A reconstruction of the history of a Holocene sand dune on Great Barrier Island, northern New Zealand, using pollen and phytolith analyses

Abstract

Aim To reconstruct the history of a Holocene sand dune using pollen and phytolith analyses, and to identify the strengths, weaknesses and compatibility of these two methods in the interpretation of Quaternary coastal environments. Location Great Barrier Island, northern New Zealand. Methods Pollen and phytolith analyses were carried out on a sequence through a Holocene sand dune containing a palaeosol. Results Phytoliths were present throughout the sequence. Grass phytoliths increased at the expense of tree phytoliths following fire disturbance. Pollen (and spores) was preserved only in the palaeosol part of the profile. Pteridium fern spores increased at the expense of tall tree pollen following the fire disturbance. Main conclusions Lack of phytolith production by many species and problems of taxonomic specificity in many others restricts the usefulness of phytolith analysis to defining only broad vegetation types. In New Zealand, gymnosperms are invisible in the phytolith record and ferns are extremely under‐represented. In contrast, pollen analysis usually provides a great deal of information regarding the composition of a particular vegetation type. The loss of microscopic charcoal fragments during the phytolith extraction process is a disadvantage in the reconstruction of environments where fires have occurred. The greater durability of phytoliths compared with pollen means that phytoliths may be found in sediments where pollen has not been preserved. The phytolith record may also provide evidence of wetter environments that are not apparent in the pollen record. Unlike grass pollen, which is widely dispersed and therefore blurs the spatial record, the presence of grass phytoliths in sediments indicates a local source. The simultaneous application of both methods potentially provides a powerful tool in ecological interpretation and the reconstruction of Quaternary coastal environments.

Phytolith transport in soil: A field study using fluorescent labelling

“…This diagnostic feature is widely applied to reconstruct the palaeovegetation. During the last decade many studies on vegetation dynamics during the Holocene and/or Pleistocene have been conducted based on the analysis of phytoliths extracted from buried soils, lake deposits, archaeological sediments, and recent soils (Fisher et al, 1995;Gol`yeva et al, 1995;Piperno and Becker, 1996;Blecker et al, 1997;Alexandre et al, 1999;Gol'eva and Aleksandrovskij, 1999;Horrocks et al, 2000;Blinnikov et al, 2002;Delhon et al, 2003;Piperno and Jones, 2003;Gallego et al, 2004;Piperno, 2006a,b, Barczi et al, 2009. In recent soils the procedure for reconstructing palaeovegetation commonly involves extracting phytoliths from different soil depths, the frequency counting of phytolith morphotypes in each depth, and interpreting the phytolith records by relating variations in phytolith morphology to past changes in the vegetation cover (Fisher et al, 1995;Gol`yeva et al, 1995;Piperno and Becker, 1996;Gallego et al, 2004).…”

Phytolith transport in sandy sediment: Experiments and modeling

“…Diagnostic features of soil phytoliths are widely used in paleoenvironmental reconstructions (Golyeva et al, 1995;Piperno and Becker, 1996;Alexandre et al, 1999;Fisher et al, 1995;Golyeva and Alexandrovskij 1999;Horrocks et al, 2000;Blinnikov et al, 2002;Delhon et al, 2003;Gallego et al, 2004;Blecker et al, 1997). Although the mean diameter of plant silica ranges from 0.1 μm to 200 μm, only phytoliths with mean diameter ≥5 μm can be used for plant diagnostics (Piperno, 1988).…”

EDF phytolith analysis from Heying site, southeast China, Shang‐Zhou dynasties

“…Currently, phytolith analysis is widely applied in archeological research. They serve as efficient clues to paleo-environmental reconstruction, paleoclimate, and human activities through analysis and interpretation of phytolith assemblages from various sediments (e.g., Ball and Brotherson, 1992;Ball et al, 1996;Bowdery et al, 2001;Horrocks et al, 2000;Pearsall, 1978Pearsall, , 2000Pearsall, , 2002Piperno, 1984Piperno, , 1988Piperno, , 2006Piperno and Pearsall, 1993;;Rosen, 1992;Rovner and Russ, 1992).…”