I began fighting fires for the US Forest Service in 1987 and over the years I’ve witnessed game-changing innovations that we now take for granted on the fireline: computers, GPS, digital cameras and mobile phones. While many of these are common in our civilian lives, we’ve often seen fireline tools arise first from military applications, from wars overseas, where the need for fast-tracked innovations have provided us our own fire-ground game changers. After World War I, this technology transfer brought the airplane to wildfire protection, which greatly increased the efficiency of fire detection across large expanses of forest reserves. World War II gave us airtankers, and the Vietnam War, helicopters.
During my time involved in the war in Afghanistan I witnessed battlefield advancements in surveillance, and it was easy to imagine how wildland firefighting might deploy this technology. In the summer of 2014, while serving as an acting Aviation Manager in Alaska, I had the opportunity to apply these technological developments. With a background in fire and military aviation, I was assigned as liaison to a US Army’s Shadow UAS Battalion detachment during their support of a large fire burning on US Military gunnery range lands. While working with these soldiers, I quickly discovered that while they were experts in the use of UASs over the tactical battlefield environment, their machines and practices were not calibrated to deal with the idiosyncrasies of wildland firefighting.
We had a lot of work and a very short time to make their mission successful. I began by coordinating between each of our leadership groups, and took key members of their tactical staff to begin training. We reviewed our firefighting organization and how it related to theirs;, our field structure as it supported initial attack and extended attack situations; our tactics compared to theirs; the background of this wildfire’s current situation; and their proposed connection for applying their UASs to the challenges of the fire ground.
The next phase was quite complex. The military’s technology focus is not in line with firefighting needs. Shadow products are primarily video with the capacity to produce still photographs. Fire mangers typically do not have the time or personnel to consume hours and hours of video feed to create useful products. Similar issues had plagued two previous UAV experiments in Alaska with the University of Alaska (a UAV, or Unmanned Aerial Vehicle, is the civilian versions of a UAS). These two experiments are said to have provided false positive and inaccurate information. For example; moose were plotted as mobile hot spots, and in one case the drone crashed.
A critical issue was to come up with useful information. I sought assistance with this aspect of the project with a qualified Situations Unit Leader and a GIS Specialist both with wildfire backgrounds. We brainstormed and came up with a goal for two products – both IR maps. One would be an IR map of the fire’s perimeter and the other, an IR map of the fire’s hotspots, overlaid with imagery. For both of these products we planned to have the UAS team geo-reference important fire locations with lat/ long coordinates for use with the mobile application Avenza PDF Maps. We also wanted to add QR codes so these maps could be quickly and easily downloaded onto smart phones and tablets from the Internet. Thinking of Google Earth, we threw in the request for “street view” photos with additional attributes to provide even more valuable information to firefighters. All of these we believed were worthy goals for this new UAS technology.
I gave the UAS Army team the suggestion for these products along with the guidance I had been provided by an Assistant Fire Management Officer. We would be evaluating their performance based on five criteria:
- Their ability to create a map of large scale fire progression with IR mapping.
- Their ability to integrate successfully into a fire management team.
- Their ease of use.
- Response time.
- Long-term goal — to enable the Army to independently fly a UAS over an unstaffed fire burning on military lands, and then send status reports and hotspot locations to fire managers.
We deployed to the fire three days later with the understanding that we would only fly the UASs at night when all fire aircraft were down (with the possible exception of a night medevac helicopter). The UAS staff developed a written plan addressing their course of action in case the medevac helicopter launched while the UAS was up.
Initially, my job was to set the UAS team up operationally; once this was done I would be reassigned to other fire tasks. Together we worked hard to figure out how to make their UAS products useful to fire managers.
Three flights, many lessons learned
We moved forward through a series of goals that were ticked off one by one on the UAS’s way to becoming useful. The first big achievement was spending several hours overflying the fire environment – but this was not without its bumps. During this first flight, the pilots were introduced to flying on a large Alaskan wildfire. They recorded a very long video and took photos at a number of locations around the fire area. The geo-referencing feature was not turned on so while we got back some interesting IR shots of flaming forest and video, these products were not of much use afterwards.
The second flight ended up being the most valuable flight during this set up phase. The UAS command cell had learned its lesson from the flight before. The Army decided I would be of the most value inside the vehicle-based cockpit with their pair of pilots flying this night’s mission. I was given a fold-up chair to place between the pilots and I joined them before their monitor screens.
These video feeds and gauges appeared not so different than a computer-based flight simulator. The UAS arrived on-station over the fire and I began my instruction with the pilots on what is important to us firefighters. I explained the fire environment as we flew about; talking about potential dangerous spot fires, our focus on the fire’s perimeter, and how to define the perimeter using our IR camera when it shows up as separate dots. I told them about slopovers and guided them to fly over an area important to the Type 3 IC (Incident Commander), where we shot photos and video of the fire’s progression.
The video feed from the UAS aircraft gives a bird’s-eye view of its flight path above the fire ground — the Alaskan black spruce forest, mixed occasionally with lakes and meadows. The IR footage displays locations of fire and remaining heat across this landscape as bright blurry spots, even if smoke obscures the image of the forest.
The Chief Warrant Officer and the Air Traffic Group Captain spent many hours perfecting our product ideas, leading to the creation of a very accurately plotted fire perimeter map. This was time stamped and dated and overlaid with information pins. They also researched and found the way to place this map data into the Avenza mobile app and mass distribute it with a QR Code.
