Suburban and exurban growth into formerly undeveloped areas frequently harm water quality. However, as planned landscapes, there is the possibility to implement measures within these suburban and exurban areas to mitigate deleterious changes to water bodies. To date, structural stormwater Best Management Practices (BMPs) such as detention ponds have demonstrated mixed potential to mitigate pollutant inputs over the long term, particularly in regards to dissolved nutrients. Additionally, such BMPs can be difficult to implement retroactively in mature, developed areas. Thus, a more fundamental understanding of sites of pollutant generation and transport pathways could lead to more refined and directed means of modifying landscapes to reduce pollutant loads to nearby surface waters. In particular, much as biogeochemical "hot spots" - areas of disproportionately high chemical and biological activity - have been surmised for natural systems, I propose analogous zones may be identified in more urban landscapes, defined by areas of small-scale landscape features that enhance pollutant generation and transport.

This dissertation will investigate spatially explicit processes that can be linked to small-scale spatial features in urban landscapes. The first chapter provides background on the current understanding of the relationship between small-scale spatial features and water quality. The second and third two chapters use a combination of theory and experiments to develop a mechanistic particulate wash-off model, investigating the role of impervious surface roughness on rates of particulate loss. The fourth chapter examines catchments from a larger scale using weather radar. This chapter relates particulate wash-off to rainfall kinetic energy (a metric traditionally ignored in urban pollutant models) while questioning the use of antecedent dry days to explain the magnitude of wash-off (the routine explanation for variations in particulate loss). Overall, this body of work reassesses standard but frequently unvalidated assumptions used in estimating non-point pollutant loads.