+C) (+B)/(+B) A/A A/A A/A
NDOHSATB 1 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/ANDOHSATW 1 (-B)/(-B) (-C)/(-C) (+B)/(+B) (A)/(A ) (+C)/(+C) (+B)/(+B) A/A A/A A/A
NDRATEB 0.002 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/ANDRATEW 0.002 (-B)/(-B) (-C)/(-C) (-B)/(-B) (A)/(A ) (+C)/(+B) (-B)/(-B) A/A A/A A/ANITROSIB 2400 (+D)/(+D) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-B)/(-B) A/A A/A A/ANITROSIW 1000 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-C)/(-C) (-B)/(-B) A/A A/A A/ANMAXGRB 0.005 (+D)/(+D) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-B)/(-B) A/A A/A A/ANMAXGRW 0.005 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-B)/(-B) A/A A/A A/ANNH4HSCB 1 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+(C) (+B)/(+B) A/A A/A A/ANNH4HSCW 1 (-B)/(-B) (-C)/(-C) (+B)(+B) A/A (+C)/(+C) (+B)/(+B) A/A (-D)/(+D) (-D)/(+D)
NITROGENBIOMCN 23.5 (+E)/(-D) (+F)/(-B) (-C)/(-C) (-E)/(-D) (-D)/(+B) (-C)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)BIOMPN 95 (+E)/(+E) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-C)/(-D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DONINITB 10000 (-C)/(+C) (-B)/(+B) (+D)/(+D) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)DONINITW 5000 (-B)/(+C) (-B)/(+B) (+D)/(+D) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTNO3YB 3.29 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTNO3YW 3.29 (-C)/(+C) (+C)/(+C) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)IMMINITB 9900000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/A (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)IMMINITW 700000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MICRONC 0.125 (-D)/(-D) (-B)/(-B) (-B)/(+B) (+D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCDON 0.00002 (-B)/(+B) (-B)/(+B) (+C)/(+C) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCNH4 0.00006 (-C)/(+B) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCNO3 0.00006 (-B)/(+C) (+C)/(+C) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NH4INITB 55000 (+E)/(+E) (+C)/(+C) (-B)/(-B) (-D)/(+D) (-C)/(-C) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NH4INITW 30000 (+E)/(+D) (+B)/(+C) (-B)/(-B) (-D)/(+D) (-C)/(-C) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NO3INITB 4000 (+C)/(-E) (+D)/(-F) (-B)/(+C) (-D)/(+D) (-B)/(+C) (-B)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)NO3INITW 400 (+B)/(+C) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSYIELDB 0.3 (-C)/(+C) (-D)/(-D) (+B)/(+B) (-D)/(+D) (+B)/(+B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSYIELDW 0.3 (-C)/(+C) (-D)/(-D) (+B)/(+B) (-D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)ONPARTF 0.6 (+D)/(+D) (+B)/(+B) (-F)/(-F) (+E)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONCOUT 0.75 (-C)/(-C) (-B)/(-B) (-B)/(+B) (+E)/(+E) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONFALL 0.25 (-B)/(+C) (-B)/(+B) (-B)/(+B) (-E)/(-E) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONINITB 300000 (+E)/(-E) (+F)/(-F) (-C)/(+C) (-D)/(+D) (-D)/(+D) (-C)/(+C) (-B)/(+B) A/A A/APONINITW 15000 (+D)/(-D) (+F)/(-F) (-B)/(+B) (+D)/(+D) (-C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONRES 0.01 (-C)/(+C) (-B)/(+B) (-B)/(+B) (+D)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONSIZE 0.05 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)REFNINIT 500000 (+E)/(-E) (+F)/(-F) (-C)/(+C) (-C)/(+C) (-D)/(+D) (-C)/(+C) (-B)/(+B) (-B)/(+B) (-B /(+B)
RESTHN (NIT) 0.01 (-C)/(+C) (-B)/(+B) (-B)/(+B) (+C)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B /(+B) (-B)/(+B)
VEGETATIONBIODENS 5000 (+B)/(-B) A/(-B) (-B)/(-B) (-B)/(-B) A/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)BIOINIT 8635000 (-B)/(-B) (-B)/(-B) (+B)/(+B) (+D)/(+D) (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B)PBIOUW 0.4 (+B)/(+B) (+B)/A (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B)
