Hydrologic Modeling with the Distributed-Hydrology- Soils- Vegetation Model (DHSVM)

Hydrologic Modeling with the Distributed-Hydrology- Soils- Vegetation Model (DHSVM) DHSVM was developed by researchers at the University of Washington and the Pacific Northwest National Lab 200 Simulated versus Measured Discharge 180 160 Recorded Flow Simulated Flow (lateral conductivity 0.01m/s) 140 GIS grid-based representation (30 m x 30 m) Discharge (cfs) 120 100 80 60 40 20 0 10/01/2001-00 11/30/2001-00 01/29/2002-00 03/30/2002-00 05/29/2002-00 07/28/2002-00 09/26/2002-00 Date

Each 30 x 30 meter grid contains information about the slope, aspect and elevation over-story vegetation under-story vegetation multi-layer soil properties stream network UTM location saturated subsurface flow

Thesis results of Katie Kelleher, Geology (2006) 200 180 160 140 Smith Creek Hydrograph (WY 2002) Recorded Flow Simulated Flow (lateral conductivity 0.01m/s) Silver Creek Carpenter Creek Discharge (cfs) 120 100 80 Olsen Creek 60 Mill Creek Euclid Creek Austin Creek Smith Creek 40 20 0 10/01/2001-00 11/30/2001-00 01/29/2002-00 03/30/2002-00 05/29/2002-00 07/28/2002-00 09/26/2002-00 Date 650 Austin Creek Hydrograph (WY 2002) Anderson Creek 600 550 Recorded Simulated (lateral conductivity.02m/s) Brannian Creek 500 450 0 2.5 5 Kilometers Discharge (cfs) 400 350 300 250 200 150 100 50 0 10/01/2001-00 11/30/2001-00 01/29/2002-00 03/30/2002-00 05/29/2002-00 07/28/2002-00 09/26/2002-00 Date

DHSVM Modeling Inputs and Outputs Digital Topographic Data Digital Land Surface Data Time Series of Meteorological Data GIS Preprocessing Precipitation Humidity Temperature Wind Speed Shortwave Radiation Longwave Radiation DEM Channel Network Vegetation Layer Soil Type Soil Thickness Meteorological Data Initial Conditions DHSVM Stream Flow Snow & Snow Melt Soil Moisture Evapotranspiration Precipitation

DHSVM calculates a water and energy budget on each grid cell for each time step inputs - outputs = change in storage

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W P = Precipitation

North Shore Meteorological (MET) Station Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Brannian Creek Rain Gauge Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Geneva Rain Gauge Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Rainfall (or snow) is distributed over the DEM Point Measurements inverse distance elevation lapse rate Distributed Rainfall (162,669 grid cells) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Rain Gauge Brannian Creek Anderson Creek Elevation (meters) High : 1024 0 2.5 5 Kilometers Low : 93 0 2 4 8 Kilometers

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W E io = overstory evaporation E to = overstory transpiration

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W E io = overstory evaporation E to = overstory transpiration E iu = understory evaporation E tu = understory transpiration

Incoming shortwave radiation Air Temperature Humidity Wind speed Incoming longwave radiation

North Shore MET Station Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Vegetation and Landcover Grid NOAA Landcover Deciduous Mixed Forest Closed Shrub Grassland Bare Urban Water Coastal Conifer 0 2 4 8 Kilometers

1 Evergreen Needleleaf 2 Evergreen Broadleaf 3 Deciduous Needleleaf 4 Deciduous Broadleaf 5 Mixed Forest 6 Woodland 7 Wooded Grassland 8 Closed Shrub 9 Open Shrub 10 Grassland 11 Cropland 12 Bare 13 Urban 14 Water 15 Coastal Conifer Forest 16 Xeric Conif Forest (Dry) 17 Mesic Conif Forest (Wet) 18 Subalpine Conif Forest 19 Alpine Meadow 20 Ice 21 Wetland

