A geospatial study of the drought impact on surface water reservoirs: study cases from Texas,USA

Satellite images have been used historically to measure and monitor fluctuations in the surface water reservoirs. This study integrates remote sensing and Geographic Information System (GIS) technologies to investigate the impact of drought on 10 selected surface water reservoirs in San Angelo and Dallas, Texas. Oscillations in summer and winter months throughout the 2005–2016 period were assessed using multispectral images from Landsat-5, ?7, and ?8, and changes in the reservoirs were characterized and correlated against local climate data of each reservoir. For quantitative comparisons of the time-series measurements, a robust density slicing approach was employed to classify the range of values of the raster cells in the near-infrared band of Landsat images for each lake into three desired classes (deep water, shallow water, and dry area) based on the natural breaks inherent in the dataset. Statistical analysis shows that the overall accuracy of the classification is about 94%, which demonstrates the efficiency of the density slicer to accurately estimate surface water area changes from an individual Landsat band. Shrinkage in the surface water area over the study period reveals the concrete impact that the drought along with other factors have on the 10 selected lakes. The San Angelo lakes located in west central Texas experienced a nearly consistent pattern of change during most of the study period; whereas the Dallas lakes in northeast Texas followed the oscillating pattern of drought and correlated closely to the local conditions. Shockingly, the extreme drought caused complete vanishing of several lakes, and consequently Texas had to remove them from its recreational plans. Our new findings can certainly help with the water resource management in Texas and our study approach can be adapted for monitoring lake oscillations in other areas across the world. This geospatial study demonstrates the societal benefits from incorporating remote sensing and GIS in investigating geo-environmental problems associated with severe climate changes.