Species distribution models (SDM) are widely used in diverse research areas because of their simple data requirements and application versatility. However, SDM outcomes are sensitive to data input and methodological choices. Such sensitivity and diverse applications mean that flexibility is necessary to create SDMs with tailored protocols for a given set of data and model use. We introduce the r package flexsdm for supporting flexible species distribution modelling workflows. flexsdm functions and their arguments serve as building blocks to construct a specific modelling protocol for user's needs. The main flexsdm features are modelling flexibility, integration with other modelling tools, simplicity of the objects returned and function speed. As an illustration, we used flexsdm to define a complete workflow for California red fir Abies magnifica. This package provides modelling flexibility by incorporating comprehensive tools structured in three steps: (a) The Pre‐modelling functions that prepare input, for example, sampling bias correction, sampling pseudo‐absences and background points, data partitioning, and reducing collinearity in predictors. (b) The Modelling functions allow fitting and evaluating different modelling approaches, including individual algorithms, tuned models, ensembles of small models and ensemble models. (c) The Post‐modelling functions include tools related to models' predictions, interpolation and overprediction correction. Because flexsdm comprises a large part of the SDM process, from outlier detection to overprediction correction, flexsdm users can delineate partial or complete workflows based on the combination functions to meet specific modelling needs.

Species distribution and ecological niche models (hereafter SDMs) are popular tools with broad applications in ecology, biodiversity conservation, and environmental science. Many SDM applications require projecting models in environmental conditions non‐analog to those used for model training (extrapolation), giving predictions that may be statistically unsupported and biologically meaningless. We introduce a novel method, Shape, a model‐agnostic approach that calculates the extrapolation degree for a given projection data point by its multivariate distance to the nearest training data point. Such distances are relativized by a factor that reflects the dispersion of the training data in environmental space. Distinct from other approaches, Shape incorporates an adjustable threshold to control the binary discrimination between acceptable and unacceptable extrapolation degrees. We compared Shape's performance to five extrapolation metrics based on their ability to detect analog environmental conditions in environmental space and improve SDMs suitability predictions. To do so, we used 760 virtual species to define different modeling conditions determined by species niche tolerance, distribution equilibrium condition, sample size, and algorithm. All algorithms had trouble predicting species niches. However, we found a substantial improvement in model predictions when model projections were truncated independently of extrapolation metrics. Shape's performance was dependent on extrapolation threshold used to truncate models. Because of this versatility, our approach showed similar or better performance than the previous approaches and could better deal with all modeling conditions and algorithms. Our extrapolation metric is simple to interpret, captures the complex shapes of the data in environmental space, and can use any extrapolation threshold to define whether model predictions are retained based on the extrapolation degrees. These properties make this approach more broadly applicable than existing methods for creating and applying SDMs. We hope this method and accompanying tools support modelers to explore, detect, and reduce extrapolation errors to achieve more reliable models. Keywords: environmental novelty, extrapolation, Mahalanobis distance, model prediction, non‐analog environmental data, transferability

Aim: Rarity and geographic aspects of species distributions mediate their vulnerability to global change. We explore the relationships between species rarity and geography and their exposure to climate and land use change in a biodiversity hotspot.Location: California, USA.Taxa: One hundred and six terrestrial plants.Methods: We estimated four rarity traits: range size, niche breadth, number of habitat patches, and patch isolation; and three geographic traits: mean elevation, topographic heterogeneity, and distance to coast. We used species distribution models to measure species exposure—predicted change in continuous habitat suitability within currently occupied habitat—under climate and land use change scenarios. Using regression models, decision‐tree models and variance partitioning, we assessed the relationships between species rarity, geography, and exposure to climate and land use change.Results: Rarity, geography and greenhouse gas emissions scenario explained >35% of variance in climate change exposure and >61% for land use change exposure. While rarity traits (range size and number of habitat patches) were most important for explaining species exposure to climate change, geographic traits (elevation and topographic heterogeneity) were more strongly associated with species' exposure to land use change.Main conclusions: Species with restricted range sizes and low topographic heterogeneity across their distributions were predicted to be the most exposed to climate change, while species at low elevations were the most exposed to habitat loss via land use change. However, even some broadly distributed species were projected to lose >70% of their currently suitable habitat due to climate and land use change if they are in geographically vulnerable areas, emphasizing the need to consider both species rarity traits and geography in vulnerability assessments.

