Sign in Subscribe
Rain

AI-Powered Tropical Cyclone Rainfall Forecasting in the Philippines

Strategic Insights

— 1 min read

Climate experts at the University of the Philippines (UP) have developed an artificial intelligence model that uses past tropical cyclone tracks to predict rainfall. The model, developed by Cris Gino Mesias and Gerry Bagtasa, can spot patterns more efficiently than traditional dynamic models, which are computationally resource-intensive. The AI model's predictive skill is comparable to dynamic models, with better skills for extreme rainfall from tropical cyclones. Parameters influencing the AI model's forecast include cyclones' distance and duration, which determine the affected areas and rainfall amounts.

Key Takeaways:

  • The AI model developed by Cris Gino Mesias and Gerry Bagtasa uses past tropical cyclone tracks to predict rainfall, showing comparable predictive skills to dynamic models.
  • The AI model's forecast is influenced by cyclones' distance and duration, determining the affected areas and rainfall amounts.
  • The model can run within minutes on a laptop, making it more efficient than traditional dynamic models.
  • Fresh data can be uploaded to the AI model, allowing it to relearn and improve its accuracy.
  • The study, titled 'AI-Based Tropical Cyclone Rainfall Forecasting in the Philippines Using Machine Learning,' was supported by the Department of Science and Technology-Accelerated Science and Technology Human Resource Development Program and DOST-Philippine Council for Industry, Energy and Emerging Technology Research and Development.
  • The AI model is not perfect but can add to the suite of rainfall forecast models available to equip disaster managers with more information on impending hazards.

Statistics:

  • The AI model takes only minutes to run on a laptop, compared to hours or days required by traditional dynamic models.
  • The model has better skills for extreme rainfall from tropical cyclones, with a predictive skill comparable to dynamic models.
  • Cyclones' distance and duration are the primary parameters influencing the AI model's forecast, determining 70% of the predicted rainfall.
  • The Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) predicts a low-pressure area east of Luzon could develop into a typhoon in the next two days.

Sources:

  • Mesias, C. G., & Bagtasa, G. (2023). AI-Based Tropical Cyclone Rainfall Forecasting in the Philippines Using Machine Learning. Meteorological Applications.
  • Department of Science and Technology-Accelerated Science and Technology Human Resource Development Program
  • DOST-Philippine Council for Industry, Energy and Emerging Technology Research and Development
  • Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA)

Read more