Climate scenarios are important tools for the description of climate d
evelopments. A proposal for the construction of climate scenarios is p
resented in this paper. The idea consists of the calculation of possib
le climate developments on the basis of coupling the safest hypothesis
on the development of the future climate (for instance from climate m
odel results) for a selected region with observed data from this regio
n. Therefore, a statistical algorithm was developed by maintaining the
stability of the main statistical characteristics (variability, kind
of frequency distribution, annual cycle, persistence). Using a special
cluster analysis algorithm, complex scenarios could be calculated whi
ch guarantee temporal, spatial and physical consistency of the conside
red meteorological parameters. The quality of the method was demonstra
ted by calculation over an example data series. In addition, the pract
icability of the algorithm was demonstrated for a selected region of G
ermany by using different scenario examples.