Watershed Assessment and Decision Support Tools: e- Estuary – A Decision-Support System for Coastal Mana gers (WQ MYP)N. Detenbeck, M. Pelletier, D. Campbell, M. ten Brink, S. Rego, M. Abdelrhman, E. Dettmann, J. Latimer, K. Ho, R. Burgess, K. Perez

Agency ProblemWatershed and coastal zone management are supported by mult iple Clean Water Act (CWA) programs and the Coastal Zone Act Reauthorization Amendments (e.g., NonPointSource program in Section 6217), and require the integration of regulatory and nonregulatory programs. EPA’s Office of Water is evolving to a more pro-active approach to manage environmental problems than is typical of the Section 303(d) process (Watershed Protection Approach Framework, 1991). Watershed and ecosystem-based management approaches are more integrative and comprehensive than the stressor-by-stressor approach supported by individual chemical water quality criteria and the tradit ional Total Maximum Daily Load process. This new approach also typically involves engagement of mult iple stakeholder groups in planning and adaptive management at all scales (township, estuary, watershed, state, regional). In response, our research and development approach will provide technical support for managers in their development and implementation of watershed and coastal management plans at mult iple scales.

The research and development activit ies described above are providing support for mult iple programs in the Office of Water and related activit ies by EPA Regions, the states, tribes, and local coastal or watershed management groups. Activit ies will provide the basis for:

1) the definit ion of estuarine list ing segments for combined 305(b)/303(d) reports the states are required to submit

2) a framework for refining aquatic life use designations in estuaries and exploring t iered aquatic life use concepts at an ecosystem scale

3) classification frameworks to support monitoring and assessment

4) priorit ization of subwatersheds for management activit ies, and water quality criteria development (particularly for nutrients)

5) diagnosis of causes of impairment in estuaries

6) partit ioning effects of local point source or contaminated sediments from upstream nonpoint sources

7) establishing restoration targets

8) ult imately supporting cost-benefit analyses of alternative management strategies for estuaries

Stressor-response models by system class

(see related posters)

Whole- and subsystem-retention time

(see related posters)

US EPA Tiered Aquatic Life Use workgroup(s): National, Estuarine

US EPA Landscape Predictive Tools Workgroup

Potential Habitat-Use Zones –We are defining and mapping habitat constraints for biotopes to be used as the basis for refining aquatic life use(s) in estuaries, starting with seagrass.

US EPA National Estuarine Experts

Workgroup(s)

Watershed Central

Which estuaries are like ly to respond to pollutant loadings or restoration actions similar ly to my estuary? How can I identify an appropriate reference estuary or habitat

for comparison with my system?

How can I segment estuaries into homogeneous units that are appropriate for assigning

designated use, report ing condition, list ing impairments, and planning management act ions, and do so in a consistent fashion?

How can I predict potentia l habitat use for different portions of an estuary and use this

information to refine designated use for aquatic life?

How can I est imate pollutant loadings

to estuaries in a consistent fashion? What data are required to parameterize s imple loading models?

How can I predict or diagnosethe effects

of mult iple stressors on estuarine systems?

What can I infer from benthic community composit ion data on the like ly nature of stressors

influencing my system of interest?

A common geospatial framework for organizing, sharing, interpreting, analyzing, and transferring

information on coastal systems to clients for environmental decision-making

Both SAV and phytoplankton light-limited

SAV light limited primarily by chlorophyll a, phytoplankton not light limited

SAV light limited primarily by TSS

From Gallegos (2001)

An initial classification of estuaries has been performed to group systems by similarity in geomorphology and tidal regime. Efforts are underway to test the assumption that systems within a class will respond similarly to a given level of nutrient loading and to refine the classification system to support

development of nutrient criteria.

