This work quantified the effects of urbanization on selected hydrologic and ecological processes in Valley Creek watershed in suburban Philadelphia. Biotic community structure was analyzed using species composition, stable isotopes and C:N:P ratios. Nitrogen isotope levels were found to be indicative of inputs of anthropogenic nitrogen sources to the stream attributed to leaking septic systems. Biotic community structure was shown to be complex and related to land use. Diversity of fish species at each of fifteen stream sampling stations showed a response to land use parameters, especially percent impervious surface coverage, but was also correlated to groundwater inputs at measured spring locations along the stream. Spatial heterogeneity of water quality (major cations, and anions, pH, and bicarbonate system) of 110 springs was evaluated using geostatistical methods. Dominating controls on spring water quality (anthropogenic versus geologic) were shown to be spatially correlated to the patterns and scales of land use and geology.

Measured anthropogenic perturbations to both stream water quality and quantity were used to determine stream-aquifer exchange parameters by calibration the USGS model OTIS to observed flow and transport data. Two sites were studied intensively to determine the impact of construction projects on local conditions in the stream. Injection of bromide above existing background levels was used to examine the temporal effect of development on solute transport in the stream both at the local (reach) scale and watershed scale.

The effectiveness of an existing system of over 100 stormwater detention basins operating at the watershed scale was also evaluated. The design objective of these detention basins is to limit a site's post-construction peak flow rate to or below its predevelopment level for 2- through 100-year storms. In order to evaluate the watershed-wide effectiveness of the network of detention basins, all basins were surveyed and included in a hydrologic model of the watershed. The HEC-HMS model was calibrated using measured rainfall and observed streamflows from a USGS stream gage. Results from modeling six measured storm events show that the detention basins reduce watershed-wide peak storm flows by an average of only 0.3%, and can potentially increase peak flow rates.