Predictive habitat models are an important tool for ecological research and conservation. A major cause of unreliable models is excessive model complexity, and regularization methods aim to improve the predictive performance by adequately constraining model complexity. We compare three regularization methods for logistic regression: variable selection, lasso, and ridge. They differ in the way model complexity is measured: variable selection uses the number of estimated parameters, the lasso uses the sum of the absolute values of the parameter estimates, and the ridge uses the sum of the squared values of the parameter estimates. We performed a simulation study with environmental data of a real landscape and artificial species occupancy data. We investigated the effect of three factors on relative model performance: (1) the number of parameters (16, 10, 6, 2) in the 'true' model that determined the distribution of the artificial species, (2) the prevalence, i.e. the proportion of sites occupied by the species, and (3) the sample size (measured in events per variable, EPV). Regularization improved model discrimination and calibration. However, no regularization method performed best under all circumstances: the ridge generally performed best in the 16-parameter scenario. The lasso generally performed best in the 10-parameter scenario. Variable selection with AIC was best at large sample sizes (EPV >= 10) when less than half of the variables influenced the species distribution. However, at low sample sizes (EPV < 10), ridge and lasso always performed best, regardless of the parameter scenario or prevalence. Overall, calibration was best in ridge models. Other methods showed overconfidence, particularly at low sample sizes. The percentage of correctly identified models was low for both lasso and variable selection. Variable selection should be used with caution. Although it can produce the best performing models under certain conditions, these situations are difficult to infer from the data. Ridge and lasso are risk-averse model strategies that can be expected to perform well under a wide range of underlying species-habitat relationships, particularly at small sample sizes.

Landslides are a hazard for humans and artificial structures. From an ecological point of view, they represent an important ecosystem disturbance, especially in tropical montane forests. Here, shallow translational landslides are a frequent natural phenomenon and one local determinant of high levels of biodiversity. In this paper, we apply weighted ensembles of advanced phenomenological models from statistics and machine learning to analyze the driving factors of natural landslides in a tropical montane forest in South Ecuador. We exclusively interpret terrain attributes, derived from a digital elevation model, as proxies to several driving factors of landslides and use them as predictors in our models which are trained on a set of five historical landslide inventories. We check the model generality by transferring them in time and use three common performance criteria (i.e. AUC, explained deviance and slope of model calibration curve) to, on the one hand, compare several state-of-the-art model approaches and on the other hand, to create weighted model ensembles. Our results suggest that it is important to consider more than one single performance criterion. Approaching our main question, we compare responses of weighted model ensembles that were trained on distinct functional units of landslides (i.e. initiation, transport and deposition zones). This way, we are able to show that it is quite possible to deduce driving factors of landslides, if the consistency between the training data and the processes is maintained. Opening the 'black box' of statistical models by interpreting univariate model response curves and relative importance of single predictors regarding their plausibility, we provide a means to verify this consistency. With the exception of classification tree analysis, all techniques performed comparably well in our case study while being outperformed by weighted model ensembles. Univariate response curves of models trained on distinct functional units of landslides exposed different shapes following our expectations. Our results indicate the occurrence of landslides to be mainly controlled by factors related to the general position along a slope (i.e. ridge, open slope or valley) while landslide initiation seems to be favored by small scale convexities on otherwise plain open slopes.

Empirical species distribution models (SDMs) constitute often the tool of choice for the assessment of rapid climate change effects on species vulnerability. Conclusions regarding extinction risks might be misleading, however, because SDMs do not explicitly incorporate dispersal or other demographic processes. Here, we supplement SDMs with a dynamic population model 1) to predict climate-induced range dynamics for black grouse in Switzerland, 2) to compare direct and indirect measures of extinction risks, and 3) to quantify uncertainty in predictions as well as the sources of that uncertainty. To this end, we linked models of habitat suitability to a spatially explicit, individual-based model. In an extensive sensitivity analysis, we quantified uncertainty in various model outputs introduced by different SDM algorithms, by different climate scenarios and by demographic model parameters. Potentially suitable habitats were predicted to shift uphill and eastwards. By the end of the 21st century, abrupt habitat losses were predicted in the western Prealps for some climate scenarios. In contrast, population size and occupied area were primarily controlled by currently negative population growth and gradually declined from the beginning of the century across all climate scenarios and SDM algorithms. However, predictions of population dynamic features were highly variable across simulations. Results indicate that inferring extinction probabilities simply from the quantity of suitable habitat may underestimate extinction risks because this may ignore important interactions between life history traits and available habitat. Also, in dynamic range predictions uncertainty in SDM algorithms and climate scenarios can become secondary to uncertainty in dynamic model components. Our study emphasises the need for principal evaluation tools like sensitivity analysis in order to assess uncertainty and robustness in dynamic range predictions. A more direct benefit of such robustness analysis is an improved mechanistic understanding of dynamic species responses to climate change.

Data limitations can lead to unrealistic fits of predictive species distribution models (SDMs) and spurious extrapolation to novel environments. Here, we want to draw attention to novel combinations of environmental predictors that are within the sampled range of individual predictors but are nevertheless outside the sample space. These tend to be overlooked when visualizing model behaviour. They may be a cause of differing model transferability and environmental change predictions between methods, a problem described in some studies but generally not well understood. We here use a simple simulated data example to illustrate the problem and provide new and complementary visualization techniques to explore model behaviour and predictions to novel environments. We then apply these in a more complex real-world example. Our results underscore the necessity of scrutinizing model fits, ecological theory and environmental novelty.

Species distribution models are useful for identifying driving environmental factors that determine earthworm distributions as well as for predicting earthworm distribution patterns and abundances at different scales. However, due to large efforts in data acquisition, studies on larger scales are rare and often focus on single species or earthworms in general. In this study, we use boosted regression tree models (BRTs) for predicting the distribution of the three functional earthworm types, i.e. anecics, endogeics and epigeics, in an agricultural area in Baden-Wurttemberg (Southwest Germany). First, we predicted presence and absence and later earthworm abundances, considering predictors depicting land management, topography, and soil conditions as well as biotic interaction by using the abundance of the other functional earthworm types. The final presence-absence models performed reasonably well, with explained deviances between 24 and 51% after crossvalidation. Models for abundances of anecics and endogeics were less successful, since the high small-scale variability and patchiness in earthworm abundance influenced the representativeness of the field measurements. This resulted in a significant model uncertainty, which is practically very difficult to overcome with earthworm sampling campaigns at the catchment scale. Results showed that management practices (i.e. disturbances), topography, soil conditions, and biotic interactions with other earthworm groups are the most relevant predictors for spatial distribution (incidence) patterns of all three functional groups. The response curves and contributions of predictors differ for the three functional earthworm types. Epigeics are also controlled by topographic features, endogeics by soil parameters.