Once inaccessible, point cloud data is being used for everyday tasks in the GIS community
A 783Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE
More and more organizations are turning to LiDAR as a source for data collection and for good reason. LiDAR produces timely, accurate and high quality data that can address a number of applications. As a result, LiDAR is a key technology for terrain mapping supported by the United States Geological Survey (USGS).
One challenge faced when using LiDAR is managing the massive amount of data that is collected. More efficient management of LiDAR point clouds can lead to making LiDAR data easily accessible for use in a wider range of tasks and projects, extending the life of the data and yielding multiple benefits.
USGS Manages LiDAR Data with GIS
"GIS technology is a great solution for LiDAR data management after acquisition and post flight processing," said Jason Stoker, USGS.
Point clouds can create a data management challenge, since a single LiDAR file can be 100 MB in size or larger, with several million LiDAR points, and typical LiDAR projects can include hundreds of files. When multiplied by many separate LiDAR projectsperhaps in different coordinate systemsthe challenge of finding data covering a region of interest can quickly become unmanageable if the data is not catalogued and indexed in a spatially aware system like GIS. Using GIS, the USGS can effectively manage large LiDAR data collections. The same system also manages and serves the products derived from LiDAR; most common are raster surfaces representing the bare earth (a digital elevation model or DEM) and first return digital surface models (DSM).
"Once points are classified, they can be converted to a more user- and computationally- friendly format, like bare-earth DEMs and TINs," said Stoker. "It is these DEMs that geologists primarily use to identify and map features like faults, alluvium, and historic landslides."
The Latest in LiDAR Manipulation and Management
Although LiDAR is often acquired primarily to provide terrain information, the 3D structural information in the point clouds can be leveraged for other applications, resulting in rich multi-use datasets that effectively reduce the net cost of the data.
Another key challenge LiDAR data managers often face is the time pressure to evaluate LiDAR quickly upon delivery to ensure data quality. Esri’s ArcGIS Platform provides the ability to work directly with LAS files, eliminating the need to convert LAS files into other formats for GIS visualization and analysis.
"Managing LAS-format LiDAR data with ArcGIS allows it to be examined and managed quickly, and in a manner that is highly scalable for managing multiple projects and massive data volumes. ArcGIS can now provide numerous useful base products such as the DEM and DSM, as well as other products derived from LiDAR point clouds, including multiple quality assurance metrics about the data," said Cody Benkelman, an Imagery Technical Product Manager at Esri.
The Mosaic Dataset Opens Up LiDAR Data
Using ArcGIS, the LiDAR-derived DEMs can be constrained using authoritative terrain data such as ridgelines, hydrological information including stream channels and lake boundaries, and terrain control points. The resulting DEMs can then be used for detailed terrain analysis and contour generation.
A Mosaic Dataset makes this possible. The Mosaic Dataset is a geodatabase structure that contains a data model with integrated processing capabilities, and has been optimized for managing, querying, viewing, and processing massive collections of raster data. In the case of elevation data, the Mosaic Dataset allows high resolution DEMs from LiDAR to be integrated with historical data sources such as the National Elevation Dataset (NED) from USGS at several different resolutions as well as lower resolution global datasets such as from the Shuttle Radar Topography Mission (SRTM). The resulting multiresolution elevation Mosaic Dataset provides the best available data to users based on their area of interest.
Using added processing functions, derived products such as hillshade, slope, and ellipsoidal height (used for orthorectification of satellite imagery, for example) may be created on-the-fly using the Mosaic Dataset. These derived products are created on demand, only within the extents and resolution requested by the client software, so additional storage space is not consumed.
"Using a Mosaic Dataset allows users to scale projects efficiently," said Benkelman. "Multiple projects and data types can be methodically integrated into a single repository to simplify management. And because it is built into ArcGIS, compatibility with other GIS functions and data layers is ensured."
Derived Products for More applications and greater Return on investment
The "first return" DSMs created from LiDAR data collections can be used for accurate viewshed or line-of-sight analysis. Another valuable product is often created by using an on-the fly function to subtract the bare earth DEM from the DSM. This creates a raster representing surface height, sometimes referred to as a Height Above Ground (HAG) model, or a normalized DSM (nDSM).
"Working with height above ground values instead of elevation values above sea level can provide great value to many different types of users," said Stoker. "Local government users can use the HAG to quantify building heights quickly. Power companies can use it to identify areas where obstacles may be in the way of power lines, and forestry companies can begin to more accurately estimate timber quantities."
Data Dissemination
With the LiDAR and other elevation data integrated into a scalable management system, the task of sharing data with other userswhether to users inside an organization or made available to the external user communitybecomes a much simpler process.
"Elevation values in the DEM and DSM can be shared directly as image services through ArcGIS Server," said Benkelman. "This enables quick and efficient access for a range of clients including desktop, web, and mobile devices."
Images are rendered and delivered to client applications at the requested scale (spatial resolution) and for the current view extent. These image services are optimized based on the data typethe DEM and DSM are delivered as 32 bit floating point data, while the functions applied to the derived Mosaic Dataset to generate the other data products configure each product for most efficient transmissionfor example, as compressed 8 bit JPG format for the hillshade (for visualization), and 8 or 16 bit PNG for the quantitative data values of slope and aspect (for analysis).
"Server side functions are also able to provide powerful analytics like viewshed analysis to lightweight client apps using limited bandwidth connections. With our newly announced World Elevation services on ArcGIS Online, we’re seeing rapid growth in use of elevation data on mobile devices," Benkelman said.
Hosting data and transmitting only analytical results allows client applications to benefit from these data and derived products without requiring significant bandwidth and significant computer power.
Karen Richardson is a senior writer at Esri. She covers stories about the use of GIS for creating maps, data and charts including 3D, LiDAR and image data.
A 783Kb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE