In this study, the performance of seven global thresholding method is evaluated, aimed to find the optimal global thresholding method for delineating water area using the Modified Normalized Difference Water Index (MNDWI). To achieve this, seven different global thresholding techniques were evaluated: Otsu, Li, Yen, ISODATA, Mean, Triangle, and Minimum thresholding. The MNDWI was calculated using high-resolution Sentinel-2 Level 2 data, image, which was captured on May 28, 2024, the Jayakwadi Dam in Maharashtra, India was used as study area. For accuracy assessment, the Support Vector Machine (SVM) classified image is used as a reference for calculating the accuracy, as SVM has high accuracy in water body discrimination. The performance of each method was evaluated using statistical measures, including overall accuracy, kappa coefficient, precision, recall, and F1 score. All thresholding methods achieved an overall accuracy of 90% and above. Notably, the highest overall accuracy and F1 score of 99.86% and 0.9986 respectively is given by the Minimum Thresholding Method, followed by Otsu, Li method. Mean thresholding method performed poorly with 90.18% overall accuracy. This thorough assessment identifies the Minimum Thresholding Method as the most effective technique for finding optimal threshold value to accurately delineate water body area.