Improving deterministic forecasts of maximum and minimum temperature using machine learning
Open Access
- 17 June 2026
- journal article
- research article
- Published by Royal Society of Chemistry (RSC) in Environmental Science: Advances
- Vol. 5 (7) , 1804-1828
Abstract
Accurately forecasting near-surface temperature is essential for heatwave and cold-wave warnings and impact-based decision support over India. Deterministic numerical weather prediction (NWP) models show systematic, regionally varying biases that increase with lead times. To improve the reliability of these forecasts, bias correction is essential. This study applies a multivariate machine-learning (ML) bias-correction framework to location specific 2 m maximum (Tmax) and minimum (Tmin) temperature forecasts from the operational NWP model at the National Centre for Medium Range Weather Forecasting (NCMRWF). Data from 179 India Meteorological Department (IMD) stations covering the period 2019–2024 were used. Four ML methods, Random Forest (RF), eXtreme Gradient Boosting (XGB), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNNs) were used for bias correction of the forecasts at the 179 stations. The ML models were assessed using continuous metrics like mean error (ME), root mean square error (RMSE), and correlation/Taylor diagnostics. Along with these categorical skills for extremes, metrics like equitable threat score (ETS) and Heidke Skill Score (HSS) (for Tmax ≥ 30/35 °C in MAMJ (March–June) and Tmin ≤ 10/15 °C in DJF (December–February)), and Relative Economic Value (REV) were used. It is found that ML post-processing substantially reduces bias and error across stations and lead times. For Tmax, RMSE improvement increases with lead time, typically ∼10–15% at Day-1, ∼20–30% by Day-5, and frequently >30–40% (locally reaching ∼50–60%) by Day-9, especially for XGB/LSTM. For Tmin, improvements are strongest: XGB improves RMSE by ∼25–40% at Day-1, increasing to ∼40–60% by Day-7 to Day-9 across many stations. Categorical verification shows consistent improvements in terms of higher ETS/HSS values after bias correction across most stations. Winter Tmin shows large gains for both thresholds, particularly for Tmin ≤ 15 °C. REV analysis indicates that ML-corrected forecasts remain economically useful over a wider range of cost–loss ratios and retain value at longer lead times compared to the raw model. Overall, XGB provides the most consistent improvement across regions and metrics, RF is generally second-best, LSTM shows competitive performance, particularly for Tmax and at longer lead times, while CNN performs worst. SHAP-based analysis links the corrections to physically meaningful drivers, with Tmax corrections dominated by boundary-layer/land-surface predictors and Tmin corrections dominated by radiative and synoptic controls.Keywords
This publication has 53 references indexed in Scilit:
- Improving the Indian Monsoon Data Assimilation and Analysis regional reanalysis maximum and minimum temperatures over India using machine‐learning techniquesQuarterly Journal of the Royal Meteorological Society, 2025
- Systematic daytime increases in atmospheric biases linked to dry soils in irrigated areas in Indian operational forecastsAtmospheric Science Letters, 2023
- A Machine Learning (ML)-Based Approach to Improve Tropical Cyclone Intensity Prediction of NCMRWF Ensemble Prediction SystemPure and Applied Geophysics, 2022
- Increasing heat waves and warm spells in India, observed from a multiaspect frameworkJournal of Geophysical Research: Atmospheres, 2017
- ENDGame: The New Dynamical Core of the Met Office Weather and Climate Prediction Model2016
- Extreme Temperature Regions in India2016
- Evaluation of statistical bias correction methods for numerical weather prediction model forecasts of maximum and minimum temperaturesNatural Hazards, 2014
- Development of a high resolution daily gridded temperature data set (1969–2005) for the Indian regionAtmospheric Science Letters, 2009
- Understanding forecast verification statisticsMeteorological Applications, 2008
- Some evidence of climate change in twentieth-century IndiaClimatic Change, 2007