1. Stand Structure in Terrestrial Ecosystems
and orientation (zenith and azimuth). Foliar orientation can be measured using a protractor, compass,
and ruler (Ross 1981, Daughtry 1990). Lang (1990)
describes a mechanical probe for measuring canopy
structure that consists of four linked arms whose
geometry is precisely measured to calculate the
three-dimensional location of the probe tip within
a sample volume of 1 m 3 • Individual leaves are
characterized by several to many point locations,
and leaf area and orientation are derived
analytically.
Martens et al. (1991) describe a method for reconstructing three-dimensional structure of tree
canopies and demonstrate its applicability in a walnut orchard. The process involves stratified sampling of stem size and orientation in different size
classes at the scale of the orchard (stand), tree, and
branch levels to yield data that can be synthesized
at the orchard level. Based on comparisons with
plumb-line measures of three-dimensional position,
the approach provides highly accurate estimates of
canopy architecture and LAD. Fournier et al.
(1996) characterized the canopy architecture of
several conifer plantations using a "tree vectorization" method that involves sampling of structural
components (trunk, branching structure, foliage
distribution, and leaf size and orientation) and application of similarity principles to simulate individual trees. These trees were then used to simulate
hemispherical photographs that closely resembled
actual photographs from the field sites and allowed
systematic investigation of effects of scale dependence, nonrandom foliar and branch distributions,
and solar zenith angle on light extinction in forest
canopies. Most recently, Sinoquet and Rivet (1997)
describe a method that utilizes a three-dimensional
digitizing device to reconstruct tree canopy architecture based on branching pattern, plant geometry,
and shoot morphology. The method worked well
when applied to a 7-m walnut tree. Needless to say,
all of the aforementioned direct measurement methods for tree canopies work best for simple stand
structures, such as plantations, and for relatively
short, open-grown (symmetrical) trees.
Indirect Methods
Ross (1981) provides a theoretical basis for characterizing vertical leaf area distribution and leaf angle distribution (LAD) based on projected leaf area
25
measured from multiple points of reference. Using
these concepts, Vanderbilt et al. (1990) employed a
laser much like a pin probe to estimate the probability of the beam being intercepted by various canopy elements as a function of height in the canopy
and view direction. Vertical foliar profile and LAD
were estimated using these probabilities.
Machine vision techniques have been successful
at reconstructing simple three-dimensional objects
and topographic surfaces from digital stereo imagery. However, these same methods have met with
only limited success when applied to complex plant
canopies, the main problem being the number of
hidden leaf surfaces. Ivanov et al. (1994) were able
to partly overcome this problem to reconstruct a
maize canopy by taking sequential photos of the
same canopy after removing successive leaflayers.
It remains to be seen whether approaches based on
machine vision will eventually evolve to handle the
very heterogeneous structures encountered in natural plant stands.
Acknowledgments. We gratefully acknowledge
John Day's contribution to the sections on radar
remote sensing, Kim Brown for LAI measurements
at Wallula and Keir Keightley for help with figures.
Helpful advice and references were provided by
Bryan Blair, Peng Gong, Joe Means, Nalini Nadkarni, and Steve Running. Data for Figures 1.1 to
1.3 were acquired with funding supplied by
WESTGECINIGEC award WEGEC 95-062A. Partial support for F. Davis was provided through a
NASA SIR-C/x-SAR project (Jet Propulsion Lab
Contract # 958468).
References
Adams, J.B.; Smith, M.D.; Gillespie, A.R. Imaging spectrometry: Interpretation based on spectral mixture
analysis. In: Pieters C.M.; Englert P., eds. Remote
Geochemical Analysis: Elemental and Mineralogical
Composition. Vol. 7. New York: Cambridge Univ. Pr.;
1993: 145-166.
Andrew, M.H.; Noble, LR.; Lange, R.T.; Johnson, A.w.
The measurement of shrub forage weight: Three methods compared. Aust. Range J. 3:74-82; 1981.
