Recurrent groups of pollen types in time

The ever-increasing taxonomic resolution of pollen and spores in pollen analysis, together with detailed macrofossil analyses, provides an opportunity for establishing groups of fossil types based on similar patterns of distribution in time and space. In delimiting such groups, patterns in dominant types prove to be as indicative for stratigraphical purposes as those of the rarer types. Arrangement of pollen and macrofossil curves has earlier been carried out by visual inspection in an attempt to derive ecologically realistic fossil groups. In this paper various multivariate statistical techniques are used in an attempt to discover which method(s) gives groups that can be used for reconstructing past vegetation. The following cases have been considered: (1) total fossil regional pollen, (2) total fossil local pollen, (3) recent local pollen in surface samples vs. coenoses in present-day vegetation, (4) fossil pollen vs. coenoses in contemporary vegetation, and (5) fossil local pollen vs. macrofossils from the same core. Only (1) and (2) are discussed here. It is concluded that, although the numerical methods should be adapted to take account of particular features of the data, general rules can be established that best serve the goal of palaeovegetational reconstruction. Canonical correspondence analysis or simple discriminant functions consistently prove to be the most suitable methods for the numerical arrangement of pollen and spore curves for the purpose of reconstructing past vegetation.

[1]  C.J.F. ter Braak.  Interpreting a hierarchical classification with simple discriminant functions: an ecological example , 1986 .

[2]  M. O. Hill,et al.  TWINSPAN: a FORTRAN program of arranging multivariate data in an ordered two way table by classification of individual and attributes , 1979 .

[3]  H. Birks,et al.  Modern pollen/land-use relationships as an aid in the reconstruction of past land-uses and cultural landscapes: an example from south Sweden , 1992 .

[4]  S. Peglar.  The development of the cultural landscape around Diss Mere, Norfolk, UK, during the past 7000 years , 1993 .

[5]  A. D. Gordon.  Numerical methods in quaternary palaeoecology. V. simultaneous graphical representation of the levels and taxa in a pollen diagram , 1982 .

[6]  Hilary H. Birks,et al.  Studies IN THE VEGETATIONAL HISTORY OF SCOTLAND. V. LATE DEVENSIAN AND EARLY FLANDRIAN POLLEN AND MACROFOSSIL STRATIGRAPHY AT ABERNETHY FOREST, INVERNESS-SHIRE , 1978 .

[7]  T. Webb,,et al.  Late- and Postglacial Climatic Change in the Northern Midwest, USA: Quantitative Estimates Derived from Fossil Pollen Spectra by Multivariate Statistical Analysis , 1972, Quaternary Research.

[8]  C. Janssen.  Modern Pollen Assemblages and Vegetation in the Myrtle Lake Peatland, Minnesota , 1984 .

[9]  Northwestern MinnesotaL,et al.  A Floristic Study of Forests and Bog Vegetation, Northwestern Minnesota. , 1967, Ecology.

[10]  C. Braak.  Canonical Correspondence Analysis: A New Eigenvector Technique for Multivariate Direct Gradient Analysis , 1986 .

[11]  L. Lefkovitch.  Hierarchical clustering from principal coordinates: an efficient method for small to very large numbers of objects , 1976 .

[12]  J. King,et al.  Pollen of the North American quaternary: The top twenty , 1976 .

[13]  J. C. Ritchie,et al.  Recent pollen assemblages from the Western Interior of Canada , 1968 .

[14]  C. Janssen.  Stevens Pond: A Postglacial Pollen Diagram from a Small Typha Swamp in Northwestern Minnesota, Interpreted from Pollen Indicators and Surface Samples , 1967 .

[15]  I. Ferguson,et al.  The Northwest European Pollen Flora, I , 1978 .

[16]  C. Janssen.  The Palaeoecology of Plant Communities in the Dommel Valley, North Brabant, Netherlands , 1972 .

[17]  H. J. B. Birks,et al.  Numerical Methods in Quaternary Pollen Analysis. , 1989 .

[18]  C.J.F. ter Braak,et al.  The analysis of vegetation-environment relationships by canonical correspondence analysis , 1987 .

[19]  E. Maarel,et al.  The Braun-Blanquet Approach , 1978 .