The
North Atlantic Oscillation (NAO) has a large influence on the European climate i.e. in the winter halfyear
(ONDJFM), explaining up to one third of the temperature and precipitation variability. However, the relation between the NAO-Index and the climate
seems to be not stable in time and space. As an example it can be shown that the first leading EOF of SLP
over the North Atlantic region is partly not identical with the NAO-pattern - although the NAO is usually expected and defined as leading EOF. In
this study we analyse changes in the relation between the atmospheric circulation and the climate with focus on changes in the spatial correlation
patterns over the Baltic Sea region in the last 1000 years.
To study changes in the relation between the atmospheric circulation indices and the near-surface climate, long historical station data,
multi-proxy reconstructions and model simulations (ECHO-G, HadCM3) are used to describe the climate anomalies and the corresponding climate
circulation giving rise to it. Using a 31year moving window for correlation and EOF-analysis, the relation is higly non-stationary in time and
partly space. Stable periods can be seperated from higly non-stationary episodes. Changing to longer multi-decadal and centennial scale, the relation
gets stationary for the last 1000 years. However, on millenial scale, the relation is expected to change again due to orbital forcing.
The scale dependent (non-)stationary in the climate-circulation relation is of great importance for understanding regional climate change and
the reconstruction of circulation indices like the NAO from proxy data. Possible physical explanations within the model as well as within direct
measurements have to be discussed together with methodical consequences for bottom-up approaches by reconstructing paleoclimate data from near-surface
climate proxies.
