SWAT Condition Data
SWAT is a watershed-scale model that applies empirical and physically based approaches to determine runoff responses to differences in land management. The model is primarily intended for application to agricultural watersheds and can simulate a wide variety of agricultural management practices and crop rotations under varying climate scenarios (Arnold et al. 1998 ; Arnold and Fohrer 2005 ; Gassman et al. 2010 ). The key inputs are soils, land cover, and topography (slope), which were intersected to generate Hydrologic Response Units (HRUs): the basic computational unit of the model. Land cover data are based on the 2008 Cropland Data Layer (National Agricultural Statistics Service, U.S. Department of Agriculture 2012 ; Han et al. 2014 ), which includes data for major and minor crop types as well as nonagricultural land classes. Pixel resolution for the 2008 Cropland Data Layer is 30 m. We reclassified some data classes (<0.25% of the original Cropland Data Layer) into available similar classes to simplify the spatial data and reduce the number of model HRUs. For example, sweet corn (0.17%) was reclassified as corn. Topography data are based on a digital elevation model (pixel size = 30 m) from the National Elevation Dataset (U.S. Geological Survey 2016 ), whereas soils data are from the county-level Soil Survey Geographic (SSURGO) database downloaded from the Web Soil Survey (Soil Survey Staff et al. 2019 ). SSURGO data for the study watershed had a minimum map unit size of 684 m 2 (excluding edge polygons). The model extent was the 12-digit watershed for headwater streams Rock and Pratt Creek, and the entire drainage area for Wolf Creek (10-digit HUC 0708020508) and Miller Creek (0708020509) (Figure 1). The final number of model HRUs ranged from 1,684 (Rock Creek) to 7,528 (Wolf Creek).
SWAT requires each day environment inputs out of rain, temperature, relative dampness, wind speed, and you will solar radiation. Weather enter in analysis was in fact based on next age group environment radar (NEXRAD) research for rain and you will Environment Prediction Program Reanalysis (CFSR) (Fuka mais aussi al. 2014 ) to possess remaining climate study. This new spatial environment research (NEXRAD and you will CSFR) is represented regarding the design while the a plastic circle install on the an effective grid (NEXRAD: 4km spacing; CSFR: around 30 kilometres spacing). SWAT instantly picks evaluate points that is actually near the centroid away from design subbasins. To own NEXRAD analysis, what number of synthetic gauges found during the watershed line ranged regarding 6 (Rock Creek) so you can 49 (Wolf Creek). To own CSFR studies, the amount of gauges ranged away from two to three. SWAT works from the a daily timestep and you may trick design outputs was streamflow including sediment and you may nutrient export. For this investigation, we utilized just the daily streamflow outputs to target the fresh new hydrologic effects of BMPs into flood wreck.
Discharge–Regularity Studies (Module step 1)
We began with a Baseline scenario to simulate current conventional agricultural practices, in which corn and soybean crops are grown in a two-year rotation typical for the Upper Midwest. Tillage, fertilizer application, and planting/harvest dates are based on farmer surveys (Minnesota Department of Agriculture 2007 ) and feedback from local stakeholders and commodity groups. We calibrated and validated SWAT against measured flow data from the Wolf Creek watershed. To do so, we ran the model for 12 years from 2002, , December 31. The first two years of model results were treated as a warm-up period and we discarded the results, leaving 10 years of model results to compare against jackd indir observed flow data. We calibrated SWAT for the five-year period from 2009, , December 31. We assessed agreement between observed and modeled flow using mean daily flow and Nash–Sutcliffe efficiency (NSE) and percent bias (PBIAS) (cf. Moriasi et al. 2015 ; Ahmadisharaf et al. 2019 ). For NSE, a value of 1 indicates perfect agreement between observed and predicted values, whereas values >0.5 are generally considered satisfactory for monthly flow. For PBIAS, values <15% are generally considered satisfactory (Moriasi et al. 2015 ; Ahmadisharaf et al. 2019 ), with 0 being the ideal PBIAS.