Reconstructing initial human settlement impacts on the vegetation of New Zealand
Principles of pollen analysis
Pollen and spores
Preserved pollen and spores are extracted from peat or sediment cores, counted under a microscope, and used to discover what plants were growing on the landscape in the past. What are pollen and spores?
Seed-producing plants (the angiosperms and gymnosperms) produce vast numbers of pollen grains, which form in the anthers and contain the male gamete part of the plant. The pollen grains need to reach the stigma, the female part of the plant, in order for fertilisation to take place to produce new offspring. Ferns, mosses, and other lower plants produce spores instead of pollen to reproduce. Spores are also released in huge numbers, but all they need to do is be dispersed to a suitable habitat and they can grow into a new plant.
Pollen and spores are extremely small. You might have seen a greenish powdery substance at various times of the year collecting in the gutters, on your car, or being blown across valleys on a windy day. Chances are that what you are seeing are millions of tiny pollen grains blowing around in the wind. On closer inspection under a microscope you can see that they vary in size (7–100 micrometers) and structure. It is the unique structuring patterns on the exterior wall of pollen grains and spores that allows us to identify most of them to species, or in some cases to their genus or family. In New Zealand our largest pollen grains are from the tall forest tree miro (Prumnopitys ferruginea); the smallest from a forget-me-not (Myosotis). Tree ferns and many ground ferns typically produce rather large and robust spores.
Pollen and spores |
||
 |
 |
 |
| Silver beech | Totara | Matai |
 |
 |
 |
| Grass | Bracken | Willow |
Pollen structure may relate to how the plants are pollinated, whether by wind, water, insects or birds. For example, the pollen of species that are insect pollinated are often sticky or spiny, whereas wind-pollinated species may produce smooth pollen (e.g. the grasses), or pollen with air sacs and a low specific gravity (e.g. podocarp trees like tōtara, miro, mataī and kahikatea) to make them more buoyant and able to blow long distances in the wind. Wind-pollinated plants tend to produce more pollen than insect- or bird-pollinated plants, as they have a harder time trying to land on receptive stigmas.
TopSediments as natural archives
Every time a plant produces pollen and spores, most will end up landing on the ground somewhere eventually, some quite close to the parent plants (like the bird-pollinated rātā or pōhutukawa trees (Metrosideros spp.) and others great distances away (like the wind-pollinated beech trees, Nothofagus spp.). While most pollen and spores will decay and rot soon after landing on the ground, in some environments such as peat bogs and lake sediments, they will be preserved and accumulate in chronological order along with the dead remains of plants, microscopic animals and possibly mineral materials for hundreds, thousands, and in some cases millions of years. The depth of sediment (and pollen and spores) therefore relates to age. Given that pollen and spores are the key to revealing what plants lived on the landscape in the past, peat bogs and lake sediments can be looked upon as important historical archives of past environments. Pollen grains and spores are usually abundant and well preserved in peat bogs, lake and pond sediments, and it is these kinds of sediments that are most often used for reconstructing past vegetation patterns.
Pollen is transported in different ways from the parent plants growing on the landscape to the sites of preservation in peat bogs and lakes. It is important to be able to consider these sources when it comes to interpreting the fossil pollen records. Local plants growing directly on a bog surface or wetland plants growing in shallow lake waters will deliver a large proportion of their pollen to their immediate surroundings, as it falls directly from the plants. In contrast, pollen from taller vegetation surrounding the peat bogs, or plants growing in the lake catchment area, can be transported great distances away from source. Pollen escaping from tall forest canopies can be lifted by thermals to high altitudes in the troposphere, which may carry the pollen many hundreds of kilometres away. For example, mataī (Prumnopitys taxifolia) pollen is found in peat deposits on New Zealand’s subantarctic Auckland and Campbell islands – a long way from the southernmost limits of their natural distribution on the mainland. Pollen-bearing soils, sediments or riverbed sediments can get eroded by water and reworked, transporting additional older pollen and spores into a mixture of contemporary sediments. This can often confuse the interpretation of a pollen record, but fortunately the surface walls of reworked pollen and spores can often be recognised by their poorer degree of preservation.
Larger lake bodies receive pollen from a much wider source area than small peat bogs, but both can be used together to piece together the regional and local vegetation history of an area. Pollen and spores are also found in lower concentrations and generally less well preserved in soils, caves (both sediments and speleothems), glacier ice, polar ice caps and ocean floor sediments. The extent to which pollen and spores are preserved in sediment deposits relies on a number of factors, the most important of which are fine grain size and a low oxygen environment from the time of deposition to minimise microbial and aerobic attack. The degree to which different types of pollen and spores are preserved is also related to the thickness of their walls; some very fine structured pollen and spores never get preserved in the fossil record.
