The ability of 5 Best Management Practice (BMP) allocation methods that consider 8 pre-selected BMPs, to control 4 Nonpoint Source (NPS) constituents in 4 watersheds with contrasting land covers, is investigated. The methods range from random selection of BMPs on randomly selected sites, to optimized selection of BMPs at optimized locations, and the land covers range from natural to ultra-urban. The optimization methods rely on Genetic Algorithms (GA), and a method that uses expert systems is also applied. Watershed hydrologic and water quality response models are developed, using the Soil Water Assessment Tool (SWAT), to compute baseline outputs from the 4 study watersheds without BMPs, and to obtain predicted reductions in NPS constituent outputs when BMPs are implemented in accordance with the 5 allocation plans. Methods used to represent BMPs in SWAT and to speed up optimization processes are also presented. Results indicate that the most computationally intensive methods produce the best results across landscape types. Results also show that opportunities exist for less intensive methods, particularly in less-built environments. For these, however, siting BMPs to hotspots remains an important requirement. The need to select the most appropriate BMP for each implementation site is observed to increase with the level of urbanization of the landscape. Results indicate that optimized selection of BMPs, sited at optimized locations, results in the highest-performing BMP allocation plans across landscape types. Also, the focus on hotspots has the advantage of resulting in BMP plans requiring involvement of fewer stakeholders than when BMPs are located in non-hotspot zones. This targeted hotspot approach could help reduce cost and increase efficiency of implementation.

Nonpoint source (NPS) pollution is a pressing issue worldwide, especially in the Chesapeake Bay, where sediment, nitrogen (N), and phosphorus (P) are the most critical water quality concerns. Despite significant efforts by federal, state, and local governments, the improvement in water quality has been limited. Investigating the spatial distribution of NPS hotspots can help understand NPS pollutant output and guide control measures. We hypothesize that as land cover changes from natural (e.g., forestland) and agricultural to suburban and ultra-urban, the distribution of NPS pollution source areas becomes increasingly spatially uniform. To test this hypothesis, we analyzed three real watersheds with varying land cover (Greensboro watershed for agriculture, Watts Branch watershed for suburban, and Watershed 263 for ultra-urban) and three synthetic watersheds developed based on the Watts Branch watershed, which ranged from forested and agricultural to ultra-urban but had the same soil, slope, and weather conditions. The Soil and Water Assessment Tool (SWAT) was selected as a phenomenological model for the analysis, and SWAT-CUP was used for model calibration and validation. The hydrologic responses of the three real and synthetic watersheds were simulated over ten years (1993–2002 or 2002–2011), and calibration and validation results indicated that SWAT could properly predict the export of runoff and three target NPS pollution constituents (sediment, total nitrogen, and total phosphorus). The results showed that the distribution of NPS pollutant outputs becomes increasingly uniform as land cover changes from agriculture to ultra-urban across watersheds. This research suggests that the spatial distribution of NPS pollution source areas is a function of the major land cover category of study watersheds, and control strategies should be adapted accordingly. If NPS pollution is distributed unevenly across a watershed, hotspot areas output a disproportionate amount of pollution and require more targeted and intensive control measures. Conversely, if the distribution of NPS pollution is more uniform across a watershed, the control strategies need to be more widespread and encompass a larger area.

Nonpoint source (NPS) pollution is a severe problem in the U.S. and worldwide. Best management practices (BMPs) have been widely used to control stormwater and reduce NPS pollution. Previous research has shown that socio-economic factors affect households’ adoption of BMPs, but few studies have quantitatively analyzed the spatio-temporal dynamics of household BMP adoption under different socio-economic conditions. In this paper, diverse regression approaches (linear, LASSO, support vector, random forest) were used on the ten-year data of household BMP adoption in socio-economically diverse areas of Washington, D.C., to model BMP adoption behaviors. The model with the best performance (random forest regression, R² = 0.67, PBIAS = 7.2) was used to simulate spatio-temporal patterns of household BMP adoption in two nearby watersheds (Watts Branch watershed between Washington, D.C., and Maryland; Watershed 263 in Baltimore), each of which are characterized by different socio-economic (population density, median household income, renter rate, average area per household, etc.) and physical attributes (total area, percentage of canopy in residential area, average distance to nearest BMPs, etc.). The BMP adoption rate was considerably higher at the Watts Branch watershed (14 BMPs per 1000 housing units) than at Watershed 263 (4 BMPs per 1000 housing units) due to distinct differences in the watershed characteristics (lower renter rate and poverty rate; higher median household income, education level, and canopy rate in residential areas). This research shows that adoption behavior tends to cluster in urban areas across socio-economic boundaries and that targeted, community-specific social interventions are needed to reach the NPS control goal.