Geoscience Reference
In-Depth Information
rather than higher indexes (areas in which there is a significant amount of biomass).
This law of spatial organization is similar to radiation models used in physics: in this
example the statistical description can be interpreted as being a causal variable.
It seems that the vegetation index is made up of a spatial structure: the index is
low in the center of Paris, the densest region of the city. This index, however,
progressively increases towards the periphery due to the fact that green spaces
become more abundant and the heavily built-up areas become increasingly sparse.
The correlation coefficient between the vegetation index and distance from town
increases to 0.39. The temperature variation according to the variable vegetation
increase (Figure 2.18) also follows this pattern.
The cities of Paris and Lyon have positive and negative coefficients for the two
correlations temperature/distance and temperature/vegetation index, respectively
(see Table 2.4). The town of Château-Thierry has low coefficients for both
correlations (temperature increases as the distance to the town center increases; the
value of r is negative because we are actually getting closer to the city of Paris).
Marseille (in particular) and Toulouse do not present the same types of data, which
shows that the statistical relationships that have been highlighted in this section
cannot be applied to all towns and cities. As far as Marseille is concerned, the
situation is clear: the city center is located near the coast, and is cooled down by the
sea breeze. This means that the urban heat island is neutralized, and the temperature
in the center of Marseille is very similar to the temperature of its periphery. As for
Toulouse, the positive r value for February and August shows that there is a positive
relationship (unexepected and which is inexplicable) between vegetation index and
temperature.
Paris
Lyon
Marseille
Toulouse
Château-Thierry
r February
-0.37
-0.48
-0.33
+0.32
-0.11
r August
-0.39
-0.37
-0.28
+0.27
-0.14
Table 2.4.
Correlation coefficients for the variables of temperature/vegetation index
recorded in four major cities in France for February and August
It can now be seen that other factors, such as distance to town and vegetation
index, also have an influence on the spatial variation of temperature. A more
detailed analysis would be able to prove this. However, the objective of our study is
to find out whether the variable distance to town can be used as an independent
variable to interpolate. Analyses have shown that the temperature variation in
relation to distance to town is only true with a large distance for certain cities, i.e. it
is true for the largest cities and not for the smaller ones. The overall negative
variation between temperature and distance is due to two facts:
- the high number of climatological stations around the larger cities, and Paris in
particular, takes an artificial reading from a spatial structure that is influenced by the
urban island effect;