Coring through time
Sediment deposits have to be sampled by coring or digging pits to extract the pollen and spores for vegetation reconstruction. When exposed sections of sediment deposits are not available, then cores have to be collected from the surface of peatlands, using a coring device. The most commonly used corer for peat is the Russian D-section corer (PIC), which has a sharp penetrating point at one end, a movable sampling chamber inside a barrel, and a revolving fin along the length of barrel.
 |
 |
| Cross-section view of Russian D-section corer | Russian D-section corer with peat core in situ |
This coring head is attached to a T-bar handle and used with attachable extension rods that allow coring down to many metres depth. In New Zealand, most of the peat bogs we have cored have been between 1 and 12 metres deep, and began accumulating peat after the end of the last glaciation (c. 18 000 –10 000 years ago).The coring chamber is either 50 cm long by 5 cm diameter, or 100 cm long bby 10 cm diameter; the larger-dimension model is useful only in softer sediments. The corer is manually pushed into the peat surface down to the required depth, the handle is turned 180 dgrees, which forces the fin to slice and seal a semicylindrical sample of peat into the sampling chamber, preserving it in an undisturbed intact state. Then the whole unit is pulled up to the surface and the core of peat transferred from the sampling chamber into a piece of plastic guttering (PIC), wrapped and taken back to the laboratory for analysis and storage.
 |
 |
| Collecting short cores from sphagnum bog in Canterbury |
Collecting a peat core using D-section corer |
Pollen analysis records past vegetation
may be subsampled (approximately a teaspoon of sediment will be removed from a horizontal slice of the core for analysis) at regular intervals down the length of the core, e.g. at intervals of 5 cm, or depending on the resolution required, this interval can be reduced to 5 mm for extra detail. The peat or lake sediments contain a matrix of dead plant remains, remains of insects, mineral materials, algal and fungal remains, along with pollen and spores. However, much time and effort is needed in the laboratory to chemically process the bulk sediment samples in order to destroy all the extraneous material and allow a pollen-rich residue to be produced. This residue should be clean and full of pollen, so that it can be mounted on a slide, allowing a representative sample of pollen and spores to be accurately identified and counted under a microscope. The number chemical analyses required relate to the removal of different components in the matrix of materials that enclose the pollen and spores. First, bulk samples are boiled in potassium hydroxide to break up the matrix and dissolve humic materials. The sample is then sieved to remove larger coarse roots and plant remains. Then hydrochloric acid is used to remove any calcium carbonate, followed by hydrofluoric acid to remove any silica from silt, sand or diatoms. Finally the residue is treated with an acetolysis procedure, which uses strong acids to remove cellulose. The tiny residue remaining after all this pretreatment is then stained to highlight and contrast the fine structures of the pollen walls, making identification easier. For each depth sampled in the core, pollen grains are counted until a sum of 250–500 pollen grains from terrestrial plants is counted. Counting is the most time consuming part of pollen analysis, and requires hours of sitting patiently at a microscope.
 |
 |
| Full protective clothing is required
when using hydrofluoric acid |
Pollen slide ready for examination |
Charcoal analysis records past fires
When plants burn during a fire, charcoal is formed. Smaller particles of charcoal are picked up in the wind and can be transported over long distances – for example, the smallest particles of charcoal (<5 microns) are often blown across the Tasman Sea during Australian bush fires if the westerly winds are strong enough. These tiny particles can be detected landing on fresh surfaces on the West Coast of New Zealand. However, most charcoal produced during a fire usually falls out in the local or regional area of the fire, and can make its way into peat or lake sediment deposits either directly from air fall, or by being water transported via rivers or surface runoff into lakes or bogs. Once it lands on the surface of a bog, or is incorporated into the surface sediments of a lake, it becomes permanently buried along with the pollen and spores produced during that year. Charcoal fragments are not destroyed by the chemical analyses employed for pollen preparations, therefore if present in the sediments analysed for pollen, it will also be preserved along with the pollen in the final residue used to mount slides. Charcoal particles can then be counted at the same time that pollen is counted, and in this way, a record of past fires can be reconstructed. The one drawback of this technique, however, is that the pollen preparation process can be mechanically damaging to the charcoal fragments, and can break them into smaller pieces. Larger fragments of charcoal (>50 microns) are used to reveal the presence of a local fire, whereas smaller particles (<50 microns) are usually derived from fires in more remote locations. In the absence of any large-sized charcoal particles, there is no way of knowing if a trace of small charcoal particles in a core represents small local fires or more widespread fires possibly thousands of kilometres away. If such information is required, then the sediments need to be gently washed through different sized sieves and the relative sizes of charcoal analysed, as an alternative to counting charcoal from the pollen slides.
 |
 |
| Pollen slide showing
pollen from intact forest cover (click to enlarge) |
Pollen slide showing charcoal fragments with bracken spores and grass pollen grains following deforestation (click to enlarge) |
 |
| Soil profile showing dark charcoal horizon from the time of forest clearance |
