IMPACT OF ATMOSPHERIC CORRECTION FOR TURBIDITY ESTIMATION IN COASTAL WATERS WITH SENTINEL-2 DATA AND AI TECHNIQUES

In recent years, Artificial Intelligence (AI) has been a key element in addressing complex issues in many fields, thanks to its ability to analyze large amounts of data and provide innovative solutions. This study focuses on the application of AI to estimate turbidity values in the coastal waters of the Veneto region in Italy using Sentinel-2 satellite data. The primary goal is to assess the impact of atmospheric correction on the reliability of derived parameters, an essential factor in monitoring water quality and safeguarding marine ecosystems. The project integrates in situ data provided by ARPAV, the regional Agency of Environment Protection, with satellite imagery accessed through Google Earth Engine (GEE), generating separate datasets based on L1C and L2A images, respectively. These datasets were employed to train and evaluate multiple Machine Learning (ML) models to identify the most effective approach for turbidity estimation and environmental monitoring optimization. The analysis highlighted that the use of atmospherically corrected data is not strictly necessary and underlined the need for strategies to improve model performance, especially in conditions of limited data availability. The results obtained demonstrate the feasibility and effectiveness of using AI algorithms for water quality monitoring, providing a concrete basis for future developments. This approach has the potential to support timely and targeted interventions for the protection of the marine environment and public health.