What question did this study set out to answer?

To evaluate surface water chemistry suitability for drinking, irrigation, and industrial applications and compare multiple machine learning algorithms for predicting water quality index.

August 14, 2026Applied Water ScienceOpen Access

Evaluation and prediction of seasonal variation in surface water chemistry for beneficial use of drinking, irrigation, and industrial in Mahanadi River of Paradip area, Odisha, India, using an GIS: integrated water quality indices, and machine learning approaches

View Full Paper

Ask AI

Bookmark

To evaluate surface water chemistry suitability for drinking, irrigation, and industrial applications and compare multiple machine learning algorithms for predicting water quality index.
Assessed 12 physicochemical parameters across nine sampling sites using multiple indices: Weighted Arithmetic Water Quality Index (WA-WQI), Numerow’s Pollution Index (NPI), Overall Index of Pollution (OIP), and Synthetic Pollution Index (SPI).
Applied multivariate statistical methods (Cluster Analysis, Principal Component Analysis) alongside Piper and Gibbs geochemical plots to trace contamination sources.
Developed and benchmarked four machine learning models (RFM, SVM, GBM, XGB) to predict WQI across six evaluation metrics including R², NSE, WI, MAE, and RMSE.
WA-WQI values spanned 40.36 to 176.13, with OIP classifying 66.66% of samples as poor quality and NPI identifying 77.78% as deteriorated, primarily driven by BOD, PO4^3-, EC, TDS, and F-.
PCA extracted three components explaining 56.65%, 19.90%, and 11.97% of total variance, identifying TDS, TH, SO4^2-, NO3^-, and PO4^3- as key drivers of chemical degradation.
Random Forest model outperformed all competing algorithms on the test set (R² = 0.93, NSE = 0.94, WI = 0.85, PBIAS = 12.02, MAE = 45.91, RMSE = 111.43), whereas SVM showed the lowest predictive performance.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Environmental modeling study reveals high surface water pollution and validates Random Forest accuracy in river basins, highlighting AI utility for water resource monitoring.

Cite This Study

Abhijeet Das (2026) studied this question.

synapsesocial.com/papers/6a7ec71db70b84ec8b913421 https://doi.org/10.1007/s13201-026-02933-8

View Full Paper

Ask AI

Bookmark

Implication

Environmental modeling study reveals high surface water pollution and validates Random Forest accuracy in river basins, highlighting AI utility for water resource monitoring.

Evaluation and prediction of seasonal variation in surface water chemistry for beneficial use of drinking, irrigation, and industrial in Mahanadi River of Paradip area, Odisha, India, using an GIS: integrated water quality indices, and machine learning approaches

Key Points

Discussion

Implication

Cite This Study

Evaluation and prediction of seasonal variation in surface water chemistry for beneficial use of drinking, irrigation, and industrial in Mahanadi River of Paradip area, Odisha, India, using an GIS: integrated water quality indices, and machine learning approaches

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider