Definition of reservoirs level from satellite images

Abstract

The monitoring water resources by in situ meters is expensive, as they depend on factors such as location, size and access to the reservoir. For this reason, in recent years the monitoring of water by remote sensing and especially by satellite has become an important objective in hydrology. That way, the purpose of the work was to use the Google Earth Engine (GEE) platform as a technology to extract the flooding area from 4 (four) reservoirs located in the state of Ceará on different dates throughout its operation and, with the respective data from the water level, extracted from the hydrological portal of the Companhia de Gestão de Recursos Hídricos – COGERH, create a linear regression equation for each of them, which makes the relation between the level and the reservoir area, in order to create a management tool that returns as a result the value of the reservoir water level, through satellite imagery. As analyzed, equations with correlation values above 0.88 were obtained, showing the efficiency of the methodology when using GEE. It can also be seen that the reservoirs with a greater quantity of good quality images showed better results of study.