Statistical downscaling of monthly mean North Atlantic air-pressure to sea level anomalies in the Baltic Sea

The term “downscaling” describes a procedure in which information about a process with a certain characteristic scale is derived from other processes with larger scales. The present paper identifies a relationship between the main components of the North Atlantic air-pressure anomalies at sea-level (characteristic length-scale > 1000 km) and the sea level anomalies at several Baltic Sea gauges (10 km-100 km) in winter. Monthly means from 20 observed winters are used to fit a statistical model that describes the dependence between both parameters. Further observations from this century are used to validate this model, which is able to estimate sea level anomalies from the air-pressure field to a good level of approximation. Sea level anomalies with periods from months to decades are reproduced well. As main forcing for the sea level anomalies, wind-stress with a strong zonal component is identified. For the past 89 years, we found that a slight decrease of mean sea level was induced by air-pressure. A slight increase is found when air-pressure from a GCM “greenhouse” experiment is downscaled.