A challenge has been issued to computer coders to help in the creation of open-source forecast models.
The European Centre for Medium-Range Weather Forecasts (ECMWF) issued the challenge as part of its third Summer of Weather Code, launched last week.
The contest connects individuals or teams with mentors and experts in weather, climate, machine-learning and cloud computing to develop open-source weather- and climate-related software. Applicants have until April 22 to submit proposals, with the best 15 selected for coding over the summer.
The results will be presented at the ECMWF in September. Coders’ proposals must address one of three areas.
The first is the creation of weather-related software and applications. These could include developing interactive analysis tools for visualising performance data from the Integrated Forecasting System (IFS), the ECMWF’s atmospheric model and data assimilation system, or creating a new set of interactive online materials for scientific training.
The second area is machine-learning and artificial intelligence and possible proposals could involve the use of AI to detect and track tropical cyclones.
The final area calls for proposals offering innovations in cloud-based weather and climate systems, with applicants invited to submit project ideas that use ECMWF or Copernicus data that need cloud-based resources to process the data.
“New this year are our partnership with Copernicus and Amazon Web Services and the increase in the number of projects we can select,” said Esperanza Cuartero, one of the program’s organisers. “The challenges proposed are very promising, with a strong emphasis on machine learning and artificial intelligence.”