Remote Sensing of Impervious Surface Area for Improved Hydrologic Modeling

Impervious surface area (ISA), one of the consequences of urban development, has been used as an important indicator for environmental impacts of urbanization. In this study, we extracted ISA information for the state of Rhode Island, USA from 1-m spatial resolution true-color digital orthophotography data by an object-based classification. As the importance of ISA in the watersheds hydrology, we developed a distributed object-oriented rainfall-runoff simulation model (DORS) with incorporation of high spatial resolution ISA. The DORS model simulates hydrologic processes of precipitation interception, infiltration, evaporation and evapotranspiration, fluctuation of soil moisture and water table, runoff routing, ground water routing and channel flow routing. The model takes objects based on land cover data as the spatial units in order to reduce data volume, increase computational efficiency, strengthen representation of watersheds and utilize the data in variable scales. We validated the temporal variation of simulated discharge using measurements from USGS gage in a study watershed. The validation results indicate that the DORS model is capable of capturing the relationship between rainfall and runoff in the study area. The precise ISA information enhances the performance of the model simulation. Finally, we analyzed the relationship between watershed characteristics and hydrology pattern in selected watersheds with various levels of urbanization using spatial regression models. The result indicates that ISA plays important role in the change of watershed hydrology and the spatial autocorrelation in observations can not be neglected in the analysis. * Corresponding author.