Absolute abundance provides a population estimate expressed as number of individuals per unit area. As previously mentioned, this method requires species-specific detection coefficients or probability functions obtained from the relationship between number of detections and distance from observer. Since differences in species-specific detectability are accounted for, absolute abundance estimates may be compared among species within the same or different habitats (see Carey et al. 1991 for examples of statistical analyses used in a comparison of cavity-nesting bird densities in stands of different age classes and moisture regimes).
The statistical analysis of spot mapping data critically depends on two factors: First, the fractional territories at the boundary of the plot must be estimated carefully, since the edge effect will be substantial on small plots. Given that many woodpeckers have territories of 10 ha or even more, very large plots (B-BBWO and B-TTWO, B-PIWO) would have to be spot mapped. Second, at least two plots will have to be spot mapped in each habitat to obtain a measure of variation for statistical tests.
Absolute abundance estimates from spot-mapping are calculated by counting the number of complete and fractional territories per grid area. To do this, superimpose all detections of each species from successive censuses on a composite map. Delineate detection clusters (centers of territorial activity) based on the locations of nests, border disputes, concurrent territorial behaviour, etc. For fractional territories (i.e., territories in which the boundaries do not lie entirely within the grid system), estimate the proportion of the total territory which is within the grid boundaries. The estimated population size (P) per grid area is calculated by P= (C + FC) x 2, where C = the number of complete territorial clusters and FC = the sum of the proportions of each fractional territory located within the grid boundaries; the factor 2 represents a breeding pair. For example, if there are five complete territories and two fractional territories detected, with 0.5 and 0.75 of the territory area located within the grid system, respectively, then add these to generate 1.25 fractional clusters (FC) and five complete territories (C), for a total of 12.5 individuals per grid area. This calculation assumes that a population consists exclusively of mated pairs. If, from the composite map, it is obvious that there were one or more unmated territorial individuals repeatedly detected during censuses, modify the calculation accordingly.
There are possibilities of using individual territories as the sampling unit for absolute abundance measurements. If territories cover most of the landscape, the reciprocal of territory size may be an estimate of population density. The advantage of this approach is that sample sizes could be increased for a given amount of effort so that higher precision could be achieved. The disadvantage is the difficult assumption that woodpecker territories are spread evenly over the landscape, and this may require larger landscape evaluations to assess the validity of this assumption. In addition, this method might be sensitive to how well a territorial area was delineated.
If encounter transect distance methods are used, then program DISTANCE can be used for the analysis of absolute abundance. Encounter transects used with distance methods might provide better statistical data then spot mapping and this would have to be checked empirically. Refer to the RIC Species Inventory Fundamentals manual, Section 5.4.1 - Distance Methods, for further explanation.