Estimativa de parâmetros da qualidade da água por meio de sensoriamento remoto no lago Igapó I em Londrina - PR

Abstract

Water bodies that suffer impacts from anthropogenic pressures require continuous environmental monitoring, especially when located in urban areas, due to the increased dynamics of water and pollutant inflow. However, this monitoring incurs high costs or limited access for sample collection, requiring study of new methodologies that facilitate this process and are accessible, such as remote sensing. In this context, the current study aims to estimate the Total Suspended Solids (TSS), Turbidity, and trophic level of Igapó Lake I, located in the urban area of the city of Londrina, thru remote sensing, using images from the Sentinel 2 satellite. For the determination of TSS and Turbidity, ten water samples from the lake were collected and analyzed, with six used for the calibration of an empirical equation thru linear regression analysis, while the other four were used to determine the validation metrics, RMSE and MAE. In the statistical analysis of these two parameters, only the results with a p-value less than 0.05 and a high R² value were considered. Regarding the estimation of Chlorophyll-a for the determination of trophic level, it was not possible to conduct in situ analyzes due to the unavailability of materials for the procedure; therefore, an empirical equation proposed for the Tietê River Basin was used, using images from the same satellite. Thus, a historical series was developed demonstrating the variation in the trophic level of the lake from January 2024 to November 2025. With the linear regression analysis, satisfactory results were not obtained for the determination of TSS, making it impossible to generate the spatial distribution map. Regarding turbidity, a linear equation was obtained with an R² of 0.71 and a p-value of 0.03, allowing the generation of a map with the turbidity estimate for the entire water body, with an RMSE of 1.55 and an MAE of 0.78 NTU. The historical series for Chlorophyll-a was estimated, indicating the evolution of the trophic level of the lake from mesotrophic and eutrophic to hyper-eutrophic, coinciding the sampling period with the most advanced state of eutrophication, which significantly impacted the collected water samples and the statistical results obtained in this work. Finally, remote sensing proved to be an efficient tool for environmental monitoring in aquatic ecosystems in urban perimeters.