Model projections of future climate change require that coupled climate-vegetation models are developed and validated, i.e. these models should be able to reproduce past climate and vegetation change. Records of pollen deposited in lake bottoms and peat bogs can provide the information needed to validate these models. The aim of this thesis was i) to explore the modern relationships between pollen and vegetation in northern and temperate China and estimate pollen productivity of major plant taxa, and ii) to use the results of i) to produce the first reconstruction of plant cover in China over the last 10 000 years for the purpose of climate modelling. A study of the modern pollen-vegetation-climate relationships was performed in northwestern China (Paper I). Pollen productivity for 18 major plants of cultural landscapes in central-eastern China was estimated (Paper II). Based on a synthesis and evaluation of all existing estimates of pollen productivity in the study region, a standard dataset of pollen productivity for 31 plant taxa is proposed (Paper III). This dataset was used to achieve pollen-based REVEALS reconstructions of plant cover over the last 10 000 years in 35 regions of northern and temperate China (Paper IV). The major findings can be summarized as follows. Paper I: Annual precipitation (Pann) is the major climatic factor influencing pollen assemblages, followed by July precipitation (PJul). The shared effect of combinations of two climatic factors explains a larger portion of the variation in pollen data than individual variables. Paper II: Of the 16 reliable pollen productivities estimated, the estimates for 8 taxa are new, Castanea, Cupressaceae, Robinia/Sophora, Anthemis type/Aster type, Cannabis/Humulus, Caryophyllaceae, Cruciferae, and Galium type. Trees have in general larger pollen productivity than herbs. Paper III: Of the total 31 taxa for which estimates of pollen productivity are available in China, 13 taxa have more than 1 value. All or most of these values are similar for Artemisia, Cyperaceae, Larix, Quercus and Pinus. Eight taxa have very variable estimates. Paper IV: The REVEALS plant percentage-cover strongly differs from the pollen percentages, and they provide new important insights on past changes in plant composition and vegetation dynamics.

The major objective of this thesis was to produce descriptions of the land vegetation-cover in Europe for selected time windows of the Holocene (6000, 3000, 500, 200, and 50 calendar years before present (BP=1950)) that can be used in climate modelling. Land vegetation is part of the climate system; its changes influence climate through biogeophysical and biogeochemical processes. Land use such as deforestation is one of the external forcings of climate change.  Reliable descriptions of vegetation cover in the past are needed to study land cover-climate interactions and understand the possible effects of present and future land-use changes on future climate. We tested and applied the REVEALS (Regional Estimates of VEgetation Abundance from Large Sites) model to estimate past vegetation in percentage cover over Europe using pollen records from lake sediments and peat bogs. The model corrects for the biases of pollen data due to intraspecific differences in pollen productivity and pollen dispersion and deposition in lakes and bogs. For the land-cover reconstructions in Europe and the Baltic Sea catchment we used 636 (grouped by 1˚x1˚ grid cells) and 339 (grouped by biogeographical regions) pollen records, respectively. The REVEALS reconstructions were performed for 25 tree, shrub and herb taxa. The grid-based REVEALS reconstructions were then interpolated using a set of statistical spatial models. We show that the choice of input parameters for the REVEALS application does not affect the ranking of the REVEALS estimates significantly, except when entomophilous taxa are included. We demonstrate that pollen data from multiple small sites provide REVEALS estimates that are comparable to those obtained with pollen data from large lakes, however with larger error estimates. The distance between the small sites does not influence the results significantly as long as the sites are at a sufficient distance from vegetation zone boundaries. The REVEALS estimates of open land for Europe and the Baltic Sea catchment indicate that the degree of landscape openness during the Holocene was significantly higher than previously interpreted from pollen percentages. The relationship between Pinus and Picea and between evergreen and summer-green taxa may also differ strongly whether it is based on REVEALS percentage cover or pollen percentages. These results provide entirely new insights on Holocene vegetation history and help understanding questions related to resource management by humans and biodiversity in the past. The statistical spatial models provide for the first time pollen-based descriptions of past land cover that can be used in climate modelling and studies of land cover-climate interactions in the past.

