Chlorophyll-a Modeling in the Sierpe River with Sentinel-2 and Google Earth Engine

Abstract

Modeling of chlorophyll-a with satellite data can be achieved by finding a best fit function between spectral bands of reflectance related to its concentration. Google Earth Engine (GEE) is a platform designed for the processing of multispectral satellite databases used for the temporary observation of terrestrial phenomena. This tool was selected to model the chlorophyll-a concentration of the Sierpe River, which is located within the Térraba Sierpe National Wetland (HNTS), Costa Rica. For the adjustment, a linear correlation function where the independent variable is the division of the red edge by the blue spectral band from Sentinel-2 level 2A images, was used. Appropriate filters for tropical conditions were applied. The modeling results were validated with laboratory data and 43 images of the river were generated showing a complete correspondence with the trend of the contents in the different seasons of the year, and the reported data of growth of aquatic plants that increase after fertilization.