IMD’s Monsoon Forecasts Remain Highly Accurate, Average Error Only 1.9% of LPA
Forecast deviations largely driven by ENSO, IOD, and intra-seasonal variability; IMD strengthens prediction systems using multi-model ensembles, high-resolution satellites.
New Delhi, February 12, 2026: The operational performance of India’s seasonal rainfall forecasts between 2023 and 2025 has demonstrated strong accuracy, with actual monsoon rainfall in each of the three years falling within forecast limits. The average absolute error during this period was just 1.9% of the Long Period Average (LPA), underscoring the reliability of the forecasting system. Deviations between forecasted and observed rainfall primarily stem from uncertainties linked to large-scale climate drivers and their representation in prediction models. Seasonal forecasts are particularly sensitive to the evolution of ENSO and its impact on the Indian summer monsoon. The Indian Ocean Dipole (IOD), along with its interaction with monsoon circulation, adds further complexity. Additionally, pronounced intra-seasonal variability and challenges in simulating synoptic-scale systems, such as Monsoon Low Pressure Systems, can affect early predictions, especially over central India. To enhance precision, the India Meteorological Department (IMD) continues integrating advanced coupled global and regional climate models through Multi-Model Ensemble (MME) frameworks, high-resolution satellite observations, improved data assimilation techniques, and high-performance computing. Emerging technologies, including artificial intelligence and machine learning under Mission Mausam, are also being adopted. This information was shared in the Rajya Sabha by Union Minister of State (Independent Charge) for Earth Sciences, Dr. Jitendra Singh.