LiDAR has a large number of applications other than the commonly known high precision sensors in autonomous cars, laser scanning, photogrammetry and bathymetry. It can also be used in detecting and analyzing the phenomenon of natural disasters such as landslides. Landslides lead to a huge loss of life, property and displace thousands from their homes. As per the USGS, in the United States landslides cause an annual loss of around $ 3 billion and the death of 30-50 people.
Landslides are often triggered by another event like an earthquake or a volcanic eruption, but they can wreak more damage than the original disaster.
LiDAR as a tool for geologists
Geologists have increasingly started using LiDAR as a way to predict and study landslides. One of the crucial steps that is being taken by governments and scientists is the integration of airborne LiDAR as a crucial component in disaster mitigation efforts. Data gathered by LiDAR provides useful insights into geographical risk zones and helps in the identification of areas vulnerable to landslides. The main advantage of LiDAR over other aerial imagery is that with the help of LiDAR we can easily see beyond thickets of vegetation and focus on otherwise hidden details like slope angle, the extent of deformation and the magnitude of erosion.
In landslide assessment, surveillance and thorough monitoring, LiDAR is a valuable tool. But what further increases its utility and makes it of utmost importance is the fact that LiDAR can help us pinpoint areas where landslides could take place.
In 2014, 42 people lost their lives in Oso landslide in Washington, LiDAR imagery and data provided by Quantum Spatial helped in analyzing the severity of the situation. The difference between LiDAR imagery and normal imagery can be very clearly discerned.
In 2018 January post the Montecito landslide that generated massive debris, a team of geologists from USGS used many methods, including LiDAR to properly analyze the situation and aid in better and faster relief and rescue operations.
One of the main reasons why landslides occur is due to change in land features like slope, terrain roughness, stream power metrics and its elevation. These changes can be either due to natural factors like wildfires or man-made factors like indiscriminate deforestation, haphazard constructions.
There are mainly two types of landslides: large and shallow. Mostly, the former, as the name suggests, includes a whole piece of a mountain, while the latter involves just the surface soil or maybe the weathered rock on top of the fresh rock. LiDAR data can be used to identify most of the large landslides that have already taken place and it can reveal shallow landslides.
Razor-sharp analysis and preemptive data
LiDAR helps in capturing the minutest details of these events and offers an unparalleled and incisive view. Landslide data from LiDAR is obtained either by analyzing the terrain surface or from change detection between surface data acquired at different times.
In the first case, surface features and related properties are analyzed to identify parameter values. The second method is aimed at detecting changes in surface movement and deformation. In both cases, the performance of the landslide prediction depends on surface representation in terms of spatial sampling and accuracy.
Both the methods are equally important, as there are dormant landslides, where there is no or undetectable surface changes, and actively developing landslides, where surface motion can be observed
Airborne LiDAR sensors have a modest spatial resolution so the identification of potential landslide areas proves to be more difficult than detecting changes between surfaces acquired at different times. But many times there is no previous data available hence filtering of data becomes crucial.
Surface data is first smoothed and then interpolated to a regular grid. Afterward, three different approaches are followed concurrently. Point-based methods try identifying areas where the distribution of surface point could have statistical parameters different from other areas. On the other hand, profile matching aims at identifying cross-sections typical of landslide areas.
Key for disaster mitigation
Combining LiDAR with GIS improves both the quality and the accuracy of geological data and enables geologists and disaster preparedness experts to identify landslide prone areas and study the underlying causes of a landslide and recurring patterns.
High-resolution LiDAR modeling reveals subtle surface features that are simply undetectable via aerial photographs or field observation. This helps with mitigation planning, landscape modeling and thus saving the lives of millions of people as well as the infrastructure.
LiDAR also increases the efficiency of observation multiple times. In 2009, a study of how many landslides had been mapped before the state of Oregthe on in US started using LiDAR found around 10,000 mapped landslides. Using LiDAR, there was more than a fourfold increase of landslides in that same database, and it covered only a small area of the state.
Recognizing the benefits of LiDAR geologists and those deal with landslide observation are updating their hazard maps with baseline data that is obtained from LiDAR images. For creating a landslide prediction model and eventually a regularly updated public landslide database, LiDAR is proving to be of great importance owing to its precision.