Possibilities of assessing the state of Lake Baikal surface water using Google Earth Engine

Abstract

The use of remote sensing (RS) data is associated with many difficulties that lead to high labor and computer resource costs. The use of cloud computing greatly simplifies this work. The aim of this study was to demonstrate the possibilities of Google Earth Engine (GEE) for assessing the state of Lake Baikal surface water. An overview of the datasets on the GEE platform was made. The available data processing algorithms were reviewed. It was noted that, in addition to standard methods of stationary processing of satellite images, the GEE platform offers quite complex machine learning methods. For example, Random Forest method significantly improves the accuracy of the results. In addition, the possibilities of constructing a composite from Sentinel-2, Landsat-8, MODIS satellite images, calculating spectral indices to determine the boundaries of water bodies, estimating surface temperature, chlorophyll-a concentration, Secchi disk transparency in selected areas of the water body are shown. It is necessary to control the accuracy of the data and compare it with field research data. The results of RS data processing can be used as additional material during monitoring the state of water bodies. Processing is carried out in the cloud using the JavaScript programming language. It is possible to create applications for interactive use. It is advisable to use GEE possibilities for operational monitoring and analysis of changes in water bodies based on large data sets available on the platform.