Agriculture Reference
In-Depth Information
not infected groups was possible as early as 1 or 2 days after inoculation for powdery
mildew and leaf rust respectively.
In short, the problems with mixed effects from nitrogen supply and from fungal
infections might be dealt with by
fusing
refl ectance and fl uorescence sensors. The
latter, however, are not yet capable to sense fungi in an on-the-go mode in fi elds.
And the multitude of different fungi and various crops still calls for much research.
11.6
Summary and Prospects
A pinpointed, site-specifi c application of fungicides is one of the most ambitious
and challenging aims in precision farming. Because extraordinary high spatial- as
well as temporal resolutions are needed. If these resolutions can be realized in an
on-the-go mode, very substantial savings in fungicides and consequently environ-
mental reliefs might become feasible.
Yet as long as such extraordinary high spatial resolutions cannot be obtained in
practice, site-specifi c application based on the varying biomass or leaf-area-indices
of crops must be regarded as the sole present alternative. This method fi ts well to a
precautionary, prophylactic approach of spraying against fungi. Though prophylac-
tic methods hardly provide options for drastic reductions in the use of fungicides,
they might even be needed in combination with pinpointed, site-specifi c applica-
tions. The timeliness of both methods might be greatly improved by local epidemic
forecasts of extension services.
The pinpointed approach might rely either on refl ectance or on fl uorescence.
Both methods still are not state of the art, and its prospects for applications in prac-
tice are diffi cult to predict. Solely refl ectance sensing for biomass and for nitrogen
is well established. This technique depends largely on leaf-area-indices and chloro-
phyll. Fluorescence can be based on either emission spectra that sense chlorophyll,
on excitation spectra that are infl uenced by phenols or on its kinetics that depends
on the photosynthesis.
The earliest detections of fungal infections theoretically can be provided when
the indications occur via phenols or via the photosynthetic process. Hence in this
respect, sensing by the blue-green fl uorescence or by the kinetics of the variable
fl uorescence is an advantage. Whenever the indications rely on biomasses, leaf-
area-indices and chlorophyll, the earliest detection is postponed by a few days until
visual symptoms like chlorophyll degradation or other canopy alterations occur.
This holds regardless of the sensing method that is used for these indicators.
Whether in practice any earliest possible detection as outlined above will be
feasible and reasonable, depends not only also on the ability of sensing but also
on control techniques that deal with the tiny initially infected loci. Since manag-
ing the spreading of fungicides with a temporal precision of 1 or 2 days might be
diffi cult in practice anyway, the use of refl ectance sensing methods should not be
excluded. For any pinpointing approach, sophisticated scanning, processing and
spraying techniques will be essential.
