Spatial Analysis, Cartography and Climate - Geographical Information and Climatology - page 68

In addition to this information, the standard deviation of the residual
autocorrelations, which is produced after the two analysis phases, is much lower in
Western France than in the Southeast of the country. This means that geographic
characteristics (such as topography, land cover, location in relation to ocean masses,
etc), which are different for each region, play a key role in determining the results of
interpolation. Interpolations that are applied to the whole of France lead to the
creation of problems that can only be overcome by the use of a local, rather than
global approach [JOL 08; JOL 09].
Through the examples that were used in this chapter it would seem that the
process of interpolation is not an easy method to use. Other examples would also
have reinforced this conclusion. Technically speaking, it is not difficult to recreate
the continuous field of any variable; all that needs to be done is to apply the
different functions of spatial analysis software to the variables in question. The real
problems arise from two main areas: from the values of the information that is
introduced into the interpolation models and from the validity of the methods that
are used. The interpolation process is not a simple operation. It requires a certain
amount of expertise, i.e. the expertise of a statistician and of a GIS specialist. These
two specialists know that the results of interpolation depend on spatial information
and on analysis methods.
2.6. Bibliography
[ARN 00] A RNAUD M., E MERY X., Estimation et interpolation spatiale: méthodes
déterministes et méthodes géostatistiques , Hermès, Paris, 2000.
[BUR 86] B URROUGH P.A., Principles of Geographical Information Systems for Land
Resources Assessment , Oxford University Press, New York, 1986.
[CAR 82] C ARREGA P., Les facteurs limitant dans le Sud des Alpes Occidentales: étude
géographique, PhD thesis, Nice, 1982.
[CAR 03] C ARREGA P., “Le climat aux échelles fines” , International Association for
Climatology , vol. 15, p. 19-30, 2003.
[COL 00] C OLLINS F.C., B OLSTAD P.V., A comparison of spatial interpolation techniques in
temperature estimation, 2000. Online at: http://www.ncgia.ucsb.edu/conf/SANTA_FE_
CD-ROM/sf_papers/ collins_fred/collins.html.
[COU 99] C OURAULT D., M ONESTIEZ P., “Spatial interpolation of air temperature according
to atmospheric circulation patterns in southeast France”, Int J Climatol , vol. 19, pp. 365-
378, 1999.
[DUB 84] D UBRULE O., “Comparing splines and kriging”, Comp. Geosci., vol. 10, no. 2-3,
pp. 327-333, 1984.
[ECK 89] E CKSTEIN B.A., “Evaluation of splines and weighted average interpolation
algorithms”, Comp. Geosci. , vol. 15, pp. 79-94, 1989.
[FEY 95] F EYT G., M AILLOUX H., D E S AINTIGNON M.F., “SIG et information climatique”,
Revue Internationale de Géomatique , vol. 8, no. 3-4, p. 361-376, 1995.