NOAA’s GOES-West satellite captured this image of the July 24, 2024 hot spot and smoke from Park Fire in northern California. The Park Fire was the largest single wildfire of California’s 2024 season and the fourth-largest in the state’s history. It destroyed over 700 structures, burned nearly 430,000 acres, and spewed smoke and hazardous particulates more than five miles into the atmosphere. Credit: NOAA Satellites
August 29, 2025 - Wildfire was once a seasonal concern primarily for the western United States and Great Plains. It has now become a year-round threat in many parts of the country, devastating communities, impacting air quality, and reshaping landscapes. In the face of the escalating risk, scientists at NOAA’s Global Systems Laboratory (GSL) in Boulder, Colorado continue to develop sophisticated tools to give forecasters, land managers, emergency response officials, and firefighters improved situational awareness of rapidly developing wildfire hazards.
One of their latest innovations is the Hourly Wildfire Potential Index (DESI ), an experimental application of NOAA’s revolutionary High-Resolution Rapid Refresh forecast model (HRRR) and the next-generation Rapid Refresh Forecast System (RRFS), developed by GSL and the National Weather Service’s Environmental Modeling Center. HWP provides an assessment of wildfire potential updated every hour based on model-predicted weather conditions. The frequent updates allow HWP to forecast sudden increases or decreases in potential fire activity due to changing weather conditions, providing more accurate and timely predictions of wildfire activity including the amount of emitted wildfire smoke.
The HWP index was created by a team of scientists from GSL and the University of Colorado cooperative institute CIRES, based on three years of measurements of radiant heat captured by satellites flying over large wildfires in the western United States. HWP incorporates predicted winds to estimate fire spread and intensity, humidity to estimate influence of atmospheric dryness on fuel moisture, and soil moisture derived from the HRRR’s advanced land surface model to predict the response of flammable vegetation to precipitation and drought.
“There are lots of existing fire weather indices, but the novel thing here is to be able to predict hourly variability in fire activity related to the weather conditions,” said Eric James, a GSL research physical scientist who played a key role in development of the HRRR. “Applying the HWP to a weather model like the HRRR provides an inexpensive way to anticipate changes in fire activity without running a computationally expensive fire behavior model.”
Credit: NOAA Global Systems Laboratory
In development since 2019, HWP is expected to be a component of the RRFS, NOAA’s next generation high-resolution weather model, which is being evaluated for use by the National Weather Service. A new paper published in the journal Weather and Forecasting describes how the index was developed, reports on test cases where it was verified, and identifies potential applications for the wildfire response community.
“We expect that this index can help forecasters anticipate sudden changes in wildfire activity, as well as predict wildfire smoke emissions, leading to improved decision support for fire management and enhanced communication with communities impacted by fire and smoke,” said James.
What sets HWP apart from other established fire indices is that it’s the first index designed specifically for use with hourly, high-resolution, storm-scale weather models. The frequent updates means the index can predict changes in wildfire activity from sunrise to sunset and through the night, as well as account for the influence of weather events such as cold fronts, windstorms, and rainfall.
Another application of the groundbreaking High Resolution Rapid Refresh model
The HRRR was initially developed to improve aviation forecasts of low-level winds and turbulence, along with hazards like icing and fog. The HRRR’s groundbreaking three-kilometer horizontal grid spacing offered dramatic improvements in the ability to simulate clouds and convective features like thunderstorms and outflow boundaries, which are too small for global or regional models that operate on 13-kilometer grids or larger, but critically important for routing air traffic safely through the sky.
HRRR has since proved to be a remarkably flexible platform capable of accurately forecasting a wide variety of weather conditions. It provides accurate and timely predictions of total precipitation from weather systems, including from atmospheric rivers that are crucial to California’s water supply, and supports water resource management decisions.
Improved predictions of lake-effect snows, which are notoriously challenging to predict, were made possible by incorporating data from the Great Lakes Operational Forecast System, developed by NOAA’s Great Lakes Environmental Research Laboratory and National Ocean Service. HRRR’s high resolution and frequent updates are especially valuable for alerting maritime users to rapidly deteriorating weather conditions, aiding in navigation safety, route optimization, and emergency preparedness.
One of the most recognizable HRRR applications is HRRR-smoke, which has become a standard tool for meteorologists seeking to produce visualizations of large wildfire smoke plumes moving from west to east across the country and south from Canada.
Who will the Hourly Wildfire Potential Index help?
Real-time HWP forecasts are now being generated from operational and experimental storm-scale models covering the contiguous U.S. and Alaska, and are available on the GSL DESI website.
Besides helping land managers and firefighters anticipate likely changes in ongoing fire activity, the HWP has several other useful applications.
It is being used in the experimental RRFS to improve predictions of the amount of smoke emitted from wildfires in the coming hours and days, as well as the likely height of the associated smoke plumes. Plume height is an important factor in predicting the downstream spread of wildfire smoke, with long-range transport generally occurring only with very deep plumes. The index can also assist with planning of prescribed burns.
The HWP Index can also help distinguish between flaming versus smoldering combustion phases in wildfires, which has important implications for the chemistry of smoke plumes including the production of ozone. Ozone is harmful for human health and has been tied to wildfire smoke plumes.
Source: NOAA
