New data on the population, distribution and habitat preferences of the Canary Islands stonechat Saxicola dacotiae

Authors: Javier SEOANE, Alexandra KOURI, Juan Carlos ILLERA, David PALOMINO, César L. ALONSO and Luis M. CARRASCAL


Published: Volume 57(2), December 2010. Pages 387-405.

Language: English

Keywords: boosted regression trees, distance sampling, Fuerteventura, habitat preferences, population size and Saxicola dacotiae


This paper updates estimates of population size, distribution and habitat preferences of the endemic Canary Islands stonechat Saxicola dacotiae on the basis of data gathered across their whole distribution range, the island of Fuerteventura. We surveyed 1,462 0.5-km line transects during the reproductive seasons in 2005 and 2006, distributed across the whole island. Results were used to estimate population size using two methods: stratified estimates of mean densities and sum of estimated abundances across strata, and sum of estimations of abundance in 1 km x 1 km UTM squares based on statistical models built by boosted regression trees (BRT). In both methods we accounted for the effects of bird detectability in transects. Overall, 490 mature individuals were recorded. The Canary Islands stonechat preferred high, steep terrain (particularly above 20% slope and 200 m a.s.l.) and selected negatively the lower and flatter areas comprising most of the island. These habitats were occupied, however, albeit at low density. The highest average densities sampled per habitat (up to 43 birds/km2) were registered on steep areas (> 11%) with scrub, although the statistical models predicted densities of 66 birds/km2 in the optimum sites (slopes higher than 22.5% with rocky ground). The population size of Canary Islands stonechat estimated with the stratified design was 20,504 individuals (CI 95%: 16,217-25,973), and the model based estimate (which we consider more reliable) was 14,436 (CI 95%: 13,376-15,492). These estimates are much higher than previous ones. We argue it is difficult to compare with Bibby and Hill’s (1987) results due to the different methodologies involved and areas covered by the monitoring programs, although we can not exclude an increase in population size during the last three decades. Discrepancies with García del Rey’s (2009) estimate appears to be due to an underestimate of the population size because of the lack of sampling in extensive areas and habitats of low bird density on the island.

Full Article:

Full Article

Enter your email and password to access the contents of the subscribers of the magazine. If you are not subscribed click here