PEATACRB 2000 (+C)/(+C) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B)PEATACRW 300 (+B)/(+B) (+B)/(+B) (-B)/(-B) (-C)/(-C) (+B)/(+B) (-B)/(-B) A/A A/A A/APEATDENS 0.7 (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (+B)/(+C) (-B)/(-B) (-B)/(-B)
245
TABLE F.1 (cont.): SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TOTHE BENTON WETLAND FOR (+/-) 10% CHANGE IN BASE VALUES
Parameter Base Value NH4 NO3 DON PON DO BOD5 TSS DP TPPRATEUP 0.7 (-C)/(-B) (+E)/(+D) (-B)/(-B) A/A (+B)/(+B) (-B)/(-B) A/A (+B)/(+B) (+B)/(+B)PSTDUW 0.4 (+B)/(+B) (+B)/A (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B)STANDIN 6000000 (-D)/(-D) (-C)/(-C) (+B)/(+B) (+D)/(+D) (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B)STDDENS 5000 (+B)/(+B) A/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)CARBON
BIOCCONT 0.47 (-D)/(-D) (-C)/(-C) (+B)/(+B) (+D)/(+E) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)BODCFRAC * 0.8 (-C)/(-C) (-C)/(-C) (+B)/(+B) (-D)/(+C) (-D)/(-D) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)BODPFRAC 0.5 (-D)/(-C) (+C)/(+C) (-B)/(-B) (-D)/(+D) (+C)/(+C) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOCINITB 60000 (-C)/(+B) (-B)/(-B) (+B)/(+B) (-C)/(+C) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOCINITW 50000 (-B)/(+C) (-B)/(-B) (+B)/(+B) (-D)/(+D) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)
LEACHR 0.01 (-C)/)(+C) (-B)/(+B) (+C)/(+C) (-D)/(+D) (-B)/(-B) (+D)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)MICROBEC 0.53 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCDOC 0.00004 (-C)/(+C) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)PEATCC * 0.8 (-C)/(+B) (-B)/(+B) (-B)/(+B) (-D)/(+C) (-B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
POCCOUT * 0.3 (-B)/(+C) (+B)/(+B) (-B)/(-B) (-D)/(+D) (+B)/(+B) (+E)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCFALL 0.45 (-D)/(-C) (+B)/(+C) (+B)/(-B) (-D)/(+D) (+C)/(+C) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCINITB 3000000 (-D)/(-D) (-C)/(-C) (+B)/(+C) (-C)/(+D) (-C)/(-C) (+B)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCINITW 200000 (-C)/(+C) (-B)/(-B) (+B)/(+B) (-D)/(+D) (-C)/(-C) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCRES 0.001 (-C)/(+B) (+B)/(+B) (+B)/(+B) (-D)/(+D) (+B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCSIZE 0.2 (-C)/(+C) (-B)/(+B) (-D)/(+D) (-C)/(+C) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)REFCINIT 15000000 (+B)/(+B) (+B)/(+B) (-B)/(+B) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DISSOLVED OXYGENDOCONCP 0.001 (-B)/(+B) (-B)/(+B) (-B/(+B) (-D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOINITB 15000 (-D)/(-D) (+C)/(+C) (+C)/(+C) (-D)/(+D) (-D)/(+D) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOINITW 15000 (-D)/(-C) (+C)/(+C) (+B)/(+C) (-D)/(+D) (-D)/(+D) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DOXYCSAT 8.5 (-E)/(-E) (+E)/(+D) (+D)/(+D) (-D)/(+D) (+F)/(+E) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTDOYB 0.15 (-E)/(NC) (+E)/(NC) (+C)/(NC) (-D)/(NC) (+E)/(NC) (+C)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTDOYW 0.15 (-D)/(-D) (+C)/(+C) (+B)/(+B) (-D)/(+D) (+D)/(+D) (+B)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTDOX 0.0001 (-E)/(NC) (+D)/(NC) (+C)/(NC) (-D)/(NC) (-C)/(NC) (+B)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)
MTFW SDOC 0.00008 (-D)/(-D) (+D)/(+D) (+C)/(+C) (-D)/(+D) (-D)/(-D) (+C)/(+C) (-B)/(-B) (-B)/(-B) (-B)/(-B)NSDOYB 0.2 (-D)/(-D) (+C)/(+C) (+B)/(+C) (-D)/(+D) (+D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSDOYW 0.2 (-C)/(-C) (+B)/(+C) (+B)/(+B) (-C)/(+C) (+C)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
