Assessment of the micro‐sieving technique for pollen analysis
Dept. EPHE Paleoclimatology and Marine Paleoenvironments, UMR EPOC, Bordeaux 1 Univ.
Our preparation technique for pollen (and spores) analysis includes a 10‐μm mesh
sieving to eliminate small non‐palynomorph particles. This technique is commonly used in
marine sediments (e.g. Heusser and Stock 1984; Leroy and Dupont 1994; Sánchez Goñi, et al.
1999; Dupont et al. 2008) to concentrate pollen grains and spores. However, micro‐sieving
has been criticized in a number of works based on experiments comparing pollen counts of
sieved and unsieved residues (e.g. Tzedakis et al. 2004; Roucoux et al. 2005; Margari et al.
2010). Sieving has been reported to result in an unacceptable level of differential pollen loss,
particularly affecting Poaceae pollen grains because they can be small in size and often
crumpled (Roucoux et al. 2005). These experiments are presented in Roucoux PhD
dissertation (Roucoux 2000) but are still unpublished. Therefore, the protocol set up for
these experiments is not available.
To test whether microsieving affects substantially the proportion and composition of
the different morphotypes included in the pollen assemblages we have performed new
experiments in our laboratory. We have analyzed sample U1386A 15H CC (IODP expedition
n°339 “Mediterranean Outflow Expedition”) collected in the southwestern Iberian margin.
The sediment was divided in two aliquots. Our routine protocol, including the 10 μm mesh
sieving (http://ephe‐paleoclimat.com/ephe/Lab%20Facilities.htm), was applied to one
aliquot, and the same protocol without sieving to the other aliquot. A known concentration
of Lycopodium spores (one tablet of 20,428 spores) was added to each aliquot to calculate
the pollen concentration of sieved and unsieved residues. Slides were prepared using a
mobile mounting medium to allow rotation of pollen grains. Pollen quantification was done
under a Zeiss Primo Star light microscope at 400× magnifications with regular use of 1000×
magnification for the identification of pollen and spores. Pollen percentages for terrestrial
taxa were calculated against the main sum of terrestrial grains, while percentages for Pinus,
aquatics and spores were calculated against the total sum of all pollen and spores.
Counting one slide from the sieved residue was enough to reach a minimum of 100
pollen grains, excluding Pinus which is overrepresented in marine sediments from the
western European margin (Turon 1984a, 1984b), and 20 pollen morphotypes. These two
criteria are needed to obtain pollen percentage assemblages that accurately reconstruct the
vegetation cover and composition (McAndrew and King 1976; Rull 1987). In contrast, the
first slide of unsieved residue (1) showed low amount of pollen grains, most of the mounted
residue being composed of non‐palynomorph particles. We were obliged to count a second
slide to get a pollen sum higher than 100 pollen grains (slides 1+2). At that point, the total
pollen concentrations of the sieved and unsieved slides from the same sample disagreed.
However, the number of Lycopodium spores counted was still low and the sum of
Lycopodium spores and pollen grains did not reach 200, which is considered the minimum
number to obtain accurate concentration values (Finsinger and Tinner 2005). For this reason,
we mounted two additional slides, 3 and 4, with higher amount of residue than that of the
previous slides. Both slides contained a high amount of non‐palynomorph particles and were
difficult to analyse. Despite that, we reached more than 100 pollen grains and 20
morphotypes in each slide and summed up more than 200 items. Concentrations of slides 3
and 4 were considered reliable and more similar to the pollen concentration of the sieved
slide. The number of identified taxa was also similar between sieved and unsieved slides,
oscillating between 18 and 22 taxa.
Table 1‐ Amount of the two treated aliquots, sieved and unsieved, from the same sample
U1386A 15H CC and pollen concentration of the different slides counted.
We observed that the concentration of the sieved slide is 25% lower than that of the
unsieved slides. However, the losses affect all types of pollen grains in the same proportion,
excluding Pinus (Figure 1). There is no a differential loss of Poaceae pollen grains contrarily
to what has been suggested in previous unpublished works. In contrast, the concentration of
Pinus pollen grains was similar in both sieved and unsieved slides. The big size of Pinus
pollen, 5 times the mean size of the pollen grains of the studied area (20 μm), would
preclude any leak of this taxon through the 10 μm mesh.
