Info

100 150 200 250

Species richness

Figure 12.2. Posterior distribution of species richness N for the avian community along BBS route 017 in New Hampshire, USA.

Figure 12.2. Posterior distribution of species richness N for the avian community along BBS route 017 in New Hampshire, USA.

The posterior summaries of this model's parameters are given in Table 12.1. Notice that the average (among species) probabilities of occurrence and detection

Figure 12.3. Posterior means of occurrence and detection probabilities for the 99 species detected in the BBS survey along route 017 in New Hampshire, USA. 'Tails' on each point indicate how each estimate changes if species richness N is not estimated (i.e., if N = n is assumed).

(on the logit scale) are smaller than those estimated by assuming that N equaled the number of species observed in the sample. This seems like a sensible result because species that were missing from the sample are likely to be those that were present in low numbers or were difficult to detect. Including these lower values of — and p would reduce the average values of the community. These lower values of — and p are also likely to be responsible for the increase in estimates of heterogeneity and correlation parameters (<r„, av, p). In other words, the species that were missed had lower values of both - and p.

The probabilities of occurrence and detection estimated for the n = 99 observed species probabilities were not dramatically different from estimates computed under the simpler model, wherein N = n was assumed (Figure 12.3). Estimates of species occurrence appear to be positively correlated with those of species detection, although, interestingly, there also appear to be a few uncommon species (with low - values) that have relatively high probabilities of detection.

Was this article helpful?

0 0

Post a comment