PHOSPHOROUSBIOMPP 300 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+D)/(+D) (+D)/(+D)BTPHOSI 250000 (+B)/(+C) (+B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)
DTPHOSIB 40000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+E)/(+E) (+E)/(+E)DTPHOSIW 38000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+E)/(+E) (+E)/(+E)MTCPHOS 0.00006 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)PMINPPC 0.05 (-C)/)(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)
PRMINBPC 0.00005 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)PSEDDEP 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)(-C)
SEDIMENTDECOMPR 0.1 (-C)/(+C) (-B)/(+B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(-B) (-B)/(-B)
MANNC (SED) 2.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+E) (-B))/(+C) (-B)/(+C)PSEDDEP 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)/(-C)
RESTHICK (SED) 0.0011 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (+E)/(+E) (+C)/(+C) (+C)/(+C)SEDFALL1 0.3 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)/(-C)SEDFALL2 0.7 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(-E) (-B)/(-B) (-B)/(-B)SEDINITB1 26000000 (-B)/(+B) (-B)/(+B) (-B)/(+B) (-C)/(+C) (-B)/(+B) (-B)/(+B) (+E)/(+E) (-B)/(+B) (-B)/(+B)SEDINITB2 10000000 (-C)/(+C) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (+B)/(+B) (-B)/(-B) (-B)/(-B)SEDINITW 1 190000 (-B)/(+B) (-B)/(+B) (-B)/(+B) (-C)/(+C) (+B)/(+B) (-B)/(+B) (+E)/(+E) (-B)/(+B) (+B)/(+B)SEDINITW 2 10000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (-B)/(+B) (-B)/(-B) (-B)/(-B)
SEDRES 0.1 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (+E)/(+E) (+C)/(+C) (+C)/(+C)SEDSIZE1 0.25 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+E) (-B)/(+B) (-B)/(+B)SEDSIZE2 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+B)/(+B) (+B)/(+B)SEDSPG1 1.1 (-C)/(+C) (-B)/(+B) (+B)/(+B) (-D)/(+D) (+B)/(+B) (+B)/(+B) (-B)/(-B) (-B)/(+B) (-B)/(-B)SEDSPG2 2.65 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (-B)/(+B) (+B)/(+B) (+B)/(+B)
- Results presented as the RS values for the +10% change in base value followed by the –10% change.- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible
results)
246
TABLE F.2: SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TO THE BENTON
WETLAND FOR (+/-) 25% CHANGE IN BASE VALUESParameter Base Value NH4 NO3 DON PON DO BOD5 TSS DP TPBACTERIA
AEMAXGRB 0.01 (NC)/(+E) (NC)/(-E) (NC)/(-D) (NC)/(-C) (NC)/(-E) (NC)/(-C) A/A A/(-D) A/AAEMAXGRW 0.05 (+D)/(+D) (-C)/(-C) (-D)/(-C) (-C)/(-C) (-D)/(-D) (-D)/(-C) A/A A/A A/AANMAXGRB 0.01 (+D)/(+D) (-D)/(-D) (-B)/(-B) (-B)/(-B) (-C)/(-C) (-B)/(-B) A/A A/A A/AANMAXGRW 0.05 (+B)/(+B) (-C)/(-C) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) A/A A/A A/A
HETROIB 240000 (NC)/(+E) (NC)/(-E) (NC)/(-C) (NC)/(-C) (NC)/(-E) (NC)/(-C) A/A A/A A/AHETEROIW 25000 (+D)/(+D) (-C)/(-C) (-C)/(-C) (-B)/(-B) (-D)/(-D) (-C)/(-C) A/A A/A A/AHNO3HSCB 0.15 (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B) A/A A/A A/AHNO3HSCW 0.15 (-B)/(-B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) A/A A/A A/AHORGHSCB 50 (-C)/(-D) (+C)/(+C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/AHORGHSCW 50 (-D)/(-D) (+C)/(+C) (+C)/(+D) (+C)/(+C) (+D)/(+D) (+C)/(+C) A/A A/A A/AHTDOHSCB 0.15 (-D)/(NC) (-C)/(NC) (+C)/(NC) (+C)/(NC) (+D)/(NC) (+C)/(NC) A/A A/A A/AHTDOHSCW 0.5 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/A
HTDRB 0.00125 (-D)/(-E) (+D)/(+D) (+B)/(+C) (+B)/(+C) (+D)/(+D) (+B)/(+C) A/A A/A A/AHTDRW 0.001 (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+C)/(+C) (+B)/(+B) A/A A/A A/A
