Quantitative Winter Ozone Forecasts
We ware developing the second generation of our wintertime ozone forecasting model, Clyfar. Current efforts focus on evaluating where and when the model did well, or where it was unhelpful.
John Lawson
Project End: Ongoing
Funding: Utah Legislature, SSD1
Project Updates
Updated: July 2026- Recent Developments:
- We recently put out a preprint on ow to mathematically evaluate Clyfar forecasts. This is important, as low-risk event have a danger of "The Boy Who Cried Wolf", and we must calibrate our forecasts to avoid this (but missing events!)
- A paper is in preparation that uses this method to learn from mistakes of 2025/2026 in developing a better model for 2026/2027.
- We recently put out a preprint on ow to mathematically evaluate Clyfar forecasts. This is important, as low-risk event have a danger of "The Boy Who Cried Wolf", and we must calibrate our forecasts to avoid this (but missing events!)
- Current and Upcoming Work:
- Development of the automated AI forecaster "Ffion".
- Adding this with improved air-quality live-data visualisation on www.basinwx.com
- Development of the automated AI forecaster "Ffion".
- Problems:
- No problems to report at this time.
More Information
Example Clyfar model output showing Optimistic, Neutral, and Pessimistic represent the low, medium, and high range of forecast ozone values, respectively.
Improve Quantitative Winter Ozone Forecasts
Winter inversions are difficult to predict due to their inherent uncertainty, such as their sensitivity to snow depth, wind speeds, and temperature profiles. Small errors in the initial state of the atmosphere will become large errors in the coming forecast. Also, issuing solely “yes” or “no” forecasts for ozone events masks the true uncertainty of the prediction and can be misleadingly confident. In our team’s past attempts to forecast winter ozone episodes, our prediction methods compared well against historical datasets but performed poorly when we used them for real-time prediction (Lyman et al., 2020; Mansfield, 2018). Because of these failures, we have thus far relied on qualitative forecasts rather than numerical prediction methods for our Ozone Alert program.
To give the forecasting team more options to guide issued Alerts, we have constructed a fuzzy logic-based ozone prediction system named Clyfar that is thus performing better than past efforts (Lawson and Lyman, 2024), in part due to the addition of risk communication. This system can predict high ozone events, including providing likelihoods of different ozone concentration thresholds, so users have more detailed information with which to make decisions. We have also created a website to display products from Clyfar: https://basinwx.com.
Over the coming year, we will continue improving forecasts for winter ozone and promote the basinwx.com website to the Uinta Basin community, current subscribers to our Ozone Alert program, and others. We will also use machine learning (a sort of artificial intelligence) to tune Clyfar for improved accuracy. Results will be published in leading, peer-reviewed scientific journals; simpler explanations will be written in blogs on the team’s side website (www.jrl.ac) and in social media short videos, with assistance from on-campus branding and outreach staff. We will investigate meteorological tipping points and other factors that hinder accurate winter ozone forecasting,. We will seek feedback from from potential users, and from those using products in real-time, to ensure that the public-facing web interface is useful to them.