Regional-scale palaeoclimate modelling on soft proxy-data basis — an example from the Upper Miocene of the Lower Rhine Embayment

A new method in palaeoclimatology, initially proposed by Mosbrugger and Schilling (1992), which makes it possible to extract a maximum of information from the available proxy-data is tested. It possesses a capacity to produce a detailed palaeoclimate reconstruction, which is consistent with both proxy-data and the physics of atmospheric models. Since proxy-data information is particularly poor on a global scale, our method focuses on smaller-scale regions of better proxy-data coverage. It consists of an iterative coupling of inverse mesoscale atmospheric modelling with proxy-based palaeoclimate reconstruction. Using this method, the palaeoclimate of the Tortonian (Upper Miocene) of the Lower Rhine Embayment (northwest Germany) is reconstructed with a high spatial resolution. The results include a physically consistent inter- and extrapolation of proxy-derived palaeoclimate data in space and onto variables not accessible from geological–palaeontological proxies alone. Moreover, the method delivers statistics of the large-scale atmospheric flows as information on the coupling of the regional climate to global fields.