Judith Curry, President Climate Forecast Applications Network (CFAN) is driven by the mission to help energy and insurance sectors mitigate weather and climate-related risks.
“With our competitive edge associated with advanced forecast calibration, ensemble interpretation techniques, forecast uncertainty characterization, and interactive web-based dashboards, we help clients interpret forecast information easily and quickly,” says Judith Curry, the president of CFAN.
A growing opportunity is in the humanitarian sector, as underdeveloped countries don’t have access to the latest weather forecast technologies. CFAN has actively been working with NGOs and Aid agencies to provide targeted weather forecasts in South Asia. Such applications of weather forecasts are important tools for operational adaptation to a changing climate.
Driving CFAN’s quest in Asia is co-founder, Peter Webster’s expertise on the Asian monsoon, which helped forecast monsoonal floods in Bangladesh.
Today, CFAN’s clients in both the public and private sectors want more accurate forecasts at longer time horizons. They are particularly interested in high-impact events such as heat and cold waves, hurricanes, severe convective weather and fire weather.
CFAN has addressed these challenges through calibrations that increase the accuracy of the global model forecasts and ensemble interpretation to increase the time horizon for skillful forecasts.
The company’s calibration methods for ensemble weather forecasts use recent and historical forecast errors along with machine learning techniques to reduce model forecast errors. Ensemble interpretation includes Monte Carlo re-sampling techniques and ensemble clustering.
CFAN offers a comprehensive suite of weather forecast products for the energy sector under its OmniCast brand that includes forecasts of temperature, solar, and wind power, as well as severe convective weather. Using synthetic tracks based on ECMWF and NOAA ensemble forecasts, the company’s tropical cyclone forecasts under the TropiCast brand offer 15-day forecasts of formation, track, and intensity probabilities. For each active system or named storm, TropiCast innovations include a dynamic cone of track forecast uncertainty, CFAN’s rapid intensification index, forecasts of integrated kinetic energy and cyclone damage potential, and calibrated highresolution forecast fields of landfall winds.
-
With our competitive edge associated with advanced forecast calibration, ensemble interpretation techniques, forecast uncertainty characterization, and interactive web-based dashboards, we can help clients interpret forecast information easily and quickly
Under the AgriCast brand, CFAN provides regional forecasts that support precision agriculture. The company’s innovations in weather forecasting based on machine learning and AI exploit ensemble forecast methods to provide more accurate probabilistic forecasts out to four weeks.
A focus of CFAN’s forecasts is extreme weather events, including probabilities of heat and cold waves and wind droughts. Visual analytics and cognitive computing techniques help clients easily and quickly interpret forecast information on web-based dashboards to make optimal and timely decisions.
CFAN continuously works with clients to develop new products and address their most challenging problems. The company has recently launched a new weather forecast platform to support smallholder farmers in India and Pakistan in partnership with Precision Development, a global nonprofit organization that builds low-cost information systems to share knowledge with the world’s poorest and most disadvantaged people.
