Richard H. W. Bradshaw

The selection of sites for paleovegetational studies

The judicious selection of sites for paleovegetational and paleoclimatic studies permits paleoecologists to answer specific research questions that go beyond primary descriptions of past vegetation. We present a model that describes the relationship between basin size and pollen source area and predicts the proportions of local, extralocal, and regional pollen sampled by lake basins of different size. The distinctive sampling properties of lakes, peats, and small hollows can be exploited to provide details of pattern in paleovegetation so long as attention is given to the limitations and problems of these types of sites. Combinations of site types in a single study most fully exploit the information contained in sediments.

Forecasting the Effects of Global Warming on Biodiversity

ABSTRACT The demand for accurate forecasting of the effects of global warming on biodiversity is growing, but current methods for forecasting have limitations. In this article, we compare and discuss the different uses of four forecasting methods: (1) models that consider species individually, (2) niche-theory models that group species by habitat (more specifically, by environmental conditions under which a species can persist or does persist), (3) general circulation models and coupled ocean–atmosphere–biosphere models, and (4) species–area curve models that consider all species or large aggregates of species. After outlining the different uses and limitations of these methods, we make eight primary suggestions for improving forecasts. We find that greater use of the fossil record and of modern genetic studies would improve forecasting methods. We note a Quaternary conundrum: While current empirical and theoretical ecological results suggest that many species could be at risk from global warming, during the recent ice ages surprisingly few species became extinct. The potential resolution of this conundrum gives insights into the requirements for more accurate and reliable forecasting. Our eight suggestions also point to constructive synergies in the solution to the different problems.

Relationships between Contemporary Pollen and Vegetation Data from Wisconsin and Michigan, USA

Scatter diagrams and regression analysis of paired pollen and tree—inventory data show how pollen percentages represent the percent basal area for the major arboreal genera in Michigan and Wisconsin. We show that the relationship between pollen and tree percentages for each taxon is generally similar for two states of comparable size and similar vegetation (Wisconsin and Michigan), but that the relationship is influenced by the size of the pollen—collecting site and the size of the area surveyed for trees around each site. These results provide information concerning the relative size of the pollen—source area for seven arboreal pollen types: Pinus, Quercus, Betula, Tsuga, Ulmus, Fagus, and Acer, listed in descending order of pollen—source area. Moderate—sized lakes (30—150 ha) accumulate significant quantities of Pinus and Quercus pollen produced farther than 30 km away, but accumulate relatively few Fagus grains from >4.5 km, and even fewer Acer grains from >2.3 km. The source areas for Betula, Tsuga, and Ulmus pollen lie within 30 km of each lake, and significant quantities of these grains travel farther than 4.5 km. Regression analysis of data from basins of different size supports the hypothesis that small basins collect their pollen from a smaller area of the surrounding vegetation than do large basins.

MODERN POLLEN-REPRESENTATION FACTORS FOR WOODS IN SOUTH-EAST ENGLAND

(1) An analysis of the relationship between pollen frequency and plant numbers in a given area is an essential step if pollen diagrams are to be used as records of past tree populations. This relationship was studied by comparing the pollen content of moss samples with the vegetation within 20-m radius of each sample site, at seventy-eight sites in the New Forest, Hampshire, The Mens and the Cut, Sussex, and Bradfield Woods, Suffolk. (2) Linear regression analysis was used to analyse the relationship between the intensity of pollen deposition and tree basal area, and between pollen percentages and percentages of total basal area per plot for the major tree species. (3) The rank-order of taxa for number of pollen grains deposited per unit of tree basal area for all sites was Betula > Pinus > Taxus > Alnus > Quercus > Fraxinus - Fagus Salix. The correlations between pollen and basal-area data were significant for all tree taxa except Fraxinus, Fagus and Salix; the correlations for these taxa were higher when the data were expressed as percentages. There was a wide range of pollen representation for different shrub and herb taxa. (4) With the exception of Quercus, a consistent order of pollen representation for the major forest trees of northern Europe can be recognized. Senescent Quercus populations are a contributory factor to the low pollen representation of this taxon in south-east England. (5) The regression coefficients derived in this study can be used to convert fossil pollen counts into estimates of tree basal area for areas of 20-m radius around very small sedimentary basins. An example is given for a small woodland pond in Suffolk.

Holocene biomass burning and global dynamics of the carbon cycle

Fire regimes have changed during the Holocene due to changes in climate, vegetation, and in human practices. Here, we hypothesise that changes in fire regime may have affected the global CO2 concentration in the atmosphere through the Holocene. Our data are based on quantitative reconstructions of biomass burning deduced from stratified charcoal records from Europe, and South-, Central- and North America, and Oceania to test the fire-carbon release hypothesis. In Europe the significant increase of fire activity is dated approximately 6000 cal. yr ago. In north-eastern North America burning activity was greatest before 7500 years ago, very low between 7500-3000 years, and has been increasing since 3000 years ago. In tropical America, the pattern is more complex and apparently latitudinally zonal. Maximum burning occurred in the southern Amazon basin and in Central America during the middle Holocene, and during the last 2000 years in the northern Amazon basin. In Oceania, biomass burning has decreased since a maximum 5000 years ago. Biomass burning has broadly increased in the Northern and Southern hemispheres throughout the second half of the Holocene associated with changes in climate and human practices. Global fire indices parallel the increase of atmospheric CO2 concentration recorded in Antarctic ice cores. Future issues on carbon dynamics relatively to biomass burning are discussed to improve the quantitative reconstructions.