NDOHSATB 1 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/ANDOHSATW 1 (-B)/(-B) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/A
NDRATEB 0.002 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A A/A A/ANDRATEW 0.002 (-B)/(-B) (-C)/(-C) (-B)/(-B) (-B)/(-B) (+C)/(+B) (-B)/(-B) A/A A/A A/ANITROSIB 2400 (+D)/(+D) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-B)/(-B) A/A A/A A/ANITROSIW 1000 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-C)/(-C) (-B)/(-B) A/A A/A A/ANMAXGRB 0.005 (+D)/(+D) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-C)/(-B) A/A A/A A/ANMAXGRW 0.005 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-B)/(-B) (-D)/(-D) (-B)/(-B) A/A A/A A/ANNH4HSCB 1 (-C)/(-C) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+(C) (+B)/(+B) A/A A/A A/ANNH4HSCW 1 (-B)/(-B) (-C)/(-C) (+B)/(+B) (+B)/(+B) (+C)/(+C) (+B)/(+B) A/A (-D)/(+D) (-D)/(+D)
NITROGENBIOMCN 23.5 (+E)/(-D) (+F)/(-B) (-C)/(-C) (-D)/(-E) (-C)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)BIOMPN 95 (+E)/(+E) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-C)/(-D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DONINITB 10000 (+B)/(+C) (-B)/(+B) (+D)/(+D) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)DONINITW 5000 (-B)/(+C) (-B)/(+B) (+D)/(+D) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTNO3YB 3.29 (+C)/(+C) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTNO3YW 3.29 (-C)/(+C) (+C)/(+C) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)IMMINITB 9900000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)IMMINITW 700000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MICRONC 0.125 (-C)/(-D) (-B)/(-B) (-B)/(+B) (+D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCDON 0.00002 (-B)/(+B) (-B)/(+B) (+C)/(+C) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCNH4 0.00006 (-C)/(-C) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCNO3 0.00006 (+C)/(+C) (+C)/(+D) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NH4INITB 55000 (+E)/(+E) (+C)/(+C) (-B)/(-B) (-D)/(+D) (-C)/(-C) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NH4INITW 30000 (+D)/(+D) (+B)/(+C) (-B)/(-B) (-D)/(+D) (-C)/(-C) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NO3INITB 4000 (+C)/(-E) (+D)/(-F) (-B)/(+C) (-D)/(+D) (-B)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NO3INITW 400 (+B)/(+C) (+D)/(+D) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSYIELDB 0.3 (-C)/(+C) (-D)/(-D) (+B)/(+B) (-D)/(+D) (+B)/(+B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSYIELDW 0.3 (-C)/(+C) (-D)/(-D) (+B)/(+B) (-D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)ONPARTF 0.6 (-C)/(-C) (+B)/(+B) (-F)/(-F) (+E)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONCOUT 0.75 (-C)/(-C) (-B)/(-B) (-B)/(+B) (+E)/(+E) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONFALL 0.25 (+C)/(+C) (-B)/(+B) (-B)/(+B) (-E)/(-F) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONINITB 300000 (+E)/(-D) (+F)/(-F) (-D)/(+C) (-D)/(+D) (-C)/(+C) (-B)/(+B) (-B)/(+B) A/A A/APONINITW 15000 (+E)/(-E) (+F)/(-F) (-B)/(+B) (+D)/(+D) (-C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONRES 0.01 (-C)/(+C) (-B)/(+B) (-B)/(+B) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)PONSIZE 0.05 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)REFNINIT 500000 (+E)/(-E) (+F)/(-F) (-B)/(+B) (-C)/(+C) (-C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B/(+B)
RESTHN (NIT) 0.01 (-C)/(+C) (-B)/(+B) (-B)/(+B) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B/(+B) (-B)/(+B)
VEGETATIONBIODENS 5000 (+B)/(-B) A/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)BIOINIT 8635000 (-B)/(-B) (-B)/(-B) (+B)/(+B) (+D)/(+D) (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B)PBIOUW 0.4 (+B)/(+B) (+B)/A (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B)
