Light Detection And Ranging

Background and Goals

LIDAR stands for Light Detection and Ranging, and is method used in remote sensing to capture three dimensional data of the characteristics of the Earths surface. LIDAR can be captured by a variety of remote sensing platforms, such as airplanes, drones, satellites (no longer used) and ground units. The data that LIDAR is referred to as a “Point Cloud”, which looks exactly like it sounds, many points of captured data in a 3d plane. LIDAR point clouds can be used to create a variety of models such as Digital Surface Models(DSM), Digital Terrain models (DTM) and can also be used in a variety of analysis about Earths man made and natural features. Here is a background video on the basics of LiDAR and some of the uses that LiDAR can have for data analysis. 

In this Lab we will look at the processing of LIDAR data into various models, such as DTMs and DSMs, as well as other images which can be derived from the LIDAR point cloud. 

Methods

First: we are given the following scenario; You are a GIS manager working on a project for the City of Eau Claire, WI, USA. You have acquired LiDAR point cloud in LAS format for a portion of the City of Eau Claire. First, you want to initiate an initial quality check on the data by looking at its area and coverage, and also verify the current classification of the LiDar.

First the LAS dataset needed to be added to ArcMap, and the properties of the LAS data set needed to be examined. When a new LAS dataset is brought into ArcMap, the statistics need to be calculated from the header information that is included in the .las data file header information. This step is important because the statistics are used for quality assurance and quality control of the overall data set and individual las files.

The next step is to assign both the vertical (Z) and horizontal (XY) coordinate systems which should be included in the LAS dataset’s metadata. The metadata also includes the units of the data which make up the point cloud.  Once the coordinate systems and units are determined are assigned, the data is now ready to be analyzed.

To make sure that the data cloud has been correctly spatially located, a shapefile can be added and used as a reference check.

Figure 1. Tiles of the LAS dataset after the statics have been calculated, the horizontal and vertical coordinate system have been assigned and the shapefile has been added to make sure the LAS dataset is spatially correct.

The reason that the LAS dataset is displayed in the form of tiles is because the point cloud has an extremely large amounts of data points and displaying them constantly would take a lot of computational power and time to redraw the points over and over as the dataset is examined (Fig 1). Instead the dataset is displayed as tiles and does not show the point cloud until a specific tile is chosen to zoom in to.

  

Below we can see an example of the point cloud (Figure 2). Here the point cloud is displaying data points organized by elevation and first return. In comparison we can also see the base map displayed of the UWEC campus (Figure 3). By comparing the two images we can see both the elevation of the buildings being represented in the point cloud, but we can also see the difference in upper and lower campus, upper campus being the orange area represented in the point cloud, and lower campus being represented by the green area, with the yellow area in between representing the hill and change in elevation between the two areas. 

Figure 2. The LAS dataset point cloud. Here the points displayed are of elevation based on first return data. This area is of the UWEC lower campus (green area with orange/yellow buildings), and the upper campus (yellow area) and the surrounding area. 

Figure 3. World Imagery Basemap. This is the same are shown in Figure 2 for reference. 

The data of the point cloud can be separated by returns. Returns occur when the LiDar pulse strikes an object and is able to return multiple pulses of information.  and the LAS dataset 3D view.

Figure 4. Number of Returns. This image depicts how a LiDar pulse can interact with an object and cause multiple returns, by interacting with various objects in the pulses path. 

The next task was to look at the LAS dataset in 3D view. By using the 3D view, a box can be created which allows the analyst to see the data in profile. The image shown below (Figure 5) is the profile view of a pedestrian foot bridge which cross the Chippewa river, connecting the UWEC campus.

Figure 5. LAS 3D Viewer. This image depicts a cross sectional view in 3D of the first returns of the LiDar data set. The Area is the pedestrian foot bridge that runs across the Chippewa river. 

Similar to being able to see specific areas in profile, we can create digital surface models using a similar technique. There are two models which we are interested in for this report. The Digital Surface Model (DSM), and the Digital Terrain Model (DTM). The tool used to create both models the same, the raster to surface tool, however the changing the returns determines which output will be achieved. 

For the DSM we want to choose the first return, because in this model we want to see the variation in terrain of the three dimensional objects on the earth’s surface (Figure 6).  

For the DTM we want to choose ground returns because we want to model ground surface features, not the three dimensional man made or natural objects which would obscure the surface

Once we have a DSM or DTM we can run tools to enhance our analysis of the LiDar data, such as creating a hillshade model. There are a variety of tools to use in the 3D analyst tools for raster in ArcGIS. 

Figure 6. Digital Surface Model with hillshade effect. The area depicted is of the confluence of the Eau Claire and Chippewa Rivers. 

Once both models are complete we can layer both models on top of each other and use the swipe tool to "peel away" the DTM to see the DSM underneath it (Figure 7). By doing this we can compare and contrast both models simultaneously. 

Figure 7. Digital Terrain Model and Digital Surface Model with hillshade effects. Here both the DSM and DTM are displayed at the same time. By using the swipe tool we can peel back the DSM and see the DTM underneath it. 

The last objective for this report was to use the LiDar returns to create a LiDar intensity image, using the LAS dataset to raster tool. 

Figure 8. LiDar Intensity Image.

 Results

LiDar is a very powerful tool that can be used in remote sensing, and is not only interesting to preform analysis on, but it is also very interesting and fun. But it must be understood how LiDar works if a robust analysis is to be preformed. LiDar point clouds hold much data and the tools we have to process this data let us explore a remotely sensed world in 3D, as well as create various models that we can use to analyze the world around us. We can create models of the earths bare surface (Figure 9), and map fluvial features or asses flood risk in areas of low elevation.  

Figure 9. Digital Terrain Model With Hillshade Effects

Or we can see the features which are both man made and natural, which dot the landscape (Figure 10). The amazing thing about a digital surface model is that it can allow us to see places we may never go, and has many possibilities in analyzing and understanding our world.  

Figure 10. Digital Surface Model With Hillshade Effects

               

Data sources

Margaret Price, 2014. Mastering ArcGIS 6th Edition. Eau Claire County Shapefile.

[The Science of Measuring Ecosystem: NEON Education]. (2014, June 18th). LiDAR – Introduction to Light Detection and Ranging. Retrieved November 10, 2016, from  https://youtu.be/EYbhNSUnIdU

"What is Light Detection and Ranging" (2016, July 30). A Complete Guide to LiDAR: Light Detection and Ranging - GIS Geography. Retrieved November 10, 2016, from http://gisgeography.com/lidar-light-detection-and-ranging/

"Basemap" ESRI 2011. ArcGIS Desktop: Release 10. Redlands, CA: Environmental Systems Research Institute.

Eau Claire County, 2013. [LIDAR Point Cloud Data and Tile Index].