No description available.

No description available.

No description available.

The principal aim of the thesis is to determine the Late Quaternary glacial history of the South Patagonian Icefield at Torres del Paine (51°S, 73°W), Chile. The secondary aim is to compare this glacial history with palaeoclimatic records from elsewhere to test the theory that climate change over the last glacial-interglacial cycle was synchronous between the two polar hemispheres. Synchronous climate change cannot be explained as an atmospheric response to insolation changes unless fundamental ocean-atmosphere reorganisations occurred (Broecker and Denton, 1990). Empirical glacial-geologic data from southern South America is therefore used to test recent models of global climate change. The global pattern of climate change over the last glacial-interglacial cycle is assessed by reviewing proxy palaeoclimate records including isotope records from polar ice cores and deep sea sediment cores, and glacial geologic records from the southern Andes. Conclusions from this review form the basis for hypotheses about what 'should' have happened at Torres del Paine. To test these hypotheses glacial geologic investigations were undertaken on site. Geomorphological evidence is used to define eight icesheet, deglaciation and valley glacier stages; ice extended ca.50km east of the modern South Patagonian Icefield margin during the last glaciation. Basal dates from peat bogs provide dating control for some glacial stages. Pumice fragments associated with glacial deposits were derived from an eruption of Volcan Reclus ca.12,000 yr BP and therefore constrain the Lateglacial depositional sequence. Models of the glacial history are constructed. Evidence that climate change at Torres del Paine was synchronous with other sites is equivocal. However, significant results are: (a) evidence of a Late-Lateglacial ('Younger Dryas') advance - the first such site in Patagonia; (b) evidence that deglaciation occurred slowly; and (c) evidence that the greatest extent of ice did not necessarily coincide with the coldest part of the last glaciation.

An initial study of the palaeoclimate signal contained within speleothem growth was undertaken by' investigating regional variations in speleothem Qrowth frequency. It was demonstrated that about 500 analyses in such a regional compilation were necessary to generate a statistically significant curve which did not suffer from sample bias. However, few such regions are likely to have such a larQe data set. That from north west Europe did provide a useful palaeoclimate record, giving evidence of multiple interstadial events within isotope stage 3, and a significantly low level of growth within stage Sa. An investigation was undertaken into the palaeoclimate signal contained in variations of speleothem growth rate, based on the theory derived by Dreybrodt (1981) and Buhmann and Dreybrodt (1985) from calcite precipitation kinetics. It was demonstrated that growth rate increases with increasing calcium ion concentration, temperature and water flux (drip rate for stalagmites, water film thickness for flowstones and seasonal variations in water availability for both speleothems); turbulent flow conditions and cave air pC02, which theoretically affect growth rate, were demonstrated to be Insi9nificant. If water flux, calcium concentration and temperature all increase with improving Climate, growth rate increases may reflect climatic improvement. In particular, it was demonstrated that stalagmites should be most sensitive to changes in calcium concentrations, temperature, and seasonal shut-off of the water feed, whilst flowstones would also be sensitive to changes in water film thickness. Theoretical growth rates were tested for recently forming speleothems in excavated caves and mines. For these, minimum growth rates were determined by knowing the date of excavation of the cave or mine, and the growth rate determining variables were measured over the course of a year. It was demonstrated that the theory accurately predicted growth rates for both stalagmites and flowstones within the 20' errors based on variations in calcium ion concentration and water film thickness. However, flowstones generally grew slower than that predicted by the theory, due to the seasonal shut-off of the water supply feeding these samples. For flowstones at Kent's Cavem, growth rate was observed to Increase with Increasing water availability, for stalagmites at Lower Cave, growth rate was shown to increase with increaSing drip rate. Assuming a good prediction of growth rate by the theory, applications to Quatemary speleothems were undertaken to determine past calcium ion concentration, temperature and water flux. Growth rates were determined by thermal ionisation mass spectrometric uraniumseries dating. The growth rate of one Holocene sample from Sutherland demonstrated that variations in growth rate over the last 7 ka did not depend on temperature variations, but either to changes in calcium concentrations due to vegetation change or a non-linear response to changes in water flow. Growth rates were also determined for two flowstones from Yorkshire which had grown over the last 200 ka. However, the very fast growth rates in these samples prevented a precise record from being obtained. Mass spectrometric dating also provided a record of the timing of growth commencement and cessation. This was shown to be more complex than previously considered; in particular the Holocene growth of the Sutherland stalagmite commenced 5 ka after glacier retreat in the region, the Yorkshire flowstone from Lancaster Hole had seven growth phases, each for only 1-3 ka, five of which correlated with solar insolation maxima. In contrast, another flowstone from Stump' Cross in Yorkshire was shown to grow in both interglacial, interstadial and glacial periods of the last 200 ka. An investigation was made into the use of 13C/1'C,(t>nM to determine the type of plant community at the time of speleothem formation, and whether a non-biogenic source of CO2 was present. 13C analyses of the Stump Cross flowstone gave elevated 13C not explicable by the plant communities present, nor were high enough to have a non-biogenic source. Further investigations are needed, but this evidence suggests caution in interpreting 13C records for flowstones. An annual signal of growth rate and growth rate variability was obtained from ultra-violet microscopic analysiS of luminescent banding within speleothems. Banding was demonstrated to be annual by mass spectrometric uranium-series dating, but was only preserved In 10% of a" samples. Variability of growth rate for the Holocene Sutherland stalagmite was compared to the theoretical annual variability of growth rate derived from annual variations in the growth rate determining variables observed today, and a good agreement was observed. Furthermore, for one period of growth, a 4-5 year period of rapid growth rate was demonstrated to correlate with the Hekla 3volcanic eruption in Iceland.

