Robert K. Hall1, Russell Watkins2, Steven B. Moore3, Sandra J. Gregory3, Daniel T. Heggem4, K. Bruce Jones4, and Phil Kaufmann5

1U.S. EPA, Region IX, ORD, NERL, ESD, LEB; 2Spectrum Mapping, LLC, 3907 SW S*1 Place, Gainesville, FL 32607; 3Bureau of Land Management, Reno, NV 89520;
4U.S. EPA, ORD, Environmental Sciences Division, Landscape Ecology Branch, Las Vegas, NV 89119; 5U.S. EPA, ORD, NHEERL, Western Ecology Division, Corvallis, OR

Figure 4 Watershed Morphology and Invasive
Species Classification: Humboldt River, NV

Structural physical habitat attributes include indices of stream size, channel gradient, substrate
size, habitat complexity and cover, riparian vegetation cover and structure, human disturbances and
channel-riparian interaction. These habitat attributes will vary naturally even in the absence of
human disturbances according to their ecological setting.

Light Detection and Ranging (LIDAR) is an airborne laser system that provides information on
4 topography, as well as height and structure of vegetation and other ground features. LIDAR-derived
digital elevation models (OEMs) at 1 meter (~39 inches) horizontal and 0.3 meter (-12 inches)
vertical resolution allow for the measuring of approximate channel dimensions (width, depth), slope,
channel complexity (residual pools, volume, morphometric complexity, hydraulic roughness),
riparian vegetation (height) dimensions of riparian zone, anthropogenic alterations and
disturbances, channel and riparian interaction, and for detailed hydrologic and sedimentation
modeling at the watershed to sub-basin levels.

Hyperspectral imagery is comprised of narrow spectrai bandwidths (1-5nm) with a continuous
spectrum in the visual to near infrared portion of the electromagnetic spectrum. Hyperspectral
imagery offers the advantages of high spectral and spatial resolution allowing for the detection and
identification of riparian vegetation and natural and anthropogenic features not possible with
satellite imagery. High resolution remote sensing provides unique capabilities in detecting a variety
of features (e.g., vegetation type, sedimentation, water column constituents, potential sources of
non-point source pollution) and indicators of environmental health and condition. When combined,
or fused, these technologies comprise a powerful geospatia! data set for assessing and monitoring
environmental characteristics and condition.

Riparian Vegetation

Canopy cover is essential for moderating stream
temperatures, providing habitat and an indicator of
potential aquatic community presence.

Figure 1. Stream Reach Transects

Hyperspectral imagery integrated with LIDAR data can
determine vegetation type (e.g., native plants, noxious,
invasive species), structure, height and distribution.

Stream Dimensions

Streams are dynamic and subject to relatively rapid
change in channel shape. Jo deteimine the integrity of the
fish habitat, it is important to measure:

width

bankfull height and width - highest flow achieved

during a major storm event

thalweg depth - deepest part of the channel

depth cross-sections, and

flow rates as indicators of stream si

The stream longitudinal profile and cross-sections (see
Figure 2 and 3) provide a measure of the stream's
dimensions.

Conclusions: High resolution RS: Advantages

There are multiple advantages to hyperspectral & LIDAR data
-100% coverage of the ground surface to approximately 1
meter (~3 feet) pixel resolution.

-	Spatially accurate within several inches using corrected

Differential Global Positioning System (DGPS).

-	Covers larger areas than a single stream reach

assessment.

-	Narrow, contiguous spectral bands in hyperspectral

imagery allows better discrimination of vegetation
types, land cover and water clarity.

-	Measures canopy height and density, riparian width and

fragmentation.

-	Can measure channel form and reveal subtle changes in

topography and bathymetry.

-	Can be used to develop quantitative water quality

indicators.

Figure 2. Stream Reach Transect A

Figure 3. Stream Reach Transect H

Major advantage is the cost to benefit relative to high density
ground sampling to cover the same land area.


-------