Th D li f Di l d O i C b f F S il H d W SThe Delivery of Dissolved Organic Carbon from Forest Soils to a Head Water StreamThe Delivery of Dissolved Organic Carbon from Forest Soils to a Head Water Streame e ve y o sso ved O ga c Ca bo o o est So s to a ead Wate St eaYi Mei1 3 George M Hornberger1 3 Lou A Kaplan2 J Denis Newbold2 Anthony Aufdenkampe2Yi Mei1,3, George M. Hornberger1,3, Lou A. Kaplan2, J. Denis Newbold2, Anthony Aufdenkampe2, g g , p , , y p

1 D f Ci il d E i l E i i V d bil U i i N h ill TN 372351.Department of Civil and Environmental Engineering, Vanderbilt University, Nashville, TN, 372351.Department of Civil and Environmental Engineering, Vanderbilt University, Nashville, TN, 372352 Stroud Water Research Center Avondale PA 193112.Stroud Water Research Center, Avondale, PA, 19311

3 Vanderbilt Institute for Energy & Environment Vanderbilt University Nashville TN 372353.Vanderbilt Institute for Energy & Environment, Vanderbilt University, Nashville, TN, 37235gy , y, , ,

INTRODUCTION OBJECTIVE IIOBJECTIVE IINTRODUCTION OBJECTIVE IIOBJECTIVE IOBJECTIVE IFIELD AND LAB METHODS 2D HILLSLOP APPROACHTh t l i ti f th d li f di l d i b (DOC) FIELD AND LAB METHODS 2D HILLSLOP APPROACHThe temporal variation of the delivery of dissolved organic carbon (DOC) FIELD AND LAB METHODS 2D HILLSLOP APPROACHp y g ( )

f hill l t th dj t t i d t i d b h d l i l d S il C E i tfrom hillslopes to the adjacent streams is determined by hydrological and Soil Core Experiments:p j y y gbi h i l th t h t b l t l tifi d I

pA li d B id d d l ti th t f i itbiogeochemical processes that have not been completely quantified. In Applied Bromide-amended solution on the top of in-situg p p y q

ti l i l i diff i th f t d t t f thpp p

il t t ll d dparticular, processes involving differences in the fate and transport of the soil core at a controlled speed.p , p g pil bi d d bl f ti f di l d i b (BDOC) d th

peasily biodegradable fraction of dissolved organic carbon (BDOC) and they g g ( )

l it t f ti f DOC (NDOC) f l i l i t C ll t d t l t th b tt t fi d timore recalcitrant fraction of DOC (NDOC) are of ecological importance. Collected water samples at the bottom at fixed time( ) g p pi t lintervals.

O h it i Whit Cl C k (WCC) t h d WCC t h d iOur research site is White Clay Creek (WCC) watershed, WCC watershed isy ( ) ,725 h 3 d d t h d l t d i th t P l i Th K t t k th t t t th t f th il d 3a 725 ha, 3rd-order watershed located in southeastern Pennsylvania. The Kept track the temperature at the top of the soil core and 3, yj l f thi h i t ti t th ti l DOC fl di t ib ti

p p pi t th ilmajor goal of this research is to estimate the spatial DOC flux distribution cm into the soil core.j g p

ithi th t h dwithin the watershed.L b i t Fig 6 Riparian zone Fig 7 Middle hillslopeLab experiments: Fig 6 Riparian zone Fig 7 Middle hillslope

OBJECTIVESp

Pl fl bi t d t t BDOC d A hill l l d h b k f h hi d d f Whi ClOBJECTIVES Plug-flow bioreactor was used to separate BDOC and non- A hillslope transect located on the east bank of the third order stream of White Clay OBJECTIVES g pBDOC b i l th h it DOC d BDOC

s ope t a sect ocated o t e east ba o t e t d o de st ea o W te C ayC k l d l h i l DOC fl l d fl hBDOC by running samples through it, DOC and BDOC Creek was selected to explore horizontal DOC flux along groundwater flow path. y g p g ,

t ti bt i d b l l ti th diffC ee was se ected to e p o e o o ta OC u a o g g ou dwate ow pat .

Objective I: Estimate vertical annual DOC flux to groundwater (One concentrations were obtained by calculating the differenceObjective I: Estimate vertical annual DOC flux to groundwater. (One y gb t th i fl d tfl l f th bi t Thi i l di l h d l ll l h fldimensional treatment) between the inflow and outflow samples of the bioreactor. This transect is nearly perpendicular to the stream and almost parallel to the flow dimensional treatment) p s t a sect s ea y pe pe d cu a to t e st ea a d a ost pa a e to t e ow

di i f h ll h i d (Fi 6 d Fi 7)direction of shallow phreatic groundwater (Fig 6 and Fig 7). SIMULATIONS RESULTS

d ect o o s a ow p eat c g ou dwate ( g 6 a d g 7). Objective II: Estimate the DOC flux to the stream during precipitation SIMULATIONS RESULTS METHOD: NUMERICAL MODEL Objective II: Estimate the DOC flux to the stream during precipitation SIMULATIONS RESULTS METHOD: NUMERICAL MODELevents along the hillslope (Two dimensional treatment) B B kth h C Si l t d Ob d

