The Fire Weather Alert System (FWAS) has been designed to warn on-the-ground firefighters of dangerous weather conditions in their area. Users set custom weather thresholds and the FWAS sends alerts via text message and email when those thresholds are exceeded. FWAS monitors several types of weather measurements (RAWS, radar, etc.) and forecasts to alert firefighters of dangerous weather such as high winds, low RH, thunderstorms, and Red Flag Warnings. FWAS is currently a prototype system with a web page-based user interface (which is accessible via a browser on a mobile device).
A short video describing the use of the FWAS can be found here.
TracPlus online article: FWAS: An Early Warning System For Wildland Firefighters.
Rationale
Many firefighters have been injured or killed due to rapid changes in weather causing unexpected fire behavior. Examples include the thunderstorm outflow that killed 19 on the Yarnell Hill Fire, the cold front passage that killed 14 on the South Canyon Fire, and the wind shift that killed 3 on the Twisp River Fire. Many of these weather events can easily be predicted and observed by tools that exist today (e.g., short-term high-resolution forecast models, radar data, nearby weather stations). Tragically, these tools and the data they provide are underutilized by on-the-ground firefighters, which leads to many preventable accidents. Some of the reasons for underutilization are that: 1) this information is difficult to access on mobile devices in the field, 2) the information is scattered among several web sites, 3) some of the information is difficult to interpret because it is intended for weather experts, and 4) firefighters do not have much time to devote to gathering this weather information. The Fire Weather Alert System has been developed to alleviate many of these issues. Users simply provide the system with a location and custom weather thresholds to watch for. The system continuously checks the many data sources that exist for exceedance of those thresholds. Thresholds can be set for wind speed, temperature, relative humidity, precipitation, thunderstorms, and National Weather Service issued Watches, Warnings, or Advisories. When one of these thresholds is reached, either from a nearby weather station, a short-term forecast, radar data, etc., the user will be alerted. The alert includes information about what threshold was met, where, and from what source (a RAWS station, forecast, etc.). Users can be alerted via text message or email. The alert system is designed to provide on-theground users with an easy-to-use custom weather delivery system that has the potential to increase safety and inform decision making on wildland fire incidents.
Implementation and usage details
The alert system is designed to be straightforward to use, lightweight, intuitive, and mobile friendly. The user interface of the FWAS is currently a web page where users enter information (in the future a mobile app may be built). To use the alert system, several parameters are requested from the user. First is the user’s location and area of interest. Location information can be provided in several ways, by manually inputting a latitude and longitude, using geolocation services or by selecting from a list of current large wildfires. This list is generated based on the National Interagency Fire Center (NIFC) Sit Report. An area of interest is also required, this is specified by indicating a radius around the location specified. Alerts are transient and as such, an expiration date must be set by the user, with a maximum duration of 7 days. Weather thresholds are set next. Weather elements that FWAS monitors includes: average wind speed, gust wind speed, temperature, relative humidity, precipitation, thunderstorm proximity, and NWS Watches, Warnings, and Advisories. Users can set threshold values for these elements which determine when they will get an alert. For example, a firefighter might want to know if wind over 15 mph occurs in their vicinity, or RH of less than 20%, or if a thunderstorm is approaching. The last input users must provide is contact information in the form of a cell phone number or email address to receive alerts. The alerts are sent as text messages and/or emails.
The FWAS gathers weather data from several different sources. These sources include: Remote Automated Weather Stations (RAWS) (including temporary incident RAWS), airport-based Automated Surface Observing Systems (ASOS), NEXRAD radar, the High-Resolution Rapid Refresh (HRRR) forecast model from the National Oceanic and Atmospheric Administration Earth System Research Laboratory (NOAA/ESRL), and National Weather Service issued Watches, Warnings, and Advisories.
Further Developments
There are several improvements and enhancements currently discussed for the FWAS. The first is improving the user interface. Implementing a mobile app interface would improve user experience and allow for future enhancements that are limited by the current web page interface. One enhancement would be in-app alert notifications which would give more flexibility on how notifications are emitted and interpreted compared to text messages and emails. Another enhancement would be the ability to save, share, and post groups of alert thresholds. So, for example, a firefighter could store the set of thresholds that work well on their home unit for future use. These could also be shared with other users, and even posted so that others could access them. This would be a similar idea to how the Fire Danger PocketCards are posted for different units across the country, except it would be build into the app. Addition of a map displaying weather would also be very useful, where users could further investigate alerts that are received.
Another enhancement could be the ability to combine thresholds, so for example, a firefighter could get an alert if RH was less than 20% AND wind was greater than 15 mph. We would like to add additional datasets as well, such as lightning detection. There is also the possibility to incorporate automated thunderstorm microburst and gust front detection algorithms that are currently used by the FAA for making real-time aircraft safety decisions. Last, alerts related to fire danger and fire behavior could be incorporated.
