Surface water dynamics are important for understanding the impact of global change and human activities on water resources. Remote sensing has many advantages over surface water monitoring. On a large scale, however, the efficiency of traditional reconnaissance methods is very low because they involve a great deal of manpower, cost, and computational resources. In this study, we propose a new method for rapidly determining the maximum and minimum annual surface water levels. The maximum and minimum water levels were calculated from 2003 to 2021 in the Razazah Lake basin in Iraq on the Google Earth Engine platform. This approach processed 4693 Landsat image frames from the data and computational benefits of the Google Earth Engine cloud platform. In the first step, based on the estimated value of cloud cover for each pixel, cloud-covered pixels were removed to eliminate cloud interference and improve computational performance. Second, the greenest and wettest annual images were mosaicized based on vegetation index (NDVI) and surface water index (NDWI), then the minimum and maximum levels of surface water were obtained by classification by maximum similarity method. The results showed that the average area of ​​Razazeh Lake in 2003 was 2786.6 hectares and from 2004 to 2021, respectively, were equal to 1824.2, 2065.4, 1760.9, 1606.9, 8327.1, respectively. / 4573, 9/6288, 7/8470, 6/8470, 4/2382, 1/3183, 9/1878, 2/9153, 7/655, 9/297, 9/232, 20/287 and 5/357 hectares Came. The accuracy of surface water classification varies from 86 to 93%.