From Wayne Bailey & Brent Hayner at Broad River Fire & Rescue.
Flight Debrief – Failures and Lessons learned from Mudslide flight incident on May 19, 2018
Sharing lessons learned from flights are meant to help others learn from our mistakes and victories.
Date: May 19, 2018
NFIRS Incident: 2018065
Location: Tryon, North Carolina
Total Flight Time
4 minutes for DJI3S (Risk analysis flight) Wayne Bailey sUAS Pilot
63 minutes for DJI Matrice 200 Brent Hayner sUAS Pilot
To provide assistance to the USAR team on locating a female that was swept away by a mudslide due to the heavy rains the area had experienced and state Geologist in assessing the slide.
What Went Well
The following items were conducted without incident:
- Working relationship between USAR Incident Command and Operations went very smoothly. Building relationships before the call is essential.
- Live data from a still photo using a DJI 3 Standard and FLIR video from a Matrice 200 using XT30 payload for FLIR and Z30 for zooming into the scene view was shared to the USAR IC, Operations, state and local Emergency Management, local law enforcement via big screen at the Command Post. We had two screens, one with the live footage and one with Exhibit 2 photo for planning purposes and to give incoming dignitaries a scene shot of the incident and what the USAR team was doing.
- The thermal camera identified an area of interest between what was left of the garage and part of the house at approximately 08:32. Upon further inspection, the ground crew found the victim at approximately 08:37. https://youtu.be/xJMMl8JZAeA Additional photographs property of Broad River Fire & Rescue can be found online here.
- We were able to stage close to the incident and away from a busy highway.
- A second Visual Observer was relocated with a handheld to keep VLOS rule in place.
- The pilots were able to capture the path of the mudslide for the state geologist. They were looking for potential secondary slides that would be an immediate threat to searchers by evaluating the color of the water and the speed of the runoff if the slide had stabilized. Note: The team will go back with the geologist and map the slide for historic purposes. The DJI UI had to have AGL limits overwritten due to max AGL 400’ but is from point of takeoff not calculating for terrain, UAS was never over 400’ AGL during flight and is noted on flight records 438-444.
- As soon as were on scene, we launched a DJI3S for a snapshot of the scene. This was crucial for the USAR team to develop a plan of action and establish a safe area in case of secondary slides. This short flight was for this purpose only and not for search and rescue. See Exhibit 2.
- IC and Ops could see live operations from the command post of the search happening from an aerial view.
What Could Have Gone Better
- A safe landing zone (LZ) was established near command got very busy as the incident progressed. Due to the closed road and limited parking, there was no place else to park. In the future, perhaps better security to keep curious onlookers and media from being in the same area as command.
- The footage or still photos could have been shared with the media to keep them from being on scene.
Recommendations From Flight
Continue to use the FLIR color designed for search and rescue. The victim was located in the white circle. Also, note the flow of water was still active around the house from up above. See Exhibit 1.
The flight crew had reports the victim had called out for help in the middle of the night. We were not sure if we were looking for a deceased victim or not. We encourage pilots to arrive on the scene as soon as possible to find the heat source from the victim. There has been no time of death given at the time of this report, however, the heat source we found was approximately 10 hours after she was sept away by the slide.
Do 360s around the structure at 2 knots to catch every angle. This victim was located 10’ down in between two structures with several layers of debris.
Fly at daybreak to take advantage of the sun not heating up the surroundings like metal objects that may give off a false reading.
When the flight crew interviewed the search crew, more than one person said it would of taken at least two 12 hour shifts to find the victim. Using the UAS, it took less than one hour. When the flight crew interviewed the state geologist, they said it would of taken days to survey the unstable mountain, instead, it took minutes.
If You Would Like to Submit a Flight Debrief
Just use to form below to reach me and please do share your lessons learned as well.