METHOD: NUMERICAL MODELevents along the hillslope. (Two dimensional treatment)

Saturated unsaturated flow DOC and BDOC transport0 48Br Breakthrough Curve: Simulated vs Observed

20DOC breakthrough curve: Simulated vs Observed Saturated-unsaturated flow DOC and BDOC transport0.48 20

Simulated

MODEL DESCRIPTION 0 46 18SimulatedData c c c h

MODEL DESCRIPTION 0.46 18 Data

b b bc c cc s D D q S

hk h R C S K k 0.44 16 b b ii ij ic s D D q St x x x x

s r sk h R C St

K k

i i j it x x x x s r sn t 0.42) 14m

) i i j i n t

ppm

(ppm

0.4

n (p 12

on (

0 38atio

n

ratio A 2-D vertical finite element model was used to solve for flow and DOC transport0.38

ntra 10

entr

NDOC

A 2 D vertical finite element model was used to solve for flow and DOC transport 0 36nc

e

8once NDOC along this 120 m hillslope in WCC watershed0.36

con 8

C C

o along this 120 m hillslope in WCC watershed.0 34B

r

6DO

C

0.34 6D SIMULATION RESULTS0.32 4 SIMULATION RESULTS10-3 DOC flux along the hillslope during rain on July 10

0.3 Simulated 2BDOC 4.5

x 10 3 DOC flux along the hillslope during rain on July 10

Data BDOC

50 100 150 200 250 300

50 100 150 200 250 3000 450 100 150 200 250 300

Time (min)50 100 150 200 250 300

Time (min)Time (min) Time (min)3.5

g

Fig 2 Calibration of one-D model from soil core experiments s / gFig 2. Calibration of one D model from soil core experiments 3

tions

2 5sect

2.5

ss s

Simulation results show that the2cr

os Simulation results show that the 2

x at

riparian zone (groundwater depth1 5flu

x

S il C DS il C Driparian zone (groundwater depth

1.5

OC

Soil Core DSoil Core D less or equal to 3m) contributes 1

DSoil Core CSoil Core C less or equal to 3m) contributes Soil Core B

Soil Core CSoil Core B

Soil Core C98% DOC to the groundwater 0.5

Soil Core BSoil Core B 98% DOC to the groundwater. and the rest of the area contributes

0 20 40 60 80 100 120 1400

Soil Core A... ...

Groundwater tableSoil Core A... ...

Groundwater tableand the rest of the area contributes

0 20 40 60 80 100 120 140Distance / mSoil Core A Groundwater tableSoil Core A Groundwater table less than 2% Distance / m

Fi 8 DOC t ti ti l di t ib ti d DOC fl l th hilll lless than 2% .

Fig 8. DOC concentration spatial distribution and DOC flux along the hilllslopeg p f g p The 2D model simulation results show that DOC flux is high near the stream and The model output of total DOC The 2D model simulation results show that DOC flux is high near the stream and The model output of total DOC

declines rapidly in the upland area (Fig 8)recharge to groundwater in 1997 isFig 3: Extend one-D model to different lengths assuming eachdeclines rapidly in the upland area (Fig 8).recharge to groundwater in 1997 is Fig 3: Extend one-D model to different lengths assuming each

about 4700 kg C/yearlength represents a soil core above groundwater table about 4700 kg C/year. length represents a soil core above groundwater table.DISCUSSIONDISCUSSIONDISCUSSIONDISCUSSIONSCUSS ON

A DOC i “fl h d” f i i l h h i i h hSCUSS ON

A DOC i “fl h d” f i i l h h i i h h Baseflow contributed about 67% As DOC is “flushed” from its terrestrial sources to the stream, the riparian zone has a muchAs DOC is “flushed” from its terrestrial sources to the stream, the riparian zone has a muchFi 1 O Di i l d T Di i l M d l St t Baseflow contributed about 67% As DOC is flushed from its terrestrial sources to the stream, the riparian zone has a much

i ifi ff h DOC d BDOC i i h dj hAs DOC is flushed from its terrestrial sources to the stream, the riparian zone has a much

i ifi ff h DOC d BDOC i i h dj hFig 1. One Dimensional and Two Dimensional Model Structure of the total DOC exported from more significant effect on the DOC and BDOC concentrations in the adjacent stream thanmore significant effect on the DOC and BDOC concentrations in the adjacent stream than

g of the total DOC exported from more significant effect on the DOC and BDOC concentrations in the adjacent stream thand f h hill l Th d l l i l h il lmore significant effect on the DOC and BDOC concentrations in the adjacent stream thand f h hill l Th d l l i l h il lWCC in 1997 (Model estimation) do sources from the upper hillslope. The groundwater level is very close to the top soil layerdo sources from the upper hillslope. The groundwater level is very close to the top soil layer1-D dual-permeability unsaturated flow and transport model: WCC in 1997.(Model estimation) do sources from the upper hillslope. The groundwater level is very close to the top soil layeri h i i d h h i idl d i h i h ddo sources from the upper hillslope. The groundwater level is very close to the top soil layeri h i i d h h i idl d i h i h d

