Wh. Qin et Yq. Xiang, ON THE HOTSPOT EFFECT OF LEAF CANOPIES - MODELING STUDY AND INFLUENCEOF LEAF SHAPE, Remote sensing of environment, 50(2), 1994, pp. 95-106
Hotspot is a prominent feature of the reflectance distribution of a ve
getation canopy consisting of finite size foliage. It depends on the g
eometric dimension and spatial organization of vegetation elements, an
d therefore has a potential for diagnosing canopy geometric structure.
In the context of this study, we first reconcile different notations
used by previous workers. This leads to geometrically quantifying the
cross-correlation function, which is essential for the hotspot modelin
g. A comprehensive formulation for the hotspot effect at both leaf and
canopy levels is then developed, by generally parameterizing some bas
ic parameters such as mean area of shadows and overlap between shadows
cast by scatterers. A rectangle model is proposed to account for the
influence of noncircular shape of scatterers on the hotspot effect, an
d explicit expressions for both the cross-correlation function and the
hotspot width are obtained. It is shown that for a leaflike object, t
he angular hotspot width progressively broadens with an increase of m,
the ratio of mean leaf width to the length. For the whole canopy, the
relative distribution of the hotspot intensity mainly depends on this
ratio. That is, the cross-correlation function decreases more rapidly
for smaller m as the viewing direction diverges from the illumination
direction. As a result, canopy reflectance increases with m, particul
arly in the region around the hotspot point where the reflectance dist
ribution strongly relies on the ratio m. For m = pi/4, the rectangle m
odel produces nearly same result as the disk model. This indicates thi
s rectangle model is more realistic and flexible than those based on c
ircle-shaped scatterers (leaves) or their shadows, which is the common
assumption underlying in most existing hotspot models.