In this work, we aim to evaluate the feasibility and operational limitations of using Sentinel-1 synthetic aperture radar (SAR) data to monitor water levels in the Poço da Cruz reservoir from September 2016–September 2020, in the semi-arid region of northeast Brazil. To segment water/non-water features, SAR backscattering thresholding was carried out via the graphical interpretation of backscatter coefficient histograms. In addition, surrounding environmental effects on SAR polarization thresholds were investigated by applying wavelet analysis, and the Landsat-8 and Sentinel-2 normalized difference water index (NDWI) and modified normalized difference water index (MNDWI) were used to compare and discuss the SAR results. The assessment of the observed and estimated water levels showed that (i) SAR accuracy was equivalent to that of NDWI/Landsat-8; (ii) optical image accuracy outperformed SAR image accuracy in inlet branches, where the complexity of water features is higher; and (iii) VV polarization outperformed VH polarization. The results confirm that SAR images can be suitable for operational reservoir monitoring, offering a similar accuracy to that of multispectral indices. SAR threshold variations were strongly correlated to the normalized difference vegetation index (NDVI), the soil moisture variations in the reservoir depletion zone, and the prior precipitation quantities, which can be used as a proxy to predict cross-polarization (VH) and co-polarization (VV) thresholds. Our findings may improve the accuracy of the algorithms designed to automate the extraction of water levels using SAR data, either in isolation or combined with multispectral images.

Although the single threshold is still considered a suitable and easy-to-do technique to extract water features in spatiotemporal analysis, it leads to unavoidable errors. This paper uses an enumerative search to optimize thresholds over satellite-derived modified normalized difference water index (MNDWI). We employed a cross-validation approach and treated accuracy as a random variable in order to: (a) investigate uncertainty related to its application; (b) estimate non-optimistic errors involving single thresholding; (c) investigate the main factors that affect the accuracy’s model, and (d) compare satellite sensors performance. We also used a high-resolution digital elevation model to extract water elevations values, making it possible to remove topographic effects and estimate non-optimistic errors exclusively from orbital imagery. Our findings evidenced that there is a region where thresholds values can vary without causing accuracy loss. Moreover, by constraining thresholds variation between these limits, accuracy is dramatically improved and outperformed the Otsu method. Finally, the number of scenes employed to optimize a single threshold drastically affects the accuracy, being not appropriate using a single scene once it leads to overfitted threshold values. More than three scenes are recommended.