The inland water bodies in the Doukkala plain are mainly water surface bodies locally known as Dayas. These Dayas are of vital socioeconomic and ecological significance. Several years of drought have resulted in a water shortage in this area. The current management of water resources lacks relevance. Sustainable water management has become a necessity and therefore must involve monitoring and mapping these Dayas. Remote sensing technologies play an important role in completing this task. In this study, we calibrated the multi-band water index (MBWI) to our study area using three weighting factors (w = 2, 3, and 4) with thresholds selected iteratively using two distinct step values (0.1 and 0.01). To make it easier to apply the indices to different situations, we utilized the average of the ideal thresholds as the single index threshold for each coefficient. The computation was carried out using Landsat images on the Google Earth Engine (GEE) platform, and then validation was carried out by collecting ground data with Google Earth Pro from very-high-resolution images. The comparison was conducted for five Landsat scenes. To assess the accuracy performance of the method, we calculated the overall accuracy (OA) and the Kappa coefficient (Kappa). The results show that the weighting coefficient (w = 4) and the threshold (-0.008) yielded better performances, with a Kappa between 0.92 and 0.97, in the five scenes.