Landscape Indicators of Watershed Impairment

Principal Investigators:

Stephen D. Prince

This project is intended to develop a watershed classification scheme based on recent, much improved, comprehensive watershed data sets to diagnose aquatic ecosystem impairment and to target resource management. Hydrologic factors, nutrient budgets incorporating point and non-point source/sinks, and landscape function metrics are being used to provide indicators of aquatic ecosystem condition. These conditions include hydrology, plant, fish, macroinvertebrates, and water quality in watersheds. To identify the watershed variables most relevant to prediction of impairment of the receiving water bodies a set of empirical classification models at multiple scales are being developed. The classifications are being developed for the mid-Atlantic then will be applied to southern New England (MA, RI, CT), and generalized for future national application.

Aquatic health is assessed using flow metrics, water quality metrics (such as sediment and nutrient concentrations), and biological indicators. Maps of land use, land cover, anthropogenic influences have been obtained using multitemporal Landsat ETM+ data. The relationships between watershed attributes, flow and water quality, and biological indicators are formalized as statistical models with associated significance and confidence metrics for each scale. The sensitivity of ecological response variables to natural and anthropogenic variations in watershed properties is an important aspect. The models will be used to develop hierarchical decision trees that specify a set of binary splits leading to a finite set of impairment categories. The rules will be tested and errors in the independent variables (e.g., land cover misclassification) will be quantified to measure classification accuracy.

As a result of the much-improved landscape data inputs and the use of metrics of watershed function, we expect to be able to predict watershed impairment and trends toward impairment. The hierarchical structure of the classification, with explicit decision rules, will be accessible to managers. The classifications will diagnose impairment of watersheds, assess ecosystem vulnerability, and provide for monitoring and prioritization for restoration activities, all at multiple scales.

UMD is collaborating in the project with the Woods Hole Ecosystem Center and the Smithsonian Environmental Research Center.