Non-native plants have invaded over 100 millions of acres of western arid land in the US and dramatically altered nutrient cycling rates. Changes in water cycling caused by invasive species are of particular interest because primary production in the Western US is typically limited by water availability and aquifer recharge reflects plant demand. Large-scale invasions can, therefore, be expected to cause large-scale changes in hydrological cycles, but until recently, there have been considerable limitations in the ability to measure the timing, location, and extent of water use. Here we injected a tracer, deuterated water (D2O), into five soil depths in two sampling periods (May and June) in two adjacent plant communities (native and non-native dominated). Plants were sampled at several distances from the tracer addition area to determine the horizontal and vertical extent of water use in native and non-native communities. The tracer injection was coupled with measurements of leaf level stomatal conductance, leaf area index, and volumetric soil water content to estimate plant transpiration. We found that both native and non-native plants transpired water from primarily the top 60 cm of the soil (>75%), with a particular emphasis (≥ 50%) on shallow soil water (<10 cm) while lateral roots did not exceed 50 cm for most species. Higher leaf area index resulted in significantly more water being transpired from the native community. Some sharp distinctions in timing and location of tracer uptake resulting from the differing phenologies of the dominant species in each community were observed and confirmed previous mechanisms thought to govern plant assemblages in these communities. In May, the non-native community dominated by annual grasses had higher tracer uptake at 10 cm than the native community but began using deep water (higher tracer uptake at 80 cm) as annual grasses senesced and tap-rooted fobs became dominant in June. The perennial native species, however, used the entire soil profile from the moment they became active until they senesced. Our approach shows promise for overcoming the lack of resolution associated with natural abundance isotopes and other enrichment approaches, and for providing detailed measurements of plant water-use space.

The aim is to gather pollen sequences and derive from them synthetic information on past species distribution and Holocene vegetation history. Lonicera nigra was selected as a model taxa. The phylogeographic hypothesis that its Alpine and Carpathian populations survived the LGM in separate refugia is supported only by the Late-Glacial record. This shrub migrated rapidly from southern to central Europe during the warm oscilations of the Late Glacial. The synthesis of vegetation history was produced by applying the REVEALS model, which can filter out following factors influencing the relationship between pollen and vegetation: pollen taphonomy, pollen productivity and pollen dispersal. It was necessary to calculate and test those parameters, so the goals were partly methodical. Pollen productivity estimates are calculated within the Relevant Source Area of Pollen, which is influenced by vegetation structure. Subsequent validation of those values in the area of the REVEALS model 10000 km2 and selection of additional values for lacking taxa created the best set of parameters for the study area. The source fossil dataset for purposes of this synthesis is the newly developed Czech Quaternary Pollen database (PALYCZ). Non-direct multivariate analysis of pollen percentages including all taxa revealed a similarity...

The assumption that prehistoric Native American land use practices had little impact on the North American landscape persists in the literature. However, recent research suggests the effects of prehistoric burning, deforestation and agriculture may potentially have been greater than previously considered. To resolve this discrepancy, quantitative estimates of changes in human population size and forest structure and composition over the course of the Holocene are needed. This thesis addresses this need by providing radiocarbon-based paleodemographic reconstructions and pollen-inferred estimates of vegetation change, as well as analyses of associations between the two at both continental and regional scales, from the late Pleistocene to the Anthropocene. One way to estimate paleodemographic change is to use the number of radiocarbon (14C) dates from a given area to study patterns of human occupation through time. A review of the literature and compilation of existing databases relevant to this method showed there is now sufficient data to study the paleodemographic history of many regions around the world. An analysis of 14C datasets from North America and Australia compared well with model-based reconstructions of past demographic growth, and provided higher frequency fluctuations in population densities that will be important for future research. Using a kernel density estimation approach, the first estimates of prehistoric population density for North America were obtained and synthesized into a series of continental-scale maps showing the distribution and frequency of 14C dates in the Canadian Archaeological Radiocarbon Database (CARD). The maps illustrated the space-time evolution of population and migration patterns, which were corroborated by independent sources of evidence. A methodology based on the statistical evaluation of cross-correlations between population and plant abundance was then developed to analyze the associations between these population estimates and plant communities derived from pollen databases. Periods of high spatial cross-correlation (positive and negative) between population and plant abundance were irregular and did not improve over time, suggesting that ancient human impacts are not discernable at a continental scale, either due to low populations or varying human land use practices. To further examine the relationship between pollen data and human land use at a regional scale, estimates of plant density and landscape openness are needed. The REVEALS (Regional Estimates of VEgetation Abundance from Large Sites) model corrects for the non-linear relationship between pollen production and plant abundance and can therefore be used to map histories of land use and land cover change. The model was applied to pollen records from lake sediments in the deciduous forest of southeastern Quebec. A preliminary analysis comparing these results to population density revealed low population during times of high Populus abundance and high population following the appearance of the mixed temperate forest suggesting a discernable human-environment association at regional scales. Overall, the results of thesis support the growing body of literature that suggests prehistoric Native Americans impacted their environments and that these impacts can be detected and quantified by integrating archaeological and paleoecological information. However, the timing, location, and intensity of human land use has changed in both space and time, suggesting regional- to local-scale analyses of human-environment interactions are most appropriate for continental North America. The methodology presented here can be used to study additional North American regions for the purpose of developing a continental history of human-environment interaction.