Agriculture Reference
In-Depth Information
or on leaf-area-indices probably holds for most crops. Belanger et al. (
2005
) as well
as Jongschaap (
2001
) published similar results for potatoes.
A reasonable conclusion from these results is to use the product of both indica-
tors of nitrogen application as the criterion for sensing. This product of the chloro-
phyll concentration within the leaves and the leaf-area-index can be defined as the
chlorophyll per unit of ground area
. The curves in Fig.
9.17
, right, indicate that
this product is sensitive to the nitrogen supply.
Using the chlorophyll concentration per unit of ground area as criterion is analo-
gous to sensing methods that are oriented at defining the yield potential of crops as
or for
nitrogen fertilizing
may not be quite the same. Because sensing for yield
potentials is oriented at detecting the photosynthetic capacity of crops per unit of
field area. And sensing for nitrogen fertilizing aims at recording the site-specific
nitrogen uptake of the crop. Strong correlations between these sensing objectives
can be expected, but differences should be looked forward too as well. Differences
can result from the fact that not all the nitrogen in crops is within its chlorophyll.
And the yield potential of crops depends on many growth factors, not only on
nitrogen.
9.4.1
Fundamentals of Nitrogen Sensing by Reflectance
Nitrogen has two effects on the reflectance of a plant canopy (Fig.
9.19
). Firstly, it
increases the chlorophyll concentration per unit area in the leaves. Thus, more light
is absorbed and consequently the reflectance decreases. However, this effect occurs
only with the visible light, the photosynthetically active radiation (PAR).
Secondly, as mentioned above, nitrogen has a very pronounced effect on the
growth of plant mass and, therefore, on the leaf-area-index of a crop. Theoretically,
the higher the leaf-area-index, the more incident solar radiation should be scattered
back by the canopy instead of by the bare soil. Yet in reality this is important only
in the near-infrared region, since in these wavebands light is barely absorbed by
plant pigments. As a result, the effect of nitrogen supply in the near-infrared range
is opposite to that in the visible range.
The rather steep slope between the red and the near infrared reflectance is gener-
ally denoted as the red edge. It has a concave and a convex part, which meet at the
red edge inflection point
. This point moves to longer wavelengths when the supply
with nitrogen is improved.
How can the information of the reflectance curves be processed for the con-
trol of the nitrogen supply of crops? Using full-spectrum methods is compli-
cated as a result of the opposite effect of the nitrogen in the visible and the
near-infrared wavelength range. Furthermore, these methods are associated with
rather high costs for the sensing instruments. Using discrete narrow wavebands
instead of these methods can simplify the sensing. The problem however is,
finding the most suitable narrow wavebands including the best index that com-
bines these mathematically.
