Development of Sediment and Nutrient Export Coefficients for US Ecoregions.

Transcription

1 Development of Sediment and Nutrient Export Coefficients for US Ecoregions. Mike White USDA/ARS Temple Texas 1

2 What are Export Coefficients? History Predates models Date to 1970 s Eutrophication linked to landuse Initiated monitoring studies for nutrients 2

3 What are Export Coefficients? 0.5 KG/HA 1.5 KG/HA 2.5 KG/HA 0.1 KG/HA 4.0 KG/HA Estimated mass loss per unit area per year By Landuse Derived from intensive edge of field monitoring Widely used.still 3

4 Example Use Watershed Load Easy to apply Area * EC = Load landuse Area (ha) Export Coeff (kg/ha/yr) Load (kg/yr) Forest Urban Corn Grassland Total Watershed

5 Problems Edge of field monitoring Expensive Relatively rare Vary regionally Climate Topography - Soils Too little measured data for each region Extreme extrapolation High uncertainty Solution Use SWAT to extrapolate limited monitoring data into a much larger dataset 5

6 SWAT Based Extrapolation SWAT Model Template Single Field Edge of Field Observations Validated SWAT Model Template New Location Calibration Validation New Edge of Field Prediction (Sample) 6

7 Export Coefficient Database Library of model predictions National basis Many many samples Consider local conditions Landuse Soils Climate Topography Management Conservation US Ecoregions Summarize by major landuses for every ecoregion in US 7

8 Existing National Data Landuse - NLCD CDL Soils - STATSGO Topography NED Seamless Climate Data 20,000 stations ( ) Irrigation & Fertilization Ag Census Conservation Practices CEAP Survey 18,000 Management US RUSLE2-20,000 templates US National Data RUSLE2 Templates 8

9 Sampling - Overview National GIS Data Basis For New SWAT Simulation 9

10 Sampling - Details Select Random Location Pick Landuse Soils Topography Defined HRU Type Location Landuse Soils Topography Ridge till 25% Conservation 15% Tillage No-till 10% Conventional 30% Strip till 20% Manure No 80% Yes 20% Management Assignment No 60% Tiles Yes 40% Conservation No 86% Yes 14% Fully Defined Management 10

11 Sampling and Simulation Select a Nearby Climate Gage 5 year Simulation Random Start Build a SWAT model Single Field Execute Model Store Output In National Database 11

12 One Down, Millions to go More samples is better Provide EC distribution Computationally Intense Windows Cluster 250 cores 5 days Current National Database 45 million simulations/samples 1 for each 22 ha in the US 12

13 How to Use it Simulation Library Extraction Tool Database Query Corn - Texas Blackland Prairie Ecoregion Runoff (mm yr-1) Sediment (Mg ha-1 yr-1) QUERY Corn Blackland Prairie Ecoregion Percent Total Nitrogen (kg ha-1 yr-1) Total Phosphorus (kg ha-1 yr-1) 13,036 Samples

14 Ecoregion Summaries Cultivated Cropland Ecoregion n Total Nitrogen Total Phosphorus (kg ha -1 yr -1 ) (kg ha -1 yr -1 ) Arizona/New Mexico ( ) ( ) Mountains Arizona/New Mexico ( ) 0 ( ) Plateau Arkansas Valley ( ) 2.15 ( ) Atlantic Coastal Pine ( ) ( ) Barrens Blue Mountains ( ) (0-2.75) Blue Ridge ( ) 2.89 ( ) Boston Mountains ( ) 2.17 ( ) Canadian Rockies ( ) ( ) Cascades ( ) 3.11 ( ) Central Appalachians ( ) 6.48 ( ) Central Basin and Range ( ) (0-1.39) Central California Valley ( ) (0-1.22) Central Corn Belt Plains ( ) 2.45 ( ) Central Great Plains ( ) ( ) Central Irregular Plains ( ) 2.69 ( ) Chihuahuan Deserts ( ) ( ) Coast Range ( ) 4.84 ( ) Colorado Plateaus ( ) ( ) Median value with range (10 th and 90 th percentiles). 14

15 Measured edge of field loads Grassland 95 observations Cropland 91 observations Looking for distributional overlap Validation Compare to Monitoring Data 15

16 Applications 16

17 Large Scale Compare with CEAP Proper distributional sampling Extract at differing spatial scales Watershed HUC 8 County Compare to CEAP Similar trends CEAP Export Coefficient 17

18 National Water Balance Areas of groundwater depletion Full Water Balance 18

19 Small Watershed nutrient Forecasting Tool Simple web based tool Predict nutrient and sediment loads Couples EC and delivery ratio concepts 19

20 20

21 Questions? 21