USGS or LiDAR-Based DEM Elevation Accuracy
* This page describes USGS 10-meter and LiDAR-Based elevation data; LiDAR-based data is shown in the website and is the preferred data for agricultural applications.
Elevation data for the United States can be acquired for free from the USGS National Map (which can be accessed in the Data Sources folder to the left). Elevation values may be based on USGS DEMs, LiDAR, or other sources. LiDAR-based elevation rasters from the National Map are 1/9 arc-second products (about 3-meter resolution) while elevation based on 10-meter USGS DEMs are 1/3 arc-second products. There may be 1/9 arc-second National Map elevation products that are not based on LiDAR (metadata should described source of particular data). LiDAR is shown and described in the About LiDAR Elevation & Availability page.
Elevation accuracy of DEMs
USGS 10-meter DEM (from Haneberg, 2006)
• Errors in 10-meter USGS DEMs based on “more than 1,600 centimeter-accurate differential GPS measurements" are reported in Haneberg (2006; pdf). Elevation uncertainty (standard deviation) is ± 1.87 meters as shown in following table ("This Study" in the table below; the Fischer  10-meter data in the table below is based on a British Ordnance Survey DEM):
(from Haneberg )
USGS DEMs from National Elevation Dataset (NED) (USGS, 2006; pdf)
• Elevation accuracy (root mean square error) for nationwide data is ± 2.44 meters.
• Elevation accuracy (root mean square error) for data ≤ 90 meters apart is ± 0.78 meters.
LiDAR 1-meter DEM (from Haneberg [2008; pdf])
• Haneberg (2008) reported that 1-meter resolution LiDAR DEMs (based on 1.5 m pulse spacing with overlap that produced about 1 pulse/m² and inverse distance squared interpolation to produce a 1-meter DEM raster) had an elevation error range from –4.88 m to +3.32 m (meters) with an average error of –0.11 m, and a standard deviation of 0.75 m.
Check the metadata for 1/9 arc second elevation products from the National Map to determine if elevation is derived from LiDAR and for the accuracy of a particular dataset; error and standard deviation can vary from that reported in Haneberg (2008).
Haneberg, W.C. 2008. Elevation errors in a LiDAR digital elevation model of West Seattle and their effects on slope stability calculations, in R.L. Baum, J. Godt, and L. Highland, editors, Landslides and Engineering Geology of the Greater Seattle Area, Washington: Geological Society of America Reviews in Engineering Geology 20, p. 55-65 (doi: 10.1130/2008.4020).
Haneberg, W.C. 2006. Effects of digital elevation model errors on spatially distributed seismic slope stability calculations: an example from Seattle, Washington: Environmental & Engineering Geoscience, v. 12, no. 3, p. 247-260.
USGS. 2006. Vertical Accuracy of the National Elevation Dataset. National Elevation Dataset. FAQ: What is the vertical accuracy of NED data? Last updated: 2006. Cited at: http://ned.usgs.gov/Ned/faq.asp#VERTICAL