Technology is becoming more prevalent in wildland fire management. Handheld GPS units, personal weather instruments and laptop computers, to name a few, are becoming ubiquitous. Combine these novel technologies with the continually increasing use of Geographic Information Systems in fire management strategies, and it becomes clear that the role of technology in wildland fire command is here to stay.

While our abilities to gather information are continuing to grow, our ability to transfer that information in a timely manner and integrate technologies lags. A perfect example of this shortcoming is the role of handheld GPS units. GPS allows field personnel to easily collect important information, such as fire perimeter and various waypoints like dip sites and drop points. However, the timely transfer of this information from the field to the managers at the incident command post is lacking. Typically, this sort of information isn't assimilated with other intelligence until field crews return to the command post in the evening. This also can be a problem with weather information, digital photography and other technologies.

This lack of ability to transfer data prompted our research group — the National Center for Landscape Fire Analysis, College of Forestry and Conservation, the University of Montana — to investigate means of getting information from the field to the command post and, where appropriate, from the command post to the field. Given the typical occurrence of wildland fire in hostile environments, complex terrain and areas unfriendly toward technology deployment, we set out to design a system that would facilitate information transfer in a self-contained, non-permanent and robust manner. Our design would require that a variety of criteria be met, including minimal maintenance, long-distance data transfer, ruggedness and weather-resistance, and modular design for ease of deployment and operations.

RADIO WAVES

In most wildland fire events, there's some geographic distance between the command post and the fire itself. Considering the geographic displacement between field personnel and managers, our most critical requirement in the data-transfer system was that the system could operate over long distances, typically tens of miles.

Radio technology was the obvious solution. We first contacted a variety of commercial vendors for design help and appropriate products selection. After considering such solutions as satellite and shortwave radio, we chose commercially available broadband internet radios manufactured by Trango Broadband and sold by Last Mile Networks, a communications engineering and consulting firm.

This solution uses line-of-sight radios that operate on the 900MHz and 5.8GHz frequencies, which occur in the unregulated industrial, scientific and medical part of the radio spectrum. Many of our initial requirements were satisfied by a multi-hop relay technique, which creates a data backbone that doesn't require FCC licensing.

These radios use industry-standard protocols and can connect over 40 miles when connected to long-distance directional antennae. Their relatively small stature facilitates the logistics of moving and deploying them as they could be carried or flown easily in helicopters. In addition, the system's moderate power requirements means that photovoltaic panels could be used.

The final package was a series of commercially available and thus proven broadband Internet radios. Their long-distance antennae and photovoltaic panels meant that we could deploy them in a variety of situations and locations to support information transfer on wildland fire events. With the help of the radio engineers at Last Mile Networks, we were convinced this would be an excellent solution.

FREEZEOUT FIRE

Coincident to our investigation of long-distance data transfer, a wildland fire began that would play a critical role in deploying and testing this new data transfer technology.

On June 24, 2004, a passing thunderstorm ignited three separate fires near the top of Freezeout Ridge in the Pasayten Wilderness on the Okanogan — Wenatchee National Forest. The fire, which was located less than a mile from the Canadian border, continued to grow and appeared on the National Situation Report on June 27, where it was listed as a wildland fire-use event. On the same day, a national fire-use management team was ordered and was reported as “assigned” on June 29.

Due to the remote location, difficult terrain and safety concerns, the fire was managed as a limited-suppression event. As the Northern Rockies #1 fire-use management team completed a long-term action plan, a number of concerns and actions were identified. In addition to the lack of escape routes and safety zones, one major concern was the proximity of the Freezeout Fire to Manning Provincial Park, just across the Canadian border. Monitoring and accessing the fire also presented a number of logistical challenges, as it was approximately 40 air miles from the command post, and suitable helispots were limited in number and size.

The long-term action plan identified remote observation as a safe and cost-effective method of monitoring this fire. Specifically, the idea of a “Web camera” was mentioned as being desirable. However, no methods for achieving this remote monitoring were identified.

CAMERA SETUP

Our group was eventually contacted because the managers of the fire had heard we were investigating methods of long-distance data transfer.

In the initial conversation we were questioned about the ability to attach a remote camera to the data backbone, thus providing the monitoring capability identified in the long-term action plan. After some consideration we agreed to deploy the equipment and test the ability to remotely monitor the Freeezout Fire through the use of a network-enabled surveillance camera connected to the radio backbone. We planned to deploy this prototype of the communications hardware while maintaining safety and cost-effectiveness and engaging in appropriate activities.

To engage in cost-effective appropriate activities, we had to identify the best locations for radio relay sites. Because we were operating in a wilderness environment, we wanted to limit the amount of helicopter incursions. To this end we chose existing helispots where possible, and we attempted to install equipment at locations such as lookouts and existing radio repeater sites. Finally, all deployments were visualized with a viewshed GIS analysis prior to any flight or equipment setup, which allowed us to remain confident that the radios' line-of-sight requirements would be maintained before embarking on logistically complex personnel and equipment transfer.

Each relay location required relatively little setup. On each terminal end of the backbone is one radio with a complete power system. The terminal system is composed of 12-volt deep-cycle batteries, a photovoltaic charging system, charge controllers, directional antennae and a tripod for attaching hardware. The intermediate relay sites differ only in that they have two radios and two antennae. The terminal end near the fire includes the camera.

