226 
the smaller CV values for the estimates of LAI. How 
ever, these latter CV values were larger than those 
for infrared reflectance. Similar conclusions could 
be drawn for other field trials (Clevers, 1986c). 
The following practical procedure has been elabo 
rated by Clevers (1986c): 
In one field trial, the regression function of LAI 
on corrected infrared reflectance was established 
by analysing a few additional plots (a training set), 
in which both LAI and reflectances were ascertained. 
The inverse of a special case of the Mitscherlich 
function was used for describing the regression 
function of LAI on the infrared reflectance correct 
ed for background. Subsequently, this regression 
function was applied for estimating LAI in the en 
tire field trial. To date there is insufficient evi 
dence that the regression curves of different crops 
or cultivars are easily transferable, or that the 
curve of one growing season can be applied in the 
following seasons, although the results of Clevers 
(1986c) pointed in that direction. So, conventional 
field measurements are still needed. 
Verhoef, W., 1984. Light scattering by leaf layers 
with application to canopy reflectance modelling 
the SAIL model. Rem. Sens. Envir. 16: 125-141. 
Youkhana, S.K., 1983. Canopy modelling studies. 
Colorado State Univ., PhD., 84 pp. 
5 CONCLUSIONS 
1. If the ratio between the reflectance factors of 
bare soil in any pair of the green, red and infra 
red spectral bands is nearly one, the corrected 
infrared reflectance may be ascertained as the dif 
ference between the measured infrared and red re 
flectance . 
2. At the vegetative stage of cereals, the inverse 
of a special case of the Mitscherlich function, 
namely the one passing the origin, was suitable for 
describing the regression function of LAI on cor 
rected infrared reflectance. 
3. If LAI was estimated by reflectance values, by 
using a regression curve of LAI on corrected infra 
red reflectance, the critical levels in testing for 
treatment differences were in general smaller than 
for the measured LAI of samples. This also applied 
to the coefficients of variation. Even at large LAI 
values (LAI 5-8) significant treatment effects could 
be distinguished by means of multispectral aerial 
photography. 
6 REFERENCES 
Bunnik, N.J.J., 1978. The multispectral reflectance 
of shortwave radiation by agricultural crops in 
relation with their morphological and optical 
properties. Thesis, Meded. Landbouwhogeschool Wa- 
geningen 78-1, 175 pp. 
Clevers, J.G.P.W., 1986a. The derivation of a sim 
plified reflectance model for the estimation of 
LAI. Proc. Seventh Int. Symp. on Remote Sensing, 
ISPRS Comm. VII, Enschede, The Netherlands. 
Clevers, J.G.P.W., 1986b. Multispectral aerial 
photography yielding well calibrated reflectance 
factors with high spectral, spatial and temporal 
resolution for crop monitoring. Proc. Third Int. 
Coll, on Spectral Signatures of Objects in Remote 
Sensing, Les Arcs, France. 
Clevers, J.G.P.W., 1986c. The application of remote 
sensing to agricultural field trials. Thesis (in 
press). 
Condit, H.R., 1970. The spectral reflectance of 
American soils. Photogram. Eng. Rem. Sens. 36: 
955-966. 
Goudriaan, J., 1977. Crop micrometeorology: a simu 
lation study. Thesis Landbouwhogeschool, Wageningen, 
249 pp. 
Stoner, E.R., M.F. Baumgardner, L.L. Biehl & B.F. 
Robinson, 1980. Atlas of soil reflectance proper 
ties. Agric. Exp. Station, Purdue Univ., W-Lafay- 
ette, Indiana, Res. Bull. 962, 75 pp.