There is a vigorous debate about the capacity of conservation biology, as a scientific discipline, to effectively contribute to actions that preserve and restore biodiversity. Various factors may be responsible for the current great divide that exists between conservation research and action. Part of the problem may be a lack of involvement by conservation scientists in actually conducting or helping implement concrete conservation actions, yet scientists' involvement can be decisive for successful implementation, as illustrated here by the rapid recovery of an endangered hoopoe population in the Swiss Alps after researchers decided to implement the corrective measures they were proposing themselves. We argue that a conceptual paradigm shift should take place in the academic conservation discipline toward more commitment on the part of researchers to turn conservation science into conservation action. Practical implementation should be regarded as an integrated part of scientific conservation activity, as it actually constitutes the ultimate assessment of the effectiveness of the recommended conservation guidelines, and should be rewarded as such.

Correlative species distribution models are frequently used to predict species’ range shifts under climate change. However, climate variables often show high collinearity and most statistical approaches require the selection of one among strongly correlated variables. When causal relationships between species presence and climate parameters are unknown, variable selection is often arbitrary, or based on predictive performance under current conditions. While this should only marginally affect current range predictions, future distributions may vary considerably when climate parameters do not change in concert. We investigated this source of uncertainty using four highly correlated climate variables together with a constant set of landscape variables in order to predict current (2010) and future (2050) distributions of four mountain bird species in central Europe. Simulating different parameterization decisions, we generated a) four models including each of the climate variables singly, b) a model taking advantage of all variables simultaneously and c) an un‐weighted average of the predictions of a). We compared model accuracy under current conditions, predicted distributions under four scenarios of climate change, and – for one species – evaluated back‐projections using historical occurrence data. Although current and future variable‐correlations remained constant, and the models’ accuracy under contemporary conditions did not differ, future range predictions varied considerably in all climate change scenarios. Averaged models and models containing all climate variables simultaneously produced intermediate predictions; the latter, however, performed best in back‐projections. This pattern, consistent across different modelling methods, indicates a benefit from including multiple climate predictors in ambiguous situations. Variable selection proved to be an important source of uncertainty for future range predictions, difficult to control using contemporary information. Small, but diverging changes of climate variables, masked by constant overall correlation patterns, can cause substantial differences between future range predictions which need to be accounted for, particularly when outcomes are intended for conservation decisions.

No abstract

Summary 1.Human outdoor recreational activities are increasing and have a significant impact on wildlife. There are few methods suitable for investigating the response of rare and endangered species to human recreational activities, although the impact can be assessed at various scales by measuring both physiological and behavioural responses to disturbance. 2. Capercaillie Tetrao urogallus are suffering strong population declines throughout central Europe. We examined the effects of ski tourism on capercaillie habitat use and adrenocortical activity, measured non-invasively in droppings. 3. During three winters, 2003-06, we radio-tracked 13 capercaillie. In the southern Black Forest in Germany, we sampled 396 droppings of these and additional individuals before and after the start of the ski season. We tested whether the intensity of human winter recreational activities affected home range location and habitat use, and we identified those factors influencing the concentration of corticosterone metabolites (CM) in droppings. 4. Capercaillie used habitats subject to ski tourism. Although the latter did not affect home range location, capercaillie preferred undisturbed forests within their home ranges and avoided areas with high recreation intensity in the ski season. Faecal CM levels of individuals in areas with low recreation intensity were significantly lower than those in areas with moderate or high recreation intensity during the entire study period. 5. We conclude that ski tourism affects both habitat use and endocrine status in capercaillie, with potential negative consequences on body condition and overall fitness. 6. Synthesis and applications. This study demonstrates the relevance of studying wildlife responses at various temporal and spatial scales, and the value of using multiple methods applied to the same individuals to monitor the impact of human recreational activities on a free-ranging species. In order to protect capercaillie populations, we recommend that managers keep forests inhabited by capercaillie free from tourism infrastructure and retain undisturbed forest patches within skiing areas.

Balancing model complexity is a key challenge of modern computational ecology, particularly so since the spread of machine learning algorithms. Species distribution models are often implemented using a wide variety of machine learning algorithms that can be fine‐tuned to achieve the best model prediction while avoiding overfitting. We have released SDMtune , a new R package that aims to facilitate training, tuning, and evaluation of species distribution models in a unified framework. The main innovations of this package are its functions to perform data‐driven variable selection, and a novel genetic algorithm to tune model hyperparameters. Real‐time and interactive charts are displayed during the execution of several functions to help users understand the effect of removing a variable or varying model hyperparameters on model performance. SDMtune supports three different metrics to evaluate model performance: the area under the receiver operating characteristic curve, the true skill statistic, and Akaike's information criterion corrected for small sample sizes. It implements four statistical methods: artificial neural networks, boosted regression trees, maximum entropy modeling, and random forest. Moreover, it includes functions to display the outputs and create a final report. SDMtune therefore represents a new, unified and user‐friendly framework for the still‐growing field of species distribution modeling.