Study Says That Machine Learning Can Make Climate Prediction More Effective

There are many areas that Machine Learning has empowered and keep empowering. According to a recently published research, climate prediction can be one of them. The study says that Machine Learning is capable of offering more effective predictions, as it can surpass one of the major challenges of now: presentation of clouds, atmospheric heat and moisture. The best part about using Machine Learning is that the process can be done with maximum level of accuracy as well. The paper was published in a journal named Geophysical Research Letters and the study was done by Pierre Gentine of Columbia Engineering.

Gentine believes that Machine Learning can be a game-changer in the field of climate prediction and overall accuracy of the process. He also adds that the uncertainties posed by greenhouse gases have made weather prediction quite complex and inaccurate. This is also because the amount of these gases keeps fluctuating on a regular basis. The relation between the fluctuating amount and the representation of atmospheric cloud structure cannot be explained using the current techniques. The study says that Machine Learning can be used to do this. According to the proof of concept, the method shows a more accurate representation and a better prediction.

According to this model, the Machine Learning technology was able to create an accurate representation of clouds and how they are connected to the increasing levels of Carbon Dioxide. The research team also adds that the model has been tested in fully-fledged climate models and that the results have been as effective as the proof of concept. Being a Machine Learning interface, it’s also clear that the effectiveness will become better in the long run. In addition to this, the Cloud Brain technology can also take care of the various aspects involved in the prediction of final climate.