On this second Shadow flight we collected lots of images of uncontrolled fireline. To process this information the Captain and I began pinning these images onto their new map along the fire perimeter. I realized that we were creating an IR map which, if we continued around the fireline overlapping these photos, could create an IR map of the fire with pins leading to images and untold other information such as lat/long, compass directions in relation to the picture’s orientation, crew names, assignments, resource lists, Division specific info, etc. Other pins could display photos of helispots for logistical planning, sling spots, jump spots, and on and on.
We also figured out a way to plot fire’s perimeter, point by point, by using geo-referenced equipment to collect lat/longs along the entire perimeter of the fire’s edge. Although this data could be turned into a map by GIS personnel, this method proved to be very labor intensive and therefore impractical.
The third flight (and my last with the Shadow team) was another important milestone. Our number five goal was to have the Army independently fly a UAS/UAV over an un-staffed fire burning on military lands, and then send a status report and hotspot locations to fire managers. We just had to work out how to do this successfully. The acting Fuels Officer requested us to fly over a huge prescribed fire area adjacent to the fire area to make sure it was out, other than where a small pocket of fire was still known to remain.
The UAS was launched. As it arrived over the target area, I was sent back into the pilot’s vehicle cockpit to figure out with the pilots how to search a huge expanse of land for something as small as a single root, burning beneath the ground, that might later creep up and ignite another expensive uncontrolled wildfire. They couldn’t miss even one tiny spot.
The pilots began by flying randomly and looking down around this expanse which I soon could tell would not be of much real use, so I began questioning the pilots for options. They came up with the idea to program a massive search area into the aircraft’s computers, which we then broke down into a series of flight grids. It took the pilots some time to figure out the process of how to program the flight computer, but this tactic proved successful. After sitting with the pilots for many hours flying back and forth in this massive but complete grid of the area, we found the remaining fire exactly where it should have been, and nowhere else and no false positives. The flight finished with us moving over the active fire and taking pictures of important unstaffed locations to document the fire’s advance for the fire managers.
My time with the Shadows was now coming to an end; these tools had now been created and fine-tuned to create products that had value to fire managers. This was at about the same time that the fire transitioned from a Type 3 organization to a larger Type 1 Incident Management Team.
The Captains in the UAS unit told me their craft would be much more efficient flying in the daytime, similar to how they were designed — to support battlefield decision makers with real time critical information. Contrary to some popular belief, the Shadow UAS is typically not used as a spy machine but more of a mobile information gatherer. They suggested that their UAS could take pictures, collect lat/longs around the fire, and be repositioned as needed by fire managers to observe important locations.
One of the Captains also spoke of soon sending his air traffic controllers home so I asked the incoming Type 1 Incident Commander if he would be interested in allowing the UAS staff move their command cell and air traffic controllers to the new ICP location and move the UAS to day operations while the air traffic control team was still in place. The Shadow UAS would remain in Military Restricted air space and still not mix with other fire aircraft. The new IC supported this idea.
My role was complete: the UAS team had accomplished so much in the short time allocated to set up and produce completely different products for civilian use.
The Shadows utilized on this fire are the US Army’s middle-of-the-range tactical battlefield information gathering devices. They are highly mobile with the option of being crated up and shipped to remote locations where they can be reassembled and operated off catapults with only a short dirt strip to land on. Within the US Army, they are the largest UASs available that are this versatile without needing an airfield.
Shadows have their uses as demonstrated, and their technology just had to be adapted to suit the needs of the fire personnel. In my experience Shadows have their limits but also unique capabilities they could add to the tool box of emergency managers.
What are some of the key lessons we may wish to consider as unmanned aerial surveillance rolls out on wlldland fire?
- A liaison needs to be embedded with the UAS on an inter-agency assignment. The UAS commanders also repeatedly stated this during their debriefings. Especially during the pioneering stages of our two organizations working together with this emerging technology, a majority of the potential problems can be more easily resolved or avoided entirely by assigning an appropriate liaison. Keeping a liaison embedded for the duration of their deployment helps ensure that concerns or issues at all phases of the deployment are addressed or resolved. This individual should have the appropriate background and skills such as solid understanding of military aviation, fire aviation and firefighting. Mapping or GIS skills would also be helpful as would direct assistance from fire GIS personnel. If the UAS is used on night shifts then liaisons should be assigned to both shifts. The night “embedded” liaison should be at least Helicopter Manager or Air Attack qualified to be an effective UAS/UAV Manager. The day liaison would represent the UAS staff at team meetings that occur outside of their normal shift while the night shift liaison actually works “embedded” with the UAS staff to assist their operational needs.
- During the early stages of this fire, the UAS night-shift command staff was expected to also attend daily Army and team organizational meetings. This decision lead to fatigue that threatened the success of the UAS operation.
- Since the UAS is flown remotely, there would also be advantages to having key members of the UAS staff take a helicopter flight of the fire to offer more encompassing viewpoints of the fire environment.
The question now is where will the use of this technology take us? It was a great opportunity to be involved with this hardworking team in these pioneer phases of UAS usage on wildfires where more building blocks were laid for future wildfire and emergency management UAS developments.
I realize some of these lessons are more applicable to military lands, but many will apply to any air space, as we roll out the use of drones/UAVs on wildfires internationally.
For more information on civilian use of unmanned aerial systems, visit the Federal Aviation Administration site: http://www.faa.gov/uas/. The Interagency Airspace Coordination site also offers a range of links regarding interagency operations and UAS: http://airspacecoordination.org/coord.shtml.
Michael Scott Hill is a U.S. Forest Service firefighter who’s worked fire and emergency aviation on four continents.