UP researchers develop AI model to predict cyclone rainfall

University of the Philippines (UP) researchers have developed an artificial intelligence (AI) model capable of predicting rainfall from tropical cyclones, a significant advancement for disaster risk reduction.

The AI model, developed by climate experts Cris Gino Mesias and Gerry Bagtasa from UP's Institute of Environmental Science and Meteorology, utilizes machine learning techniques on historical tropical cyclone data.

Unlike difficult-to-run dynamic models, this AI model can run within minutes on a laptop, offering comparable predictive skill, especially for extreme rainfall.

Researchers calibrated the model using ground rainfall observations and validated its accuracy with 10 tropical cyclones that caused extreme rainfall between 2016 and 2020.

The study highlights that a cyclone's distance and duration are key factors influencing the AI model's rainfall predictions and the extent of areas affected.

While not perfect, the AI model can add to the suite of rainfall forecast models available to equip disaster managers with more information on impending hazards.