Khan et al. (2026) One-month-ahead river discharge forecasting using machine learning and limited hydrometeorological data: the Karatoya-Atrai River basin, Bangladesh

Identification

• Journal: Frontiers in Water
• Year: 2026
• Date: 2026-09-22
• Authors: Khaled Mahamud Khan, Md. Nazrul Islam, Bo Wang
• DOI: 10.3389/frwa.2026.1917644

Research Groups

• Department of Earth, Environmental and Geospatial Science, Saint Louis University, St. Louis, MO, United States
• Department of Geography and Environment, Jahangirnagar University, Dhaka, Bangladesh
• Bangladesh Meteorological Department (BMD)
• Bangladesh Water Development Board (BWDB)

Short Summary

This study investigates the capacity of various Machine Learning (ML) models to forecast monthly discharge of the Karatoya-Atrai River 1 month ahead using only rainfall and temperature, together with antecedent (lagged) discharge. The Random Forest (RF) model offers superior discharge forecasting accuracy.

Objective

• Investigate the capacity of ML models to forecast monthly discharge of the Karatoya-Atrai River 1 month ahead.
• Compare the performance of different ML models in predicting discharge using rainfall, temperature, and antecedent discharge as predictors.

Study Configuration

• Spatial Scale: The study focuses on the Karatoya-Atrai River basin in northwestern Bangladesh.
• Temporal Scale: Monthly data from 1964 to 1994 were used for training and testing the models.

Methodology and Data

• Models used:
  • Support Vector Regression (SVR)
  • Random Forest (RF)
  • Linear Regression (LR)
  • Polynomial Regression (PR)
  • Ridge Regression (RR)
  • Extreme Gradient Boosting (XGBoost)
• Data sources:
  • Hydrological and meteorological data from the Bangladesh Meteorological Department (BMD) and the Bangladesh Water Development Board (BWDB)

Main Results

• The RF model gives the best one-month-ahead forecasts (NSE 0.86, RMSE 69.71 m^3 s^-1, MAE 43.97 m^3 s^-1, MAPE 40.26%, KGE 0.89, NRMSE_range 0.11).
• The RF model outperforms persistence and seasonal-climatology baselines.
• Antecedent-discharge memory contributes most to the forecast skill.

Contributions

• This study addresses gaps in existing research by developing an operational, one-month-ahead discharge forecasting framework at the gauging-station scale using only available variables (rainfall, temperature, and discharge).
• The study provides new insight into the runoff–discharge relationship in a monsoonal, transboundary catchment.

Funding

• This research was not funded by any specific project or program.

Citation

@article{Khan2026Onemonthahead,
  author = {Khan, Khaled Mahamud and Islam, Md. Nazrul and Wang, Bo},
  title = {One-month-ahead river discharge forecasting using machine learning and limited hydrometeorological data: the Karatoya-Atrai River basin, Bangladesh},
  journal = {Frontiers in Water},
  year = {2026},
  doi = {10.3389/frwa.2026.1917644},
  url = {https://doi.org/10.3389/frwa.2026.1917644}
}