Vegetation Description 15 = Coastal Conifer Forest Impervious Fraction 15 = 0.0 Overstory Present 15 = TRUE Understory Present 15 = TRUE Fractional Coverage 15 = 0.9 Hemi Fract Coverage 15 = 0.9 Trunk Space 15 =.5 Aerodynamic Attenuation 15 = 2.0 Radiation Attenuation 15 = 0.15 Max Snow Int Capacity 15 = 0.040 Snow Interception Eff 15 = 0.6 Mass Release Drip Ratio 15 = 0.4 Height 15 = 50.0 0.5 Overstory Monthly LAI 15 = 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 12.0 Understory Monthly LAI 15 = 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 Maximum Resistance 15 = 5000.0 3000.0 Minimum Resistance 15 = 666.6 200.0 Moisture Threshold 15 = 0.33 0.13 Vapor Pressure Deficit 15 = 4000 4000 Rpc 15 =.108.108 Overstory Monthly Alb 15 = 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 Understory Monthly Alb 15 = 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 Number of Root Zones 15 = 3 Root Zone Depths 15 = 0.10 0.25 0.40 Overstory Root Fraction 15 = 0.20 0.40 0.40 Understory Root Fraction 15 = 0.40 0.60 0.00

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W S io = change in overstory interception storage S iu = change in understory interception storage

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W S io = change in overstory interception storage S iu = change in understory interception storage W = change in snow-water equivalent

Inputs - Outputs = Change in Storage P E io E iu E s E to E tu Q s = S s1 + S s2 + S io + S iu + W Q s = discharge leaving the lower soil S s1 = change in upper soil storage S s2 = change in lower soil storage

Soil Coverage (USDA) USDA Soils Silty Loam Loam Water Muck 0 2 4 8 Kilometers

Soil Depth (estimated) Soil Depth (meters) High : 2.91583 Low : 0.933977 0 2 4 8 Kilometers

USDA Soil Classifications 1 = SAND 2 = LOAMY SAND 3 = SANDY LOAM 4 = SILTY LOAM 5 = SILT 6 = LOAM 7 = SANDY CLAY LOAM 8 = SILTY CLAY LOAM 9 = CLAY LOAM 10 = SANDY CLAY 11 = SILTY CLAY 12 = CLAY 13 = ORGANIC (as loam) 14 = WATER (as clay) 15 = BEDROCK 16 = OTHER (as SCL) 17 = MUCK 18 = TALUS

SOIL 6 Soil Description 6 = LOAM Lateral Conductivity 6 = 0.01 Exponential Decrease 6 = 3.0 Maximum Infiltration 6 = 1e-5 Surface Albedo 6 = 0.1 Number of Soil Layers 6 = 3 Porosity 6 =.43.43.43 Pore Size Distribution 6 =.19.19.19 Bubbling Pressure 6 =.11.11.11 Field Capacity 6 =.29.29.29 Wilting Point 6 =.14.14.14 Bulk Density 6 = 1485. 1485. 1485. Vertical Conductivity 6 = 0.01 0.01 0.01 Thermal Conductivity 6 = 7.114 6.923 7.0 Thermal Capacity 6 = 1.4e6 1.4e6 1.4e6

Surface and Subsurface Flow

Stream Network Segment length Channel width Channel depth Stream Network 0 2 4 8 Kilometers

DHSVM Modeling Inputs and Outputs Digital Topographic Data Digital Land Surface Data Time Series of Meteorological Data GIS Preprocessing Precipitation Humidity Temperature Wind Speed Shortwave Radiation Longwave Radiation DEM Channel Network Vegetation Layer Soil Type Soil Thickness Meteorological Data Initial Conditions DHSVM Sreamflow Snow & Snow Melt Soil Moisture Evapotranspiration Precipitation

