Quantitative evaluations to optimize complexity have become standard for avoiding overfitting of ecological niche models (ENMs) that estimate species’ potential geographic distributions. ENMeval was the first R package to make such evaluations (often termed model tuning) widely accessible for the Maxent algorithm. It also provided multiple methods for partitioning occurrence data and reported various performance metrics. Requests by users, recent developments in the field, and needs for software compatibility led to a major redesign and expansion. We additionally conducted a literature review to investigate trends in ENMeval use (2015–2019). ENMeval 2.0 has a new object‐oriented structure for adding other algorithms, enables customizing algorithmic settings and performance metrics, generates extensive metadata, implements a null‐model approach to quantify significance and effect sizes, and includes features to increase the breadth of analyses and visualizations. In our literature review, we found insufficient reporting of model performance and parameterization, heavy reliance on model selection with AICc and low utilization of spatial cross‐validation; we explain how ENMeval 2.0 can help address these issues. This redesigned and expanded version can promote progress in the field and improve the information available for decision‐making. ​

Invertebrates constitute the majority of animal species and are critical for ecosystem functioning and services. Nonetheless, global invertebrate biodiversity patterns and their congruences with vertebrates remain largely unknown. We resolve the first high-resolution (~20-km) global diversity map for a major invertebrate clade, ants, using biodiversity informatics, range modeling, and machine learning to synthesize existing knowledge and predict the distribution of undiscovered diversity. We find that ants and different vertebrate groups have distinct features in their patterns of richness and rarity, underscoring the need to consider a diversity of taxa in conservation. However, despite their phylogenetic and physiological divergence, ant distributions are not highly anomalous relative to variation among vertebrate clades. Furthermore, our models predict that rarity centers largely overlap (78%), suggesting that general forces shape endemism patterns across taxa. This raises confidence that conservation of areas important for small-ranged vertebrates will benefit invertebrates while providing a “treasure map” to guide future discovery.

Scientific research increasingly calls for open‐source software that is flexible, interactive, and expandable, while providing methodological guidance and reproducibility. Currently, many analyses in ecology are implemented with “black box” graphical user interfaces (GUIs) that lack flexibility or command‐line interfaces that are infrequently used by non‐specialists. To help remedy this situation in the context of species distribution modeling, we created Wallace, an open and modular application with a richly documented GUI with underlying R scripts that is flexible and highly interactive. Wallace guides users from acquiring and processing data to building models and examining predictions. Additionally, it is designed to grow via community contributions of new modules to expand functionality. All results are downloadable, along with code to reproduce the analysis. Wallace provides an example of an innovative platform to increase access to cutting‐edge methods and encourage plurality in science and collaboration in software development.

Species distribution models (SDMs) are widely used in ecology, biogeography and conservation biology to estimate relationships between environmental variables and species occurrence data and make predictions of how their distributions vary in space and time. During the past two decades, the field has increasingly made use of machine learning approaches for constructing and validating SDMs. Model accuracy has steadily increased as a result, but the interpretability of the fitted models, for example the relative importance of predictor variables or their causal effects on focal species, has not always kept pace. Here we draw attention to an emerging subdiscipline of artificial intelligence, explainable AI (xAI), as a toolbox for better interpreting SDMs. xAI aims at deciphering the behavior of complex statistical or machine learning models (e.g. neural networks, random forests, boosted regression trees), and can produce more transparent and understandable SDM predictions. We describe the rationale behind xAI and provide a list of tools that can be used to help ecological modelers better understand complex model behavior at different scales. As an example, we perform a reproducible SDM analysis in R on the African elephant and showcase some xAI tools such as local interpretable model‐agnostic explanation (LIME) to help interpret local‐scale behavior of the model. We conclude with what we see as the benefits and caveats of these techniques and advocate for their use to improve the interpretability of machine learning SDMs.