A Method for Chlorophyll-a and Suspended Solids Prediction through Remote Sensing and Machine Learning.

Area of Science:

• Environmental Science
• Remote Sensing
• Machine Learning

Background:

• Water quality monitoring is crucial for environmental health.
• Traditional water sampling for parameters like Total Suspended Solids (TSS) and chlorophyll-a is costly and time-consuming.
• Optically active water components offer potential for remote sensing estimation.

Purpose of the Study:

• To develop and validate a cost-effective methodology for estimating TSS and chlorophyll-a concentrations.
• To leverage remote sensing and Machine Learning (ML) for water quality assessment.
• To demonstrate the applicability of the methodology across different water bodies.

Main Methods:

• Utilized supervised Machine Learning (ML) algorithms for prediction.
• Employed remote sensing data from Sentinel-2 spectral images and unmanned aerial vehicles (UAVs).
• Integrated laboratory analysis data for model training and validation across two distinct water bodies.

Main Results:

• Developed ML models capable of predicting TSS and chlorophyll-a concentrations.
• Achieved high model performance with R-squared values exceeding 0.8 in both study areas.
• Demonstrated the effectiveness of remote sensing data for water quality parameter estimation.

Conclusions:

• Remote sensing combined with ML provides an accurate and efficient alternative to traditional water quality monitoring.
• The proposed methodology is transferable to different aquatic environments.
• This approach offers significant cost and time savings for water quality assessment.