Output and tools: Map of estuarine classes and suiteof nutrient-response relationships by class

02

04

06

0810

03

05

07

11

0802

01Interbasin HUCs/ segments

Basin HUCs/ segments

Load = a*Areaurb+ b*Areaag- c*Areawtld+…

Protection of water clarity in estuaries requires management of multiple water quality parameters, i.e., nutrients, turbidity, and water color. Graphical tools have been developed and applied to a small set of estuaries. With appropriate calibration, these

approaches could be extended to a much larger suite of systems.

Estuaries can be segmented by local residence time, and segments coded in a hierarchical fashion to efficiently represent upstream-downstream relationships and connections with adjacent watersheds.

Point coverages of habitat features can be interpolated to create continuous coverages

which can be combined to predict zones of potential habitat use.

Look-up tables can be provided with loading coefficients by land-use type for different coastal regions.

Open Water (fish & shellfish use)

Deep water (seasonal fish & shellfish use)

Deep channel (seasonal refuge use)

Shallow water (bay grass use)

suspensionmoderately

biginfaunasabellid

polychaetePotamilla

renifor mes

omni voremoderately

biginfauna-deeperpolychaete

Nephtysbucera

brood/direct developmentsuspension

infauna-surfaceamphipod

Ampeliscaabdita

brood/direct developmentsuspensionepifauna

mysidshrimp

Americamysisbahia

ReproductionFeeding

TypeSizeBenthic

TypeCommon

NameSpecies

Information on benthic invertebrate life history and tolerance values can be used to interpret assessment data and to develop meaningful benthic indices

Science Inputs

Stakeholder Groups

Tools

Watershed and Estuarine Classification

Risk Assessment: Impact to Outcome –Watershed classification provided the principles behind watershed-based survey designs, the subject of a special session in the EMAP 2003 symposium. Principles were applied through technical support to Office of Pesticide Programs in their development of a national atrazine monitoring program as part of the 2003 re-registration process for atrazine.

Nutrient criteria guidance – Classification concepts have been incorporated into draft nutrient criteria guidance for wetlands and are being included in draft guidance on approaches to nutrient criteria for estuaries. In collaboration with an interagency workgroup, we will be comparing three approaches to classification of estuaries: a conceptual hydrogeomorphic model, EPA/NOAA cluster analysis, and a new Bayesian Classification and Regression Tree approach.

Extending STAR Grant ResultsSite-Specific Diagnostic Tools

(see related posters)

Supporting Science BaseAbdelrhman, M.A. 2005. Simplified modeling of flushing and residence t imes in 42 embayments in

New England, USA, with special attention to Greenwich Bay, Rhode Island. Estuarine, Coastal and Shelf Science 62: 339–351

Abdelrhman, M.A. 2007. Embayment characterist ic t ime and biology via t idal prism model. Estuarine Coastal and Shelf Science 74:643-56.

Brown, B, N Detenbeck, and R Eskin (2005) Integrating 305(b) and 303(d): How EMAP Aids in Monitoring and Assessment of State Waters. Environmental Monitoring and Assessment (special issue). Environmental Monitoring and Assessment 103:41-57. (DOI: 10.1007/s10661-005-6854-0)

Campbell DE, Detenbeck NE, Ho KT, Hill BH, Kurtz JC, Burgess RM, Engle VD, Perez KT, Pelletier MC, Snarski V. 2006. Conceptual models and methods for diagnosing the causes of impairments to aquatic systems. U. S. Environmental Protection Agency, Atlantic Ecology Division, Narragansett, RI. 600/R-06/024.

Detenbeck NE, Batterman SL, Brady VJ, Brazner JC, Snarski VM, Taylor DL and Thompson JA (2000) A test of watershed classification systems for ecological risk assessment. Environmental Toxicology and Chemistry 19:1174-1181.

Detenbeck NE, Elonen CM, Taylor, DL, Anderson, LE, Jicha, TM, and Batterman SL. (2003a) Effects of hydrogeomorphic region, watershed storage and mature forest on baseflow and snowmelt stream water quality in second-order Lake Superior Basin tributaries. Freshwater Biology 48(5):911-27.

