Geoscience Reference
In-Depth Information
- the variable distance to town is masked by another, more causal variables that
associate temperature with the land cover.
The effect that distance to the town center has on temperature has been clearly
highlighted for the region of Paris. This effect exists for other major cities in France,
but to a lesser extent when compared to Paris, because the density of the built-up
areas in these other cities is lower than is the case for Paris. In addition, the density
of the climatological stations located near the town centers of the other large cities
does not make it possible to provide an accurate measurement of this variable. It is,
in fact, the vegetation index that should be kept and used as the estimation variable
in the regression and interpolation models. The distance from town variable, could
be kept if the same level of heat as that generated in the town center is transferred by
advection to the town's periphery. However, this is not the case.
2.4.4.
Spatial and scale variables
Spatial resolution, which is fixed as soon as all data has been collected and
which is uniform for all of the information layers that are held within the GIS, is
another aspect that needs to be considered. Image resolution is normally chosen
depending on the extent of the area to be analyzed. Depending on the actual scale of
interpolation to be used, a resolution of 100, 250, and 500 m is used on a national
level, whereas a resolution of 25 or 50 m is used on a regional level. As far as the
study of micro-climates is concerned, this resolution may be as small as 1 or 5 m
[JOL 02]. This means that, theoretically, it is possible to use different analysis and
interpolation scales.
In addition, for studies that are to be carried out on only one area, France for
example, several DEMs are available and they include DEMs that have resolutions
of 1,000, 500, 250, 100 and 50 m. It would be a major error to confuse resolution
with scale level. Different resolutions for the same piece of information provide us
with an interesting method that can be used to better understand scale levels
(topographic forms that are produced by a DEM with a resolution of 50 m are part
of a larger scale level than those produced by a DEM with a resolution of 1 km).
To make this information easier to understand, the examples of the urban area of
Dijon and the region of Franche-Comté will be used. These are two neighboring
areas for which a DEM with a resolution of 50 m is available (Figure 2.19). The
objective here is to examine the response generated by each resolution that estimates
different climate variables [JOL 07]. This will be shown by using the example of the
spatial variation of rainfall amounts.
2.4.4.1.
Available data
The 50 m resolution DEM is available at the French National Institute for
Geography. New data layers that characterize the area surrounding each
climatological station are derived by computing, as has already been explained in
