The Basis Of Pollen Analysis

Paleoclimatic reconstruction by pollen analysis is possible thanks to four basic attributes of pollen grains: (1) they possess morphological characteristics that are specific to a particular genus or species of plant; (2) they are produced in vast quantities by wind-pollinated plants, and are distributed widely from their sources;

(3) they are extremely resistant to decay in certain sedimentary environments; and

(4) they reflect the natural vegetation at the time of pollen deposition, which (if viewed at the right scale) can yield information about past climatic conditions.

9.2.1 Pollen Grain Characteristics

Pollen grains range in size from 10-150 jjim and are protected by a chemically resistant outer layer, the exine. Because pollen grains of many plant families are different morphologically, they can be recognized by their distinct shape, size, sculpturing, and number of apertures (Fig. 9.2). In some cases, identification to species level may be possible (Faegri and Iversen, 1975; Moore and Webb, 1978). The exine is made of sporopollenin, a complex polymer resistant to all but the most extreme oxidizing or reducing agents. Thus, the organic or inorganic matrix in which the pollen grains are trapped can be removed by chemical means without de-

Annually laminated


Mor humus

Climatic Change Soil Development


Clearance patterns



Succession Competition

Fire, Disease, Disturbance

: Buried SMALL HOLLOWS * soils__

10 10 Time (years)

FIGURE 9.1 Scale is important in defining the information that can be obtained from pollen analysis. Information about climate is obtained at the larger temporal and spatial scales, mainly from lakes and peatlands.At smaller spatial and temporal scales, nondimatic effects dominate the pollen signal (Bradshaw, 1994).

stroying the pollen itself. There is some evidence, however, that in certain sedimentary environments not all pollen grains will be equally well preserved (Cushing, 1967). For example, pollen grains are more subject to corrosion in moss peat than in silt deposits and this may be due to the activities of phycomycetes, bacteria, and other micro-organisms. Furthermore, the pollen of some species (e.g., Populus) may begin to disintegrate even before reaching a deposition site (Davis, 1973).

9.2.2 Pollen Productivity and Dispersal: the Pollen Rain

All plants that participate in sexual reproduction produce pollen grains, dispersing them by various mechanisms in an endeavor to reach and fertilize the female reproductive organs of other plants. The amount of pollen produced is generally inversely proportional to the probability of success in fertilization; thus plants using insects or animals as a dispersal agent (entomophilous or zoophilous species) produce orders of magnitude less pollen than those dispersing pollen by wind (anemophilous species). By the same token, plants that are self-fertilizing (autogamous or cleistogamous species) produce only minute quantities of pollen compared to anemophilous species. Because of these factors, even though the vast majority of flowering plants are insect-pollinated, the accumulation of pollen grains at any given site will usually be dominated by pollen of anemophilous species. A single oak tree may produce and disperse by wind more than 108 pollen grains per yr, and hence the pollen from an entire forest (the pollen rain) assumes astronomical proportions. Pollen accumulation in a

Pollen Types
FIGURE 9.2 Some of the principal pollen types in British Holocene deposits, drawn to the same scale. Common names of plants are given in Table 9.1 (Godwin, 1956).

TABLE 9.1 Some Important Plant Taxa in North American and European Quaternary Palynology (see Fig. 9.1).

Genus Family Common name

Abies Fir






































Sweet gum














Douglas fir






Bald cypress









" Ostrya and Carpinus pollen are indistinguishable and are generally considered together. ''Exotic pollen added to samples for pollen influx calculations (see Section 9.2.4).

northern hardwood forest may reach 80 kg ha"1 a"1 (Faegri and Iversen, 1975). Pollen production by entomophilous species is generally several orders of magnitude lower and autogamous species produce even less. In some cases, an entomophilous species, such as Tilia, may produce fairly large amounts of pollen but the relatively efficient dispersal mechanism (via insects) means that pollen grains are rarely found in large numbers, even in forests where Tilia is abundant (Janssen, 1966).

9.2.3 Sources of Fossil Pollen

As pollen is an aeolian sediment, pollen falling on sites where organic or inorganic sediments are accumulating will become part of the stratigraphic record (Traverse, 1994). Pollen has thus been recovered from peat, lake sediments, alluvial deposits, estuarine and marine sediments, and glacial ice. Pollen has also been recovered from archaeological sites (Dimbleby, 1985), rat middens (King and Van Devender, 1977), and coprolites (fossilized fecal matter of animals; Martin et al., 1961). In Quaternary palynology, the principal sources of paleoclimatic information are peat from bogs and marshes, and sediments from relatively shallow lakes (Jacobson and Bradshaw, 1981). In many lakes, sedimentation rates are often quite high, providing an opportunity to recover samples with a high temporal resolution, generally an order of magnitude greater than in marine sediments. Terrestrial pollen studies can thus provide a temporal perspective on climatic changes rarely possible in a marine setting.

The vast majority of pollen grains dispersed by wind are not carried more than 0.5 km beyond their source. Dispersal by wind is a function of grain size, the larger and heavier grains falling to the ground sooner than the smaller, lighter grains (Dyakowska, 1936). Pollen grains of beech (Fagus) and larch (Larix) for example, are relatively heavy and settle out close to their source. Consequently, the occurrence of fossil beech or larch grains in a deposit would indicate the former growth of a species in the immediate vicinity of the site. Field measurements of pollen dispersal from artificial sources and from isolated stands of vegetation indicate that pollen produced by an individual plant is not identifiable above background levels (the regional pollen rain) beyond a few hundred meters. This is also indicated by theoretical dispersal models (Tauber, 1965). Many investigators thus favor the analysis of sediments from fairly large lakes (>1 km2) because they act as catchment basins for the regional pollen rain and are not unduly influenced by vegetation in the immediate vicinity of the sampling site (Prentice, 1985).

Much work has been conducted on the problems of pollen transport and sedimentation in lake basins (Pennington, 1973; Holmes, 1994). Just as in the atmosphere, differential settling of pollen grains occurs in water also, with the result that the original ratios in which pollen enters the lake from the air may be distorted, with the lighter pollen grains preferentially deposited in the littoral zone. Pollen is also concentrated in lake basins by inflowing streams, especially during periods of heavy runoff. Furthermore, resuspension and redeposition of pollen grains during periods of turbulent mixing, particularly in shallow water, also occur, thereby smoothing out yearly variations in pollen and sediment inputs to the lake. Further smoothing may result from the activities of burrowing worms and other mud-dwellers

(Davis, 1974). Thus, very high resolution studies (annual to decadal) are not practicable except perhaps in annually laminated (varved) sediments (Swain, 1978). In any case, this temporal scale is not appropriate for climatic reconstruction from pollen, as the climatic signal (recorded through changes in vegetation) will not be strong at this scale. Other nonclimatic factors would likely overwhelm any climatic signal. As the sampling timescale increases, the climatic signal begins to predominate over nonclimatic noise (Bradshaw, 1994). By careful interpretation of the record at the appropriate temporal scale and aggregation of data at the regional scale, the fundamental climatic signal can be distilled from the array of other factors affecting the accumulation of pollen in lakes.

9.2.4 Preparation of the Samples

In order to isolate pollen grains and spores from the matrix of organic or inorganic sediment, rigorous chemical treatment by hydrochloric, sulfuric, and hydrofluoric acid is generally required, as well as acetolysis by a mixture of acetic anhydride and sulfuric acid (for details, see Moore and Webb, 1978; or Faegri et al., 1989). Removal of the matrix enables the remaining pollen grains and spores to be seen clearly when stained and mounted on slides for microscopic analysis. Generally, the original core is sampled at intervals of a few centimeters (depending on the sedimentation rate) and slides are prepared of pollen and spores at each level. These are then examined and the number of different grains in each sample are noted. Although the total number of grains counted at each level would depend on the purpose of the study and the source of material being studied (Moore and Webb, 1978), at least 200 grains are usually counted.

For pollen flux density calculations (see Section 9.4), the number of pollen grains counted on each slide must be related to the total pollen content of the sample from the level being considered. The most widely used method is to add a known quantity of exotic pollen or spores (e.g., Eucalyptus or Lycopodium) to the sample initially and then to count the number of these grains that occur on the final slide preparation. The ratio of exotic pollen counted to the number of exotic pollen added originally can be used to estimate the total pollen content of the original sample (Stockmarr, 1971; Bonny, 1972).

9.2.5 Pollen Rain as a Representation of Vegetation Composition and Climate

Differences in pollen productivity and dispersal rates pose a significant problem for the reconstruction of vegetation composition because the relative abundance of pollen grains in a deposit cannot be directly interpreted in terms of species abundance in the area. It is necessary to know the relationship between plant frequency in an area and the total pollen rain from that plant species in order to use pollen data to calculate the actual composition of the surrounding vegetation. For example, a vegetation community composed of 10% pine, 35% maple, and 65% beech may be represented in a deposit by approximately equal amounts of pine, maple, and beech pollen, because of the differences in their pollen productivity and disper sal rates. However, for paleoclimatic reconstruction, these matters are of less significance. What the paleoclimatologist needs to know is whether there are patterns in the pollen data that can be calibrated in terms of climate. Again, the question comes down to the appropriate scale of analysis. At the smallest spatial scale, and the shortest temporal scale, the signal represented by pollen will be dominated on the one hand by short-term (synoptic scale) factors affecting dispersal of pollen, and on the other by local factors affecting plant growth. Stepping back from this complexity reveals the desired climate information.

Palynologists adopt the uniformitarian principle: the present is the key to the past. By using spatial relationships in modern pollen distribution and their relationship to modern climate as a guide to interpreting pollen patterns recorded in the past, paleoclimatic reconstructions can be made. Modern vegetation communities can be considered as analogs of former vegetation cover; if pollen assemblages in the modern pollen rain resemble fossil pollen assemblages, the former vegetation and its associated climate is assumed to be similar to that in the analog region today. If similar pollen assemblages cannot be found today, presumably no modern analog for the former vegetation cover and climate exists (Ritchie, 1976). The main problem with this approach is the difficulty of sampling the vast array of possible vegetation assemblages in the modern landscape in order to find a good analog, and the fact that much "natural vegetation" in both the Old and New Worlds has been destroyed or greatly modified. Nevertheless, even in such modified environments, on the regional scale (~103 km) modern pollen still contains a climatic signal that allows differentiation of regional-scale climatic conditions.

Pollen rain in mountainous regions poses particular problems of interpretation, because vegetation communities may be confined to narrow climatic zones along mountainsides (Maher, 1963). In parts of South America, for example, vegetation may grade all the way from equatorial rainforest below 500 m to tundralike páramo (or the drier puña) above 3500 m, over a horizontal distance of less than 100 km. Nevertheless, modern pollen rain studies demonstrate that good discrimination is possible between different vegetation zones, even where complex spatial interfingering of different vegetation communities occurs (Salgado-Labouriau, 1979; Gaudreau et al., 1989; Lynch, 1996). Pollen from lower vegetation zones is commonly carried upwards to higher elevations by daytime upslope winds. However, with increasing elevation, gradient winds predominate, so that there may be an upper limit to upslope pollen transport (Markgraf, 1980). Pollen from high elevations is not dispersed very far beyond the high-altitude vegetation zone (Hamilton and Perrott, 1980). Characteristic pollen rain assemblages corresponding to different elevations can thus be identified and used to reconstruct altitudinal changes in vegetation through time (Salgado-Labouriau et al., 1978). Such changes may be converted to paleoclimatic estimates if modern climatic controls on the altitudinal limits of different taxa are known, although consideration must be given to constraints imposed on vegetation by changing C02 levels during glacial periods (Jolly and Haxeltine, 1997). At those times, vegetation may have changed more in response to C02 levels than to a simple drop in temperature.

Was this article helpful?

+1 -1


  • jonas
    How to calculate pollen influx?
    1 year ago
  • Celio
    Which form of paleoclimatology is the study of pollen grains?
    1 year ago

Post a comment