Agriculture Reference
In-Depth Information
indicate the distance to the soil surface as well as to the top of the canopy. Hence
from the difference of these distances, it can sense the apparent crop height, which
in turn is well correlated to the biomass dry matter. In experiments with winter-
wheat and growth stages between EC or BBCH 21 and stepwise up to EC or
BBCH 69, the biomass was indicated with coefficients of determination (r 2 )
between 0.79 and 0.94. Contrary to sensing via canopy resistance against bend-
ing, the biomass was also precisely recorded in early developmental stages
between BBCH 21 and 32. Hence this rather simple physical method could be
used even for the first top dressing of small cereals. However, practical experi-
ences in this direction do not yet exist.
9.4.7
Cell Sizes or Resolution
An important question is the spatial resolution in site-specific nitrogen application.
In case the resolution that is obtained is too low, and consequently the cell sizes are
too large, any site-specific application might be useless (see Chap. 2 ) . However, it is
essential to differentiate between
the cell sizes for site-specific distributing (spreading or spraying)
the cell sizes for site-specific sensing of the crop status during fertilizing
the cell sizes when recording the heterogeneity of crops is done.
An adequate cell size in site-specific distributing is the final aim. But having a
distributing cell size that is below that of sensing is useless. To be effective, the cell
sizes that hold for site-specific distributing must be as large or larger than those that
are sensed. Usually the cell sizes in proximal sensing are much smaller and aver-
aged during processing because of the speed and frequency with most optical meth-
ods. Yet a prerequisite of having adequate cell sizes both for distributing and for
sensing during fertilizing is knowledge about the heterogeneity of crops.
Recording the heterogeneity of crops in respect to the nitrogen supply can easily
be done by proximal reflectance sensing of crop canopies with a high frequency.
This method can provide more than 100 basic signals per second for small sampling
areas and the field-transects thus recorded or mapped are very accurate if a precise
georeferencing method is used (Thiessen 2002 ). The semivariograms that are
obtained from the transects can supply the upper limits of cell sizes or of cell side
lengths as outlined in Fig. 2.5 . Exceeding this upper limit of cell side length dete-
riorates the precision in site-specific management. The upper limit of cell side
length sometimes is called the mean correlated distance.
Results from this method of recording the heterogeneity by sensing of red
edge inflection indices and hence of the nitrogen supply of small cereal crops in
Schleswig-Holstein, Germany are listed in Table 9.6 .
The upper limits of cell side lengths vary between 5.9 and 139.0 m. The reasons
for this wide span are difficult to find out. The geographic areas in Schleswig-
Holstein that are involved are
Search WWH ::




Custom Search