To investigate the consequences of climate change on the water budget in sm
all catchments, it is necessary to know the change of local precipitation a
nd temperature. General Circulation Models (GCM) cannot provide regional cl
imate parameters yet, because of their coarse resolution and imprecise mode
lling of precipitation. Therefore downscaling of precipitation and temperat
ure has to be carried out from the GCM grids to a small scale of a Few squa
re kilometres. Daily rainfall and temperature are modelled as processes con
ditioned on atmospheric circulation. Rainfall is linked to the circulation
patterns (CPs) using conditional probabilities and conditional rainfall amo
unt distribution. Both temperature and precipitation are downscaled to seve
ral locations simultaneously taking into account the CP dependent spatial c
orrelation. Temperature is modelled using a simple autoregressive approach,
conditioned on atmospheric circulation and local areal precipitation. The
model uses the classification scheme of the German Weather Service and a fu
zzy rule-based classification. It was applied in the Aller catchment for va
lidation using observed rainfall and temperature, and observed classified g
eopotential pressure heights. GCM scenarios of the ECHAM model were used to
make climate change predictions (using classified GCM geopotential heights
); simulated values agree fairly well with historical data. Results for dif
ferent GCM scenarios are shown.