 This study models the spatiotemporal changes of Lake Urmia using satellite data from Landsat 5TM, 7 ETM+, and 8 OLI images. The applicability of different satellite-derived indexes is investigated for extracting surface water from Landsat data. The Normalized Difference Water Index, NDWI, is found to be superior and is used to model the spatiotemporal changes of the lake. A new approach based on principal components of multi-temporal NDWI, NDWI-PCS, is proposed and evaluated for surface water change detection. The results indicate a decreasing trend in Lake Urmia's surface area, with a loss of about one-third of its surface area between 2010 and 2013 compared to the year 2000. The NDWI-PCS approach is effective for detecting changes between two and three different times simultaneously. This article was authored by Komail Rakhni, Anuramaad, Ali Selamat, and others.