Figure 1‐ Pollen concentrations of the main taxa (pollen grains/g) in the sieved and sum
We calculated the pollen percentages in the sieved and unsieved slides to evaluate if
the observed loss of pollen grains in the sieved residue produces a substantial difference in
the pollen percentages between the sieved and unsieved residues (Figures 2 and 3).
Figure 2‐ Pollen percentages of the main taxa in the sieved and sum unsieved slides.
Figure 2 shows that the difference of the pollen percentages between sieved and
unsieved slides of each morphotype, excluding Pinus and corroded grains, is less than 5%
that is within the counting error estimated for pollen sums of 100 grains (Fletcher and
Sanchez Goñi 2008). Therefore, our results showed that Poaceae pollen percentages were
not differentially affected by the application of the 10 μm microsieving technique. In
contrast, the comparison between the pollen assemblages from the sieved slide and the
unsieved slides 1+2 highlights the distortion produced by the counting of low amount of
palynomorphs in unsieved slides (Figure 3). The difference in the pollen percentages of each
taxon is of 50% in average between the sieved and unsieved slides.
Figure 3‐ Pollen percentages of the main taxa in the sieved and unsieved slides 1+2.
In conclusion, the microsieving technique at 10 μm mesh produces a loss of pollen
grains, which affects similarly all the morphotypes excluding Pinus. This lost does not
produce, however, any distortion in the percentages of the different morphotypes included
in the pollen assemblage. Therefore, we recommend the inclusion of microsieving in the
pollen preparation protocol applied to marine sediments. This technique is a valuable
technique to concentrate the pollen grains and produce clear residues that help the
identification and quantification of pollen grains. This test demonstrates, in turn, that
counting only 100 pollen grains in unsieved sediment does not give a fair representation of
the sample pollen assemblages.
Dupont, L. M., H. Behling, et al. (2008). "Thirty thousand years of vegetation development
and climate change in Angola (Ocean Drilling Program Site 1078). Climate of the Past 4: 107‐124.
Finsinger, W. and W. Tinner (2005). Minimum count sums for charcoal‐concentration
estimates in pollen slides: reliability and potential errors. The Holocene 15: 293‐297.
Fletcher, W. J. and M. F. Sanchez Goñi (2008). Orbital‐ and sub‐orbital‐scale climate impacts
on vegetation of the western Mediterranean basin over the last 48,000 yr.
Quaternary Research 70 451‐464.
Heusser, L. E. and C. E. Stock (1984). Preparation techniques for concentrating pollen from
marine sediments and other sediments with low pollen density. Palynology 8: 225‐
Leroy, S. A. G. and L. Dupont (1994). Development of vegetation and continental aridity in
northwestern Africa during the Late Pliocene: the pollen record of ODP Site 658.
Palaeogeography, Palaeoclimatology, Palaeoecology 109: 295‐316.
Margari, V., L. C. Skinner, et al. (2010). The nature of millennial‐scale climate variability
during the past two glacial periods. Nature Geoscience 3: 127‐133.
McAndrew, J. H. and J. E. King (1976). Pollen of the North American Quaternary: The top
twenty. Geoscience and Man 15: 41‐49.
Roucoux, K. H. (2000). Millennial scale vegetation and climate variability in north‐west Iberia
during the last glacial stage. Cambridge, University of Cambridge. PhD Thesis: 267.
Roucoux, K. H., L. de Abreu, et al. (2005). The response of NW Iberian vegetation to North
Atlantic climate oscillations during the last 65 kyr. Quaternary Science Reviews 24:
Rull, V. (1987). A note on pollen counting in paleoecology. Pollen et Spores 29(4): 471‐480.
Sánchez Goñi M.F., Eynaud F., Turon J.L., Shackleton N.J.,1999. High resolution palynological
record off the Iberian margin: Direct land‐sea correlation for the Last Interglacial
complex. Earth and Planetary Science Letters, 171(1): 123‐137.
Turon, J.‐L. (1984a). Direct land/sea correlations in the last interglacial complex. Nature 309: 673‐676.
Turon, J.‐L. (1984b). Le palynoplancton dans l'environnement actuel de l'Atlantique nordoriental.
Evolution climatique et hydrologique depuis le dernier maximum glaciaire.
Bordeaux, Université de Bordeaux I.
Tzedakis, P. C., K. H. Roucoux, et al. (2004). The duration of forest stages in southern Europe
and interglacial climate variability. Science DOI:10.1126/science.1102398.