You are here

Study helps beat biases in monsoon prediction

“When will the monsoon arrive?” is a million dollar question in India as a sizeable population, practising agriculture, depends on the rains for its livelihood. In spite of the recent advances in monsoon prediction, accurately forecasting the monsoon rains and their intensity is still a challenge considering the complexity of the weather system. In a recent study, researchers at the University of Aizu, Japan, and Indian Institute of Tropical Meteorology, Pune, have proposed some improvements to existing monsoon-predicting models that could help precise prediction. An accurate monsoon forecast can help farmers plan agricultural activities and have early warning systems in place to prevent the loss of life and property by floods.

Studies have suggested that Coupled General Circulation Models (CGCM)—mathematical models that take into account the circulation of currents in atmospheric and oceanic systems—could help best predict the Indian Summer Monsoon (ISM). However, since the ISM is dependent on a vast number of regional and global factors, these models can be severely biased and consistently predict a higher or lower rainfall over a region. In this study, published in the journal Scientific Reports, the researchers have addressed one such bias with an improved model.

The researchers studied the bias in the rainfall predictions made by the National Centre for Environmental Prediction’s (NCEP) Climate Forecast System version-2 CGCM. This climate forecast system has been operational in India since 2011. In many climatic models over the years, there has been an underestimation of rainfall, also called a dry bias, over parts of central India, the northern Bay of Bengal and the Western Ghats, which hinders accurate predictions. This dry bias over central India results in the underestimation of rainfall by about 2–8 mm per day.

Through observations and models that focus on smaller regions, the researchers have analysed the effects of the warm coastal sea surface temperatures in the Bay of Bengal on this bias. In one of their experiments, the system in the ocean for June to September was studied using Mixed Layer Depth (MLD). Mixed layers in oceanic systems, are caused due to turbulence and lead to the mixing of water at different depths in the ocean. This mixing results in homogenisation of factors like salinity and temperature at different depths in the ocean, which otherwise would be different. It is essential to study these phenomena as they play a significant role in the climate.

The realistic representation of mixed layer structures, present in the study, has produced a clearer picture of the sea surface temperatures in the Bay of Bengal over a large area. By including this understanding in the climate prediction models, the researchers showed that the model captures the Indian Summer Monsoon rainfall well, particularly in central India, addressing the existing bias. The study provides a fresh perspective to improve the prediction system of the Indian Summer Monsoon.