Climate
On the sidelines of the Conference of the Parties to the United Nations Framework Convention on Climate Change (UNFCCC) or more commonly known as "COP". COP28 was held from November 30, 2023 to December 12, 2023 inclusive, hosted earlier this month by the United Arab Emirates.
The editors of Muse™ would like to complement COP28 with a look at how the development of prediction models based on artificial intelligence (AI), more specifically machine learning (ML) and deep learning (DL), can help to improve weather predictions and to combat the consequences of climate change disruption.
AI may not be the magical wand capable of combating climate change. It can, however, improve weather forecasting results by revolutionizing current calculation methods.
There are several Numerical Weather Prediction (NWP) models, such as :
The European global forecasting system, the Integrated Forecasting System (IFS)
The American Global Forecast System (GFS);
Canada's Global Environmental Multiscale Model (GEM);
British Meteorological Office (Met Office) unified model.
The version of the IFS run at the European Centre for Medium-Range Weather Forecasts (ECMWF) is often referred to as the "ECMWF" or "European model" in North America. Which distinguishes it from the American Global Forecast System (GFS) model.
In addition to its role in weather forecasting, AI is also used in the prevention and management of meteorological threats, by calculating the risk of damage caused by meteorological phenomena (snow, storms, rain, etc.) or their consequences (flooding, icing, etc.).
All this is not without reminding us that the use of AI also has its drawbacks. But that's another story. Feel free to comment if you'd like us to tackle the subject of AI's environmental footprint in a future issue.