The Indian Summer Monsoon is the backbone of socioeconomic life in the subcontinent. It influences agriculture, water resources, energy production and the country’s economy. Excess monsoon rainfall can trigger devastating floods and damage infrastructure, while too little can lead to droughts, crop failures and water scarcity. At the same time, the design of roads, bridges, dams, coastal infrastructure and even urban drainage systems increasingly depends on how monsoon wind and rainfall are expected to change over the coming decades. After all, these structures are built to last for many decades, not just a few years. Understanding how the monsoon will evolve in a warming climate is therefore essential for planning a climate-resilient future.
To answer this question, climate scientists often rely on global climate models developed under the Coupled Model Intercomparison Project (CMIP). These models simulate the atmosphere, ocean and land together, allowing them to interact continuously. For example, stronger winds over the ocean enhance mixing and cool the sea surface. In turn, the cooler ocean modifies the winds, creating a feedback that shapes weather and climate. Such interactions are essential for realistically simulating the monsoon. Yet these models are not perfect. Errors in representing the ocean, atmosphere or land can propagate through the climate system and affect the simulated monsoon. This is particularly important for India because monsoon winds spend most of their journey over the Indian Ocean before reaching the subcontinent. Thus, if the ocean is represented incorrectly, the monsoon can be too.
Previous studies using the latest generation of climate models from the sixth phase of CMIP project (CMIP6) suggest that the Indian Summer Monsoon is likely to become wetter in the future. However, this projection appears puzzling because the observed monsoon has weakened over the past two decades despite global warming. This raises an important question: How reliable are these future projections? One way to answer this is to examine how well climate models reproduce present-day conditions over the Indian Ocean. A recent study from the Centre for Atmospheric Sciences, IIT Delhi, published in Environmental Research Letters (https://doi.org/10.1088/1748-9326/ae8460), addresses this question. The study found that most CMIP6 models simulate the northern Arabian Sea to be 2-3°C colder than the observed sea surface temperature during the pre-monsoon season. Although this may appear to be a modest difference, warming the ocean by even one degree requires an enormous amount of heat because seawater has a very high heat capacity as compared to atmosphere. Such a bias therefore represents a substantial error in the amount of heat stored in the ocean. Additionally, the pre-monsoon warming of the northern Arabian Sea plays a crucial role in the monsoon because it strengthens the land-sea temperature contrast and helps draw moisture-laden monsoon winds towards the Indian subcontinent. To understand the consequences of this cold sea surface temperature bias, the researchers used a regional coupled atmosphere-ocean numerical model, which was configured and run on the high-performance computing facilities at IIT Delhi. They first introduced the cold sea surface temperatures from CMIP6, including the cold Arabian Sea bias, into the regional model. They then repeated the experiment after correcting the temperature bias.
The results were striking. The cold Arabian Sea weakened the southwesterly monsoon winds, delaying the onset of the monsoon over Kerala by about a week and reducing rainfall over India by nearly 30 per cent, with the strongest impacts observed along the Western Ghats and parts of northeast India. Once the monsoon reaches Kerala, the moisture-laden winds from the Arabian Sea ascend into the upper atmosphere over the west coast. This upper-level moisture is subsequently transported northeastward, driving the advance of the monsoon across the Indian subcontinent. However, the cold Arabian Sea surface temperature substantially weakened this upward transport of moisture, thereby slowing the monsoon’s progression across the country (Figure 1).
These findings suggest that even though global climate models are the best tools available for projecting future climate, their simulations must be interpreted carefully. Small regional biases in the Arabian Sea can substantially alter the timing and strength of the Indian monsoon. Improving the representation of these regional ocean processes is therefore essential for increasing confidence in future monsoon projections.
Sankar Prasad Lahiri was a Prime Minister's Research Fellow at the Centre for Atmospheric Sciences, IIT Delhi, where he recently completed his Ph.D. His research focuses on Indian Ocean climate variability and tropical air-sea interactions during the Indian Summer Monsoon using regional coupled numerical models and global climate models.
Vimlesh Pant is a Professor at the Centre for Atmospheric Sciences, IIT Delhi. His research interests include physical and coastal oceanography, air-sea interactions, and polar oceanography.
Views expressed are the authors’ own and don’t necessarily reflect those of Down To Earth