Agriculture Reference
In-Depth Information
frequencies and the real permittivities or dielectric constants of the respective soils.
Equally important are factors that have to do with the site-specific surface that is hit
by the radiation such as
the slope of the soil relative to the incident radar wave
the roughness of the soil surface at scales relative to the wavelength
the structure of any vegetation.
The last factor will be dealt with in Chap. 6 since it is important for sensing of
crops by means of radar waves. When it comes to water sensing in bare fields, only
the first two factors apply (Fig. 3.5 ) . But the effect of these factors on the signals
that are obtained with active radar sensing can be equal of even greater than the
effects of soil moisture (Engman and Chauhan 1995 ). So methods are needed that
make it possible to eliminate these disturbing effects.
Two approaches to cope with these problems seem to be feasible. The first
approach is based on the fact that the error or noise caused by varying slopes or by
changing roughness of the soil surface depends largely on the incidence angle with
which the radiation coming from the satellite hits the earth. This angle varies
between satellites and can often be adjusted. By systematically using different inci-
dence angles, the effects of varying slopes as well as different surface roughnesses
can be detected and be taken care of while the sensed data are processed (Baghdadi
et al. 2008 and Srivastava et al. 2003 ). However, up to now neither signals nor maps
that are corrected in such a way are available commercially.
The second approach relies on “ change detection ”. Its theoretical background is
that the factors which define the permittivity change differently on a time basis . The
soil moisture varies almost constantly and sometimes even rather fast. Contrary to
this, the slope of the soil always is about the same. And the roughness of the soil
surface as well as the structure of the vegetation do change, however, in most cases
much more slowly than the soil water content does. Hence repeated sensing within
defined time spans combined with sophisticated processing of the signals allows to
separate the effects of soil water from those of site-specific variations in slope, sur-
face roughness and even vegetation (Moran et al. 2000 ; Kim and van Zyl 2009 ). But
this method of “change detection by means of multi-temporal sensing” too is not yet
state of the art. So the future will have to show whether using several incidence
angles or change detection will provide a breakthrough for more precision in the
sensing of soil moisture.
However, if bare areas are flat and have smooth surfaces, it is even possible to do
without these special sensing and processing techniques as shown in Fig. 5.18 for
fields from a farming region in Western Turkey.
Summing up the outlook for soil water sensing by radar satellites: the perspec-
tives are encouraging, especially when taking into consideration that in the future
operating with various incidence angles and change detection might remove still
existing obstacles. Low frequencies and hence long waves will be needed in order
to obtain sufficient sensing depth. Perhaps it might be feasible in the future to pro-
vide farm machinery that is operating in fields in an online and on-the-go manner
with signals or maps about the respective site-specific soil water situation in a
Search WWH ::




Custom Search