PEATACRB 2000 (+C)/(+C) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B) (-B)/(-B) (+B)/(+B) (+B)/(+B)PEATACRW 300 (+B)/(+B) (+B)/(+B) (-B)/(-B) (-C)/(-C) (+B)/(+B) (-B)/(-B) A/A A/A A/APEATDENS 0.7 (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (+B)/(+C) (-B)/(-B) (-B)/(-B)
247
TABLE F.2 (cont.): SENSITIVITY ANALYSIS RESULTS OF SET-WET MODEL AS APPLIED TOTHE BENTON WETLAND FOR (+/-) 25% CHANGE IN BASE VALUES
Parameter Base Value NH4 NO3 DON PON DO BOD5 TSS DP TPPRATEUP 0.7 (-C)/(-B) (+E)/(+D) (-B)/(-B) (-B)/(-B) (+B)/(+B) (-B)/(-B) A/A (+B)/(+B) (+B)/(+B)PSTDUW 0.4 (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B) (+B)/(+B)STANDIN 6000000 (-D)/(-D) (-C)/(-C) (+B)/(+B) (+D)/(+D) (-C)/(-C) (+D)/(+D) (+B)/(+B) (+B)/(+B) (+B)/(+B)STDDENS 5000 (+B)/(+B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B) (-B)/(-B)CARBON
BIOCCONT 0.47 (-D)/(-D) (-C)/(-C) (+B)/(+B) (+D)/(+E) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)BODCFRAC * 0.8 (-C)/(-C) (-C)/(-C) (+B)/(+B) (-D)/(+C) (-D)/(-D) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)BODPFRAC 0.5 (-D)/(-C) (+C)/(-C) (-B)/(-B) (-D)/(+D) (+D)/(-D) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOCINITB 60000 (-B)/(+B) (-B)/(-B) (+B)/(+B) (-C)/(+C) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOCINITW 50000 (-B)/(+C) (-B)/(-B) (+B)/(+B) (-D)/(+D) (-C)/(-C) (+D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B)
LEACHR 0.01 (-C)/)(+C) (-B)/(+B) (+C)/(+C) (-D)/(+D) (-B)/(-B) (+D)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)MICROBEC 0.53 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTCDOC 0.00004 (-C)/(+C) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)PEATCC * 0.8 (-C)/(+B) (-B)/(+B) (-B)/(+B) (-D)/(+C) (-B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
POCCOUT * 0.3 (+B)/(+C) (+B)/(+B) (-B)/(-B) (-D)/(+D) (+B)/(+B) (+E)/(+E) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCFALL 0.45 (-C)/(-C) (+B)/(+C) (+B)/(-B) (-D)/(+D) (+B)/(+C) (-E)/(-E) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCINITB 3000000 (-D)/(-D) (-C)/(-C) (+B)/(+C) (-C)/(+D) (-C)/(-C) (+B)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCINITW 200000 (-C)/(-C) (-B)/(-B) (+B)/(+B) (-D)/(+D) (-C)/(-C) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCRES 0.001 (-C)/(+C) (+B)/(+B) (+B)/(+B) (-D)/(+D) (+B)/(-B) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)POCSIZE 0.2 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(-D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)REFCINIT 15000000 (+B)/(+B) (+B)/(+B) (-B)/(+B) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DISSOLVED OXYGENDOCONCP 0.001 (-B)/(+B) (-B)/(+B) (-B/(+B) (-D)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOINITB 15000 (-D)/(-D) (+C)/(+C) (+C)/(+C) (-D)/(+D) (+D)/(+D) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)DOINITW 15000 (-C)/(-C) (+C)/(+C) (+B)/(+C) (-D)/(+D) (+D)/(+D) (+C)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)
DOXYCSAT 8.5 (-E)/(NC) (+D)/(NC) (+C)/(NC) (-D)/(NC) (+E)/(NC) (+C)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTDOYB 0.15 (-E)/(NC) (+E)/(NC) (+C)/(NC) (-D)/(NC) (+E)/(NC) (+C)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)HTDOYW 0.15 (-C)/(-D) (+C)/(+C) (+B)/(+C) (-D)/(+D) (+D)/(+D) (+B)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)MTDOX 0.0001 (-E)/(NC) (+D)/(NC) (+C)/(NC) (-D)/(NC) (-C)/(NC) (+B)/(NC) (-B)/(+B) (-B)/(+B) (-B)/(+B)
MTFWSDOC 0.00008 (-D)/(-D) (+D)/(+D) (+C)/(+C) (-D)/(+D) (-D)/(-D) (+C)/(+C) (-B)/(-B) (-B)/(-B) (-B)/(-B)NSDOYB 0.2 (+D)/(+D) (+C)/(+C) (+B)/(+C) (-D)/(+D) (+D)/(+D) (+B)/(+C) (-B)/(+B) (-B)/(+B) (-B)/(+B)NSDOYW 0.2 (-C)/(-C) (+B)/(+C) (+B)/(+B) (-C)/(+C) (+C)/(+D) (+B)/(+B) (-B)/(+B) (-B)/(+B) (-B)/(+B)
