Why in News?

The India Meteorological Department (IMD) has predicted above-normal rainfall of 105% of the long-period average (LPA) for the June-September southwest monsoon season.

Key drivers like the El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) are favorable for the monsoon.

 The April 15 forecast is the first of the year, with an updated forecast expected in the last week of May. Long-range forecasts can extend from 30 days to two years.

What’s in Today’s Article?

• Early Efforts in Monsoon Forecasting
• Post-Independence Challenges and the IMD’s Early Forecasting Models
• Recent Improvements in Monsoon Forecasting

Early Efforts in Monsoon Forecasting

• A systematic effort to forecast monsoon rainfall began in 1877, following the establishment of the India Meteorological Department (IMD) in 1875.
• The impetus for this was the Great Famine of 1876-78, which highlighted the critical need to understand monsoon patterns for agriculture, revenue, and public health.
• British colonial interests, including agricultural production and shipping, relied heavily on the monsoon.
• Blanford's Contribution (1882-1885)
  • The first tentative forecasts were made by Henry Francis Blanford, who analyzed the relationship between Himalayan snow cover and monsoon rainfall.
  • Blanford’s theory suggested that the extent and thickness of snow in the Himalayas influenced rainfall patterns over India, particularly in northwest regions.
• Eliot's Advances (1889)
  • Sir John Eliot succeeded Blanford as the first Director General of Indian Observatories in 1889.
  • Eliot expanded on Blanford's work by incorporating data on Himalayan snow, local weather conditions, and factors from the Indian Ocean and Australia.
  • Despite these advancements, Eliot’s forecasts were still unable to predict droughts or famines, such as the devastating Indian Famine of 1899-1900.
• Sir Gilbert Walker and Global Influences (1904)
  • In 1904, Sir Gilbert Walker succeeded Eliot and made significant advancements by incorporating global atmospheric, land, and ocean parameters.
  • Walker identified 28 predictors with stable historical correlations to the Indian monsoon and identified the Southern Oscillation (SO) as a key global pressure pattern influencing India's climate.
  • SO was later linked to El Niño, which was identified by Jacob Bjerknes in the 1960s.
  • Walker also divided India into three subregions—Peninsula, Northeast, and Northwest—for more accurate forecasts.

Post-Independence Challenges and the IMD’s Early Forecasting Models

• After India’s independence, the IMD continued using Walker’s monsoon forecasting model until 1987, but the forecasts were not very accurate.
• From 1932 to 1987, the average error in predictions was significant, with errors of 12.33 cm for the peninsula and 9.9 cm for Northwest India.
• The primary issue was that many of Walker’s parameters had lost their relevance over time, leading to poor accuracy despite attempts to improve the model.
• Introduction of the Gowariker Model (1988)
  • In 1988, the IMD adopted a new model based on power regression, developed by Vasant R Gowariker and his team.
  • This model used 16 atmospheric variables as predictors in statistical relationships with total rainfall.
  • Forecasts for the entire country replaced regional forecasts, though regional predictions were reintroduced in 1999 with modified geographical boundaries.
  • However, the new model still faced issues, and by 2000, four of the 16 parameters had lost their correlation with the monsoon, requiring adjustments.
• Failures and Re-evaluation (2000s)
  • The Gowariker model faced significant challenges, including its failure to predict the drought of 2002, which followed 14 years of good monsoons.
  • This failure led to a re-evaluation of the model.
  • In 2003, the IMD introduced two new models based on 8 and 10 parameters, along with a two-stage forecast strategy.
  • While the 2003 forecast was accurate, the models again failed to predict the 2004 drought, prompting further refinement.
• Development of the Statistical Forecasting System (2007)
  • In 2007, the IMD introduced a Statistical Ensemble Forecasting System (SEFS) to support its two-stage forecasting strategy.
  • This new system reduced the number of parameters in the models, replacing the eight-parameter model with a five-parameter model for the first forecast and the ten-parameter model with a six-parameter model for the update.
  • The aim was to avoid "overfitting," ensuring the models could accurately predict new data.
  • The IMD also implemented ensemble forecasting, which combined all possible forecasting models based on different predictor combinations to generate a more robust prediction.
  • This new approach significantly improved the accuracy of monsoon forecasts, with the average error decreasing from 7.94% of the long-period average (LPA) between 1995 and 2006 to 5.95% of LPA between 2007 and 2018.

Recent Improvements in Monsoon Forecasting

• Launch of the Monsoon Mission Coupled Forecasting System (MMCFS) - 2012
  • The introduction of the MMCFS in 2012 marked a significant advancement in monsoon prediction.
  • This coupled dynamic model combined data from the ocean, atmosphere, and land to provide more accurate forecasts.
  • The IMD used MMCFS alongside the SEFS for improved predictions.
• Multi-Model Ensemble (MME) Approach - 2021
  • In 2021, the IMD further enhanced its forecasting accuracy with the introduction of an MME system.
  • This approach incorporated coupled global climate models (CGCMs) from various global climate prediction and research centers, including India’s own MMCFS.
  • The MME system has significantly improved the accuracy of monsoon predictions.
• Notable Improvements in Forecast Accuracy
  • Since the introduction of SEFS in 2007 and MME in 2021, the IMD's operational forecasts have shown marked improvement.
  • The absolute forecast error in India's seasonal rainfall has decreased by about 21% between 2007 and 2024 compared to 1989-2006.
  • IMD's April forecasts have also become more precise, with deviations of only 2.27 percentage points in the actual rainfall from 2021-2024, well within the forecast range of 4%.
• Scope for Further Improvement
  • Despite these advancements, there is still room for further refinement.
  • Experts have suggested that the IMD should improve its dynamical models by addressing systematic errors and biases, as well as enhancing teleconnectivity with global climate modes such as the ENSO.
  • This could further enhance the precision of the IMD's monsoon forecasts.