DHSVM Modeling Inputs and Outputs Digital Topographic Data Digital Land Surface Data Time Series of Meteorological Data GIS Preprocessing Precipitation Humidity Temperature Wind Speed Shortwave Radiation Longwave Radiation DEM Channel Network Vegetation Layer Soil Type Soil Thickness Meteorological Data Initial Conditions DHSVM Streamflow Snow & Snow Melt Soil Moisture Evapotranspiration Precipitation

Modeling Methods Collect and format MET and spatial data Calibrate the model to recorded streamflow Validate the model

Modeling Assumptions Recorded streamflow adequately captures actual streamflow. Recorded MET data provide an adequate representation of MET conditions in the basin. Model parameters, other than calibration parameters, provide an acceptable physical representation of the basin. Simulations were performed on a Dell Precision having two, 3-GHz processors, 2 GB of RAM and a Linux platform.

Model Calibration Alter model parameters until there is a sufficient match between the simulated streamflow and recorded streamflow. The most influential parameters are the: precipitation lapse rate soil thicknesses lateral hydraulic conductivity

Runoff Estimate: Lake Whatcom Water Budget Method inputs - outputs = change in storage outputs change in storage inputs

Rating curve generated from the lake level -capacity data developed by Ferrari and Nuanes (2001) Lake Volume 258000 256000 Millions of Gallons 254000 252000 250000 248000 246000 310 311 312 313 314 315 Feet

Water Budget Runoff Estimate inputs outputs = change in storage Water Budget for water year (WY) 2003-2004 Inputs Volume (MG) % of total Direct Precipitation 7612 18.6 Diversion 5095 12.4 Runoff 28288 69.0 Outputs Whatcom Creek 26948 71.2 Hatchery 1278 3.4 Georgia Pacific 2053 5.4 City of Bellingham 4449 11.8 Water District 10 204 0.5 Evaporation 2924 7.7 Change in Storage +3139

Water Budget Runoff Estimate Runoff = change in storage + outputs direct precipitation diversion flow Water Budget for water year (WY) 2003-2004 Inputs Volume (MG) % of total Direct Precipitation 7612 18.6 Diversion 5095 12.4 Runoff 28288 69.0 Outputs Whatcom Creek 26948 71.2 Hatchery 1278 3.4 Georgia Pacific 2053 5.4 City of Bellingham 4449 11.8 Water District 10 204 0.5 Evaporation 2924 7.7 Change in Storage +3139

9/6/04 8/6/04 7/6/04 Runoff: Water Budget Approach 6/5/04 5/5/04 4/4/04 3/4/04 2/2/04 1/2/04 12/2/03 11/1/03 10/1/03 1800 1300 800 300-200 Millions of Gallons

Simulated Whatcom Creek Outflow Q Stream Network 0 2 4 8 Kilometers

9/6/04 8/6/04 7/6/04 6/5/04 Runoff: Simulated 5/5/04 4/4/04 3/4/04 2/2/04 1/2/04 12/2/03 11/1/03 10/1/03 2000 1500 1000 500 0 Millions of Gallons

Runoff: Water Budget versus Simulated Water Budget Simulated 9/6/04 8/6/04 7/6/04 6/5/04 5/5/04 4/4/04 3/4/04 2/2/04 1/2/04 12/2/03 11/1/03 10/1/03 1800 1300 800 300-200 Millions of Gallons

Model Validation If the model is calibrated, it should adequately simulate streamflow for a different water year (e.g., different MET and recorded streamflow).

Runoff into Lake Whatcom (WY2001 to 2005) 1800 DHSVM (modeled) Water Budget Estimate Millions of Gallons 1300 800 300 10/1/01 2/1/02 6/1/02 10/1/02 2/1/03 6/1/03 10/1/03 2/1/04 6/1/04 10/1/04 2/1/05 6/1/05-200

What next? Constant model refinement. Determine the groundwater input to the lake. Focus on specific basins and include the effect of roads and impervious services. Examine sediment inputs to streams. Model the effect of glacial recession on diversion magnitudes. 0 3 6 Kilometers Simulate drought influences.

