Geoscience Reference
In-Depth Information
simultaneously recording reflectances in the visible to short-wave region of the electromagnetic
spectrum, the canopy reflectance associated with these spatial characteristics may be used to provide
information on the biophysical characteristics of vegetation (Goudriaan, 1977). To predict the
vegetation response to external stresses, it is essential to identify biophysical characteristics observ-
able by remote sensing techniques that have well-defined connections to vegetative community
type and condition.
In complex vegetation communities, canopy structure and leaf spectral properties are biophysical
characteristics that can vary in response to changes in vegetation type, environmental conditions,
and vegetation health. These changes can modify the spectrum of light reflected from the canopy
and thus directly influence the remotely sensed signal. Transformed into reflectance, variations in
the image are directly related to changes in the canopy properties broadly defined by the leaf
composition, canopy structure, and background reflectance. Direct links, however, cannot be inferred
unless vegetation type covaries directly and uniquely with these canopy parameters, or when one
canopy property dominates the canopy reflectance (e.g., leaf reflectance). Historically, limited
ground-based observations circumvented the need for directly incorporating variation in canopy
properties into the remote sensing classification by defining reflectance ranges (e.g., class ranges)
that incorporate within-type canopy variability and acceptable between-vegetation-type classification
errors. Currently, the trend is to transform the temporal patterns revealed in the remote sensing data
into quantitative rate determinations to support qualitative judgments of external effects on these
resources (Lulla and Mausel, 1983). As we strive to extract more detailed and accurate information
about vegetation class variability, a greater understanding is needed of how each canopy property
(e.g., canopy structure) influences the canopy reflectance portion of the remotely sensed signal.
Leaf spectral properties have been directly related to vegetation type and stress and are general
indicators of the leaf chlorophyll, water content, and leaf biomass. Numerous studies have related
the canopy structure variable of leaf area index (LAI) to vegetation type, health, and phenology
(Goudriaan, 1977). In essence, to map vegetation type, and especially to monitor status, it is
necessary to relate, both individually and in aggregate, changes in leaf spectral properties and
structural and background parameters to changes in the canopy reflectance. In the pursuit of
extracting more detailed and accurate information about vegetation type and status from remote
sensing data, our goal is to provide an accurate assessment of canopy structure that will not covary
with leaf spectral and background properties with respect to location or time. As part of this goal,
the canopy structure indicator must be ultimately linkable to the remote sensing signal in complex
wetland and adjacent upland forest environments. Our challenge is to provide this information
based upon routine measurements that are cost-effective and easily implemented into operational
resource management and verified and calibrated with current operational ground-based measure-
ments (Teuber, 1990; Nielsen and Werle, 1993).
This chapter will examine light attenuation profiling as an indicator of changes in marsh canopy
structures. Reported here are techniques that were tested and implemented to gain a useful measure
of canopy light attenuation over space and time. Within the constraints of the data collected, the
consistency, reliability, and comparability of the collected light attenuation data are related to the
(1) area sampling frequency (horizontal spacing between profile samples), (2) canopy profile
(vertical) sampling frequencies, (3) exclusion of atypical canopy structures, and (4) collections at
different sun elevations. In addition, we present some relationships observed between and within
coastal wetland types and changes in the canopy structural properties. These relationships are
presented to indicate the spatial and temporal stability of these biophysical indicators as related to
mapping and monitoring with remote sensing imaging.
5.1.1
Marsh Canopy Descriptions
Measurements of canopy light attenuation and canopy reflectance spectra were collected at 20
marsh sites (30
¥
30 m) in coastal Louisiana and at 15 marsh sites in the Big Bend area of coastal
