WEPP—A Process-Based Hydrology and Erosion Model for Watershed Assessment and Restoration

WEPP—A Process-Based Hydrology WEPP—A Process-Based Hydrology and Erosion Model for Watershed and Erosion Model for Watershed

Assessment and RestorationAssessment and Restoration

Joan Q. Wu, Markus Flury, Shuhui Dun, R. Cory GreerJoan Q. Wu, Markus Flury, Shuhui Dun, R. Cory GreerWashington State University, Pullman, WAWashington State University, Pullman, WA

Donald K. McCoolDonald K. McCoolUSDA ARS PWA, Pullman, WAUSDA ARS PWA, Pullman, WA

William J. ElliotWilliam J. ElliotUSDA FS RMRS, Moscow, IDUSDA FS RMRS, Moscow, ID

Dennis C. FlanaganDennis C. FlanaganUSDA ARS NSERL, West Lafayette, INUSDA ARS NSERL, West Lafayette, IN

Major Funding AgenciesMajor Funding Agencies

USDA National Research Initiatives (NRI) Programs

US Forest Service Rocky Mountain Research Station (RMRS)

US Geological Survey/State of Washington Water Research Center

In-house funding from various collaborating research In-house funding from various collaborating research institutesinstitutes

The NeedsThe Needs

Protecting and improving water quality in agricultural watersheds are major goals of the USDA NWQ and NRI Programs

For many watersheds, sediment is the greatest pollutant

In watershed assessment, it is crucial to understand sedimentation processes and their impacts on water quality

To successfully implement erosion control practices, it is necessary to determine the spatiotemporal distribution of sediment sources and potential long-term effectiveness of sediment reduction by these practices

Surface runoff and erosion from undisturbed forests are negligible

Stream formed due to subsurface flow has low sediment

Both surface runoff and erosion can increase dramatically due to disturbances

Models are needed as a tool for forest resource management

The WEPP ModelThe WEPP Model

WEPP: Water Erosion Prediction Project

a process-based erosion prediction model developed by the USDA ARS

built on fundamentals of hydrology, plant science, hydraulics, and erosion mechanics

WEPP’s unique advantage: it models watershed-scale spatial and temporal distributions of soil detachment and deposition on event or continuous basis

Equipped with a geospatial processing interface, WEPP has great potential as a reliable and efficient tool for watershed assessment

The WEPP Model cont’d

WEPP Windows Interface

WEPP Internet Interface

GeoWEPP

Long-term Research EffortsLong-term Research Efforts

Goal Continuously refine and apply the WEPP model for watershed

assessment and restoration under different land-use, climatic and hydrologic conditions

Objectives Improve the subsurface hydrology routines so that WEPP can be

used under both infiltration-excess and saturation-excess runoff conditions in crop-, range- and forestlands

Improve the winter hydrology and erosion routines through combined experimentation and modeling so that WEPP can be used for quantifying water erosion in the US PNW and other areas where winter hydrology is important

Continually test the suitability of WEPP using data available from different localities across the world

ProgressesProgresses

Numerous modifications to WEPP have been made to Correct the hydraulic structure routines

Improve the water balance algorithms

Incorporate the Penman-Monteith ET method (FAO standard)

Improve the subsurface runoff routines

Expand and improve winter hydrology routines to better simulate

Freeze-thaw processes

Snow redistribution processes

WEPP newest release accessible at NSERL’s website http://topsoil.nserl.purdue.edu/nserlweb/index.html

Comparison of ProcessesComparison of Processes

* Previous version of WEPP typically overestimated Dp

RedistributionRedistributionof Infiltration Water in WEPPof Infiltration Water in WEPP

E v a p o ra tio na n d

T ra n sp ira tio n

D e e pP e rc o la tio n

S u b su rfa c eL a te ra l

F lo w

In filtra ted W a ter

22 3311

Code ModificationCode Modification

Provide options for different applicationsProvide options for different applications a flag added to the soil input filea flag added to the soil input file

User-specified vertical hydraulic conductivity User-specified vertical hydraulic conductivity KK for the added restrictive layerfor the added restrictive layer

e.g., 0.005 mm/hre.g., 0.005 mm/hr basalt (basalt (Domenico and SchwartzDomenico and Schwartz, 1998), 1998)

User-specified anisotropy ratio for soil saturated User-specified anisotropy ratio for soil saturated hydraulic conductivityhydraulic conductivity

horizontal horizontal KKh h vertical vertical KKvv, e.g., K, e.g., Khh/K/Kvv = 25= 25

Code ModificationCode Modification cont’dcont’d

Subroutines modified to properly Subroutines modified to properly write the “pass” fileswrite the “pass” files

WEPP’s approach to passing outputsWEPP’s approach to passing outputs subsurface flow not “passed” previouslysubsurface flow not “passed” previously

Simplified hillslope-channel relationSimplified hillslope-channel relation all subsurface runoff from hillslopes assumed to all subsurface runoff from hillslopes assumed to

enter the channelenter the channel flow added and sediment neglectedflow added and sediment neglected

A Case Application:Modeling Forest Runoff and Erosion

Study Site: Hermada WatershedStudy Site: Hermada Watershed

Physical SettingPhysical Setting Located in the Boise National Forest, SE Lowman, ID

