Evaluation of the impacts of vegetative buffers and conservation grasslands on total phosphorus loads using hydrological modeling with remote sensing techniques

Over the past three decades, the Conservation Reserve Program (CRP) has led to the conversion of millions of environmentally sensitive farmlands to grasslands, as well as the implementation of vegetative buffers (VBS) to prevent agrochemicals from entering water systems. The CRP has demonstrated through site-specific studies that it is effective in reducing contaminants in surface waters. However, the ability to successfully assess the impacts of the program on water quality at the watershed level requires techniques that can detect VBS and conservation grasslands. The areas and locations of these practices can then be used in hydrologic models to understand their functionality in the environment. This study assessed the impacts of conservation buffers and grasslands on discharge and Total Phosphorus (TP) loads using the Soil and Water Assessment Tool (SWAT) model with land cover data developed from an Object- Based Image Analysis (OBIA) approach along with different precipitation datasets from the National Climatic Data Center (NCDC), the National Soil Erosion Research Laboratory (NSERL), and the Next-Generation Radar (NEXRAD) in the St. Joseph River Watershed (SJRW) and the Cedar Creek Watershed (CCW) located in northeast Indiana. The OBIA land cover data showed greater than 50% of the stream area consisted of VBS of 30.5 m and 61.0 m widths. Comparison of the conservation grasslands identified by the OBIA with those reported in the Common Land Unit (CLU) inventory for the year 2005 showed the area depicted by the OBIA was less than reported by the CLU. A comparison of the radar-derived precipitation and rain gauge precipitation showed that increased rain gauge density reduced the overall bias by 3.3% for data collected between 2005 and 2008. To gain a greater understanding of how different precipitation datasets affect the hydrology of the watershed, simulated streamflow values from NSERL+NCDC, NEXRAD precipitation datasets were compared with that generated using the NCDC data alone. SWAT simulated discharge with NEXRAD-derived precipitation data under-estimated the observed streamflow, while discharge simulated from the rain gauge over-estimated the observed streamflow. Calibration of the model with each precipitation dataset increased model efficiency and decreased the bias in the simulated streamflow. Although the results showed that simulated discharge with the NCDC+NSERL showed less bias, either precipitation dataset can be a valuable input for hydrological processes simulation. The NCDC+NSERL precipitation dataset was used in SWAT combined with the OBIA land cover to assess the impact of conservation buffers and grasslands on TP loads. All land cover categories were considered to define the Hydrologic Response Units such that small vegetation patches would be represented in the model.The model was run with and without conservation grasslands, VBS of 30.5 m and 61.0 m. It was found that greater than 7.0% of the TP load reduction was achieved with all the practices combined, while VBS of 61.0 m alone decreased the TP load by more than 3.0%. Furthermore, the combination of these practices with the SWAT edge-of-field buffer module achieved more that 30% reduction of TP loads. This research provides a better means of monitoring existing programs and developing more effective management strategies that can considerably reduce nutrient inputs into water systems. Because the OBIA rules can be applied to different image acquisition dates, it should help policy makers and natural resource managers to rapidly quantify areas under the vegetative buffers and conservation grasslands or other similar conservation practices. This information can then be used to assess their impact on the environment