Bird-habitat models are frequently used us predictive modeling tools--for example, to predict how a species will respond to habitat modifications. We investigated the generality of thepredictionsfrom this type of model. Multivariate models were developed for Golden Eagle (Aquila chrysaetos), Raven (Corvus corax), and Buzzard (Buteo buteo) living in northwest Scotland. Data were obtained for all habitat and nest locations within an area of 2349 km 2. This assemblage of species is relatively static with respect to both occupancy and spatial posttioning. The area was split into five geographic subregions: two on the mainland and three on the adjacent Island of Mull, which has one of United Kingdom's richest raptor fauna assemblages. Because data were collected for all nest locations and habitats, it was possible to build models that did not incorporate sampling error. A range of predicttve models was developed using discriminant analysis and logistic regression. The models differed with respect to the geographical origin of the data used for model development. The predictive success of these models was then assessed by applying them to validation data. The models showed a wide range of predictive success, ranging from only 6% of nest sites correctly predicted to 100% correctly predicted. Model validation techniques were used to ensure that the models'predictions were not statistical artefacts. The variability in prediction success seemed to result from methodological and ecological processes, including the data recording scheme and interregional differences in nesting habitat. The resuits from this study suggest that conservation biologists must be very careful about making predictions from such studies because we may be working with systems that are inherently unpredictable.Probando la generalidad de los modelos de hfibitat para aves Resumen: Los modelos de h~bitatpara aves han sldo usados frecuentemente como herramientas predictivas de modelaje, por ejemplo, para predecir como una especie va a responder a modtficaciones en el hd~bitat. En el presente estudio investigamos la generalidad de las predicciones hechas por este ttpo de modelos. Modelos multivariados fueron desarrollados para las ~guilas doradas (Aquila chrysaetos), los cuervos (Corvus corax) y los buitres (Buteo buteo) que habitan el noroeste de Escocia. Se obtuvieron datos para todos los h~bitat y sittos con nidos dentro de un drea de 2349 km 2. Este conjunto de especies es relativamente estd~ttco con respectio a su posesi6n y postci6n espacial. E1 d~rea fue dividida en cinco subregtones geogrdficas; dos en tierra flrme y tres en las islas adyacentes de Mull que poseen una de las asociactones de fauna de ayes de rapifla mds ricas del Reino Unido. Debldo a que se recolectaron datos de todos los sitios con nidos y h~bitats, fue posible construir modelos que no incorporaron errores de muestreo. Se desarroll6 una serie de modelos predtctivos usando an~llsis discrlminante y regresiones logisticas. Los modelos difirieron en 1o que respecta al origen geo-gr~fi...

No abstract

No abstract

Amongst raptor species, individuals with specialized diets are commonly observed to have higher reproductive output than those with general diets. A suggested cause is that foraging efficiency benefits accrue to diet specialists. This diet specificity hypothesis thus predicts that diet breadth and reproductive success should be inversely related within species. We highlight, however, that a prey availability hypothesis also makes the same prediction in some circumstances. Hence, when high diet specificity results from high encounter rates with an abundant, preferred prey, then prey availability may affect reproductive success, with diet specialization as an incidental correlate. Using three insular study areas in western Scotland, we examine diet specificity and reproductive success in Golden Eagles Aquila chrysaetos. Diet breadth and breeding productivity were not negatively related in any of our study areas, even though birds with specific diets did tend to have a higher incidence of preferred prey (grouse and lagomorphs) in the diet. Indeed, in two study areas there was evidence that diet generalists had higher breeding productivity. Our results therefore failed to support the diet specificity hypothesis but were consistent with the prey availability hypothesis. We highlight that although many other studies are superficially consistent with the diet specificity hypothesis, our study is not alone in failing to provide support and that the hypothesis does not provide a generic explanation for all relevant results. Diet specificity in predators can be at least partially a response to prey diversity, availability and distribution, and benefits associated with different prey types, so that being a generalist is not necessarily intrinsically disadvantageous. We suggest that the available evidence is more consistent with variation in prey abundance and availability as a more influential factor explaining spatial and temporal variation in breeding productivity of ‘generalist’ species such as the Golden Eagle. Under this argument, prey abundance and availability are the main drivers of variation in reproductive output. Diet specificity is a consequence of variation in prey availability, rather than a substantial cause of variation in reproductive success.

No abstract