Instrumented during 1995−2000 to collect whether, runoff, and erosion data

5-yr observed data showing an average annual precipitation of 860 mm, among which nearly 20% was runoff

Temperature

-60

-40

-20

0

20

40

60

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200

days

Tem

pera

ture

, c

Tmax

Tmin

Tdew

Precipitation

0

10

20

30

40

50

60

70

1 55 109

163

217

271

325

379

433

487

541

595

649

703

757

811

865

919

973

1027

1081

1135

1189

1243

1297

1351

1405

1459

1513

1567

1621

1675

1729

1783

1837

1891

1945

1999

2053

2107

2161

days

Pre

cipi

tatio

n,m

m

Watershed DiscretizationWatershed Discretization

Model InputsModel Inputs

TopographyTopography Derived from 30-m DEMs using GeoWEPPDerived from 30-m DEMs using GeoWEPP 10-ha in area, 3 hillslopes and 1 channel10-ha in area, 3 hillslopes and 1 channel 40−60% slope40−60% slope

SoilSoil Typic Cryumbrept loamy sand 500 mm in depthTypic Cryumbrept loamy sand 500 mm in depth

underlying weathered graniteunderlying weathered granite

ManagementManagement 1992 cable-yarding harvest1992 cable-yarding harvest 1995 prescribed fire on W and N slopes1995 prescribed fire on W and N slopes

ClimateClimate 11/1995−09/2000 observed data11/1995−09/2000 observed data

Results

Living Biomass and Ground Cover

(a)

1995 1996 1997 1998 1999 2000

Liv

ing

Bio

mas

s k

g m

-2

0

1

2

3

4

5

(b)

Year

1995 1996 1997 1998 1999 2000

Liv

ing

Bio

mas

s k

g m

-2

0

1

2

3

4

5

(a)

1995 1996 1997 1998 1999 2000

Gro

un

d C

ove

r

0.85

0.90

0.95

1.00

(b)

Year

1995 1996 1997 1998 1999 2000

Gro

un

d C

ove

r

0.85

0.90

0.95

1.00

PredictedObserved

* (a) unburned, (b) burned

Runoff and Erosion: Obs vs PreRunoff and Erosion: Obs vs Pre

* Observation Period: 11/3/1995−9/30/2000

Water Year

Precipitation (mm)

Observed Watershed Discharge

(mm)

Observed Sediment

(t/ha)

Simulated Hillslope Runoff (mm)

Simulated Hillslope

Lateral Flow (mm)

Simulated Watershed Discharge

(mm)

Simulated Watershed Sediment

(t/ha)1995–1996 966 87 0 0 436 433 0.11996–1997 884 89 0 0 253 250 01997–1998 1085 322 0 28 288 313 01998–1999 690 172 0 0 164 162 01999–2000 689 142 0 0 150 147 0

Average 863 162 0 7 258 261 0.0

(b)

Month

M J S D M J S D M J S D M J S D M J S D M J S

Ob

serv

ed R

un

off

, mm

0

2

4

6

8

10

12

(a)

M J S D M J S D M J S D M J S D M J S D M J S

Pre

dic

ted

Ru

no

ff, m

m

0

2

4

6

8

10

12 Cu

mu

lative P &

RM

, mm0

200

400

600

800

1000

1200

Runoff Cum. P Cum. RM

(b)

Month

O N D J F M A M J J A S

Ob

serv

ed

Ru

no

ff, m

m

0

2

4

6

8

10

12

Prec

ipita

tion

, mm

0

20

40

60

80

100

(a)

O N D J F M A M J J A S

Pre

dic

ted

Ru

no

ff, m

m

0

2

4

6

8

10

12 Cu

mu

lative P &

RM

, mm0

200

400

600

800

1000

1200

Runoff Cum. P Cum. RM

Water year 1997–1998

SummarySummary

Modifications were made in the approach to, and algorithms for modeling deep percolation of soil water and subsurface lateral flow

The refined model has the ability to more properly partition infiltration water between deep percolation and subsurface lateral flow

For the Hermada forest watershed Vegetation growth and ground cover were described realistically

WEPP-simulated watershed discharge for 1998–2000 was compatible with field observation; however, the agreement was poor for first two years

Overall, predicted annual watershed discharge and sediment yield were not significantly different from the observed (paired t-test)

Nash-Sutcliffe model efficiency coefficient for daily runoff was –1.7, suggesting improvement needed

Ongoing Efforts

Palouse Conservation Field Station Palouse Conservation Field Station (PCFS), Pullman, WA, USA(PCFS), Pullman, WA, USA

Laboratory and field experimentation on runoff and erosion as affected by freezing and thawing of soils

Tilting flume at PCFS

Experimental plots at PCFS

Collaborating Research UnitsCollaborating Research Units Testing WEPP using data collected at

USDA ARS CPCRC, Pendleton, OR, USA (Dr. John Williams) Ag Exp Farm, University of Bologna, Italy (Dr. Paola Rossi Pisa)

Other PNW Watersheds Testing and applying WEPP for evaluating DEM effect on

soil erosion prediction Paradise Creek Watershed, ID, USA (Dr. Jan Boll) Mica Creek Watershed, ID, USA (Dr. Tim Link)

Thank You!

Questions?

Important Parameters

Parameters Values

Surface Soil Effective K, mm/hr

16.6

Bedrock K, mm/hr 1.0E−2

Anisotropy Ratio 25