A new framework to model the distributed transfer and retention of nutrients by incorporating topology structure of small water bodies.

Small water bodies such as interval water-flooded ditches, ponds, and streams serve as important nutrient sinks in many landscapes, especially in the multi-water continuum system. Yet watershed nutrient cycling models often fail to or insufficiently capture these waters, resulting in great uncertainty in quantifying the distributed transfer and retention of nutrients across diverse landscapes in a watershed. In this study, we present a network-based predictive framework of the nutrient transport process in nested small water bodies, which incorporates topology structure, hydrological and biogeochemical processes, and connectivity to perform a nonlinear and distributed scaling of nutrient transfer and retention. The framework was validated and applied to N transport in a multi-water continuum watershed in the Yangtze River basin. We show that the importance of N loading and retention depends on the spatial context of grid source and water bodies because of the great variation in location, connectivity, and water types. Our results demonstrate that hotspots in nutrient loading and retention could be accurately and efficiently identified through hierarchical network effects and spatial interactions. This offers an effective approach for the reduction of watershed-scale nutrient loads. This framework can be used in modeling to identify where and how to restore small water bodies for reduced non-point pollution from agricultural watersheds.