Monitoring Surface Water Areas Using Satellite Remote Sensing and Cloud Computing: A Case Study of Lake Burdur, Turkey

Abstract

Currently, there are barriers and limited tools for researchers and stakeholders to use remote sensing data and space-borne imagery to track ecological changes in bodies of water. This research explored the use of satellite images stored on the cloud and leverages online services to extract and use open-source software to estimate surface water parameters. Specifically, the research provided evidence for estimating lake water area using satellite data available in the Amazon Web Services (AWS) Open Data Program, open-source code and open-source software. A new methodology was developed to extract green spectral band and near-infrared (NIR) spectral band from satellite images to calculate the Normalized Water Index (NDWI). This methodology was applied to six satellite images of Lake Burdur, Turkey, from two different satellite systems, Sentinel and Landsat, in order to determine decreasing water area trends previously identified in time interval studies. The data and the NDWI analysis of the sample satellite images demonstrated a reduction of the water surface of Lake Burdur. This methodology yielded similar results for both satellite imagery sets despite different image attributes. I concluded that the water surface reduction first identified by a previous study continued beyond the period of the study, with additional water surface reduction between 2015 and 2019. The study achieved its main goal of validating a methodology to monitor surface water changes in lakes over time, while removing research access barriers to any parties in academia or outside of it.