2018). These five variables were not strongly intercorrelated (r ≤ 0.72; Table S1). We combined these two approaches to demonstrate the fine‐scale thermal mechanisms driving broad‐scale population responses to climate in the Ethiopian Bush‐crow. Observations on the biology of the Ethiopian Bush Crow, Interspecific competition and the geographical distribution of Red and Arctic Foxes, Very high resolution interpolated climate surfaces for global land areas, Handbook of the Birds of the World, Volume 14, Potential impacts of climatic change on southern African birds of fynbos and grassland biodiversity hotspots, A Climatic Atlas of European Breeding Birds, Climate Change 2014: Impacts, Adaptation and Vulnerability, Examining the behavioural costs of heat in a climatically range‐restricted arid zone bird; the case of the Ethiopian Bush‐crow Zavattariornis stresemanni (Masters thesis), Varied diet and opportunistic foraging in the Ethiopian Bush‐crow, Mechanistic niche modelling: combining physiological and spatial data to predict species’ ranges. 2011). 2008 revisions) (v.1)
Sibley and Monroe (1993, including corrections up to 1998) (v.1)
2012). Please check your email for instructions on resetting your password. 2012, Cunningham et al. 2000). Spreo Starlings have bred regularly at Paradise Park, collecting grass to build their nest within a box. In the second study, foraging adults were approached to a distance of 5–10 m (at which none of the species showed signs of disturbance such as alarm calling or retreating from the observer) and one bird was selected at random and watched continuously through binoculars for as long as contact could be maintained (‘watches’). Clements 6th edition (version 6.9 incl. We collated geo‐referenced sightings of Bush‐crows made by various observers between 2005 and 2011 (Gedeon 2006, Mellanby et al. Group identity was modelled as a random effect, with a random effect of observation period nested within it, as the data were over‐dispersed. Model variants were fitted with all possible combinations of each of the parameters being the same or different among the three species and with either temperature or time of day as the independent variable (Table S2). For the Bush‐crow, co‐ordinates of all presence records used to fit the species distribution models are shown. For the Superb Starling SDM, multiple adaptive regression splines (MARS) were the best performing algorithm (AUC = 0.971) and precipitation of the wettest quarter again had the highest ΔAUC score (0.142). IOC World Bird Names, version 2.1 (v.1)
We found that the effects of high temperatures on Ethiopian Bush‐crow behaviour were considerably more pronounced than those for two sympatric starling species in the same habitat. IOC World Bird Names, version 1.6 (v.1)
IOC World Bird Names, version 6.4 (v.1)
Morony, Bock and Farrand (v.1)
For the effect of time of day, we wished to allow for the probable effect to be hump‐shaped, as suggested by inspection of scatter diagrams. 2009). Visual inspection of scatter plots indicated that it was not straightforward to analyse the data using ordinary least squares regression with the proportion of the watch spent panting (Ppant) as the dependent variable and temperature in full sunlight or time of day as the independent variable. 2015) or influences of temperature on habitat, foraging ability or prey abundance (du Plessis et al. For the two starling Focal observations that occurred under cloud cover were excluded. Learn more. These problems resemble those addressed by the statistical model of avian primary feather moult (Underhill & Zucchini 1988).