shifts to the longer wavelength, it does not immediately reach the 726nm level. At the beech site, canopy cover 
and position of maximum slope appear to shift between two extremes simultaneously. At both sites the large 
variations in the position of maximum slope are caused by changes in the particular association of scene 
elements presented to the sensor. While the trees are in-leaf, leaves are the dominant scene elements, and the 
maximum slope is positioned at around 726nm. During leafless periods, when the major scene elements are 
bark and litter, the red edge moves to a shorter wavelength domain, and becomes a less distinct spectral feature. 
In previous studies, red edge position has been shown to be dependent on the concentration of chlorophyll and 
other photosynthetic pigments in leaves and canopies (Horler et al., 1983, Curran et al., 1991). This 
experiment suggests that the dev elopment of tree leaves causes a substantial shift in the location of the red 
edge, to where it remains fixed at 726nm, and only deviates from this while leaf pigment concentrations are 
known to be changing, during early leaf development and late leaf senescence. 
ASH 
Red-edge 
Cover 
90 
80 
70 b 
60 8 
50 £ 
40 g 
30 u 
20 £ 
10 
Figure 6. Seasonal variations in the position of maximum slope (in region 680-750nm) and canopy cover. 
4. CONCLUSION 
The beech and ash canopies differ in terms of timing, magnitude and rate of change, of the seasonal variations 
in mean visible and near-infrared reflectance. These differences can largely be explained by contrasts between 
the sites in the characteristics of percentage canopy cover, canopy structure and leaf physiology . 
Variations in NDVI are strongly related to canopy cover over time, and NDVI remains high throughout leaf 
senescence, up until abscission. Visible and near-infrared reflectance and percentage canopy cover all exhibit 
different values at the start and end of the sampling year, this suggests that it is necessary to account for inter- 
annual variability by undertaking similar future studies over a number of years. 
The position of maximum slope in the reflectance spectra, was related to canopy cover over the sampling year. 
The position of maximum slope varied over a 25nm range, depending on the particular association of scene 
elements presented to the sensor, with ash and beech canopies differing in the timing of the shifts between the 
two domains. 
ACKNOWLEDGEMENTS 
The UK Natural Environment Research Council supported this work through Research Studentship 
GT/90/TLS/58, and the loan of spectroradiometers from the Equipment Pool for Field Spectroscopy. We 
gratefully acknowledge the assistance of Dr E. M. Rollin, R. Treves, W. Damon, and H. Hurd (Department of 
Geography, University of Southampton), E. Adams (Department of Human Morphology), and Dr J. M. Roberts 
(Institute of Hydrology, Wallingford). 
REFERENCES 
BOYER, M., MILLER, J., BELANGER. M. and HARE, E. (1988), Senescence and spectral reflectance in 
leaves of Northern Pin Oak (Quercus palusris Muenchh). Remote Sensing of the Environment, 25, 71-87.