Related Experiment Video
Updated: May 7, 2026
12:44
Watershed Planning within a Quantitative Scenario Analysis Framework
Published on: July 24, 2016
7.9K
AQuA-P: A machine learning-based tool for water quality assessment
L Díaz-González1, R A Aguilar-Rodríguez2, J C Pérez-Sansalvador3
1Centro de Investigación en Ciencias, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, Mexico..
Journal of Contaminant Hydrology
|January 9, 2025
Summary
Machine learning models, including decision trees and XGBoost, effectively assess groundwater and surface water quality. The developed AQuA-P software offers a novel, probability-based water classification system for informed management.
Area of Science:
- Environmental Science
- Data Science
- Water Resource Management
Background:
- Assessing water quality is crucial for societal needs.
- Existing water quality classification systems are limited.
- Machine learning offers advanced analytical capabilities.
Purpose of the Study:
- To apply machine learning models for evaluating water quality.
- To develop a novel water quality classification system.
- To integrate models into user-friendly software.
Main Methods:
- Utilized a national Mexican database encompassing groundwater and surface water quality parameters.
- Employed five machine learning techniques: decision trees, extreme gradient boosting (XGB), support vector machines, K-nearest neighbors, and multinomial logistic regression.
- Evaluated model performance using accuracy, precision, and F1 score.
Main Results:
- Decision tree models demonstrated the highest effectiveness across all water body types.
- XGBoost models showed strong performance, closely following decision trees.
- The AQuA-P software, incorporating decision tree models, was developed as a unique water classification tool.
Conclusions:
- Decision tree and XGBoost models are highly effective for water quality assessment.
- The AQuA-P software provides a valuable, probability-based system for water quality management.
- Open-access data and code facilitate global adoption and adaptation of the methodology.
