Modern pollen samples collected from 80 locations and representing a wide array of mature habitats in Panama and Costa Rica provide analogs to assist in the interpretation of fossil pollen records. Pollen spectra accurately reflect changes in actual forest types. Upslope transport of pollen of anemophilous species is evident in the sparsely vegetated montane samples. However, the corresponding downslope transport of these prolific pollen producers is masked by local pollen production. Mean pollen representation across gradients of mean annual temperature (MAT; 4°C increments) and mean annual precipitation (MAP; 500 mm increments) for 17 pollen types are presented as response matrices. Although preliminary in nature, these response matrices present a clearer image of pollen representation than can be obtained by considering gradients of MAT or MAP alone.

A diatom-based calibration model for predicting summer temperatures was developed using climatically sensitive subarctic lakes in northern Fennoscandia. The model was applied to a sediment core from a treeline lake to infer trends in Holocene climate. The record exhibits long-term variations, as well as a series of shorter-term fluctuations on a time scale of centuries. Summers were warmest in the area about 6200 cal yr B.P. and featured distinct cooling episodes around 8300, 7200, 4200, 3000, and 400 cal yr B.P., most of these coinciding with some known climate events (e.g., the 8200 cal yr B.P. event and the Little Ice Age). The similarity of the observed shifts with the pacings of climate events from marine and ice-core records represents evidence for coupled ocean–atmosphere forcing of the regional climate.

A procedure to calculate probability limits for the inference that the absence of a taxon from a woodrat (Neotoma) midden implies absence from the paleolandscape uses paired samples of modern vegetation communities and paired samples of paleocommunities. Assumptions are: (1) each member of a sample pair is an independent measure of the same vegetation assemblage; (2) behavioral patterns of woodrats are the same as each midden in a paired sample is constructed; and (3) the probability of fossilization is zero when a taxon is absent from the vegetation. The procedure provides a logical test of data consistency: the upper probability limit of making a false inference should be greater than the lower limit. Averaged over 140 plant taxa, the upper and lower probability limits for a false inference were 11 and 7%, respectively. More than 70% of taxa passed the logical test, indicating a reliable procedure. For many taxa that failed the logical test, four potential explanations account for this failure, two of which can be solved by simply increasing sample sizes. Using analogous assumptions, the procedures are applicable to other types of stratigraphic sampling such as macrofossils from sediment cores or fossils from biostratigraphic units.

A natural ∼1450-yr global Holocene climate periodicity underlies a portion of the present global warming trend. Calibrated basal radiocarbon dates from 71 paludified peatlands across the western interior of Canada demonstrate that this periodicity regulated western Canadian peatland initiation. Peatlands, the largest terrestrial carbon pool, and their carbon-budgets are sensitive to hydrological fluctuations. The global atmospheric carbon-budget experienced corresponding fluctuations, as recorded in the Holocene atmospheric CO2 record from Taylor Dome, Antarctica. While the climate changes following this ∼1450-yr periodicity were sufficient to affect the global carbon-budget, the resultant atmospheric CO2 fluctuations did not cause a runaway climate–CO2 feedback loop. This demonstrates that global carbon-budgets are sensitive to small climatic fluctuations; thus international agreements on greenhouse gasses need to take into account the natural carbon-budget imbalance of regions with large climatically sensitive carbon pools.