Vegetation Mapping

It is important that the vegetation surrounding each trap is recorded in a manner which allows distance weighted abundance to be calculated for each plant taxon. It is recommended that this is done in two stages, one, in the field and the second using digitised maps, air photos or forest inventories.
Pollen deposited in the traps has its origin in sources which may be at a whole range of distances. However, modelling has demonstrated that the pollen source area can be divided into two, the relevant source area (within which different mosaics of vegetation communities are separately reflected in the pollen assemblage) and the area beyond this, from which the pollen signal is homogenous (whatever the mosaic of vegetation communities). This latter can be classed as background pollen (Sugita 1994, Davis 2000). For this reason it is necessary to map vegetation out to several hundred metres (even kilometres) from the trap but the degree of detail of the mapping needs to be highest closest to the trap.
We suggest three scale categories:
  1. Within 10.5 m of the trap. Within this area it will be primarily the herbaceous vegetation which is recorded and the plants will be identified to species. For this mapping we recommend the walking in circles method described below.
  2. For the area between 10.5 and 500 m from the trap the main focus will be on the abundance and distribution of the trees, while the herbaceous and shrub vegetation will be mapped as vegetation units for which, as far as possible, the average species composition and abundance is known. Depending upon availability of data and the nature of the trapping locality, these vegetation data may be obtained from existing forest inventories or remote sensing of air photos. In some instances it may be most practical to map this intermediate area using the Bitterlich method described below.
  3. For the area from 500 m out to 1 - 2km it is significant to know the patchiness of the vegetation, the relative distribution of forested and unforested land and, where possible, the species composition of distinguishable vegetation units. Such information may be obtained from forest inventories, remote sensing of air photos or basic topographical maps, depending upon availability.

Recording vegetation in the field by walking in circles

The mapping method described here has been developed primarily by the POLLANDCAL (Pollen Land-use Calibration) group and we are grateful to them for inspiring discussions about pollen dispersal and for permission to adapt their mapping strategy for PMP.

Figure 1

Figure 1: Plan for vegetation analyses to give distance weighted plant abundance.

A series of concentric rings are considered around each pollen trap. The innermost of these is a circle, centred on the trap with a radius of 50 cm. Beyond this circle, rings of 1 m diameter are considered out to a distance of 6.5 m from the trap. Beyond this the width of the ring is increased to 2 m, out to a distance of 10.5 m. This gives an inner circle and 8 surrounding concentric rings (Figure 1). These are located in the field using a series of wooden stakes and a rope, marked off at 0.5m and then in 1 m intervals. The rope is anchored at the trap and the person making the analysis literally walks in circles with it around the trap, recording the percentage cover of each species in each individual concentric ring, on a pre-prepared form (Figure 2).

Figure 2: Vegetation record form.

Figure 2
The innermost circle may be used to facilitate estimation of plant coverage during a vegetation survey using the walking in circles approach (Figure 3). Make a ring or hoop in a light material e.g. plastic or metal, with a radius of 0.5 m. Use coloured tape to mark percentage sectors around the circle. When walking in circles wind the walking rope around the centre pole (cp). Let one end of the rope remain fixed, aligned to the north, out until 10 m and the other end moveable around the circle. The two rope ends allow you to divide your circle into cake slices of various percentage size. This enables you to train your perception for percentage coverage in a circle before starting the actual survey. This is particularly important each time you move out to a circle with larger radius.

Figure 3

Figure 3: Vegetation analysis of the innermost circle when using the walking in circles approach.

Bearing in mind the pollen taxonomic resolution, grasses and sedges may be recorded as Gramineae and Cyperaceae respectively and not identified to species. Note that with overlapping vegetation the total coverage in one ring may exceed 100%. Mosses/rocks/bare ground should also be given a percentage cover value. The position (compass orientation and distance) of individual trees occurring within the concentric rings is also marked on a paper copy cf. Figure 1. This field method has been developed by Anna Broström in consultation with Shinya Sugita. See also Broström 2002.

