As athletes worldwide prepare for the Paris 2024 Olympics, the best of the best in weather research are also gearing up for a friendly competition. Several international weather research and forecasting organizations will collaborate to provide high-resolution forecast information for Olympics organizers as well as test the capabilities of experimental air quality and weather forecasting models. 

The collective forecasts from the participating nations may guide the organizers’ decisions about when actions should be taken to protect athletes from conditions like extreme heat and/or air pollution. A substantial number of weather data-collecting instruments will be deployed across Paris during the Olympics and Paralympics for these forecasts. The extensive data collection also provides researchers with a unique opportunity to test the accuracy of models being developed to predict heat, thunderstorms, and air quality in complex urban areas.

Researchers from the U.S. National Science Foundation National Center for Atmospheric Research (NSF NCAR) will participate in this research demonstration project, which was proposed by Météo-France, France’s national weather service, and approved by the World Meteorological Organization (WMO). 

“All the different organizations will compare notes and learn from one another and maybe it will be a little competitive too,” said Scott Swerdlin, an NSF NCAR program director who is overseeing the organization’s role in the project. “People feel passionate about their models and there are fierce debates about whether different model attributes are superior to others, but in the end this will be a learning experience for everyone.”

Complexities of urban forecasting

Modeling and predicting weather in urban areas is complicated. Buildings interact with the natural environment to create microclimates. For example, the way the sun hits a building can make it hotter on one side of a building, which in turn affects the flow of winds and air quality. Capturing the effects of all the structures in a heavily populated city is challenging. 

To address this, NSF NCAR scientists will use an ensemble modeling approach, which averages multiple forecasts to provide predictions that are statistically more likely to represent the real world. The ensemble is computationally intensive, requiring the use of NSF NCAR’s supercomputer, Derecho. The scientists will utilize the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) for their weather and air quality predictions. 

“We’re excited to be contributing to this international collaboration, and we’re definitely pushing ourselves beyond what is required for the model intercomparison exercise,” said NSF NCAR scientist Hailey Shin. “We are really curious about answering our own research questions on the prediction accuracy and uncertainty of our high-resolution coupled weather-air quality models as we test the sensitivity of our models to our algorithms that represent the real effects of urban buildings, air pollutant emissions, etc.” 

For the project, each international organization will provide 36-hour weather forecasts modeled at a prescribed resolution of 100 meters so there is a common base for comparison of results. This means that, for every 100 meters, the participating agencies will provide a prediction for temperature, humidity, winds, and pressure.

For air quality forecasts, most of the participating groups will model at 3-kilometer resolution, but NSF NCAR scientists will zoom in to provide more detailed information. They will model primary fine particulate matter (PM2.5), which are tiny particles found in smoke and haze that can wreak havoc on the lungs and respiratory system, and carbon monoxide at 111-meter resolution, which is a very fine scale for running WRF. 

“Air quality is an issue for elite athletes, especially since they are breathing the air at a faster rate when they compete. Hot and polluted air can lead to dehydration and there could also be combined effects of heat and humidity,” said NSF NCAR scientist Rajesh Kumar. “Beyond the Olympics, nearly 80 percent of the world's population will be living in urban areas by the 2050s. It is very important for us to have modeling capabilities that work well in these areas so that they can play a part in ensuring urban sustainability.” 

See all News