Streams in urban watersheds are typically warmer at low flows and can experience higher temperatures during storm flows when run-off passes over hot paved surfaces. It's known that high temperatures can stress in-stream organisms, however our understanding of the relationship between development and altered stream temperature remains imprecise. Water temperature will be monitored at multiple locations to distinguish the importance of the spatial configuration of development, contrasting above-ground connectivity (e.g., impervious patch size, shape, location, shading) and below-ground connectivity (e.g., stormwater infrastructure, especially pipes). This will improve our understanding of the impact of urban development on heat pulses in streams, and more generally on stream ecosystems. Because these metrics will be linked more explicitly to hydrologic behaviors than many development proxies that are now commonly used, this research will help to better understand how urban watersheds function and how policies can ameliorate negative effects of urbanization on stream ecosystems.
Collaborators in this research will include faculty and graduate student researchers at Duke University, University of North Carolina at Chapel Hill, and North Carolina State University, the Triangle J Council of Governments, and watershed managers working for Chapel Hill, Durham, Carrboro, and Raleigh. A local eighth grade class will be involved through the use of an outdoor classroom and the presentation of results during class times and members of Girl Scout Troop 514 will participate in fieldwork and a day of learning about thermal pollution through hands-on activities.
In this project we explored the relationship between the spatial configuration of urban development infrastructure, such as impervious surfaces and stormwater pipes, and the thermal regime of urban streams in the Piedmont of North Carolina (framed by Durham, Chapel Hill, and Raleigh). Our approach was to choose study watersheds that showed similar levels of development intensity (approximately 50% developed land cover) while exploring variation in the spatial configuration of this development. In this, we asked: To what extent can the arrangement of development compensate for developed area? Importantly, the amount of variablity in spatial configuration within this narrow range of development intensity was similar to that observed over the full range of variation in developement intensity (2-94% developed) in our study area. We collected field measurements on stream temperature, and developed statistical models to explain the magnitude and timing of temperature pulses following storm events. We used hierarchical linear regression models to show how development infrastructure mediates the effects of antecedent air temperature and stream flow on thermal pulses. Stream-scale variable such as channel incision and stormwater pipe length were the best predictors of baseflow stream temperatures, while watershed-scale variables related to impervious surface area, the proximity of stormwater pipe outlets, and the proportion of the watershed in forest cover best explained stormflow heat pulses. Our approach represents a novel contrast to the prevailing model based on gradients of development intensity, and offers substantial promise for finding ways to mitigate developement impacts on stream ecosystems at a given level of development.
