History and Evolution of Monsoon Forecasting in India

General Studies Paper-1

Context: The India Meteorological Department (IMD) predicted  ‘above normal’ rainfall (105% of the long-period average) during the 2025 southwest monsoon season (June-September).

• The monsoon is crucial for agriculture, economy, and water resources, providing around 70% of India’s annual rainfall.
• Since 2007, the accuracy of monsoon forecasts has improved significantly, with the absolute error in rainfall reducing by 21% from 1989-2006 to 2007-2024.

History of Monsoon Forecasting

• The IMD began forecasting the monsoon in 1877, driven by the need to understand rainfall patterns after the devastating 1876-78 Great Famine.
• Henry Francis Blanford, in the late 1800s, studied the relationship between Himalayan snow cover and monsoon rainfall.
  • He made the first long-range forecast in 1886.
• Sir John Eliot took Blanford’s work forward by incorporating local weather conditions and data from the Indian Ocean and Australia, although his predictions were still limited in accuracy.
• Sir Gilbert Walker in 1904, introduced statistical models using 28 parameters, identifying the Southern Oscillation (SO) as a key influence on the Indian monsoon.
  • He divided India into three subregions for forecasting.

Scenario After Independence

• IMD continued using Walker’s model until 1987, but it became less effective due to changes in climate patterns and loss of correlation with key parameters.
• In 1988, The IMD shifted to a new regression model (Gowariker Model) using 16 variables, but issues persisted with the accuracy of regional forecasts.

New Models and Strategies

• In 2003, IMD introduced two new models based on 8 and 10 parameters.
  • The two-stage forecast strategy was also implemented, although it had mixed results.
• In 2007, IMD developed a Statistical Ensemble Forecasting System, reducing the number of parameters to improve accuracy and introduced ensemble forecasts to increase robustness.
• In 2012, the Monsoon Mission Coupled Forecasting System (MMCFS) was launched to combine ocean, atmospheric, and land data for better forecasts.
• In 2021, the Multi-Model Ensemble system further improved forecast accuracy by combining global climate models, including MMCFS.