the returning laser-waveform that corresponds to the water surface from that of the
seabed. For coastal mapping, this problem may be resolved through the use of a
temporally shortened transmitted pulse (\2 ns), or may be obviated by combining
successive flights at low-tide with a topographic LiDAR, and at high-tide with a
bathymetric LiDAR (Pastol et al. 2007). Modern LiDAR systems address this
problem by employing ‘‘shallow-water’’ algorithms and waveform deconvolution
techniques. Though experimental, the shallow-water problem can alternatively be
tackled using a statistical relationship that links water depth to the Raman
waveform shape (Pe’eri and Philpot 2007). Since reef-tops tend to be considerably
shallower than 2 m, these advances are very relevant for the successful application
of LiDAR as a complete-system for coral reef mapping.
State-of-the-art LiDAR systems are calibrated and have the capability to capture reflectance data from the returning pulse, in addition to the three-dimensional
coordinates of the laser returns (Lillesand et al. 2004; Tuell and Park 2004; Tuell
et al. 2005). Like the strength of radar returns, the intensity of LiDAR ‘‘echoes’’
varies with the wavelength of the source energy and the reflectance spectrum of
the material that reflects the downwelling laser. For bathymetric applications, such
radiometric information, termed LiDAR ‘‘intensity’’, can aid in the identification
of seabed character. Intensity values are extracted from the returning laser
waveform and represent the magnitude of the bottom return. Via interpolation, an
intensity image of the seafloor can be created from spot soundings. To capture a
true reflectance image, it is necessary to employ a calibrated system and all
environmental and system response parameters must be appropriately modelled.
The reflectance image is captured at the wavelength of the transmitting laser
within the LiDAR system. The interpolated image, while still only single-band
(i.e., monochromatic), can be considered spectrally distinct. Accordingly, there
exists the opportunity of seabed mapping based upon blue-green reflectance.
Bathymetric sensor-packages that collect LiDAR intensity include the Optech
SHOALS system (Fig. 5.4), the Tenix LADS ADS Mk II (Fig. 5.5), and HawkEyeII. For SHOALS and CZMIL, first-generation versions of these systems did not
return a spectrally calibrated signal, and therefore the resulting soundings could
not be used in a classification that relied upon spectral albedo. It was therefore
common that the data were interpolated manually, with a user digitizing reef
features (Walker et al. 2008), or alternatively, a classifier was devised that operates
on the uncalibrated laser backscatter values (Filin 2004; Arefi and Hahn 2005;
Collin et al. 2008). In either case, good ground-truth was demanded in the form of
seabed descriptions, photographs, or video.
Neither airborne topographic nor bathymetric LiDAR can be collected through
clouds or dense haze/smoke; however, the data are acquired at relatively low
altitudes, often below cloud-level. This is unlikely to remain the case as the nextgeneration LiDAR, which now boasts much higher pulse repetition frequencies
than today’s instruments, and are able to acquire data at altitudes in excess of
5,000 m, while still maintaining acceptable point spacing. Providing that visiblespectrum imagery is not being captured concurrently and that aircraft safety
considerations can be adequately addressed, LiDAR surveys can also be flown at
5 LiDAR Overview
123
seabed. For coastal mapping, this problem may be resolved through the use of a
temporally shortened transmitted pulse (\2 ns), or may be obviated by combining
successive flights at low-tide with a topographic LiDAR, and at high-tide with a
bathymetric LiDAR (Pastol et al. 2007). Modern LiDAR systems address this
problem by employing ‘‘shallow-water’’ algorithms and waveform deconvolution
techniques. Though experimental, the shallow-water problem can alternatively be
tackled using a statistical relationship that links water depth to the Raman
waveform shape (Pe’eri and Philpot 2007). Since reef-tops tend to be considerably
shallower than 2 m, these advances are very relevant for the successful application
of LiDAR as a complete-system for coral reef mapping.
State-of-the-art LiDAR systems are calibrated and have the capability to capture reflectance data from the returning pulse, in addition to the three-dimensional
coordinates of the laser returns (Lillesand et al. 2004; Tuell and Park 2004; Tuell
et al. 2005). Like the strength of radar returns, the intensity of LiDAR ‘‘echoes’’
varies with the wavelength of the source energy and the reflectance spectrum of
the material that reflects the downwelling laser. For bathymetric applications, such
radiometric information, termed LiDAR ‘‘intensity’’, can aid in the identification
of seabed character. Intensity values are extracted from the returning laser
waveform and represent the magnitude of the bottom return. Via interpolation, an
intensity image of the seafloor can be created from spot soundings. To capture a
true reflectance image, it is necessary to employ a calibrated system and all
environmental and system response parameters must be appropriately modelled.
The reflectance image is captured at the wavelength of the transmitting laser
within the LiDAR system. The interpolated image, while still only single-band
(i.e., monochromatic), can be considered spectrally distinct. Accordingly, there
exists the opportunity of seabed mapping based upon blue-green reflectance.
Bathymetric sensor-packages that collect LiDAR intensity include the Optech
SHOALS system (Fig. 5.4), the Tenix LADS ADS Mk II (Fig. 5.5), and HawkEyeII. For SHOALS and CZMIL, first-generation versions of these systems did not
return a spectrally calibrated signal, and therefore the resulting soundings could
not be used in a classification that relied upon spectral albedo. It was therefore
common that the data were interpolated manually, with a user digitizing reef
features (Walker et al. 2008), or alternatively, a classifier was devised that operates
on the uncalibrated laser backscatter values (Filin 2004; Arefi and Hahn 2005;
Collin et al. 2008). In either case, good ground-truth was demanded in the form of
seabed descriptions, photographs, or video.
Neither airborne topographic nor bathymetric LiDAR can be collected through
clouds or dense haze/smoke; however, the data are acquired at relatively low
altitudes, often below cloud-level. This is unlikely to remain the case as the nextgeneration LiDAR, which now boasts much higher pulse repetition frequencies
than today’s instruments, and are able to acquire data at altitudes in excess of
5,000 m, while still maintaining acceptable point spacing. Providing that visiblespectrum imagery is not being captured concurrently and that aircraft safety
considerations can be adequately addressed, LiDAR surveys can also be flown at
5 LiDAR Overview
123