The Bitterlich method of estimating tree abundance

Figure 4

Figure 4: Bitterlich stick for estimating tree abundance. The stick is held to the eye and pointed horizontally with the crosspiece end to each tree surrounding the sample point. If the tree appears wider than the crosspiece it is counted otherwise it is excluded.

The Bitterlich method provides an easy, fast, simple and inexpensive way of estimating tree abundance as basal area per hectare. With the use of an angle-gauge (Bitterlich stick see Figure 4) all trees that are larger in diameter than a specified angle are counted in a circle from a central sampling point. The angle is set by the configuration of the Bitterlich stick. For simplcity using a stick of 1 m length with a 2 cm wide crosspiece made of cardboard, plastic or metal is suggested. The crosspiece is mounted to one end of the stick while a notch or peephole can be fixed to the other side. When using a round stick the notch may not be necessary. The trees should be measured at breast height and those that are seen as wide as the crosspiece may be counted as half.

If a Bitterlich stick with the above suggested configuration is used the number of trees counted from the sample point is a direct estimate of abundance as basal area per hectare (m2 * ha-1). Because this way of estimating tree abundance is relatively quick it is possible to survey larger areas around the pollen trap. In order to later distance weight the abundance of different trees it is important to record the positions at which the abundance was estimated. Two ways are suggested here, but different approaches may be followed.

Figure 5

Figure 5: Suggested format for a table containing the results of estimating tree abundance using a hand held GPS. UTM coordinates should be given to 1 metre; the UTM zone needs to be indicated.

  1. A hand held GPS can be used to record the positions at which the count was conducted. This way the surveyor can move around freely and a large number of points can be used. Sampling points should be more frequent closer to the pollen trap and a very homogenous forest may be covered with a lower amount of points than a patchy forest. The coordinates can later be saved in a table as UTM coordinates and combined with the abundance of the different trees in m2 * ha-1 (Figure 5).
  2. If no hand-held GPS is available the position of sampling points can be recorded with a compass and some means of estimating distance (e.g. measuring tape, counting steps, topographical map). Along the compass directions sampling points can be set up at increasing intervals (Figure 6). At or near these points the tree abundance can be estimated a few times and averaged. The results should be presented in a table with the direction and distance from the trap as coordinates (Figure 7).
    Figure 6

    Figure 6: Suggested sampling design for estimating tree abundance around a pollen trap in a forest opening of about 100 metre in diameter.

In both cases the exact sample design should be adopted to the specific situation. If the pollen trap site is in a forest opening the first ring of sampling points should be arranged near the forest edge. If the pollen trap is positioned more or less inside the forest, the first estimate should be made at the position of the pollen trap. The size of the opening should be estimated and recorded separately.

Figure 7

Figure 7: Suggested format for a table containing the results of estimating tree abundance when the sapling points are described by direction and distance measurements.

The description of the Bitterlich method of estimating tree abundance is largely based on the relevant chapters in the text book Aims and Methods of Vegetation Ecology by Mueller-Dombois, and Ellenberg (1974). Jackson and Kearsley (1998) make use of the method and present another sampling setup that enables the presentation of the sampling points in a table format.

Vegetation mapping by remote sensing of digitised air photographs (ground resolution of 1m)

Precise details of this will be provided at a later date.


Broström, A. 2002. Estimating source area of pollen and pollen productivity in the cultural landscapes of southern Sweden - developing a palynological tool for quantifying past plant cover. Doctoral Thesis, Lund University, Lund.
Davis, M.B. 2000. Palynology after Y2K - understanding the source area of pollen in sediments. Annu. Rev. Earth and Planet Sciences 28:1-18.
Sugita, S. 1994. Pollen representation of vegetation in Quaternary sediments: Theory and method in patchy vegetation. Journal of Ecology 82:881-897.
Mueller-Dombois, D. and Ellenberg, H. 1974. Aims and Methods of Vegetation Ecology. J. Wiley and Sons.
Jackson, S.T. and Kearsley J.B. 1998. Quantitative representation of local forest composition in forest-floor pollen assemblages. Journal of Ecology 86(3):474-490.

11 November 2003