No abstract

No abstract

No abstract

A study of past lake-level changes in Lake Xinias covering the last 40 000 years was presented in an earlier paper. In that paper, changes during the Holocene were only briefly dealt with, because of poor sedimentary representation. In this paper a supplementary transect of cores is described, in which Holocene lake-level changes are better recorded. The correlation of the cores was based on pollen and lithostratigraphic analyses, and a curve was constructed showing recorded lake-level changes. The curve from Lake Xinias was compared with an earlier compilation of lake-level changes from the Balkans, mainly from Greece. The compilation shows rising lake levels during the earlier part of the Holocene, and successively lowered levels during the later part. Most lake levels were high between about 8000 and 5500 cal. yr BP. The reconstruction from Lake Xinias indcates similar changes during the early and late Holocene. However, it differs by indicating a lowering in lake level, culminating around 7000 cal. yr BP, interrupting and dividing the period of mid-Holocene high lake-level stand. Further studies are required to show if this lowering in Lake Xinias was regionally representative and caused by climate change, or was caused by tectonic disturbance locally affecting lake level.

The hitherto longest found lake sediment sequence on Byers Peninsula, Livingston Island, South Shetland Islands, was analysed with respect to lithology, chronology, diatoms, Pediastrum, pollen and spores, mosses, mineralogy, and sediment chemistry. During the ca. 5000 year long development the sediments were influenced by frequent tephra fall-outs. This volcanic impact played a major role in the lake's history during two periods, 4700-4600 and 2800-2500 BP, but was of importance during the lake's entire history with considerable influence on many of the palaeoenvironmentally significant indicators. The large and complex data set was analysed and zonated with different types of multivariate analysis. This resulted in a subdivision of the sequence into 8 time periods and 21 variables. Redundancy analysis (RDA) of this data set, both without and with the tephra periods, and with 4-6 of the variables as explanatory environmental variables, reveal that climatic/environmental signals are detectable. The palaeoclimatic picture that emerged out of the tephra 'noise' suggests that the first 100 years were characterized by mild, humid conditions. This was followed by a less mild and humid climate until ca. 4000 BP when a gradual warming seems to have started, coupled with increased humidity. These mild and humid conditions seem to have reached an optimum slightly after 3000 BP. At ca. 2500 BP a distinct climatic deterioration occurred with colder and drier conditions and long seasons with ice cover. This arid, cold phase probably reached its optimum conditions at ca. 1500 BP, when slightly warmer conditions might have prevailed for a while. Except for the modern sample with rather mild climate, the last 1400 years seem to have been fairly arid and cold, and the effects of the frequent volcanic activity during this period is only vaguely seen in the records.