New UH climate tool delivers custom data to support wildfire, drought planning
A new tool developed by University of Hawaiʻi researchers allows anyone in the state to generate custom, site-specific climate reports to support decisions related to drought, wildfire and land management—a major step forward in Hawaiʻi’s climate resilience efforts.
The Climate Change, Climate Variability and Drought Portfolio tool was unveiled at a May 2025 meeting of the Hawaiʻi Climate Data Portal (HCDP) User Group, which drew lawmakers, emergency managers and wildfire officials. The new system lets users select or draw an area of interest, enter basic details and receive a tailored climate portfolio by email. It is free to use and portfolios are generated and delivered in less than an hour.
“We used to generate these portfolios one at a time, but the demand for them was so great that we decided to automate the entire process,” said Ryan Longman, lead researcher on the HCDP project and the UH consortium director of the Pacific Islands Climate Adaptation Science Center.
The tool is part of the larger Change Hawaiʻi project, which is working to transform how the state tracks weather, drought and wildfire risk using advanced climate modeling, real-time data systems and artificial intelligence. Change Hawaiʻi is part of the Established Program to Stimulate Competitive Research or Hawaiʻi EPSCoR.
Real-time dashboard
At the meeting, researchers also rolled out a newly developed, open-source dashboard for the Hawaiʻi Mesonet—a growing network of 110 weather monitoring stations statewide, 66 of which are now operational. Each station collects 21 environmental variables and generates more than a million data points daily that feed into predictive models for wildfire and drought.
“This dashboard puts real-time data at the fingertips of the people who need it most,” said Tom Giambelluca, Hawaiʻi Mesonet project lead, Change Hawaiʻi co-principal investigator and former director of the Water Resources Research Center at UH. “High-quality data has never been easier to access, and future applications with the data are limitless.”
Wildfire system
The team has developed daily wildfire probability maps and forecasts statewide through the use of optimized machine learning models for more accurate fire behavior forecasting and drought assessment. The user-friendly, real-time data is accessed through the HCDP, which is publicly available. Read more on UH News.
The goal is to provide early warning to fire managers, emergency responders and landowners so they can deploy resources, issue public advisories and reduce risks through more informed planning. This wildfire system can identify critical wildfire ignition factors, including relative humidity, temperature, rainfall, normalized difference vegetation index (components of wildfire fuel) and land cover.
Future efforts will leverage robust cyberinfrastructure, advanced data visualization, and innovative AI and machine learning applications, including computer vision and edge AI systems, to create a more resilient Hawaiʻi.