Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Monitoring water contaminants in coastal areas through ML algorithms leveraging atmospherically corrected Sentinel-2 data

Francesca Razzano, Francesco Mauro, Pietro Di Stasio, Gabriele Meoni, Marco Esposito, Gilda Schirinzi, Silvia Liberata Ullo

TL;DR

This study tackles turbidity monitoring in coastal waters by integrating atmospheric correction-aware Sentinel-2 Level-2A imagery with CatBoost regression, powered by data processed in Google Earth Engine for scalable monitoring. The authors construct a tabular dataset from 660 atmospherically corrected Sentinel-2 samples around in-situ turbidity dates in Hong Kong and train a CatBoost model to predict turbidity from image-band features, achieving strong predictive performance and favorable comparisons to prior ANN-based work. The findings demonstrate that this RS-based, ML-driven approach can deliver accurate, scalable water quality insights across coastal regions, highlighting its potential for global applications. Future work will extend the framework to additional chemical contaminants, further enhancing coastal ecosystem preservation and public health protection.

Abstract

Monitoring water contaminants is of paramount importance, ensuring public health and environmental well-being. Turbidity, a key parameter, poses a significant problem, affecting water quality. Its accurate assessment is crucial for safeguarding ecosystems and human consumption, demanding meticulous attention and action. For this, our study pioneers a novel approach to monitor the Turbidity contaminant, integrating CatBoost Machine Learning (ML) with high-resolution data from Sentinel-2 Level-2A. Traditional methods are labor-intensive while CatBoost offers an efficient solution, excelling in predictive accuracy. Leveraging atmospherically corrected Sentinel-2 data through the Google Earth Engine (GEE), our study contributes to scalable and precise Turbidity monitoring. A specific tabular dataset derived from Hong Kong contaminants monitoring stations enriches our study, providing region-specific insights. Results showcase the viability of this integrated approach, laying the foundation for adopting advanced techniques in global water quality management.

Monitoring water contaminants in coastal areas through ML algorithms leveraging atmospherically corrected Sentinel-2 data

TL;DR

This study tackles turbidity monitoring in coastal waters by integrating atmospheric correction-aware Sentinel-2 Level-2A imagery with CatBoost regression, powered by data processed in Google Earth Engine for scalable monitoring. The authors construct a tabular dataset from 660 atmospherically corrected Sentinel-2 samples around in-situ turbidity dates in Hong Kong and train a CatBoost model to predict turbidity from image-band features, achieving strong predictive performance and favorable comparisons to prior ANN-based work. The findings demonstrate that this RS-based, ML-driven approach can deliver accurate, scalable water quality insights across coastal regions, highlighting its potential for global applications. Future work will extend the framework to additional chemical contaminants, further enhancing coastal ecosystem preservation and public health protection.

Abstract

Monitoring water contaminants is of paramount importance, ensuring public health and environmental well-being. Turbidity, a key parameter, poses a significant problem, affecting water quality. Its accurate assessment is crucial for safeguarding ecosystems and human consumption, demanding meticulous attention and action. For this, our study pioneers a novel approach to monitor the Turbidity contaminant, integrating CatBoost Machine Learning (ML) with high-resolution data from Sentinel-2 Level-2A. Traditional methods are labor-intensive while CatBoost offers an efficient solution, excelling in predictive accuracy. Leveraging atmospherically corrected Sentinel-2 data through the Google Earth Engine (GEE), our study contributes to scalable and precise Turbidity monitoring. A specific tabular dataset derived from Hong Kong contaminants monitoring stations enriches our study, providing region-specific insights. Results showcase the viability of this integrated approach, laying the foundation for adopting advanced techniques in global water quality management.
Paper Structure (11 sections, 4 figures, 3 tables)

This paper contains 11 sections, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. GEE ENVIRONMENT AND DATA SOURCES
  3. GEE environment
  4. Sentinel-2 Level-2A
  5. AREA OF INTEREST
  6. DATASET CREATION AND METHODOLOGY
  7. Dataset creation
  8. The ML algorithm: CatBoost
  9. Methodology applied
  10. RESULTS
  11. CONCLUSIONS

Figures (4)

  • Figure 1: Map of the 76 monitoring stations in coastal areas and open sea of Hong Kong region - the Area of Interest (AOI)
  • Figure 2: Workflow of the proposed work
  • Figure 3: Comparison of expected (red) and predicted (blue) values for Training, Validation, and Testing sets
  • Figure 4: Comparison of expected (red) and predicted (blue) values for Training, Validation, and Testing sets