Assessment of water quality and evaluation of best management practices in a small agricultural watershed adjacent to Coral Reef area in Japan

The predominantly agricultural coastal watershed located in tropical–subtropical Island has served as a pathway for nonpoint source (NPS) pollution loading into the downstream coastal ecosystems. The present study aims to assess watershed hydrology and water quality by applying a watershed model to long–term high–resolution monitoring data, and to evaluate the reduction of nonpoint–source pollution by implementing best management practices (BMPs) in a small agricultural watershed located in southwestern Japan. The Soil Water and Assessment Tool (SWAT) was coupled to a groundwater model (MODFLOW) and applied to simulate stream flow, sediment, and nutrients discharges. The results revealed that the model successfully predicted hourly stream flow, sediment, and nutrients concentrations in the study watershed and that the model could be further used to evaluate BMP scenarios. This is the first attempt to quantify the impacts of multiple agricultural mitigation measures on long-term water quality for sediment, nitrate, and phosphate at an hourly resolution. From the model results, the land cover conversion and mulching scenarios achieved a significant reduction in sediment, followed by the cover crop, alternative tillage, and filter strip scenarios. Moreover, conversion of all summer sugarcane to pasture farmland and fertilizer control scenarios gave the superior reduction in nitrate load, while other scenarios were minimally effective. On the other hand, most of the scenarios significantly achieved the reduction of phosphate load. The results further indicated that the BMPs recommended in this study are appropriately combined scenarios, which achieved the significant reduction in all NPS loads. Although this study revealed some effective mitigation measures to improve the water quality in the study watershed, actual implementation of correct ones may be practically challenging unless other environmental and economic risks (e.g. the increase of other pollutants and reduction in agricultural productivity) are also taken into considerations.