Ground reference data of at least three times better accuracy than the estimated IFSAR accuracy are used to validate the error model. Height error estimates are assumed valid for the entire dataset if at least 90 percent of the test points are within 20 percent of the values shown on the error map.
IFSAR has matured to the point where there are a limited number of commercially operated systems and products available from private sector vendors. Investment in an IFSAR mapping system is substantially more than that required for photogrammetry or lidar, and the data processing procedures and work flows overlap very little with these other mapping technologies in terms of hardware, software, and technical staff.
Future developments in IFSAR promise improved elevation and image products. Finer resolutions, increased height accuracy, and improved surface characterization have been achieved with experimental airborne systems. Differentiating heights of various physical surfaces, from treetops to bare earth, remains a significant research challenge. Fully polarimetric interferometers at multiple frequencies are the next technological leap to be made (Hensley et al., 2007). Development of new systems requires large research investments. Further innovation may require continued federal funding in support of scientific or defense-related programs of national interest, in addition to ongoing commercially funded research and technology projects.
Numerous spaceborne system concepts have been proposed to build on the success of STRM, but none have been approved for funding. Geosynchronous systems pointing continuously at a site of interest could measure very small changes in the surface conditions on an hour-by-hour basis, applications of which are diverse and important but of little benefit to the development of seamless bare-earth elevation models for the nation.
The purpose of this chapter is to provide the fundamental concepts of remote sensing technologies for creation of elevation datasets and other base map products of interest to a national floodplain mapping program. The three technologies discussed in detail are photogrammetry, lidar, and IFSAR.
Photogrammetry is flexible in terms of the number of products that can be made from a single source: aerial photography. The technology is mature; sources of error are well understood and rigorously modeled. Photogrammetry has made a successful transition into the digital age with large-format digital aerial cameras and soft copy processing work flows. Many processes are automated, but detailed feature extraction tasks still require significant human involvement. Bare-earth elevation models are particularly time consuming