Micro‐scale patterns of modern pollen deposition within three alpine plant communities*

Abstract

SUMMARYFew studies have examined patterns of pollen deposition or the relationship between vegetation and pollen assemblages at the small scale. This paper examines the representation of both individual taxa and whole plant communities in modern pollen spectra. Micro-scale variations in modern pollen deposition were studied in three alpine plant communities on Storbreen glacier foreland, southern Norway: a pioneer community; a 5a/ij:-shrub community; and an ericaceous heath community. The relationships between vegetation and modern pollen deposition and the variability of these relationships at the local scale were investigated within 16 >: I m^-quadrats. The data were classified using two-wa}-indicator-species ana]>'sis (TWINSPAN) and ordinated usin^ dftrendedcorrespondence analysis (DCA). In addition, the pollen and vegetation data were ordinated simultaneously, using canonical correspondence analysis (CCA). The efTect of removing arboreal pollen from the pollen sum was examined. The three plant communities were distinguished successfully by multivariate statistical techniques, using both pollen data and vegetation data. However, within the communities there was high variability in pollen deposition, which was most pronounced in the 5i3//s'-shrub community. The abundance of indi\ idual taxa in the vegetation was compared with their representation in the pollen spectra. The strongest relationships were shown by Salix, Ericales, Empetritm and Lycopodium. Certain taxa showed consistent, reproducible patterns of behaviour. These patterns provided valuable evidence for separating contrasting plant communities. Comparison of modern poiien-vegetation relationships of this kind with pollen data from Holocene and Pleistocene sites is important to improve the reconstruction of past plant communities. Statistical analyses showed that for the majority of taxa a moss polster consisting of 20 subsamples was necessary to reduce the effect of microscaie variability in this environment to acceptable levels.

Comparing pollen spectra from modified Tauber traps and moss samples: examples from a selection of woodlands across Europe

“…For example, the best modern analogue of a fossil sample might be a single moss sample, since fossil samples from a peat core are also from a single location. However, previous studies such as Pardoe (1996) have shown a high degree of local variability in pollen assemblages collected in moss samples over a small area. It might, therefore, be preferable to combine several subsamples since this will be more representative of the local vegetation.…”

“…The collection of surface pollen in moss samples for calibration datasets or modern analogues is now a wellestablished technique (see for example Hicks 1977;Hjelle 1998;Caseldine 1989;Pardoe 1996Pardoe , 2001Pardoe , 2006Tonkov et al 2001). Moss samples are also used extensively in studies of pollen productivity and dispersal (Räsänen et al 2004), such as those published by Broström et al (2004Broström et al ( , 2008.…”

Herb pollen representation in surface moss samples from mown meadows and pastures in western Norway

“…These occurrences are easily detected in the scatter plots, and may argue against the common use of the amalgamation of pollen samples in surface-sample investigations (e.g. Adam and Mehringer 1975;Berglund et al 1986;Pardoe 1996). Combinations of pollen samples will result in a lower mean value, but nevertheless a value higher than that obtained by natural dispersal (cf.…”

Replicability of data collected for empirical estimation of relative pollen productivity

“…Micro-scale variation in pollen assemblages has been little studied though it may be significant (Pardoe, 1996), especially where there is micro-topographical variation within the sampling site (Kitcher, 2014). Earlier studies have recommended an amalgam of mosses from multiple points within a small area (Broström et al, 2004;Räsänen et al, 2004; et al, 2008), and although this has its own limitations (for example, in areas where mosses are sparse such as meadows multiple mosses will not occur in close proximity; ensuring sampling is unbiased both in the field and when subsampling for extraction requires extra time and care; a single outlying assemblage will raise the average of the taxa concerned in the mixed sample, and is less likely to be detected) it may give more representative results.…”

“…Taking multiple samples within a small area of known radius (e.g. Broström et al, 2004;Mazier et al, 2008) to mitigate against micro-scale variations in pollen influx has been shown to be effective for both moss polsters (Pardoe, 1996) and surface soil samples (Adam and Mehringer, 1975).…”