CAN BIRD ATLAS DATA BE USED TO ESTIMATE POPULATION-SIZE - A CASE-STUDY USING NAMIBIAN ENDEMICS

Citation
A. Robertson et al., CAN BIRD ATLAS DATA BE USED TO ESTIMATE POPULATION-SIZE - A CASE-STUDY USING NAMIBIAN ENDEMICS, Biological Conservation, 71(1), 1995, pp. 87-95
Citations number
25
Categorie Soggetti
Ecology
Journal title
ISSN journal
00063207
Volume
71
Issue
1
Year of publication
1995
Pages
87 - 95
Database
ISI
SICI code
0006-3207(1995)71:1<87:CBADBU>2.0.ZU;2-3
Abstract
An assessment of population size is a fundamental stage in the conserv ation of any species. The recent availability of data from the Souther n African Bird Atlas Project (SABAP) provides a tool to investigate th e status and distribution of approximately 900 bird species covered by this work. It is often assumed, but rarely shown, that bird atlas rep orting rates provide a measure of relative abundance. Here, we go one step further to assess whether or not reporting rate can be used to es tablish absolute abundance and thereby produce estimates of population size. To do so, one must establish consistent relationships between r eporting rates and bird densities from samples throughout the species' range, and then extrapolate those densities to all areas where the sp ecies occurs. Field transects were undertaken in the austral winter of 1993, across the distributions of most near-endemic birds in north-we stern Namibia. Density values obtained from the main riverine and upla nd zones inhabited by these species were used in combination with a Ge ographic Information System (GIS) and overlays of three spatial layers (vegetation, altitude and rainfall) to produce bird estimates for eac h sampled square. The relationship between reporting rate and abundanc e was investigated using log-linear regression for four species differ ing in size and habitat specialisation. We found significant relations hips (p < 0.005) for three of the four species and results marginally significant (p = 0.051) for the fourth, most cryptic, species. We conc lude that reporting rates can be used to give reasonable estimates of abundance for some species, but careful consideration must be given to factors such as field sampling effort, detail of GIS overlays and par ticularly the quality of the original atlas data.