Austin Creek Stream Gauge (IWS) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Austin Creek Sreamflow: WY 2004 800 700 600 Simulated Recorded (IWS) Discharge (CFS) 500 400 300 200 100 0 Oct Mar Sept

16 Rainfall (inches) 14 12 10 8 6 4 WY2004 (63.5 inches total) Drought (29.8 inches total) 2 0 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

1600 1400 1200 1000 800 600 400 200 0 Drought Year Millions of Gallons 10/1/03 11/1/03 12/1/03 1/1/04 2/1/04 3/1/04 4/1/04 5/1/04 6/1/04 7/1/04 8/1/04 9/1/04 WY2004 Drought

Deming glacier is shrinking! Mt. Baker 10,772 ft Nooksack River Middle Fork Basin (260 sq-km) Deming Glacier Diversion Photo by Joe Wood 0 5 10 20 Kilometers

Geology graduate student Carrie Donnell is using DHSVM to predict how Deming glacier recession changes Middle Fork streamflow. Photo by Joe Wood

Middle Fork Nooksack calibration results: 2005 water year 14000 Calibration period (WY 2005) 12000 Measured Simulated 10000 Discharge (cfs) 8000 6000 4000 2000 0 09/30/2004-23:00:00 11/29/2004-23:00:00 01/28/2005-23:00:00 03/29/2005-23:00:00 05/28/2005-23:00:00 07/27/2005-23:00:00 09/25/2005-23:00:00 Date

Questions? Lake Whatcom looking west toward the Puget Sound Photo by Margaret Landis

Brannian Creek Stream Gauge (USGS) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Anderson Creek Stream Gauge (IWS) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Smith Creek Stream Gauge (IWS) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Olsen Creek Stream Gauge (USGS) Silver Creek Carpenter Creek Olsen Creek Euclid Creek Smith Creek Mill Creek Austin Creek Anderson Creek Brannian Creek 0 2.5 5 Kilometers

Major causes for water degradation in Washington Lake Whatcom listed parameters 300 Water Bodies 250 200 150 100 Bacteria Temperature Toxics Dissolved Oxygen ph Flow Nutrients 50 0 1 Causes http://www.ecy.wa.gov/programs/wq/tmdl/index.html

Development of the Lake Whatcom TMDL The objectives of Ecology and researchers at Portland State are to: 1) Develop and calibrate a hydrodynamic and water quality model of Lake Whatcom using CE-QUAL-W2 2) Use the model to evaluate strategies for water quality improvement http://www.ce.pdx.edu/w2/projects_lake_whatcom.html

Development of the Lake Whatcom TMDL The objectives of Ecology and researchers at Portland State are to: 1) Develop and calibrate a hydrodynamic and water quality model of Lake Whatcom using CE-QUAL-W2 2) Use the model to evaluate strategies for water quality improvement runoff magnitudes are required inputs to the model http://www.ce.pdx.edu/w2/projects_lake_whatcom.html

Middle Fork Diversion Canada USA Nooksack River Bellingham Middle Fork (Nooksack) Diversion Pipeline Anderson Creek Basin Western Whatcom County

Diversion Off Diversion On (max of 65 cfs)

Diversion Flow into Mirror Lake Mirror Lake 0 1.5 3 6 Kilometers Anderson Creek Basin

Recorded Streamflow Anderson Creek Stream Gauge Anderson Creek Rating Curve 10 9 0 1.5 3 6 Kilometers Anderson Creek Basin Sq.Rt. Discharge 8 7 6 5 4 3 2 y = 5.4587x - 0.5524 R 2 = 0.9924 1 0 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Gage Height (feet)

Motivation 1) The lake serves as Bellingham s drinking water supply. 2) Runoff is the largest input to the lake. Runoff = streamflow and groundwater discharge to the lake Lake Whatcom 20 to 30% of the total runoff is groundwater (Pitz, 2005)