Detenbeck, NE, LA Jagger, SL Stark, and MA Starry (2003b WV REMAP Final Report: Watershed Classification Framework for the State of West Virginia. US Environmental Protection Agency, NHEERL, Mid-Continent Ecology Division, Duluth, MN, EPA 600/R-03/141. 55 p.

Detenbeck, Naomi E., Colleen M. Elonen, Debra L. Taylor, Leroy E. Anderson, Terri M. Jicha, and Sharon L. Batterman. (2004). Region, Landscape, and Scale Effects on Lake Superior Tributary Water Quality. Journal of the American Water Resources Association (JAWRA) 40(3):705-720.

Detenbeck, N.E., V.J. Brady1, D.L. Taylor, V.M. Snarski and S.L. Batterman. 2005a. Relationship of stream flow regime in the western Lake Superior basin to watershed type characterist ics. Journal of Hydrology. 309(1-4): 258-276 (doi:10.1016/j.jhydrol.2004.11.024).

Detenbeck NE, D Cincotta, JM Denver, SK Greenlee, AR Olsen, and A Pitchford. (2005b) Watershed- based survey designs. Environmental Monitoring and Assessment (special issue). 103:59-81. (DOI: 10.1007/s10661-005-4774-7)

Detenbeck, NE, MM Moffett, M Pearson, TP Simon, and D Albert. 2007. Flow and nutrient-based classification for Lake Michigan coastal riverine wetlands. Chapter 4 in Coastal Wetlands of the Laurentian Great Lakes, TP Simon and PM Stewart (eds.), AuthorHouse. Bloomington, IN. pp. 115-136.

Gallegos C.L. 2001. Calculating optical water quality targets to restore and protect submerged aquatic vegetation: Overcoming problems in partit ioning the diffuse attenuation coefficient for photosynthetically active radiation. Estuaries 24(3):381-397.

Hagy, J. D., L. P. Sanford, and W. R. Boynton. 2000. Estimation of net physical transport and hydraulic residence t imes for a coastal plain estuary using box models. Estuaries 23(3): 328-340.

Kurtz, J.C., N.D. Detenbeck, V.D. Engle, K. Ho, L.M. Smith, S.J. Jordan and D.Campbell. 2006. Classifying Coastal Waters: Current Necessity and Historical Perspective. Estuaries and Coasts 29: 107-123.

U.S. Environmental Protection Agency, 2004, Classification Framework for Coastal Systems,. U.S. Environmental Protection Agency, National Health and Environmental Effects Research Laboratory. EPA Report EPA-600/R-04/061.

Water quality

Hydrology

Habitat

Algal biomass and community

Macroinver-tebratecommunity

Fish community

Translate to common codes and units

Filter, aggregate by t ime

Watershed variables influencing nutrient-response

Calculator for existing biological metrics

Mult ivariate analysis to derive new diagnostic indicators

Historic data

Regional Methods Initiative Project with EPA Region 5Nutrient Regional Technical Advisory Group (RTAG)(states, tribes, interstate commissions, USGS, USDA)

node 2, r2=0.83, p<0.0001)log10(sestonic chl a)

= 0.88 + 1.21log10(TN)

We have successfully used watershed classification by hydrologic regime as the basis for a statewide watershed monitoring and assessment design for WV.

350 - 1000----1.8 -> 4725 -> 35T. testudinum

88m mol /m2----1.8 -> 28.820 -> 35H. wrightii

0-1--4.7 -> 900 -> 15R. maritima

400 (max)2.3 -> .4 14 -> 295 -> 35Z. marina

Sulfide Content (mM)

Wave tolerance (m)

Current Velocity (m/s)

Fine sediment range (%)

S Range(ppt)

15.0634.2 ly/hr 46.7 ly/hr4025 -> 32T. testudinum

9.5746.2 ly/hr56.6 ly/hr22 -> 23520 -> 30H. wrightii

12370011 -> 8820 -> 30R. maritima

331403000 -> 355 -> 20Z. marina

Min. l t. requirement (%)

Half saturation constant (µµµµE/m2/s)

Lt. Sat. (µµµµE/m2/s)

CompensationPoint (µµµµmol/m2/s)

Temp.Range (°C)

Table 2: Data for the four most commonly monitored North American estuarine seagrass species: Zostera marina, Ruppia maritima, Halodule wrightii, and Thalassia testudinum .

