On March 22, 2014, a large landslide in Oso, Washington killed 43 people and destroyed dozens of homes. Heavy rainfall in the previous days and weeks had saturated the soil on a hillside over State Road 530 and the Stillaguamish River. This eventually triggered the slide, burying approximately 1 acre below in 30 to 70 feet of mud, soil, and rock debris.
Shortly after the mudslide, the Field Innovation Team (FIT) sent a team to support first-responders in finding innovative solutions to expedite recovery efforts. FIT is a non-profit organization that responds to crises, and also works on disaster preparedness and disaster risk reduction. Although based in Utah, FIT has staff and volunteers across the country and around the world that can quickly respond to disasters. After the Oso mudslide, FIT used Reality Computing—capture, compute, and create—to develop and print a 3D surface model of the mudslide area, giving relief teams more understanding of the topography of the disaster.
FIT volunteers who traveled to the site began by meeting with the disaster management personnel at the local Incident Command Post (ICP) to discuss response and recovery needs and offer assistance for technology shortfalls. FIT learned that the disaster relief personnel needed better situational awareness of the mudslide terrain itself, which was partially inaccessible due to the debris piles and the flooding of the Stillaguamish River from the mudslide.
Aerial imagery of the valley from manned helicopter flights along with U.S. Geological Survey (USGS) LiDAR data could have been obtained. But helicopter flights were expensive, and obtaining the LiDAR data would take two to four days. So FIT decided that using a digital camera attached to a drone was the quickest, most efficient way to capture data of the mudslide terrain. FIT worked with ICP to have Roboticists Without Borders (RWB) perform the flights. RWB is part of Texas A&M University’s Center for Robot-Assisted Search and Rescue(CRASAR).
Initial data capture and modeling
FIT’s team, which also included volunteers from Autodesk and PrecisionHawk, used an AirRobot AR100Brotorcraft for the data capture. After receiving an emergency Certificate of Authorization (COA) from the FAA for the flights, 40 acres of the mudslide were captured with seven flights with a Canon digital camera fixed to the AirRobot rotorcraft.
The images were processed by the PrecisionHawk team using their own software to quickly create a 3D surface model of the area. This model was then shown to ICP personnel who were impressed by the team’s ability to quickly capture data and generate a model. More importantly, they immediately recognized the value of a 3D digital model to help them make decisions regarding recovery efforts such as where to place excavating equipment.
The model did not include geotagging for geospatial reference and the camera used was not capable of obtaining high-resolution photogrammetry, which was reflected in the resolution of the resulting model (5 centimeters per pixel). But from this initial effort, it was decided to move forward with a higher-resolution survey of the mudslide debris field to produce a higher-quality digital model.
High resolution data capture and modeling
The team had hoped to use a fixed-wing PrecisionHawk drone (offering higher-resolution data capture and geospatial sensing), but could not obtain an emergency COA. Instead, the team used the government LiDAR data mentioned earlier. This data was processed using ReCap and converted to a digital terrain model using AutoCAD Civil 3D.
The model’s higher level of resolution displayed detailed topographic features of the mudslide area and gave the relief team a better understanding of the topography and the soil conditions. In addition, the model detailed the placement of debris piles, helping hydrologists understand river flow and flood patterns of the Stillaguamish River, which dammed and flooded the valley as a result of the mudslide.
The full process of how this LiDAR data was processed & at the end a 3D printed model was made, is fully described in this article.
The Civil 3D model could also be used for 3D printing. When FIT first approached ICP, the potential of producing a 3D printed model of the surveyed mudslide interested the engineers handling the mudslide and river reconstruction. Autodesk volunteers 3D-printed a portion of the mudslide area on an Objet 500 Connex 3D for the Snohomish County Department of Public Works.
For more information on FIT and how it used Reality Computing’s capture, compute, and create technology for disaster recovery efforts after the Oso mudslide, check out these articles:
- Forbes article: Nonprofit Brings Technology To Disaster Recovery
- FIT paper: Osa, WA Mudslide After Action Report
For more similar news around Reality Capture technologies, make sure to subscribe to our blog.