As with bathymetry, LiDAR is not the only means of measuring watershed
topography, but it is the most accurate. For applications where a regional-scale but
coarse-resolution watershed map is sufficient, there are three relevant remote
sensing programs; NASA’s Shuttle Radar Topography Mission (SRTM), the
ASTER Global Digital Elevation Model (GDEM), and the TerraSAR-X/TanDEMX topography program. The first two of these are accessible at no-cost, while the
latter is for sale through the German consortium that operates the mission. If a finer
resolution product is required, as is often the case for reef studies, a topographic
LiDAR survey is appropriate. To provide maximum return from the financial
outlay, mapping of the watershed should be combined with a bathymetric survey
through use of an instrument such as SHOALS or EAARL, which can acquire both
in a single mission. However, if the topographic area is very large relative to the
bathymetric area, it may be more cost effective to use a topo-only sensor for their
faster coverage rates and therefore lower cost.
5.4 Processing and Validation Requirements
Discrete-return LiDAR data requires several complex processing steps to develop
useable products; the workflow is usually divided into two phases:
• Preprocessing is the preparation of the raw data, merging of the GPS/IMU with
the laser ranges to produce a point cloud, correcting for errors induced by flight
geometry, removing overlaps between flight lines (particularly if the LiDAR
system is internally inconsistent), surfacing operations, and accuracy assessment
of the raw point locations. Preprocessing is typically undertaken by the LiDAR
contractor.
• Postprocessing is the development of usable information from the point clouds,
including development of TINs, DEMs, and other products like canopy height
maps.
Because of the great volume of data acquired during a laser survey, validation is
laborious and every effort is taken to reduce error at source, so as to lessen the need
for post hoc validation. Indeed, a considerable portion of the error associated with
a LiDAR survey can be mitigated by thorough validation of the instrument
package before flight. Some of this, such as laser scanner calibration, laser beam
alignment, noise reduction of the LiDAR signal, and so forth, are carried out by the
instrument design company (Adams 2000; Fang and Huang 2004; Latypov 2005;
Wagner et al. 2006). These parameters may drift through time and so the instrument-package must be returned periodically for factory-service. In an effort to
further reduce errors, other calibrations are conducted on-site by the operator prior
to take-off. These include measurement of the position-shift between laser scanner,
IMU, and GPS. This shift distance will be applied to the sounding data during
post-processing georectification. Also, the laser scan rate may not be in concert
with the data acquisition rate of the GPS and the IMU, and must be synchronised
136
S. J. Purkis and J. C. Brock
topography, but it is the most accurate. For applications where a regional-scale but
coarse-resolution watershed map is sufficient, there are three relevant remote
sensing programs; NASA’s Shuttle Radar Topography Mission (SRTM), the
ASTER Global Digital Elevation Model (GDEM), and the TerraSAR-X/TanDEMX topography program. The first two of these are accessible at no-cost, while the
latter is for sale through the German consortium that operates the mission. If a finer
resolution product is required, as is often the case for reef studies, a topographic
LiDAR survey is appropriate. To provide maximum return from the financial
outlay, mapping of the watershed should be combined with a bathymetric survey
through use of an instrument such as SHOALS or EAARL, which can acquire both
in a single mission. However, if the topographic area is very large relative to the
bathymetric area, it may be more cost effective to use a topo-only sensor for their
faster coverage rates and therefore lower cost.
5.4 Processing and Validation Requirements
Discrete-return LiDAR data requires several complex processing steps to develop
useable products; the workflow is usually divided into two phases:
• Preprocessing is the preparation of the raw data, merging of the GPS/IMU with
the laser ranges to produce a point cloud, correcting for errors induced by flight
geometry, removing overlaps between flight lines (particularly if the LiDAR
system is internally inconsistent), surfacing operations, and accuracy assessment
of the raw point locations. Preprocessing is typically undertaken by the LiDAR
contractor.
• Postprocessing is the development of usable information from the point clouds,
including development of TINs, DEMs, and other products like canopy height
maps.
Because of the great volume of data acquired during a laser survey, validation is
laborious and every effort is taken to reduce error at source, so as to lessen the need
for post hoc validation. Indeed, a considerable portion of the error associated with
a LiDAR survey can be mitigated by thorough validation of the instrument
package before flight. Some of this, such as laser scanner calibration, laser beam
alignment, noise reduction of the LiDAR signal, and so forth, are carried out by the
instrument design company (Adams 2000; Fang and Huang 2004; Latypov 2005;
Wagner et al. 2006). These parameters may drift through time and so the instrument-package must be returned periodically for factory-service. In an effort to
further reduce errors, other calibrations are conducted on-site by the operator prior
to take-off. These include measurement of the position-shift between laser scanner,
IMU, and GPS. This shift distance will be applied to the sounding data during
post-processing georectification. Also, the laser scan rate may not be in concert
with the data acquisition rate of the GPS and the IMU, and must be synchronised
136
S. J. Purkis and J. C. Brock