Geoscience Reference
In-Depth Information
In agricultural regions drought can have serious repercussions on the region's
economy. To monitor how vegetation cover is affected by drought it is possible to
use data from NOAA satellites, from SPOT-Vegetation and from near-infrared
images, in particular. The reflectance of the vegetation depends on the structure of
the cells of the individual plants as well as on their water content. If a vegetation
index (a combination of information produced by different spectral bands) is
calculated for different dates of the year it becomes possible to compare vegetation
cover for different times of the year. Amongst the most commonly used vegetation
indexes used, the NDVI (normalized difference vegetation index) [ROU 74] is used
for chlorophyllous vegetation (i.e. green plants) by using the red and near-infrared
bands of the satellite's sensors:
NDVI = (Near infrared - Red) / (Near infrared + Red)
[3.3]
In addition to cloud cover and the problems associated with atmospheric
absorption [BAR 87], images have been synthesized by using a method known as
the Maximum Value Composite (MVC) method. The most commonly used
synthesis method is the method that synthesizes NDVI maximum values, and this
usually takes place over a period of 10 days or over a monthly period. The MVC
minimizes the presence of clouds and reduces the effects of sighting and solar
angles. In addition, the MVC minimizes atmospheric effects caused by the presence
of aerosols and water vapor [HOL 86]. At the moment MVC NDVI syntheses are
commonly used to analyze vegetation cover on a global or continental scale.
However, even though the MVC NDVI minimizes some effects that interfere with
the radiometric values of the pixels, the effects are not completely removed from the
images.
It is difficult to interpret the NOAA images due to the spatial resolution (1 km at
nadir) of the AVHRR sensor. Monitoring vegetation cover using the NDVI is not
always a straightforward process. The NDVI can sometimes be low due to a lack of
vegetation (the vegetation may be stubble or ploughed) or due to the stage of the
vegetation's life cycle. The NDVI can also be low because of the lack of water that
the plants receive and this is reflected in their chlorophyll content. Therefore, it is
necessary to have ground knowledge of the vegetation cover to be able to
distinguish between the examples mentioned. This means that the NDVI cannot be
considered as a drought map on its own, as a water deficits map could be.
Conversely, the combined study of images taken during a year where there was no
water deficit (calculated by the Turc's method and with an average of several years),
compared with maps showing areas where there is a water deficit, shows that there
are many different regions in France that experience abnormally dry periods. This is
where images from low-resolution spatial satellites become important. These images
make it possible to compare the spatial distribution of the regions by studying the
vegetation cover of each region. This is made possible by using the water-deficit
maps that have been created for each region.
