489 
Symposium on Remote Sensing for Resources Development and Environmental Management / Enschede / August 1986 
Application of multispectral scanning remote sensing 
in agricultural water management problems 
G.J.A.Nieuwenhuis 
Institute for Land and Water Management Research (ICW), Wageningen, Netherlands 
J.M.M.Bouwmans 
Government Service for Land and Water Use (LD), Utrecht, Netherlands 
ABSTRACT: A method has been developed for the automatical mapping of évapotranspiration from digitally taken 
reflection and thermal images. This method has been tested in combination with field measurements and agro- 
hydrological model calculations. We found that an important improvement of the hydrological description of an 
area can be achieved by combining the remote sensing approach with conventional methods. Especially the spatial 
variation in soil physical characteristics can be determined more accurately by applying remote sensing 
techniques. Satellite and airplane images acquired after a dry period supply information about the occurrence 
of drought damage. Because of the fast changing weather conditions in humid areas like the Netherlands remote 
sensing in water management will mainly depend on scanning techniques from airplanes. 
1 INTRODUCTION 
Remote sensing techniques can be applied to obtain 
information about the hydrological characteristics 
of an area. In a remote sensing study project per 
formed in the eastern part of the Netherlands (East 
Gelderland) we developed a method to automatically 
derive an évapotranspiration map from reflection and 
thermal infrared multispectral scanning images. In 
formation about regional évapotranspiration of crops 
is important for optimal water management in agri 
culture and for the determination of the effect of 
man-made changes in the overall hydrological situa 
tion. 
In the framework of the mentioned study project 
experience is obtained especially with the detection 
of drought damage. From an economical point of view 
prevention of damage caused by excess of water is 
also important in the Netherlands. Untili now only 
Figure 1. Location of the three study areas in the 
Netherlands 
limited experience has been obtained with remote 
sensing like MSS techniques and radar. Probably remote 
sensing images taken directly after the winter period 
can supply information about the regional distribution 
of areas with water excess. At the beginning of the 
growth stage information about excess of water might 
be obtained through differences in germination. 
We are currently testing the usefulness of remote 
sensing to solve water management problems. In several 
projects we are investigating whether the additional 
use of remote sensing is technically and financially 
advantageous in determining the effects of hydrolog 
ical measures. 
In this paper we give an overview of the MSS re 
search. The concerning study areas are indicated in 
Figure 1. 
2 THEORY AND METHODS 
2.1 Relation between instantaneous crop temperature 
and 24 hour évapotranspiration rate 
For the translation of instantaneous to daily values 
the TERGRA-model has been developed (Soer 1977). In 
terpretation of thermal images with the aid of the 
TERGRA-model is rather complicated, however, because 
of the large number of input parameters that are re 
quired. For an operational use of remote sensing 
techniques simplified procedures are of importance. 
For this aim Jackson et al. (1977) related midday 
surface-air temperature differences linearly to 24 
hour évapotranspiration and net radiation values. To 
estimate the slope of this relationship a crop depen 
dent analytical expression has been derived by Seguin 
and Itier (1983). 
With the approach of Jackson absolute estimates of 
crop évapotranspiration are obtained. As heat images 
are mainly applied to determine reductions in évapo 
transpiration this approach has been adjusted 
(Nieuwenhuis et al. 1985). 
The surface air temperature difference in the ap 
proach of Jackson has been replaced by the temperature 
difference that exists between the crop that is 
transpiring under the actual restriction of the soil 
moisture condition and that transpiring under optimal 
soil moisture conditions (T C ~T*). The net radiation 
term was replaced by the 24 hour potential évapotrans 
piration rate of the crop. With this adjustment they 
obtained: