Modern Pollen Sampling

Some basis for interpretation of Quaternary palynological records is derived from examination of patterns and processes of modern pollen deposition recorded in pollen traps and surface sediments. Such sampling examines deposition both within the rainforest and outside the rainforest in lake, swamp, and marine environments used for reconstructions of vegetation history.

The first quantitative study of pollen deposition in rainforest was by Flenley (1973). In the lowland rainforest of Malaysia he found significant pollen influx (between 800 and 2,020 grains/cm2/annum) and relatively high pollen diversity (60 to 62 taxa) although representation within taxa through time was very variable. Similar results were found by Kershaw and Strickland (1990) in a north Queensland rainforest. They also found, from a knowledge of the distribution of trees surrounding the traps, that two-thirds of the pollen could have been derived from within 30 m of the traps. An examination of traps situated less than 100 m outside rainforest, in a small crater lake on the Atherton Tableland in north Queensland, demonstrated an enormous reduction in pollen deposition and substantial sifting out of pollen of local producers (Kershaw and Hyland, 1975). Pollen influx values dropped to below 200 grains/cm2/annum and spectra were dominated by a relatively small number of taxa with significant regional pollen dispersal. It was determined that there was about equal representation of pollen from above canopy and rainout components. Any trunk space component was small and the high degree of correspondence between trap assemblages and those derived from the topmost part of a sediment core from the lake (Kershaw, 1970) suggested also that there was little inwash of pollen, though this component may have been trapped by marginal swamp.

Despite the great variability of pollen deposition within rainforest, patterns of representation appear to reflect systematic vegetation variation on a regional scale. Numerical analysis of a number of surface litter samples from throughout the lowland and sub-montane forests of northeast Queensland (Kershaw, 1973; Kershaw and Bulman, 1994) revealed a similar pattern to floristic analysis of forest plots from which the samples were derived. Although there was little in common between taxon representation and abundance in the two groups, it suggested that pollen assemblages could be used to characterize the broad environmental features of the landscape, including the vegetation. A similar result was achieved with the use of percentages of only those taxa that had been identified from lake-trapping and existing fossil pollen records as regionally important. This finding indicated the potential for analysis of pollen diagrams from tropical rainforest in a similar manner to those from other vegetation types where variation in abundance of a small number of taxa provides the basis for interpretation. Bioclimatic estimates for such "common taxa" in northeastern Queensland (Moss and Kershaw, 2000) demonstrate their potential for quantitative paleoclimatic reconstruction (Figure 4.2). The presence of numerous other taxa can allow refinement of interpretation (Kershaw and Nix, 1989) although insufficient pollen may be present in samples to allow counts of a sufficient size to demonstrate presence or absence in potential source vegetation. Figure 4.2 also demonstrates the degree of penetration into rainforest of pollen from the dominants of surrounding sclerophyll vegetation, Eucalyptus and Casuarina, that have generally wider pollen dispersal than rainforest taxa.

The heterogeneous nature of lowland tropical rainforest is an impediment to determination of the actual sources of "common" taxa and, therefore, their relative degree of dispersal. This complication is reduced at higher altitudes where widely dispersed taxa, many of which are clearly wind-pollinated, make up significant and identifiable components of the vegetation. The compilation of Flenley (1979) provides an excellent summary of variation in pollen representation along an altitudinal transect in New Guinea (Figure 4.3). Above the highly human-modified vegetation, clearly recognized by high values of Poaceae or Casuarina, the montane zones of oak and beech forest are dominated by pollen of their dominant taxa: Lithocarpus/ Castanopsis and Nothofagus, respectively. Upper montane mixed forest is characterized by Quintinia while alpine vegetation is recognized by the only occurrences of "alpine pollen taxa". The bare ground on the mountain summit has a unique pollen signature that clearly identifies those taxa, Nothofagus and Casuarina, which have wide pollen dispersal. Flenley (1979) remarks on the tendency for pollen to be carried uphill and suggests it is due to the fact that pollen is released during the day when anabatic winds are active.

A much broader indication of pollen transport, including a potentially major water-transported component, is provided by recent analyses of suites of core-top pollen samples from the Indonesian-Australian region (van der Kaars, 2001; van der Kaars and De Deckker, 2003; van der Kaars, new data) and the South China Sea (Sun et al., 1999). Isopolls interpolated from samples along the steep precipitation gradient from east Indonesia to northwest Australia are shown for major pollen groupings based on a dryland pollen sum, excluding pteridophytes (Figure 4.4). This gradient is clearly reflected in the pollen with predominantly rainforest taxa including pterido-phytes showing high values in the rainforested Indonesian region and then progressively declining relative to the predominantly sclerophyll taxa of Myrtaceae (attributable mainly to Eucalyptus) and Poaceae that dominate Australian vegetation. Compared with other pollen types, the pollen of rainforest angiosperms reflect most faithfully the distribution of rainforest. Rainforest conifers are much better represented than angiosperms considering their almost total restriction to montane forests, a feature no doubt due to obligate wind dispersal of pollen and greater opportunity for wind transport from higher altitudes. The major concentration of montane pollen types between Sulawesi and New Guinea reflects also the proximity to mountainous areas within the study area. Pteridophyte spores have a very similar distribution to the rainforest conifers and, although this pattern can be accounted for—to some degree—by the fact that they are most abundant in wet tropical and often montane forest, transport is facilitated also by water. The fact that percentages of pteridophytes are so much higher than those of pollen is probably the result of effective water transport.

26 1

An =


Ca =


Cu =

Cu non i aceite

El =


Ma =

Maca ranga

Ol =


Po =


Tr =


LC =

Lynch's Cráter

Modern pollen

8«) 1000 1200 1400 1600 1800 2000 2200 2400 2600

Mean annual precipitation (mm)






Figure 4.2. Climatic ranges for highest representation of major rainforest taxa in relation to bioclimatic estimates for modern pollen samples from northeast Queensland rainforests. The extent of penetration of high values for the sclerophyll woodland taxon Casuarinaceae is also shown (adapted from Moss and Kershaw, 2000).

oc CD


Xcc o


Was this article helpful?

0 0
Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook

Post a comment