Do McKinnon lists provide reliable data in bird species frequency? A comparison with transect-based data

Although occurrence-based listing methods could provide reliable lists of species composition for a site, the effective reliability of this method to provide more detailed information about species frequency (and abundance) has been rarely tested. In this paper, we compared the species frequencies obtained for the same set of species-rich sites (wetlands of central Italy) from two different methods: McKinnon lists and line transects. In all sites we observed: (i) rapid cumulating curves of line transect abundance frequencies toward the asymptote represented by the maximum value in McKinnon occurrence frequency; (ii) a large amount of species having a low frequency with line transect method showing a high range of variation in frequency obtained by McKinnon lists; (iii) a set of species having a subdominant (>0.02-<0.05) and dominant species (>0.05) frequency with line transect showed all the highest value in McKinnon frequency. McKinnon lists provides only a coarse-grained proxy of species frequency of individuals distinguishing only between common species (having the highest values of McKinnon frequency) and rare species (all the other species). Although McKinnon lists have some points of strength, this method does not discriminate the frequencies inside the subset of common species (sub-dominant and dominant species). Therefore, we suggest a cautionary approach when McKinnon frequencies should be used to obtain complex univariate metrics of diversity.