Prediction of Water Quality Parameters in the Paraopeba River Basin Using Remote Sensing Products and Machine Learning.

This project develops and tests predictive models for water‑quality parameters in the Paraopeba River Basin. It combines satellite‑based remote‑sensing data with machine‑learning algorithms to link spectral reflectance measurements to on‑site water‑quality observations. The workflow starts by preprocessing multi‑spectral satellite images, extracting useful environmental indices, and matching those data with historical ground‑truth records. Several machine‑learning methods are tried, including Random Forests, Support Vector Regression, and Gradient Boosting. These models predict key variables such as turbidity, total suspended solids, and chlorophyll‑a concentration. Model performance is measured with the coefficient of determination (R²) and root‑mean‑square error (RMSE). The approach gives clear guidance for modeling how water quality changes over space and time, and

it can be scaled up for watershed management and environmental monitoring. By merging remote sensing with advanced computation, the research offers a non‑intrusive, low‑cost way to monitor water quality continuously in complex river basins.

Message: Hi FE, I need help with “Prediction of Water Quality Parameters in the Paraopeba River Basin Using Remote Sensing Products and Machine Learning.” in “Civil Engineering”