DATE2018-05-17 10:11:23
AUTHORSM Orlic (1), M Pasaric (1), Z Pasaric (1)
  1. University of Zagreb, Zagreb, Croatia
ABSTRACTSea level data collected from 1930 to 2015 in the Atlantic off Gibraltar and in the West Mediterranean and Adriatic Seas are analyzed. Time series of regional sea level are constructed and are subjected to several methods that aim to separate trends from multidecadal variability. The results point to a steady sea level rise in the Atlantic but also to a deceleration of the rise in the late 1950s and early 1960s and to an acceleration of the rise in the late 1980s and early 1990s in the Mediterranean Sea. This strongly suggests that the Mediterranean sea level was considerably influenced by multidecadal variability, which was characterized by a period of ca 60 years. The finding also suggests that not only the Atlantic sea levels but also the Mediterranean sea levels could be reasonably well approximated by linear trends over the 1960-1990 interval. It is shown that the trends thus obtained could mostly be ascribed to (1) the regional trends of the air pressure and wind forcing, (2) the regional trends of thermosteric and halosteric sea levels, and (3) the increase of mass of seawater that was similar in the whole area considered. In the Atlantic the three contributions resulted in a sea level rise similar to the global one between 1960 and 1990, whereas in the Mediterranean Sea the increase of mass of seawater was nearly compensated by the atmospheric loading and steric effects during the thirty-year interval. Finally, it is tested whether a variant of semi-empirical method, which allows for both inertial and equilibrium response of sea level to temperature forcing and which has been previously successfully applied to global data, could be used to analyze and project the Mediterranean sea levels. The underlying theory suggests that the method is applicable to regional sea levels if regional temperatures are linearly related to global temperatures and if regional response times are similar to global response times. The test shows that the method is successful at reproducing past sea levels and sea level trends in the Mediterranean Sea but not in the Atlantic off Gibraltar. Consequently, semi-empirical projections of sea levels and related trends are prepared for the former basin.