A
National
Science Foundation Major Research Instrumentation Program
grant has provided funding for the purchase of a High Accuracy/Resolution
Landscape and Structure Characterization System (HARLS-CS). The HARLS-CS
is an integrated suite of instruments designed to rapidly and accurately
characterize the topographic and multi-spectral properties of the surfaces
of relatively small areas (e.g. 100s of square meters to 10s of hectares)
and structures (such as building facades) at very high resolution (25
mm - 0.5 m). The system provides coordinated three dimensional, multispectral
and metric photographic-based measurements necessary for a wide range
of mensuration, classification, and quantitative characterization analyses.
The system dramatically expands existing research activities in a number
of disciplines and multidisciplinary areas. It serves as a shared resource
in the fields of anthropology, archeology, architecture, biological
and agricultural engineering, community planning/policy, entomology,
geography, geology, and water resource management. It also plays a significant
role in expanding research training by providing advanced undergraduates
and graduate students access to critically important data as well as
new capabilities for the development of valuable methodological expertise.
The system is
comprised of an Optech
ILRIS 3D laser profiler, ASDI
field spectroradiometer, Duncan
MS4100 multispectral camera, Nikon
digital cameras, Trimble
5600/5700/5800 total station with GPS, supporting software (e.g.
Innovmetric PolyWorks
and EOS PhotoModeler) and towable
lift boom. This suite integrates seamlessly into a well-developed
set of existing research resources including a Raytheon
250 thermal imager, extensive multispectral and photogrammetric
software and hardware as well as existing research programs and institutional
structures. The HARLS-CS instruments are suitable for close range, terrestrial
and boom uses and the multispectral and photographic components can
be mounted in light aircraft, if needed.
Integrating
laser range data and image data (both panchromatic and multispectral)
is a growing area of research (see, for example, McIntosh et al. 2000)
and we expect to contribute to this research with the data collected
by the HARLS-CS. The University of Arkansas will closely collaborate
with the Department of
Civil and Environmental Engineering and Geodetic Sciences and the Byrd
Polar Research Center at the Ohio State University to develop new
and innovative techniques for fusing multi-source data collected by
HARLS-CS.
Multi-Scalar,
Multi-Sensor Research
It is well known
that the investigation of a wide range of physical, biological and human
processes requires investigations at a range of scales (c.f.
Ehleringer at al 1993, Goodchild 1992, Robinson 1950, Turner et al 1989,
Woodcock and Strahler 1987), where the term scale implies both the
magnitude of the study (geographic extent) and degree of detail (resolution)
(Goodchild and Quattrochi 1997). In
many projects research objectives are driven by the ready accessibility
of moderate resolution (e.g. landscape scale 1:12,000-1:100,000 scale
1 - 250 m pixel) sources such as MODIS, Landsat ETM, IKONOS, QuickBird,
USGS DEMs and/or aircraft sensors such as AVRIS and others.
These systems
provide multispectral measurements and three-dimensional (topographic)
characteristics derived from photogrammetric processing of stereo images
that have well understood properties and low cost. These multispectral
and topographic inputs have been used in a variety of analyses by the
research team members in areas that include crop characterization and
precision agriculture (Bajwa and Vories
2002), soil characterization (Bajwa
and Tian. 2001), water quality (Pionke
et al., 1997, Gburek and Sharpley 1998), forest entomology (Stephen
et al. 1997; Stephen and Browne, 2000), bathymetry and sedimentation
(Brown and Boss 2002), archeological
site characterization (Kvamme 2001),
architectural visualization, early hominid site analysis and visualization
(Nigro et al., in press, Lacruz et al.,
in press), urban morphological analysis (Burian
et al. 2002a), and urban air quality and dispersion modeling (Burian
et al. 2002b).
While studies
utilizing data from aircraft or satellite based sensors are very effective
they are substantially limited in the level of detail they can achieve.
Studies using such data are also constrained by researchers' inability
to control the scheduling of data acquisition from sources such as satellites.
It is also clear that there is a wide range of significant processes
that occur at a much finer scale (resolution) than that supported by
standard sources, frequently defined as site, structure, plot or field
levels. In the past, acquisition of measurements at this scale involved
manual measurements, traditional survey, or other time consuming approaches.
The HARLS-CS instrumentation is designed to fill the current gap in
readily accessible measurements needed to provide accurate, dense characterization
at the site/plot/field level of analysis that will dramatically expand
research analyses of processes at this scale. It takes advantage of
recent commercialization of both high resolution laser scanning and
multi-spectral cameras and software to process and integrate them. Measurements
are provided at large to very large scale (e.g. 1:2,400 - 1:50), with
small to very small pixel resolution (e.g. 50 - 0.25 cm pixel) and at
times appropriate to the specific study - providing critically needed
research data not otherwise available.
The HARLS-CS
will be initially used in research and research training by 13 faculty
and students in nine units across three colleges as well as by the Arkansas
Archeological Survey (a UA systems level unit). The University units
are currently using the system include The Center for Advanced Spatial
Technologies, The Arkansas Archeological Survey, The Archeo-Imaging
Lab, and the Departments of Anthropology, Architecture, Biological Sciences,
Biological and Agricultural Engineering, Civil Engineering, Entomology,
Geosciences, and Environmental Dynamics.
