Machine Learning and the GR2M Model for Monthly Runoff Forecasting - Comparative Study for Rainfall-Runoff Analysis

Aug 2, 2026

Upload by Eden

This article evaluates monthly rainfall-to-runoff analysis using machine learning algorithms—Multiple Linear Regression, Multilayer Perceptron, and Support Vector Machine—and compares them with the GR2M hydrologic model. The study targets watersheds in Thailand’s lower southern region near the Thailand–Malaysia border, where rainfall measurements exist only on the Thai side, creating uncertainty and incomplete cross-border data. Across sub-basins, Support Vector Machine delivers the highest accuracy and reliability, outperforming GR2M. Results support using SVM as an optimal framework for runoff prediction where data coverage is partial.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Rainfall-Runoff modeling challenges in border watersheds

Hydrologic models and machine learning approaches

Study purpose and application context

Machine Learning and the GR2M Model for Monthly Runoff Forecasting Natapon Kaewthong 1, Torlap Kanplumjit 1, Naras Kwanthong 2, Kritsana Sureeya 3, Chayanat Buathongkhue 4* 1 Department of Civil Engineering, Faculty of Engineering Rajamangala University of Technology Srivijaya, Songkhla 90000, Thailand. 2 Faculty of Engineering and Technology Rajamangala University of Technology Srivijaya, Trang 92150, Thailand. 3 Research assistant, Rajamangala University of Technology Srivijaya, Songkhla 90000, Thailand. 4 College of Industrial Technology and Management, Rajamangala University of Technology Srivijaya, Nakhon Si Thammarat 80210, Thailand. Received 04 August 2024; Revised 09 December 2024; Accepted 15 December 2024; Published 01 January 2025 Abstract This article presents the results of an analysis of monthly rainfall into monthly runoff using Machine Learning algorithms, including Multiple Linear Regression, Multilayer Perceptron, and Support Vector Machine, which were compared with the GR2M hydrologic model to identify the most suitable approach for rainfall-runoff analysis in watersheds in the lower southern region of Thailand. This region is characterized by its unique geographic location at the border between Thailand and Malaysia. It faces challenges due to uncertainty in rainfall data, measured only on the Thai side, leading to a lack of corresponding data from Malaysia. The analysis found that the Machine Learning Support Vector Machine algorithm consistently provided the most accurate results across all sub-basins. Sub-basin TU02 achieved an MAE of 2.63 mm/month, while sub-basin X. 119Ahad an MAE of 68.10 mm/month, sub-basin X. 184 had an MAE of 145.05 mm/month, and sub-basin X.274 had an MAE of 66.08 mm/month. This research demonstrated the utility of advanced algorithms in rainfall-runoff analysis for areas with partial or incomplete data coverage. The findings confirm that the Machine Learning Support Vector Machine algorithm outperformed the Hydrologic Model (GR2M) in terms of accuracy and reliability. Therefore, this study concludes that applying the Machine Learning Support Vector Machine algorithm is an optimal approach for runoff prediction in the southern region of Thailand and provides a framework for potential applications in other areas with similar data and geographic challenges. Keywords: Hydrologic Model; Runoff Forecasting; Machine Learning; GR2M; Thailand. 1. Introduction The hydrologic assessment of rainfall-runoff is a complex process, especially in watersheds with national boundaries, where the constraints of rainfall measurements due to such boundaries can affect the accuracy of runoff estimates. Key factors contributing to errors include the lack of comprehensive temporal data for model calibration [1, 2] and the incompleteness of constants used in analytical processes [3, 4] . Furthermore, variability in inconsistent rainfall data increases the risk of errors, especially in regions with diverse geographical features, such as surface runoff loss and streamflow changes due to land-use modifications [5], soil permeability variations [6, 7], and evapotranspiration [8, 9] . Another issue to consider is climate and land use changes, which affect the water balance in river basins, locally and globally. Such changes are likely to increase the intensity of heavy rainfall and the frequency of flood events. For * Corresponding author: [chayanat.b@rmutsv.ac.th](chayanat.b@rmutsv.ac.th) [http://dx.doi.org/10.28991/CEJ-2025-011-01-022](http://dx.doi.org/10.28991/CEJ-2025-011-01-022) © 2025 by the authors. Licensee C.E.J, Tehran, Iran. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). example, studies in the Yellow River basin [10], Kunhar [11], and Fujiang [12] found that changes in rainfall and land

FAQs

Which machine learning methods are compared with GR2M in this study?

The study compares Multiple Linear Regression, Multilayer Perceptron, and Support Vector Machine against the GR2M hydrologic model for monthly rainfall-runoff analysis.

Why is runoff analysis more difficult in the Thailand–Malaysia border region?

Rainfall data are measured only on the Thai side, creating uncertainty and incomplete datasets from the Malaysia side, which increases estimation errors.

What model performed best for monthly runoff prediction?

Support Vector Machine consistently provided the most accurate results across all sub-basins and outperformed the GR2M hydrologic model in accuracy and reliability.

Aug 2, 2026

Upload by Eden