×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Remote sensing for resources development and environmental management
Author
Damen, M. C. J.

very special conditions to satisfying results /McDo
nald, Hall 1978/. To improve the performance of the
remote sensing technique for yield estimation some
more detailed and complex investigations are necessa
ry.
2 THE CONCEPTION OF THE PROJECT
Main topic of the project is the improvement of the
power of remote sensing techniques for yield estima
tion of agronomic targets. That means the improvement
of accuracy and reliability of the yield estimation
and, in contrast to already existing methods, the
application onto the typical ground cover characte
ristics in Europe with relatively small fields and
frequent changes in the cultivated plants.
The spectral data which will found the base in a
future stage of application will originate from
spaceborne satellite sensors as they are used today.
However, the spectral bands of the future systems
should be stronger designated to the agronomic appli
cation.
It is aimed to approach the results for relations
between reflected radiation and plant characteristics
as they were found in the laboratory. The realization
of this demanding aim needs in a first step a very
extensive investigation of the spectral and agronomic
parameters and their interaction. This will be done
by a very broad data collection procedure in the
field using sophisticated equipment.
The main aspects of the project which should help to
indicate how to realize the improvement and which
should give better understanding of the relationships
between plants and radiation read as follows:
1. Taking a great number of frequently repeated
field measurements allowing to record the influence
of the yield relevant and plant typical phenology
/Doerfel, 1978/.
2. Collection of a broad spectrum of agronomic
parameters like physiology, morphology and other
informations describing the stage of the plant
growth and health.
3. Realization of spectral field measurements cove
ring a broad window of the electromagnetic spectrum
ranging from 0.4 - 2.2 jjm.
4. High spectral resolution in the measurements
(0.002 - 0.004pm) assuring good sensitivity in the
detection of qualitative changes in the plant re
flection.
5. Extensive compensation of effects originating
from exterior influences which may reduce the qua
lity of the measurments.
One major influence exists in the atmospheric situa
tion during measurement which has to be taken into
account, because very different illumination condi
tions have to be expected for the high frequent
repititions. Other important parameters are the mi
croclimatic situation, the surface topography of the
plant targets and the geometric relations between
light source, targets and measuring device.
3 ORGANIZATION OF THE MEASUREMENTS
3.1 Plant cultivation
The investigations will concentrate on two different
plant species of different phenotype. The first one,
way
Figure 1 : Arrangement of cultivated plant species
winter wheat, is an often used plant for which alrea
dy successful yield estimation could be attained
/McDonald, Hall 1978/. The second one, sugar beets,
is of heterogenous shape and serves as touchstone for
the improvement to be expected.
The two plants are cultivated in a great field with
six different spots. In each spot we have one of
three plant species (sugar beet, winter wheat, bar
ley), which are cultivated following the rotation
principle (cf. fig.l).
For each plant in consideration we have three indé
pendant cultivation parameter. These parameters are
plant density (D^,D 2 ), seed time (T^»T 2 ) and nitrogen
fertilization (NpN 2 ) for sugar beets and fungizid
treatment (F^,F 2 ), wheat species (S^,S 2 ) and nitrogen
fertilization (N^,N 2 ) for winter wheat. These parame
ters were chosen to produce distinct differences in
plant growth, health and plant production in order to
have good pre-conditions to find typical changes in
the reflection characteristic.
All parameters are varied with each other so that we
have in total 8 plots of different cultivation (cf.
fig. 2)
Each plot has a special organization for the collec
tion of agronomic and radiometric data. It is divided
into 7 s
sides a
the adja
laborato
be harv
determin
Figure 3
3.2 Orga
All meas
as possi
plants
laborate
The met
spectral
the pie
tures j
situatic
videotaf:
cycle tc
In eact
three di
approxin
for the
positior
the gre
reflecti
one tare
about tt
3.3 Meas
The col 1
- 1. t
These
devele
logic
- 2. Í
This
the n
graph)
of sir
- 3. I
The r
situai
ronmer
energj
avail«
portar
wind,