Watershed Classes in Western A lleghany Pl ateau

M ap l ayo ut prod uc ed u nd er th e FAIR I I Co ntrac t 6 8-W0 1-03 2 T ask Ord er #0 24

0 8040Ki lom eters

LegendState Boundaries

Level III Ecoreg ions

Sample Points

Low Im pac t/Low Storage

Low Im pac t/H igh Storage

H igh Urbaniz ation/Low Storage

H igh Urbaniz ation/High Storage

M oderate Agric ul ture /Low Storage

M oderate Agric ul ture /High Storage

H igh Agricu lture /Low Storage

H igh Agricu lture /H igh Storage

Mid-Continent Ec ology Division6201 Congdon BlvdDuluth, MN 55804Apri l 25, 2003

0

1

2

3

4

5

Pre

dic

ted

Q2

/wa

ters

he

d a

rea

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

PRQ2A

Fraction watershed storage

(lake + wetland area/watershed area)

(1)

(1)

Georeferencing to NHD stream reach IDs

QA’d, documented, matched data

Flow Responsiveness for N Atlantic States

Impact and Outcomes:

The basic conceptual model includes three elements: Residence t ime, Type A modifying factors that determine bioavailability, and Type B modifying factors that alter the relationship between the stressor and the response. Four canonical forms of this model are shown above.

A. Unidirectional flow B. Bidirectional flow C . Unidirectional flow, stratified D. Bidirectional flow, stratified

W ater Storage

Proces sing

Inf low

Outflow

Residence Time

Bio effe ct iv eConce ntra tio n

M

Wat er

P X

P = Pollut ant M = Modifying Factors

T ype A and B

X

Po llu tan t

W ate r Sto rage

Proc es sing

Inf low

Outflow

Residence Time

Bi o ef fe ct iv eC once nt ra ti o n

M

Wat er

P X

P = Pollut ant M = Mod ify ing Fac to rs

T ype A a nd B

X

Po llu tan t

Water Storage

Proces sing

Inf low

Net Export orImport

Resid en ce Time

Bio effect iv eConc entration

M

Water

PX

P = Pollutant M = Modifying Factors

Type A and B

M

Water

PWater Storage

Processing

Inf low

Net Export orImport

Resid en ce Time

B io ef fect iv eC onc ent rati on

M

Water

PM

Water

PX

P = Pollutant M = Modifying Factors

Type A and B

M

Water

PM

Water

P

M = M odif ying FactorsType A and B

Water StorageI nfl ow

Outf low

Residence Time 1

Process ing 1Bio effec tiv e

Con centra tio nM

Water

P X

P = Pollutant

Pr ocess ing 2B io effec ti v e

C onc ent rat i on

Residence Time 2

X

M = M odif ying FactorsType A and B

Water StorageInfl ow

Outflow

Residence Time 1

Process ing 1B io effec ti v e

C on cent ra ti o n

Process ing 1B io effec ti v e

C on cent ra ti o nM

Water

P X

P = Pollutant

Pr ocess ing 2B io effec ti v e

C onc ent rat i on

Pr ocess ing 2B io effec ti v e

C onc ent rat i on

Residence Time 2

X

Wa ter Stora geI nflow

Net E xport orImport

Residence Time 1

Process ing 1Bioe ffect i ve

Concent rat ion

M

Water

P X

P = Pollutant M = Modifying Factors

Type A and B.

