Future Estimation of Hydrometeorological Variables Using Machine Learning Techniques: A Comparative Approach
Authors:
- Jean Firmino CardosoJean Firmino Cardoso
- Erickson Johny Galindo da SilvaESErickson Johny Galindo da Silva
- Ialy Rayane de Aguiar CostaICIaly Rayane de Aguiar Costa
- Andreia Azevedo Abrantes de OliveiraAOAndreia Azevedo Abrantes de Oliveira
- Artur Paiva CoutinhoACArtur Paiva Coutinho
- Saulo de Tarso Marques BezerraSBSaulo de Tarso Marques Bezerra
Open Access
- 25 March 2024
- journal article
- Published by RGSA- Revista de Gestao Social e Ambiental in Revista de Gestão Social e Ambiental
- Vol. 18 (6) , e08267
Abstract
Objective: The objective of the research was to analyze and compare different machine learning models to identify which technique presents the best performance in predicting hydrometeorological variables. Theoretical Framework: This section presents the main concepts that underpin the work. Machine learning techniques such as support vector machines, decision trees, random forests, artificial neural networks, and gradient boosting are presented, providing a solid foundation for understanding the context of the investigation. Method: The study uses a comparative methodology by applying machine learning techniques to predict hydrometeorological variables based on data collected in Petrolina-PE. Various machine learning techniques were employed and compared. Data normalization was performed through logarithms, and the treatment included filling or excluding inconsistent records. The effectiveness of the models is evaluated using metrics such as the Nash-Sutcliffe efficiency coefficient, Willmott index, and Pearson correlation coefficient. Results and Discussion: The obtained results showed good predictability, ranging from 50 to 70% efficiency. The comparative analysis of the results allowed identifying patterns and relationships between variables and initial configurations of the algorithms, contributing to a better understanding of hydrometeorological processes and their predictability. Research Implications: By providing more accurate and reliable forecasts, the models presented can assist managers in making decisions about the sustainable use of water and the mitigation of natural disasters such as floods. Originality/Value: This study contributes to the literature by advancing the estimation of hydrometeorological variables, improving existing techniques, and providing more accurate data for water resource management. Its impact extends from mitigating risks associated with extreme hydrological events to promoting efficiency in the use of water resources, contributing to the sustainability and resilience of aquatic ecosystems, essential in the face of climate change and environmental challenges.Keywords
This publication has 20 references indexed in Scilit:
- Exploring the feasibility of Support Vector Machine for short-term hydrological forecasting in South Tyrol: challenges and prospectsDiscover Applied Sciences, 2024
- Machine-learning-based short-term forecasting of daily precipitation in different climate regions across the contiguous United StatesExpert Systems with Applications, 2023
- A review of deep learning and machine learning techniques for hydrological inflow forecastingEnvironment, Development and Sustainability, 2023
- Rapid Prediction Model for Urban Floods Based on a Light Gradient Boosting Machine Approach and Hydrological–Hydraulic ModelInternational Journal of Disaster Risk Science, 2023
- Estimating wheat yields in Australia using climate records, satellite image time series and machine learning methodsISPRS Journal of Photogrammetry and Remote Sensing, 2020
- Modelling of the mechanical properties of concrete with cement ratio partially replaced by aluminium waste and sawdust ash using artificial neural networkSN Applied Sciences, 2019
- Combination of Multiple Data-Driven Models for Long-Term Monthly Runoff Predictions Based on Bayesian Model AveragingWater Resources Management, 2019
- Building Resource Auto-scaler with Functional-Link Neural Network and Adaptive Bacterial Foraging Optimization2019
- Multi-objective groundwater management strategy under uncertainties for sustainable control of saltwater intrusion: Solution for an island country in the South PacificJournal of Environmental Management, 2019
- Pearson Correlation Coefficient2009