Considerable advances have been made over the past 200 years in reconstructing the temporal and spatial patterns of variation in Holocene climate, in quantifying the magnitude and rate of climate change, and in assessing the role of forcing factors such as orbital forcing and solar variability on Holocene climate. These advances have mainly occurred as a result of improved methodologies, of studying many different proxies in a wide range of geographic areas, of improved project design, of increasingly finer resolution studies, of greater concern for data quality, detail, and interpretation, and of increasing interactions between paleoecologists, earth scientists, climatologists, and climate earth-system modelers. Despite these enormous advances, several critical problems continue to pervade Holocene climate research. These problems relate to dating, data interpretation, basic gaps in our knowledge, and data availability.

The single most critical problem in establishing the temporal and spatial patterns of Holocene climate changes is chronology. An absolute chronology, mainly provided in the Holocene by radiocarbon dating, is the major means for deriving age-depth models for individual sequences and for almost all correlations between sequences. Many Holocene paleoproxy sequences (e.g. pollen), particularly those studied from lake sediments before about 1990, have poor chronologies, often based on only four to eight radiocarbon-dates from bulk sediment. Such dates may be influenced by unknown hardwater errors and other factors that can result in erroneous ages. The small number of radiocarbon-dates per sequence greatly limits the reliability of the resulting age-depth models (Telford et al. 2004). Such chronologic problems inevitably limit the value of many paleoclimatic reconstructions in detailed data synthesis, correlation, and synoptic mapping.

In the interpretation of Holocene paleoecologic data (e.g. pollen), there are often several possible interpretations for the observed changes (e.g. climatic shifts, human impact, biologic factors). Multi-proxy studies (Birks and Birks 2006) where several paleoenvironmental proxies are studied on the same stratigraphic sequence are particularly important as they permit the testing of alternative hypotheses to explain the observed changes. The elegant study by Shuman et al. (2004) on the interpretation of New England vegetational history in terms of climate changes deduced independently of the pollen stratigraphy from stable hydrogen isotope ratios and lake-level changes shows the value of multi-proxy approaches in Holocene climate research. The use of ecologic simulation models (e.g. Anderson et al. 2006) is another important approach to testing alternative hypotheses about underlying factors for the observed changes (e.g. Heiri et al. 2006).

A further problem that arises in the use of paleoecologic data as a source for paleoclimatic reconstructions concerns the climatic sensitivity of different types of organisms. Such data are from a wide range of ecologic settings that may result in different thresholds, sensitivities to climate change, and responses of individual species and biotic assemblages. It can thus be difficult to know what changes in particular proxies (e.g. diatoms, chironomids) may mean in climatic terms when local limnologic variables (e.g. lake-water pH, nutrient status) also change with time (Battarbee 2000). A related problem concerns the quantification of realistic uncertainties in paleoclimatic reconstruction that take account of the strong temporal autocorrelation in paleoclimatic time-series and the inherent spatial autocorrelation of climate variables used in organism-climate transfer functions (Telford and Birks 2005).

In addition to these paleoecologic problems, there are problems in understanding and quantifying interactions between external forcing factors such as orbital forcing, solar variability, volcanic activity, and land-atmosphere-ocean-ice feedbacks and in evaluating and quantifying the roles of land-use and vegetation cover in influencing Holocene climate (Oldfield 2005, this volume).

A challenge facing Holocene climate research is how to reconstruct "modes of variability" (Oldfield 2005). Climatologists commonly assess natural climate variability in terms of climate modes such as the El Nino-Southern Oscillation

(ENSO), the North Atlantic and Arctic Oscillations, the Pacific Decadal Oscillation, and the Atlantic Multi-decadal Oscillation (e.g. Diaz and Markgraf 2000; Hurrell et al. 2003). These modes are most relevant in the late Holocene when ice-extent, sea-level, orbital forcing, and atmospheric greenhouse-gas concentrations have changed relatively little (Bradley et al. 2003; Oldfield 2005). Late Holocene climate with almost constant external and internal boundary conditions is likely to be close to the pre-industrial patterns under those boundary conditions. It is the natural variability in this late Holocene climate system and its temporal and spatial patterns that recent climate change can be compared to identify anthropogenic forcing on climate (Oldfield 2005). Climate modes capture a large amount of the variance in climate variability and are ideal as composite indices for comparisons in time and space. The reconstruction of the temporal variation in such indices for the past 2000-3000 years requires not only fine-resolution studies but also novel ways of identifying climate modes from proxy-climate data (e.g. Thompson et al. 2000; Bradley et al. 2003; Gray et al. 2004; Dykoski et al. 2005).

Ruddiman's (2003, 2005b) Anthropocene hypothesis has highlighted basic gaps in knowledge about the extent of past land-use and vegetation-cover changes and about methane sources. Recent reports that terrestrial plants growing in an aerobic environment may emit methane (Keppler et al. 2006; see also Parsons et al. 2006) emphasize a major gap in our understanding of methane production and hence the global carbon budget. Much remains to be discovered about methane production and the causes of the observed changes in methane concentrations.

A final problem that is particularly critical in future data syntheses, mapping of paleoclimate proxies and reconstructed climate estimates, and data-model comparisons is the continuing need for international collaboration to ensure effective storage and management of the ever increasing amounts and diversity of primary paleoecologic, paleoceanographic, ice-core, geologic, and archaeologic data, modeling-experiment outputs, and paleoclimatic reconstructions (Eakin et al. 2003). These and many other data types must be carefully stored, managed, and made available to all researchers as these data may well be very relevant to furthering our understanding of Holocene climate history (Alverson and Eakin 2001; Anon 2001; Dittert et al. 2001).

0 0

Post a comment