The radios use industry-standard protocols and are fully TCP/IP-compatible. In other words, they can be connected to a computer on the command post end for a closed network, or they can be routed through an existing Internet connection, which allows access to the cameras from anywhere in the world with an Internet connection. Security features are integrated in the cameras and radio relays so that camera access can be controlled and inappropriate use of network resources is prevented.

Once installed on the Freezeout Fire, the temporary data network consisted of six relay sites. The longest leg was about 17 miles. On the terminal end near the fire, a surveillance camera provided streaming video (up to 29.92 frames per second) through the data network. Additionally, the Web interface provided approximately 300° pan and 180° tilt capabilities, as well as a 26x optical zoom.

On the opposite terminal end, the network was connected to the Internet via a local service provider so anyone with the proper address and password could access the cameras. Varying levels of camera control permission were instituted via user names and passwords: Some were allowed full drive capabilities of the camera, while others were allowed to view only what the camera was recording.

Overall, the camera proved very useful for fire managers and interested third parties, including managers at the forest and regional level as well as Canadian fire officials. Managers could monitor fire behavior from the command post whenever they wished, and the fact that third parties could monitor fire activity without having to speak with members of the management team was very useful to everyone.

While the camera installation garnered a lot of interest, the real utility of this experiment was the deployment and testing of the data backbone. Because the radios use standard communications protocols, the ability to attach other network-enabled devices is possible. This network is essentially the same as any other computer network and, with adequate power, should be adaptable to most network devices. In fact, the use of the radio network should be nearly transparent to most users.

NETWORK USES

During the summer of 2005, we were again summoned to deploy the data network on a fire: the Selway — Salmon Complex on the Bitterroot National Forest. The Selway — Salmon Complex comprised more than 40 individual fires, all of which were managed under a wildland fire-use strategy.

Our deployment on this incident was similar to the Freezeout Fire in that part of the goal was to install remote surveillance capability. However, an additional goal was to use more of the network's data capabilities, including the transfer of GPS information to the command post from the field; the transfer of data such as the incident action plan from the command post to field personnel in spike camps; the demonstration of Voice over Internet Protocol, or VoIP, for remote telephony; and the testing of new remote weather sensors.

For the deployment of the network on the Selway — Salmon Complex, we went through the same line-of-sight analysis as on the Freezeout Fire. This included the computer visualization and assurance that our selected sites would be adequate before deploying equipment in the field. By doing computer analysis of selected deployment sites, we were able to set up five relays with two cameras in different locations in three days. On this deployment, the longest hop was 24 miles, and terminal ends were set to support both the command post and local district office for camera viewing.

In addition to the dual camera setup on the Selway — Salmon Complex, we were able to test additional technologies attached to the network. The first was a proof of concept using VoIP technology. The goal here was to provide the capability for free network-based phone calls between the command post and spike camps. The VoIP hardware allows standard telephones to be connected to the network. An obvious benefit of this technology would be remote briefings for field personnel in areas where telephony doesn't exist. It would also provide cost savings over satellite telephones.

The second technology test with the network was the installation of a wireless “hot spot” to facilitate the data transfer between the command post and field personnel. This was constructed by connecting a standard wireless router to a broadband radio that was part of the larger network. Field crews then could upload GPS information and digital photographs to wireless-enabled laptops before sending them to the command post via a local network Web interface. Field crews also were able to receive electronic copies of critical documents, such as the incident action plan and maps, that they could turn into hard copies with portable printers.

A third demonstration project was a series of autonomous weather sensors placed in transects across the landscape to provide supplemental localized weather information. These sensors have their own radios and pass data between themselves until the data arrive at the interface with the broadband network. Once there, those data are passed and archived on computers at the command post.

The last demonstration project was the use of ArcPad software to allow field GIS to run on handheld PDAs with custom fire forms programmed into the software. This allows for weather and fire observations to be made in a spatially explicit manner that can be directly incorporated into a GIS map.

OTHER BENEFITS

While we have no direct information on cost-containment provided by this system, we certainly have anecdotal information alluding to its value and usefulness.

One example from the Freezeout Fire was the inversions the fire often experienced, inversions that didn't affect the Methow Valley where the command post was located. Without field personnel on a radio or the remote camera, there would be little way to know the fire was experiencing inversions, thus limiting the ability of airborne personnel to monitor it. With the camera, managers were able to better manage monitoring flights. On several occasions they prevented flights where observation wouldn't have been possible due to smoke obscuration caused by inversions.

Other benefits of this network technology include the ability to send digital information, including GPS, digital photos and fire observation forms, from the field to the command post in a more timely fashion and in a form more immediately useful by fire staff, such as the GIS technicians. The information that can be provided by the network, cameras and miscellaneous widgets provides increased situational awareness for fire personnel by helping to create a more complete picture of the fire.

Historically, our research group has provided a module of experienced firefighters to deploy, manage and demonstrate the data capabilities of the network. We will continue to do so in the near term and will continue to test new technologies for use in fire management. To that end, we are eager to hear from appropriate interested parties wishing to use this technology for themselves.

Jim Riddering is a remote sensing specialist with the National Center for Landscape Fire Analysis at the University of Montana in Missoula.