Bibliography: p. 155-175. / Palynological analysis of organic sediments from a freshwater lake near Port Elizabeth (34°S,25°30'E) has provided a high- resolution vegetation history of the area for the last 2200 years. Detailed identification and counting of the fossil pollen resulted in the generation of a pollen diagram. Changing frequencies in fossil pollen over time are represented, and inferences are made regarding environmental conditions which influenced the vegetation. A detailed narrative of vegetation history in response to environmental change is presented, and this is compared to results from related studies. The significance of the Lake Farm study site has been noted in terms of its location as a 'zone of convergence' for a variety of vegetation types. Results of fossil pollen analysis indicate that environmental conditions prior to 1 500BP were drier than at present. Forest and fynbos vegetation were not well-represented in the pollen spectrum at this time, and it is suggested that they were not favoured by these conditions. Environmental conditions ameliorated after 1500BP, becoming more mesic, which favoured the proliferation of both forest and fynbos vegetation types. At present xeric and grassland elements are declining, while shrubs increase, indicating an enhanced human-induced disturbance regime. It is suggested that the partial decline in forest elements at present 1s most likely attributable to human-induced disturbance of the environment. The introduction of exotic trees has been noted (approx. 280BP) and is seen to have coincided with the influx of european settlers to the region. Principal Components Analysis has revealed that the vegetation distribution in the area has been most heavily influenced by human activity and moisture availability. The necessarily subjective interpretation of the statistical results, however, casts some doubt on the validity of the conclusions drawn. The validity of the conclusions drawn from this study becomes apparent not only in terms of what is learned about the history of forests, but also the form any future management should take.

No description available.

No description available.

A high quality, historical surface air temperature data set is essential for the reliable investigation of climate change and variability. In this study, such a data set has been prepared for Australia by adjusting raw mean annual temperature data for inhomogeneities associated with station relocations, changes in exposure, and other problems. Temperature records from long-term stations were collaborated from the set of all raw data held by the Australian Bureau of Meteorology. These long-term records were extended by combining stations and manually entering previously unused archived temperature measurements. An objective procedure was developed to determine the necessary adjustments, in conjunction with complementary statistical methods and station history documentation. The objective procedure involved creating a reference time series for each long-term station, from the median values at surrounding, well-correlated stations. Time series of annual mean maximum and mean minimum temperatures have been produced for 224 stations, and the adjusted dataset has been made available to the research community. The adjusted data are likely to be more representative of real climatic variations than raw data due to the removal of discontinuities. The adjusted data set has been compared with previously used temperature data sets, and data sets of other parameters. The adjusted data set provides adequate spatial coverage of Australia back to 1910. Additional adjusted data are available prior to this date at many stations. Trends in annual mean maximum, minimum, the mean of the maximum and minimum, and the range between the maximum and minimum, have been calculated at each site. Maximum and minimum temperatures have increased since about 1950, with minimum temperatures increasing faster than maximum temperatures.

A high quality, historical surface air temperature data set is essential for the reliable investigation of climate change and variability. In this study, such a data set has been prepared for Australia by adjusting raw mean annual temperature data for inhomogeneities associated with station relocations, changes in exposure, and other problems. Temperature records from long-term stations were collaborated from the set of all raw data held by the Australian Bureau of Meteorology. These long-term records were extended by combining stations and manually entering previously unused archived temperature measurements. An objective procedure was developed to determine the necessary adjustments, in conjunction with complementary statistical methods and station history documentation. The objective procedure involved creating a reference time series for each long-term station, from the median values at surrounding, well-correlated stations. Time series of annual mean maximum and mean minimum temperatures have been produced for 224 stations, and the adjusted dataset has been made available to the research community. The adjusted data are likely to be more representative of real climatic variations than raw data due to the removal of discontinuities. The adjusted data set has been compared with previously used temperature data sets, and data sets of other parameters. The adjusted data set provides adequate spatial coverage of Australia back to 1910. Additional adjusted data are available prior to this date at many stations. Trends in annual mean maximum, minimum, the mean of the maximum and minimum, and the range between the maximum and minimum, have been calculated at each site. Maximum and minimum temperatures have increased since about 1950, with minimum temperatures increasing faster than maximum temperatures.