Baltimore Ecosystem Study LTER

Tracking Nutrients Through an Urban Watershed

Assessing macrinvertebrates to determine stream health

The spread and intensification of urban land uses is among the most conspicuous of global changes. Because urban systems are so dynamic, long-term studies are crucial to understanding their ecological implications. The Baltimore Ecosystem Study (BES LTER) has been under way for only a short time. But already the importance of long-term studies of effects and processes of urbanization are becoming clear. Nowhere is this more apparent than in the loading of pollutants in Baltimore's streams. We chose to study stream nutrient loading as one of the key aspects of urban ecosystems because streams integrate many of the complex interactions between biota, physical processes, socioeconomic factors, and infrastructure. Our stream studies have focused so far on the 17,150 ha Gwynns Falls watershed.

The BES stream record began in 1998 and already, some clear temporal and spatial trends are emerging. First, any site that supports urban or suburban development, or is managed by humans has higher nitrate and phosphate loading than the forested reference watershed. The small watersheds representing urban and suburban development have similar nitrate concentrations, but the highest nitrate loadings are from the agricultural watershed.

Comparing sites along the main stem of the Gwynns Falls stream showed that phosphate and nitrate were highest in the headwaters in suburban areas, but decreased downstream as the degree of urbanization increased. The very high levels of nutrient loading in the suburban site was unexpected, which prompted us to look not only at contemporary factors such as roads and lawn fertilization, but also to explore lags from prior agricultural land uses in areas that are now suburban. In addition, the pattern requires that the complex of factors that make up urbanization be examined through spatial comparisons of contrasting land covers, infrastructure, and human use, as well as measured into the future. This pattern is counter to conventional wisdom about urban streams and raises questions of infrastructural connections with the streams. Without long-term data, this pattern would not have appeared. Retrospective studies are adding temporal depth to the stream quality data. Stream channel morphology is being measured and will be related to contemporary re-sampled biotic and chemical stream data from 23 sites in addition to those main stem and small watershed stations presented above.

Although analysis of these data is not yet complete, the approach illustrates the power of long-term stream data to integrate urban ecosystem function. This research has been conducted by a team of BES researchers representing several institutions: Lawrence Band, University of North Carolina, Chapel Hill; Kenneth Belt, USDA Forest Service, Northeastern Research Station; Mark Colosimo, US Army Corps of Engineers; Gart Fisher, US Geological Survey;Peter Groffman, Institute of Ecosystem Studies.