M

Water

P

Pr ocess ing 2B io ef fec ti ve

C oncen tr at i on

Res idence Time 2

Water StorageI nflow

Net E xport orImport

Residence Time 1

Process ing 1Bi oe ff ect i ve

C oncent r at i on

Process ing 1Bi oe ff ect i ve

C oncent r at i on

M

Water

PM

Water

P X

P = Pollutant M = Modifying Factors

Type A and B.

M

Water

PM

Water

P

Pr ocess ing 2B io ef fec ti ve

C oncen tr at i on

Pr ocess ing 2B io ef fec ti ve

C oncen tr at i on

Res idence Time 2

Conceptual Models

Past Ongoing

Research GoalsOur framework builds upon existing conceptual models to explain differences in response across systems using classification strategies and site-specific diagnostic tools. Our specific objectives are to:

• Develop a supporting information management framework for estuaries and coastal watersheds that is compatible with ongoing init iat ives within the Agency

• Develop and test classification frameworks for coastal watersheds, estuaries, and estuarine segments that facilitate extrapolation of results across systems

• Test segmentation schemes for improved list ing, assessment, and management of impaired waters in estuaries

• Provide scientific basis for refining aquatic life designated use(s) and developing the t iered aquaticlife use framework for estuaries

• Partit ion effects of mult iple stressors to prioritize management actions for regulatory vs. nonregulatoryprograms

• Support technical transfer of information and tools to coastal management groups, including collaborative development of tools with local or regional pilot projects.

Region 1: New EnglandServing Connecticut, Maine, Massachusetts, New Hamp shire,

Rhode Island, Vermont & 10 Tribal Nations

Web-based Interface

Analytical tools

Metadata

Bibliography

Link to experts

Extract

Interpolate

Extrapolate

Links to geospatial data

NWIS

Georeferencedrelational database

Future Directions• Geodatabase

• Development of a geodatabase structure suitable for 4-D estuarine and associated watershed data• Incorporation of historic estuarine data (NCA, NWIS, STORET, NERRS, NEPs, etc) and real-t ime data

streams (COOS, NOAA)

• Classification

• Comparison of three classification approaches to support derivation of nutrient criteria for estuaries• Addit ion of temporal component with atmospheric-ocean teleconnection indices + hydroclimatic zones

• Merger of watershed and estuarine classification approaches

• Site-specific diagnostic tools

• Completion of database of benthic community attributes for diagnostic analysis

• TOC-grain size relationships• Quantile regression to partit ion effects of mutliple stressors

• Outreach

• ERF 2007 special session and panel discussion on decision-support tools for coastal management (NGOs, NEPs, NOAA, states, regions, academia)

• Coordination with other EPA init iat ives• Watershed Central (watershed management support tools)

• Landscape Predictive Tools workgroup products• Coordination with sister agency init iat ives

• NOAA NowCOAST

• NOAA NE Integrated Ecosystem Assessments• NatureServe Ecosystem-based management tools

Compare nutrient-response curves

Classify systems by response- CART (average)- Bayesian CART (model)

Classify systems (indep. vars)- Cluster analysis

Filter using common data quality standards

Detenbeck_WatershedAssessPeer07.ppt 9/17/07

Monitoring, information mgt.,and visualization tools(see related posters)

ME

-2

ME

-1

SC

-2

DE

-2

NC

-1

CT-

1

SC

-4

NH

-1

CT-

2

NH

-2

CT-

4

SC

-3

DE

-3

NC

-2

CT-

3

SC

-1

MD

-2

NH

-3

NC

-4

NC

-3

DE

-1

NC

-5

MD

-1

State-Class

0

10

20

30

2-ye

ar p

eak

disc

harg

e/w

ater

shed

are

a (m3 . s

ec-1

. km

-2 )

Using data in USGS reports, we have described the wide range of flow responsiveness of all coastal watersheds in the conterminous U.S.

Watershed classification by Bayesian CART analysis significantly improved variance explained for sestonicchlorophyll a, from 26% for all combined to up to a maximum of 83% for one watershed class.