- SPECIAL REPORTS
- THE MAGAZINE
The U.S. Geological Survey (USGS) has a long and impressive history in mapping the topography of the country. In 1879, the USGS began the process of mapping the nation by conventional surveying means. In the 1930s, the USGS, in cooperation with the Tennessee Valley Authority (TVA), began using aerial photography and photogrammetry to continue this mapping effort. In 1975, the USGS created the first digital elevation model (DEM), and by 1990, all 1:24,000-scale quadrangle maps had been completed for the conterminous United States. In 1997, the seamless National Elevation Dataset (NED) was completed, and in 2003, LiDAR data began replacing less-accurate, lower-resolution elevation data in the NED.1
While the NED provides an excellent source of elevation data for the country, much of the data was collected 30 or more years ago. Since that time, considerable development has taken place in many parts of the country, and this development has obviously reduced the accuracy and value of elevation data captured prior to that activity. Moreover, the original elevation data was captured from higher altitude photography and prior to many of the recent improvements in photogrammetric cameras, collection hardware and processing software. Therefore, the accuracy of this data is considerably less than what is achievable with current LiDAR technology. Finally, elevation data collected 30 or more years ago does not reflect the considerable subsidence experienced in some parts of the country. It is estimated that 17,000 square miles of land--much of it in California, Texas, Louisiana and Florida--is sinking at a rate of inches per year.2 Elevation data collected 30 or more years ago can have changes that are measured in feet compared to today’s elevations.
And there is one more dynamic in play: Many potential users of elevation data need more than just a bare-earth model currently available from the NED. For example, the U.S. Forest Service uses LiDAR to estimate forest inventory information, including tree (or stand) height, stand density, volume and biomass. This information is not attainable from an elevation model of the bare-earth surface alone but can be estimated from multi-return LiDAR data sets. Multi-return LiDAR data is also useful in urban planning, line-of-sight communication planning and hydraulic modeling.
National LiDAR InitiativeThe first meeting for the national LiDAR initiative was held Feb. 14-16, 2007, at the USGS National Center in Reston, Va., and included representatives of state and federal government, universities and private-sector representatives from surveying, engineering and mapping firms. The goal of the meeting was to develop a national initiative to collect seamless, high-resolution, high-accuracy elevation data for all 50 states.
The idea of this new elevation data set, now commonly referred to as “Elevation for the Nation,” is gaining traction. LiDAR is the enabling technology that makes such an effort possible, and the report of this meeting states this is the preferred technology for this elevation collection effort. Such a program would not have been possible−or affordable−10 years ago at the onset of commercial LiDAR data technology. But over the last 10 years, LiDAR units and processing software have improved dramatically, and the ability to collect high-accuracy elevation data with dense postings has become a reality. At the same time, the cost of capturing and processing elevation data on a per-unit basis has dropped significantly.
The second National LiDAR Initiative meeting will be held at the USGS headquarters in Reston, Va., on May 21-22. You can register for this event at http://lidar.cr.usgs.gov/.
Changing TechnologyIn the last 10 years, we have seen LiDAR sensors become faster, more powerful, more capable and more accurate. The number of pulses per second from the lasers within the sensors, commonly quoted in terms of kilohertz (kHz), which is equivalent to laser pulses per second, has grown from 2 to 167 kHz. These rates are equivalent to maximum collections of 2,000 to 167,000 points per second, respectively. These increased collection rates allow for the capture of higher density data and faster aircraft speeds during airborne acquisition.
More powerful lasers also allow higher flight altitudes and, therefore, wider swaths of coverage at these increased heights. Since acquisition cost is directly proportional to the area of ground covered during each pass, higher altitudes and wider swaths equate to lower unit costs.
In terms of increased capabilities, one of the more important additions to LiDAR sensors came along in 1995: the ability to record multiple returns from a single outgoing laser pulse. This allows the recording, for example, of both the tops of trees and the ground beneath from a single laser pulse. Beyond discrete multiple returns, LiDAR waveform modeling is somewhat in its infancy. As such, it is a promising addition that will eventually provide significant improvements in the data, particularly for forest and vegetation applications.
For numerous reasons, the accuracy of today’s sensors is significantly better than the units of a few short years ago. More GPS satellites are on orbit, which improves the ability to position the aircraft during acquisition. The inertial measurement units (IMU), which provide the 3D rotation of the LiDAR sensors, are more accurate and more stable than earlier models. We have a better understanding of the calibration of the LiDAR units and more efficient software routines to help us in that respect. Finally, we have improved geoid models that allow us to determine orthometric heights more accurately than before. All of these add up to provide accuracies in LiDAR elevation data suitable for 1-foot contours and beyond.
Also significant are improvements in LiDAR processing software. Filtering techniques that allow the automated classification of LiDAR data have improved greatly. These techniques are used, for example, in the classification of non-ground features like vegetation and buildings, which can be very important for many uses, such as forest inventories, 3D visualizations and hydraulic modeling.
What Does All This Mean?
Ultimately, Elevation for the Nation will provide the professional community with a seamless elevation model that is much more useful than the existing NED due to improvements in both accuracy and resolution that are available today. This improved NED will be invaluable to a diverse group of users and will benefit an almost limitless number of applications beyond traditional land development, transportation planning, emergency response, hydraulic modeling, etc.
LiDAR data is making its way into the hands of more professionals. Luckily, software solutions that allow exploitation of this data are growing considerably. Companies like QCoherent Software, which developed LP360, a LiDAR software extension for ESRI ArcGIS, and Applied Imagery, which produces Quick Terrain Modeler, provide affordable solutions for this data exploitation.
Finally, a number of very good recommendations for Elevation for the Nation were made in a National Research Council Report titled “Elevation Data for Floodplain Mapping.” These recommendations included:
1. Elevation for the Nation should employ LiDAR as the primary technology for digital elevation data acquisition. LiDAR is capable of producing a bare-earth elevation model with 2-foot equivalent contour accuracy in most terrain and land cover types; a 4-foot equivalent contour accuracy is more cost-effective in mountainous terrain, and a 1-foot equivalent contour accuracy can be achieved in very flat coastal or inland floodplains.
2. A seamless nationwide elevation model has application beyond the FEMA Map Modernization program.
3. The new data collected in Elevation for the Nation should be disseminated to the public as part of an updated National Elevation Dataset.
4. The Elevation for the Nation database should contain the original LiDAR mass points and edited bare-earth surface, as well as any breaklines required to define essential linear features.
5. In addition to the elements proposed for the national database, secondary products including triangulated irregular networks, hydrologically corrected digital elevation models, and hydrologically corrected stream networks and shorelines should be created to support FEMA floodplain mapping.3
References1 Report of the First National LiDAR Initiative Meeting, Feb. 14-16, 2007, Reston, Va., U.S. Department of the Interior, U.S. Geological Survey, page 26.
2 Elevation Data for Floodplain Mapping, Report in Brief, The National Academies, January 2007, page 3.
3 Condensed from the free executive summary of “Elevation Data for Floodplain Mapping,” Committee on Floodplain Mapping Technologies, National Research Council, 2007, http://books.nap.edu/execsumm_pdf/11829.pdf.