Machine Learning may help mitigate consequences of erratic monsoon

According to the scientists, longer seasonal forecasts based on machine learning may be a way to mitigate the consequences of an erratic monsoon system under future global warming. The advanced predictions might help farmers and others depending on the Indian monsoon to plan ahead and mitigate the consequences of erratic monsoon.

As global warming is changing the monsoon system, climate scientists from the Potsdam Institute of Climate Impact Research (PIK Potsdam), Germany have developed a new machine learning model to provide an improved, highly precise three-month preseasonal forecast of the Indian summer monsoon.

According to the scientists, longer seasonal forecasts based on machine learning may be a way to mitigate the consequences of an erratic monsoon system under future global warming. The advanced predictions might help farmers and others depending on the Indian monsoon to plan ahead and mitigate the consequences of erratic monsoon.

The results of the study were published by Dr Takahito Mitsui and Dr Niklas Boers of PIK Potsdam in the scientific journal Environmental Research Letters. The work is part of the European TiPES project, Coordinated from The Niels Bohr Institute at the University of Copenhagen, Denmark and PIK Potsdam.

We can confirm with a level of optimism that it should be possible to predict the onsets of future monsoons even as global climate change accelerates in the coming decades. Indeed, our prediction method works well for the last 40 years, during which gradual global warming already has taken place.

Takahito Mitsui

The predictions utilized data since 1948, hence they cover the climate changes of the latest decennia. The scientists compared reconstructed data of tropospheric temperatures over the Indian Ocean and the Indian subcontinent and used a shift in temperature balance between two areas to predict the monsoon onset, achieving an accuracy of +/- 4.8 days as compared to earlier attempts using traditional weather prediction models.

"Our study reveals the large potential of machine learning methods in forecasting climate phenomena such as the monsoon onset. Ultimately, our goal is to combine traditional weather prediction models with machine learning models such as the one proposed here, which will hopefully lead to even more skilful forecasts," says Niklas Boers.

The result provides a promising basis for further research toward predicting the onset of the Indian summer monsoon in the coming decades. However, the possibility of accurate predictions in a world with a much higher global warming is yet to be investigated.

(With inputs from EurekAlert)