Asrar, G.; Myneni, R.B.; Kanemasu, E.T. Estimation of
plant-canopy attributes from spectral reflectance measurements. In: Asrar, G., ed. Theory and Applications
and orientation (zenith and azimuth). Foliar orientation can be measured using a protractor, compass,
and ruler (Ross 1981, Daughtry 1990). Lang (1990)
describes a mechanical probe for measuring canopy
structure that consists of four linked arms whose
geometry is precisely measured to calculate the
three-dimensional location of the probe tip within
a sample volume of 1 m 3 • Individual leaves are
characterized by several to many point locations,
and leaf area and orientation are derived
analytically.
Martens et al. (1991) describe a method for reconstructing three-dimensional structure of tree
canopies and demonstrate its applicability in a walnut orchard. The process involves stratified sampling of stem size and orientation in different size
classes at the scale of the orchard (stand), tree, and
branch levels to yield data that can be synthesized
at the orchard level. Based on comparisons with
plumb-line measures of three-dimensional position,
the approach provides highly accurate estimates of
canopy architecture and LAD. Fournier et al.
(1996) characterized the canopy architecture of
several conifer plantations using a "tree vectorization" method that involves sampling of structural
components (trunk, branching structure, foliage
distribution, and leaf size and orientation) and application of similarity principles to simulate individual trees. These trees were then used to simulate
hemispherical photographs that closely resembled
actual photographs from the field sites and allowed
systematic investigation of effects of scale dependence, nonrandom foliar and branch distributions,
and solar zenith angle on light extinction in forest
canopies. Most recently, Sinoquet and Rivet (1997)
describe a method that utilizes a three-dimensional
digitizing device to reconstruct tree canopy architecture based on branching pattern, plant geometry,
and shoot morphology. The method worked well
when applied to a 7-m walnut tree. Needless to say,
all of the aforementioned direct measurement methods for tree canopies work best for simple stand
structures, such as plantations, and for relatively
short, open-grown (symmetrical) trees.
Indirect Methods
Ross (1981) provides a theoretical basis for characterizing vertical leaf area distribution and leaf angle distribution (LAD) based on projected leaf area
25
measured from multiple points of reference. Using
these concepts, Vanderbilt et al. (1990) employed a
laser much like a pin probe to estimate the probability of the beam being intercepted by various canopy elements as a function of height in the canopy
and view direction. Vertical foliar profile and LAD
were estimated using these probabilities.
Machine vision techniques have been successful
at reconstructing simple three-dimensional objects
and topographic surfaces from digital stereo imagery. However, these same methods have met with
only limited success when applied to complex plant
canopies, the main problem being the number of
hidden leaf surfaces. Ivanov et al. (1994) were able
to partly overcome this problem to reconstruct a
maize canopy by taking sequential photos of the
same canopy after removing successive leaflayers.
It remains to be seen whether approaches based on
machine vision will eventually evolve to handle the
very heterogeneous structures encountered in natural plant stands.
Acknowledgments. We gratefully acknowledge
John Day's contribution to the sections on radar
remote sensing, Kim Brown for LAI measurements
at Wallula and Keir Keightley for help with figures.
Helpful advice and references were provided by
Bryan Blair, Peng Gong, Joe Means, Nalini Nadkarni, and Steve Running. Data for Figures 1.1 to
1.3 were acquired with funding supplied by
WESTGECINIGEC award WEGEC 95-062A. Partial support for F. Davis was provided through a
NASA SIR-C/x-SAR project (Jet Propulsion Lab
Contract # 958468).
References
Adams, J.B.; Smith, M.D.; Gillespie, A.R. Imaging spectrometry: Interpretation based on spectral mixture
analysis. In: Pieters C.M.; Englert P., eds. Remote
Geochemical Analysis: Elemental and Mineralogical
Composition. Vol. 7. New York: Cambridge Univ. Pr.;
1993: 145-166.
Andrew, M.H.; Noble, LR.; Lange, R.T.; Johnson, A.w.
The measurement of shrub forage weight: Three methods compared. Aust. Range J. 3:74-82; 1981.
Asrar, G.; Myneni, R.B.; Kanemasu, E.T. Estimation of
plant-canopy attributes from spectral reflectance measurements. In: Asrar, G., ed. Theory and Applications