Search for: All records

Abstract Ecological niche models (ENMs) are commonly used to predict species' potential distributions across space and time, allowing researchers and practitioners to assess invasion risk, estimate the impacts of future climate change, and reconstruct past distributions. These predictions, however, often require extrapolation into environmental conditions that fall outside the range of values used for model training (i.e. ‘non‐analog environments'). Two primary strategies exist for extrapolating into new conditions for some modeling algorithms: unconstrained extrapolation (allowing the modeled response curve to maintain its trajectory) and clamping (holding the response constant at the values observed at the limits of the training data). As these can lead to considerable differences in model predictions, it is important to inspect the behavior of modeled responses under different extrapolation strategies, which is rarely done in ENM studies. Such examination can help promote biologically realistic model transfers, for example by avoiding unconstrained extrapolation when the relevant tail of the modeled response has a sharply increasing trend. Here, we present a proposed workflow and new visualization tools to guide researchers in choosing appropriate extrapolation strategies, including plots for marginal response curves and non‐analog pixel densities. As an illustration, we used this workflow and tools to compare extrapolation strategies for two hindcast scenarios at the Last Glacial Maximum (~ 21 kya) for the black‐eared mousePeromyscus melanotisin Mexico, which currently occurs in montane forests. Making extrapolation choices individually for each tail per variable (‘custom clamping') helped avoid unrealistically high suitability predictions for non‐analog environments. This study emphasizes the importance of both inspecting response curves and assessing the total area with non‐analog conditions. By making these choices by variable and tail – a first for ENMs – this study and associated code empower researchers to understand how individual extrapolation strategies affect results and to make decisions that should lead to more realistic predictions. 

ABSTRACT Species distribution modeling can be used to predict environmental suitability, and removing areas currently lacking appropriate vegetation can refine range estimates for conservation assessments. However, the uncertainty around geographic coordinates can exceed the fine resolution of remotely sensed habitat data. Here, we present a novel methodological approach to reflect this reality by processing habitat data to maintain its fine resolution, but with new values characterizing a larger surrounding area (the “neighborhood”). We implement its use for a forest‐dwelling species (Handleyomys chapmani) considered threatened by the IUCN. We determined deforestation tolerance threshold values by matching occurrence records with forest cover data using two methods: (1) extracting the exact pixel value where a record fell; and (2) using the neighborhood value (more likely to characterize conditions within the radius of actual sampling). We removed regions below these thresholds from the climatic suitability prediction, identifying areas of inferred habitat loss. We calculated Extent of Occurrence (EOO) and Area of Occupancy (AOO), two metrics used by the IUCN for threat level categorization. The values estimated here suggest removing the species from threatened categories. However, the results highlight spatial patterns of loss throughout the range not reflected in these metrics, illustrating drawbacks of EOO and showing how localized losses largely disappeared when resampling to the 2 × 2 km grid required for AOO. The neighborhood approach can be applied to various data sources (NDVI, soils, marine, etc.) to calculate trends over time and should prove useful to many terrestrial and aquatic species. It is particularly useful for species having high coordinate uncertainty in regions of low spatial autocorrelation (where small georeferencing errors can lead to great differences in habitat, misguiding conservation assessments used in policy decisions). More generally, this study illustrates and enhances the practicality of using habitat‐refined distribution maps for biogeography and conservation. 

Species distribution models, also known as ecological niche models or habitat suitability models, have become commonplace for addressing fundamental and applied biodiversity questions. Although the field has progressed rapidly regarding theory and implementation, key assumptions are still frequently violated and recommendations inadvertently overlooked. This leads to poor models being published and used in real‐world applications. In a structured, didactic treatment, we summarize what in our view constitute the ten most problematic issues, or hazards, negatively affecting implementation of correlative approaches to species distribution modeling (specifically those that model suitability by comparing the environments of a species' occurrence records with those of a background or pseudoabsence sample). For each hazard, we state relevant assumptions, detail problems that arise when violating them, and convey straightforward existing recommendations. We also discuss five major outstanding questions of active current research. We hope this contribution will promote more rigorous implementation of these valuable models and stimulate further advancements. 

Copious questions in global change biology require estimates of climatic suitability for species in the past or future, often via transfers of ecological niche models (ENMs) using outputs from global circulation models (GCMs). However, available GCMs differ markedly, affecting hindcasts and forecasts of species potential distributions. We propose using demographic inferences based on genetic data (indicative of either population‐level continuous occupation or postglacial colonization) to test which GCM leads to a better match with reality for ENM hindcasting. We implement an intuitive worked example for four isolated focal populations of a montane shrewCryptotis mexicanusin central‐eastern Mexico, by comparing suitability maps at the Last Glacial Maximum (LGM) and today. We built an optimized Maxent niche model and transferred it to the LGM based on four GCMs (CCSM4, IPSL‐CM5A‐LR, MIROC‐ESM, MPI‐ESM‐P), followed by phylogeographic analyses to test hypotheses of changes in distribution according to each GCM. CCSM4 and IPSL‐CM5A‐LR indicated an LGM suitability area forC. mexicanusmainly in the southern portion of its range, suggesting that extant focal populations to the north result from postglacial colonization. In contrast, MIROC‐ESM and MPI‐ESM‐P indicated LGM suitability for three or all the populations, respectively. Genetic results for the four focal populations showed high genetic diversity and signals of constant population size. Because only the hindcast based on MPI‐ESM‐P generated the prediction of stable occupation for all four sites, we interpret that its estimate (a cold and wet LGM climate) best approximates reality for this system. Future studies can apply this framework using more extensive genetic or genomic data and finer temporal resolutions, also exploring differences in the assumptions and methodologies underlying the various GCMs. 

Abstract Quaternary climatic oscillations affected species distributions worldwide, creating cycles of connectivity and isolation that impacted population demography and promoted lineage divergence. These effects have been well studied in temperate regions. Taxa inhabiting mesic montane habitats in tropical ecosystems show high levels of endemism and diversification in the distinct mountain ranges they inhabit; such a pattern has commonly been ascribed to past climatic oscillations, but few phylogeographic studies have tested this hypothesis. Here, we combine ecological niche models of species distributions with molecular data to study phylogeographic patterns in two rodents endemic to the highlands of Costa Rica and western Panama (Reithrodontomys creper and Nephelomys devius). In so doing, we apply a novel approach that incorporates a basic ecological principle: the expected positive relationship between environmental suitability and population abundance. Specifically, we use niche models to predict potential patterns of population connectivity and stability of different suitability levels during climatic extremes of the last glacial–interglacial cycle; we then test these predictions with population genetic analyses of a mitochondrial and a nuclear marker. The detailed predictions arising from the different levels of suitability were moderately to highly congruent with the molecular data depending on the species. Overall, results suggest that in these tropical montane ecosystems, cycles of population connectivity and isolation followed a pattern opposite to that typically described for temperate or lowland tropical ecosystems: namely, higher connectivity during the colder glacials, with isolation in montane refugia during the interglacials, including today. Nevertheless, the individualistic patterns for each species indicate a potentially wide gamut of phylogeographic histories reflecting particularities of their niches. Taken together, this study illustrates how phylogeographic inferences may benefit from niche model outputs that provide more detailed predictions of connectivity and finer characterizations of potential refugia through time.