Estimating plant abundances from pollen percentages: The use of regression analysis

Abstract After describing the theory developed for estimating plant abundances from pollen numbers, we introduce a general model for linear regression of pollen percentages against plant percentages and describe the mathematical and empirical conditions under which this model is appropriate. The slope term derived from this model accounts for the over- and under-representation of plant percentages by pollen percentages, and the y -intercept term accounts for pollen from plants not growing in the area of vegetation sampled about each pollen site. The model is evaluated empirically by applying it to 181 paired pollen and tree samples from Wisconsin and western Upper Michigan. The pollen data are from the surficial sediments of lakes, and the tree data are estimated from tree-inventory plots within 30 km of each lake. A sum of the sixteen most important tree taxa is used to calculate the percentages. The evaluation begins with an examination of scatter diagrams to check whether the pollen percentages are linearly related to tree percentages for each taxon. These diagrams show a wide range in the patterns of scatter and allow identification of over-, equally-, and under-represented types. The calculation of regression parameters for eleven taxa simplifies the description of how the pollen/tree relationship differs among taxa. Because measurement errors affect both the pollen and tree values, a geometric-mean regression procedure is used. Pinus , the most over-represented type, for example, has the highest slope and intercept, whereas Acer has the lowest. Quercus is equally represented with a statistically significant positive y -intercept and a correlation coefficient between pollen and tree percentages of 0.9. Regression of the data from two subsets of 65 small-sized and 28 moderate-sized lakes shows the effects of basin size on how pollen percentages represent the tree percentages. These results show that caution is needed before any one set of regression results is used as “correction factors” to convert past values of pollen into estimates of past tree percentages. Such conversions, when possible, are only a first step in describing the past vegetation. Completing the description of past vegetation for several different spatial scales requires an integrated research design including study of pollen in basins of different size and the use of floristic data as well as quantitative paleoecological data. Full interpretation of these data will also require the matching of fossil spectra to modern spectra and the identification of key indicator species in fossil samples.

REGIONAL SPREAD AND STAND‐SCALE ESTABLISHMENT OF FAGUS SYLVATICA AND PICEA ABIES IN SCANDINAVIA

Palaeoecological studies are yielding fresh insights into slow forest ecosystem processes that are rarely observed using standard ecological methods, yet have major impacts on ecosystem function. Regional pollen data describe the broad features of the regional spread of trees but yield few insights into the processes of stand invasion and the facilitating role of disturbance. Pollen and charcoal data from small forest hollows are used to complement regional data in the study of the spread of Picea abies and Fagus sylvatica into southern Scandinavia during the last 4000 years. P. abies spread as a migrating front and preferentially invaded successional Betula stands, which had become particularly widespread in the region during the last 1000 years as a result of human activity. The spread of P. abies also closely tracked the changing area of suitable regional climate. The spread of F. sylvatica was more directly linked to anthropogenic activities and disturbance by fire prior to stand establishment. F. sylvatica preferentially invaded rich deciduous stand types that had declined in abundance during the last 2000 years. A recent range reduction of F. sylvatica can also be ascribed to human activity. The stand-scale palaeoecological data show how site conditions and disturbance are more important rate-limiting factors for F. sylvatica than for P. abies and help explain why F. sylvatica spread shows a patchy dynamic rather than the smoother migrating front of P. abies.

A long-term perspective on ungulate–vegetation interactions

The fossil record of vegetation and ungulates places present conditions and trends in a temporal perspective. Ungulate–vegetation interactions during the last 500 000 years were primarily driven by the climatic variation of the glacial–interglacial cycle. There were distinctive faunas associated with each temperate period and a loss of species diversity only in the present interglacial. Climate change and human activities have interacted during the most recent glacial cycle, accelerating extinction rates. This unique course of events has the consequence that no stable, ‘base-line’ conditions can be recognised. A review of the full-glacial ‘mammoth-steppe’ debate suggests that ungulate populations were limited by available forage, but a mosaic of habitat supported a diverse fauna in Beringia. In the debate over early–mid Holocene ‘wood pasture’, past ungulate populations are one of a range of disturbance factors, including burning, that influenced regional vegetation composition and structure in northern Europe. These debates concerning the scale and impacts of past ungulate–vegetation interactions will not be fully resolved until more is known about past ungulate population sizes. Modelling past scenarios would enhance the value of retrospective studies and help provide goals for management of near-natural ecosystems.

A two thousand year history of a northern Swedish boreal forest stand

. Successional processes within northern Swedish boreal forest are investigated for the last 2000 years by analysis of pollen, charcoal fragments and insect remains preserved in a deep mor humus layer on a small island in a large lake. Frequent disturbances by fire, blow-down, insect attack and flood have been the chief influence on vegetation dynamics. The gradual climatically induced replacement of Betula pubescens by Picea abies is superimposed on the disturbance dynamics. Picea first became established during the Little Ice Age, which was the longest period without fire in the record. A thin layer of volcanic ash from Iceland confirms the chronology at the onset of the Little Ice Age. The present importance of Pinus sylvestris and Ericaceae on the island is a product of a recent increase in disturbance frequency. The disturbance regime ensures that the communities stay in equilibrium with prevailing climate.

Climatic change, human influence and disturbance regime in the control of vegetation dynamics within Fiby Forest, Sweden

1. The immigration and rise to dominance of Picea abies within Fiby Forest, Sweden, was investigated by pollen and charcoal analysis of tiny (c. 1-m 2 ) bogs which contained a 4000-year record. 2. A mixed deciduous forest that existed 4000 years ago was replaced in a stepwise manner by a boreal community. The detailed dynamics of the changeover were controlled by catastrophic disturbances 2500 and 2200 years ago. 3. Two long periods of disturbance by burning and grazing kept the woodland open in structure. Eventual abandonment of this management practice initiated a succession that resulted in the present dominance of Picea, over 2000 years after its immigration into Fiby