Geoscience Reference
In-Depth Information
summit means that the cloud itself is very thick, and is the origin behind convection
rain that falls on the Earth's surface. The aim of this section of the chapter is to
show how the infrared data, which is produced by geostationary GOES satellites,
can be used to compensate for the poor observation network that exists in Mato
Grosso. The data are then used to create a regional scale precipitation map for the
climatological year 2000-2001.
The method used for the purposes of our study is made up of two datasets:
- The first dataset records the maximum temperature observed for each pixel on
a daily basis. By analyzing such data over a period of 10 days or over a monthly
period, elements such as atmospheric variables and clouds are eliminated, meaning
it becomes possible to focus solely on the temperature on the ground. The
temperature on the ground varies according to the type and amount of vegetation
cover there is, and on the amount of water that has been absorbed by the vegetation
(in other words the rain that has fallen). Studying variations in maximum brightness
temperature comes back to the idea of measuring the quantities of water that have
accumulated at one measurement site [GUI 94]. This method has been used in
Africa as part of the Satellite Rainfall Estimation program. This program showed
that there is a significant link between maximum brightness temperature and
precipitation.
- A second dataset was created by thresholding the same images at a minimum
temperature of -40°C: this meant that clouds with cold summits (convection clouds
that produce rain) could also be monitored on a monthly basis (Figure 3.7). This
method is an adaptation of the pioneering work carried out by Arkin on the Goes
Precipitation Index, however, according to the results obtained by Cadet and Guillot
in 1991 the threshold temperature is lower (-40°C in comparison to -38°C)
[CAD 91].
The images in Figure 3.7 are used to monitor the development of clouds with a
cold summit. The images are also used to see whether there is a link between these
types of clouds and the quantity of precipitation that falls in the Mato Grosso region.
The monthly correlation coefficients (Table 3.2) calculated for each month show
that there is a link between satellite data and the quantity of precipitation that can be
found on the ground.
Apart from this relationship, the link between precipitation and satellite data are
not always very strong. The correlations are at their strongest at the beginning and
the end of the rainy season, they are also at their strongest when the value of the
maximum brightness temperature is at its highest, and this occurs half-way through
the rainy season. Precipitation has a stronger correlation with clouds that have a cold
summit than with maximum surface temperatures. If two satellite parameters are
used to calculate multiple regressions and to predict precipitation, the correlation
coefficients are above average for 7 months of the year (r >0.61, which is 50% of
the dependent variance). For 2 months of the year, however, the coefficients are just
average.
