Using high resolution images from UAV and satellite remote sensing for best management practice analyses

Abstract

Best Management Practices (BMPs) are commonly adopted to ameliorate the quality of runoff and reduce the frequency and intensity of flash floods in urban areas. To date, many of the BMP studies are conducted using coarse resolution data. However, the accuracy of such studies may be compromised due to the shortcomings inherent in the input data; as such, the evaluation of the BMP cost-effectiveness may not be accurate. The objective of this paper is to demonstrate the improvements of higher resolution images over coarse resolution data in BMP analyses. An unmanned aerial vehicle (UAV) was used to collect a more detailed and accurate picture of the digital surface model and digital elevation model. Landsat 8 multi-spectral imagery was classified by object-oriented classification to generate a land use/land cover map. The method used in this study provided more detailed and accurate information of the physical conditions of the study area, an improved subwatershed delineation, a more comprehensive list of the suitable locations for BMPs, and a more reliable estimate of the cost-effectiveness of the BMP ensembles than that generated using coarse resolution data. Using the fine resolution data, this study further determined the utility of the selected BMP ensembles under a changed future climate regime and identified the best BMP and BMP ensemble in reducing urban surface runoff. This method can be especially useful in areas without quality topography and land use data.