Search

Species distribution modeling (SDM) is a fundamental tool in theoretical and applied ecology. However, relatively little is known about the performance of different approaches for scale optimization, model selection, and algorithmic prediction in the context of nonlinear, multiscale and interactive relationships between environmental variables and species occurrence. Modelers often struggle to optimize a tradeoff between ecological relevance, model robustness, complexity, and overfitting. In this paper, we investigated several methods designed to optimize spatial scale and variable selection in SDMs, in each case evaluating model fitness, parsimony and predictive performance. We used a simulation approach to produce a large pool of alternative underlying habitat relationships that reflect a broad range of realistic habitat associations. We also compared several different modeling algorithms, including logistic regression with a generalized linear model (GLM), Lasso and Elastic-Net Regularized GLMs (GLMNet), and random forest (RF), as well as alternative variable and scale selection methods. We found that GLM methods employing all-subsets dredge routines for variable selection were consistently the best predictors based on all criteria of our model performance assessment and across all attributes of the simulated underlying relationship, including nonlinearity and interaction. We had expected machine learning approaches, such as random forest, to perform better in these more complex forms of species-environment relationships. GLM using dredge variable selection was also the method that included the fewest spurious covariates and included the most correct predictors as a proportion of all predictors. We found that univariate scaling was the most robust method of variable and scale selection, along with Minimal Redundancy Maximal Relevancy (MRMR) which performed equivalently. The simulation experiment presented here provides a robust assessment of simulated multi-species distribution model performance, complexity and fidelity. By simulating a large range of potential habitat relationships with varying spatial scale, effect sizes, linearity, and interactions, we comprehensively evaluated model performance across gradients of complexity of the underlying relationships and violations of classical statistical assumptions. This study provides a valuable assessment and a broader example of the power and utility of controlled simulation experiments in habitat relationships and other ecological spatial predictive modeling.

Abstract Habitat selection is strongly scale‐dependent, and inferring the characteristic scale at which an organism responds to environmental variation is necessary to obtain reliable predictions. The occupancy framework is frequently used to model species distribution with the advantage of accounting for imperfect observation, but occupancy studies typically do not define the characteristic scale of the modelled variables. We used camera trap data from winter wildlife surveys in the Swiss part of the Jura Mountains to model occupancy of wild boar (Sus scrofa) and roe deer (Capreolus capreolus). We used a three‐step approach: (1) first, we identified factors influencing detectability; (2) second, we optimized the characteristic scale of each candidate explanatory variable; and (3) third, we fit multivariable, multiscale occupancy models in relation to land cover, human presence and topography. Wild boar occupancy was mainly influenced by the interaction between elevation within 2500 m and the proportion of forested areas within a 2500 m, with a nonsignificant additional effect of the interaction between ruggedness within 1900 m and the proportion of forested areas within 2500 m as well as the distance to urban areas. Roe deer occupancy was mainly associated with the interaction between ruggedness within 900 m and the proportion of open landscape within 900 m, with an additional nonsignificant effect of the interaction between elevation within 1500 m and the proportion of open landscape within 900 m as well as the distance to urban areas. Incorporating scale optimization in occupancy modelling of camera trap data can greatly improve the understanding of species‐environment relationships by combining the possibility of occupancy models to correct for detection bias and simultaneously allowing to infer the characteristic scale at which certain factors influence the distribution of the organisms studied.