Holocene Climate Optimum and Last Glacial Maximum in the Mediterranean: the marine oxygen isotope record

Reconstructions with comprehensive estimates of confidence intervals are pr esented of changes in the W-E stable oxygen isotope gradient in Mediterrane an surface waters between the Holocene Climate Optimum and the Present, and between the Last Glacial Maximum and the Present. Rigorous statistical ass essment is made of the significances of the mean geographic trends observed in these reconstructions. Firstly, it is concluded that any reconstruction should strictly be based on values obtained by analyses of one single fora miniferal species throughout the basin, as different species are found to r espond with isotopic variations of different amplitudes to climatic/hydrogr aphic change. This difference is tentatively related to differences between the habitats and seasons of growth of the various species. Secondly, a sig nificant increase of roughly a factor 3 is found in the Mediterranean W-E o xygen isotope gradient during the Last Glacial Maximum, relative to the Pre sent. This difference is almost entirely due to increased glacial values in the Levantine Sea, which are considered to be a result of a combination of increased evaporation rates and/or somewhat cooler than anticipated surfac e water conditions. Thirdly an eastward increase of roughly half the presen t-day magnitude is found for the W-E oxygen isotope gradient during the Hol ocene Climate Optimum. Values in the Levantine Sea appear to have undergone up to 0.3 parts per thousand more depletion than those elsewhere in the ea stern Mediterranean. However, no significant trends are found between the e astern and western parts of the Levantine Sea, nor between values near the Nile delta and those from elsewhere in the Levantine Sea. The Holocene Clim ate Optimum's eastward increase in the Mediterranean oxygen isotope gradien t, although weaker than the Present, suggests that the Mediterranean contin ued to function as a concentration basin, albeit in a less vigorous way tha n today. Finally, simple mixing arguments are used to argue that inferred o xygen isotope ratios of surface waters may not be used as an indication of conservative property (e.g. salinity) distribution on geological timescales , but instead show amplitudes of response to climatic/hydrographic changes that likely are >2 times larger than the corresponding amplitudes of respon se for truly conservative properties. (C) 1999 Elsevier Science B.V. All ri ghts reserved.