Google earth engine application for estimating changes in water surface area of Lake Toba

Abstract

The change in the water surface area is essential to be studied as it can affect ecosystem health and function of lakes to meet human needs. The use of remote sensing techniques is to analyze such changes in the Lake Toba is considered effective and efficient ways in terms of time and cost, moreover the application of Google Earth Engine (GEE) as an internet-based data processing platform (cloud computing) that provides various kinds of free satellite image data can simplify the process. In this study, the author aims to estimate the rate of change in the surface water area of Lake Toba in four periods, that are the period of 1990 - 1997, 1998 - 2004, 2005 - 2012, and 2013 - 2018, by using Landsat 5 TM and 8 OLI/TIRS imagery. Cloud masking methods is used to obtain cloud-free images and then using CART method in making land cover maps. After that, calculating the accuracy of CART method by using the confusion matrix to determine the Cohen's Kappa coefficient value. The results of this study show that the highest lake surface water area is in the period of 1990 - 1997 within the area of 1133.88 km2, and the lowest one is in the periode of 1998 - 2004 within the area of 1119.20 km2. The correlation of land cover classes with a surface water area of Lake Toba shows that the open land has high correlation value of -0.77.