The Role of AI in Climate, Meteorology and Air Quality

The application of artificial intelligence in meteorology, climate and air quality started not long ago. Initially, machine-learning techniques sparked interest in the Earth observation satellite community due to the huge number of images available to recognize relevant features of the Earth system. Soon after, applications began to emerge in the reformulation of physical and chemical processes in weather and climate models (for example, to reduce the cost of atmospheric radiation calculations), in the processing of weather forecasts and air quality, and in the prediction of extreme climatic events such as droughts and heat waves.

Recently, weather prediction systems based solely on deep learning have been developed with great success, exploiting the large number of observations of the atmosphere and the Earth’s surface that are available. Added to this success is the emergence of automatic interpretation systems for both weather forecasts and air quality, and climate projections based on language models. An important aspect is that various of these developments are in the hands of private bodies, though they use publicly funded data sources such as meteorological observations. This situation has led to significant concern on the part of the national and international agencies responsible for providing society with environmental information and alerts.

 

Beta Sessions

This session forms part of a cycle of six informative talks open to the public, in which research teams from the Barcelona Supercomputing Center (BSC) explain the state of their investigation and projects, which also form part of the exhibition AI: Artificial Intelligence. The aim is to learn about the present and future of AI and supercomputing with the help of the scientific community.