PHOSPHOROUSBIOMPP 300 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+D)/(+D) (+D)/(+D)BTPHOSI 250000 (+C)/(+C) (+B)/(+B) (-B)/(+B) (-C)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)
DTPHOSIB 40000 (-B)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+E)/(+E) (+E)/(+E)DTPHOSIW 38000 (-B)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (+E)/(+E) (+E)/(+E)MTCPHOS 0.00006 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)PMINPPC 0.05 (-C)/)(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)
PRMINBPC 0.00005 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+C)/(+C) (+C)/(+C)PSEDDEP 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)(-C)
SEDIMENTDECOMPR 0.1 (-C)/(+C) (-B)/(+B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+B) (-B)/(-B) (-B)/(-B)
MANNC (SED) 2.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(+E) (-B))/(+C) (-B)/(+C)PSEDDEP 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)/(-C)
RESTHICK (SED) 0.0011 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (+E)/(+E) (+C)/(+C) (+C)/(+C)SEDFALL1 0.3 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-F)/(-F) (-C)/(-C) (-C)/(-C)SEDFALL2 0.7 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(-B) (-B)/(+B) (-B)/(-F) (-B)/(-B) (-B)/(-B)SEDINITB1 26000000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (+E)/(+E) (-B)/(+B) (-B)/(+B)SEDINITB2 10000000 (-C)/(+C) (-B)/(-B) (-B)/(-B) (-D)/(+D) (-B)/(-B) (-B)/(-B) (+B)/(+B) (-B)/(-B) (-B)/(-B)SEDINITW1 190000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (+B)/(+B) (-B)/(+B) (+E)/(+E) (-B)/(+B) (+B)/(+B)SEDINITW2 10000 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (-B)/(+B) (-B)/(-B) (-B)/(-B)
SEDRES 0.1 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (+E)/(+E) (+C)/(+C) (+C)/(+C)SEDSIZE1 0.25 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+E) (-B)/(+B) (-B)/(+B)SEDSIZE2 0.5 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/(+B) (-B)/(+B) (-B)/(+B) (+B)/(+B) (+B)/(+B)SEDSPG1 1.1 (-C)/(+C) (-B)/(+B) (+B)/(+B) (-D)/(+D) (+B)/(+B) (+B)/(+B) (-B)/(-B) (-B)/(+B) (-B)/(-B)SEDSPG2 2.65 (-C)/(+C) (-B)/(+B) (-B)/(+B) (-D)/(+D) (-B)/((+B) (-B)/(+B) (-B)/(+B) (+B)/(+B) (+B)/(+B)
- Results presented as the RS values for the +10% change in base value followed by the –10% change.- (+/-) sign represents whether direct (+) or indirect (-) relationship between parameter and output change.- RS value equals: A (0); B (0 to .001); C (.001 to .01); D (.01 to .1); E (.1 to 1.0); F (>1.0); NC (incomprehensible
results)
248
VITA
Erik Ryan Lee
Erik Ryan Lee was born and mostly raised his entire life in San Francisco, California.
He attended Lowell High School in San Francisco, where he was involved in government, choir,
and track. He then attended the University of California at Davis and received his bachelor of
science in Biological Systems Engineering with an emphasis on ecology in 1997.
He made a drastic change by moving to what he considers the country in Virginia, and for
the past two years has attended Virginia Polytechnic Institute and State University. Here he has
learned to appreciate research, having seasons, and carrying an umbrella. In September of 1999,
he finished his thesis after taking a couple of wrong turns, and realizing that there is a
tremendous amount of research that needs to be done.
At the dawn of a new millenium, his future plans include seeing the world, living a full
and content life, and cleaning his native San Francisco Bay. The past is in the rear view mirror,
only the future lies ahead.
The ultimate measure of a man is not where he stands in moments of comfort and conveniencebut where he stands in times of challenge and controversy.
- Martin Luther King Jr.