1 D dual permeability unsaturated flow and transport model:

ear in the riparian zone, and when the stream stage rises rapidly during the rain, the groundin the riparian zone, and when the stream stage rises rapidly during the rain, the ground The one dimensional model was used to estimate vertical flux through

g/ye

in the riparian zone, and when the stream stage rises rapidly during the rain, the groundl l i h i i i di l d h j DOC i h

in the riparian zone, and when the stream stage rises rapidly during the rain, the groundl l i h i i i di l d h j DOC i h

The one dimensional model was used to estimate vertical flux through

W k

g water level in the riparian zone rises accordingly and reaches major DOC sources in thewater level in the riparian zone rises accordingly and reaches major DOC sources in thesoil column4000G

W

water level in the riparian zone rises accordingly and reaches major DOC sources in theil l

water level in the riparian zone rises accordingly and reaches major DOC sources in theil l

soil column.4000

he G upper soil layers.upper soil layers.

3000o th upper soil layers.upper soil layers.

2000ux to FUTURE RESEARCHFUTURE RESEARCH The model was first calibrated using the experimental data (Fig 2) 2000

C fl

u FUTURE RESEARCHFUTURE RESEARCH The model was first calibrated using the experimental data (Fig 2), 1000O

Cthen 1997 meteorology data was used to run the calibrated model for 1000

BD Collaboration with the PennState CZO group has been initiated The PennState IntegratedCollaboration with the PennState CZO group has been initiated The PennState Integratedthen 1997 meteorology data was used to run the calibrated model for

0C& Collaboration with the PennState CZO group has been initiated. The PennState Integrated Collaboration with the PennState CZO group has been initiated. The PennState Integrated the whole year

(0,1]DO

C

Hydrological Modeling System (PIHM) a physically-based fully distributed hydrologyHydrological Modeling System (PIHM) a physically-based fully distributed hydrologythe whole year.(0,1]

(1,3]DOCua

l D Hydrological Modeling System (PIHM), a physically based fully distributed hydrology Hydrological Modeling System (PIHM), a physically based fully distributed hydrology ( ]

(3,6](6 11]

DOC

nnu model will be extended to simulate the DOC dynamics at a whole catchment scale Themodel will be extended to simulate the DOC dynamics at a whole catchment scale The(6,11]

(11 18] BDOCAn model, will be extended to simulate the DOC dynamics at a whole catchment scale. The model, will be extended to simulate the DOC dynamics at a whole catchment scale. The 2-D saturated and unsaturated flow and transport model: (11,18]

(18,29] Christina Basin will be the test bed for the modelChristina Basin will be the test bed for the model2 D saturated and unsaturated flow and transport model:(18,29]

Different GroundWater Depth range/ mChristina Basin will be the test bed for the model. Christina Basin will be the test bed for the model.

Different GroundWater Depth range/ m ACKNOWLEDMENTSACKNOWLEDMENTS ACKNOWLEDMENTSACKNOWLEDMENTS The two dimensional model was used to calculate DOC flux throughThis work was supported by the National Science Foundation project EAR-0450331This work was supported by the National Science Foundation project EAR-0450331

The two dimensional model was used to calculate DOC flux through Fig 5 Calculation results of NDOC and BDOC flux This work was supported by the National Science Foundation project EAR-0450331 This work was supported by the National Science Foundation project EAR-0450331 the hillslope for a precipitation event Fig 5. Calculation results of NDOC and BDOC flux

“Hydrologic regulation of dissolved organic matter biogeochemistry from forests through river“Hydrologic regulation of dissolved organic matter biogeochemistry from forests through riverthe hillslope for a precipitation event. to groundwater in each specific groundwater depthFi 4 G d t d th di t ib ti Hydrologic regulation of dissolved organic matter biogeochemistry from forests through river

k d l b h i l i d i j h i i iHydrologic regulation of dissolved organic matter biogeochemistry from forests through river

k d l b h i l i d i j h i i ito groundwater in each specific groundwater depth Fig 4. Groundwater depth distribution

networks” and also by the National Science Foundation project EAR-0724971 “Christina Rivernetworks” and also by the National Science Foundation project EAR-0724971 “Christina Riverregiong p

across the watershed networks and also by the National Science Foundation project EAR 0724971 Christina River B i C i i l Z Ob (CRB CZO)”networks and also by the National Science Foundation project EAR 0724971 Christina River B i C i i l Z Ob (CRB CZO)”*One Dimensional Model Equations are Available on Handout

regionacross the watershedBasin Critical Zone Observatory (CRB-CZO)”.Basin Critical Zone Observatory (CRB-CZO)”.*One Dimensional Model Equations are Available on Handout (Each region uses a representative soil core length) y ( )y ( )q (Each region uses a representative soil core length)