Aim: Variation in spatial predictions of species' ranges made by various models has been recognized as a significant source of uncertainty for modelling species distributions. Consensus approaches that combine the results of multiple models have been employed to reduce the uncertainty introduced by different algorithms. We evaluate how estimates of habitat suitability, projected using species distribution models (SDMs), varied among different consensus methods relative to the variation introduced by different global climate models (GCMs) and representative concentration pathways (RCPs) used for projection.Location: California Floristic Province (California, US portion).Methods: We modelled the current and future potential distributions of 82 terrestrial plant species, developing model predictions under different combinations of GCMs, RCPs, time periods, dispersal assumptions and SDM consensus methods commonly used to combine different species distribution modelling algorithms. We assessed how each of these factors contributed to the variability in future predictions of species habitat suitability change and aggregate measures of proportional change in species richness. We also related variability in species‐level habitat change to species' attributes.Results: Assuming full dispersal capacity, the variability between habitat predictions made by different consensus methods was higher than the variability introduced by different RCPs and GCMs. The relationships between species' attributes and variability in future habitat predictions depended on the source of uncertainty and dispersal assumptions. However, species with small ranges or low prevalence tended to be associated with high variability in range change forecasts.Main Conclusions: Our results support exploring multiple consensus approaches when considering changes in habitat suitability outside of species' current distributions, especially when projecting species with low prevalence and small range sizes, as these species tend to be of the greatest conservation concern yet produce highly variable model outputs. Differences in vulnerability between diverging greenhouse gas concentration scenarios are most readily observed for end‐of‐century time periods and within species' currently occupied habitats (no dispersal).

An increasing amount of California’s landscape has burned in wildfires in recent decades, in conjunction with increasing temperatures and vapor pressure deficit due to climate change. As the wildland–urban interface expands, more people are exposed to and harmed by these extensive wildfires, which are also eroding the resilience of terrestrial ecosystems. With future wildfire activity expected to increase, there is an urgent demand for solutions that sustain healthy ecosystems and wildfire-resilient human communities. Those who manage disaster response, landscapes, and biodiversity rely on mapped projections of how fire activity may respond to climate change and other human factors. California wildfire is complex, however, and climate–fire relationships vary across the state. Given known geographical variability in drivers of fire activity, we asked whether the geographical extent of fire models used to create these projections may alter the interpretation of predictions. We compared models of fire occurrence spanning the entire state of California to models developed for individual ecoregions and then projected end-of-century future fire patterns under climate change scenarios. We trained a Maximum Entropy model with fire records and hydroclimatological variables from recent decades (1981 to 2010) as well as topographic and human infrastructure predictors. Results showed substantial variation in predictors of fire probability and mapped future projections of fire depending upon geographical extents of model boundaries. Only the ecoregion models, accounting for the unique patterns of vegetation, climate, and human infrastructure, projected an increase in fire in most forested regions of the state, congruent with predictions from other studies.

Landsat is among the most popular satellites used for forest change assessments. Traditionally, Landsat data users relied on annual or biennial images to measure forest recovery after disturbance, a process that is difficult to monitor at broad scales. With the availability of free Landsat data, intra-annual change analyses are now possible. Phenology, the timing of cyclical vegetation events, can be estimated using indices derived from intra-annual remote sensing data and used to classify different vegetation types after a disturbance. We used a smoothed harmonic modelling approach to estimate NDVI and NBR phenology patterns in pre- and post-fire Landsat sample pixels for two forest groups in South Carolina, using nearby unburned samples as an approximate control group. These methods take advantage of all available images collected by Landsat 5, 7, and 8 for the study area. We found that within burned samples, there were differences in phenology for the two forest groups, while the unburned samples showed no forest group differences. Phenology patterns also differed based on fire severity. These methods take advantage of the freely available Landsat archive and can be used to characterize intra-annual fluctuations in vegetation following a variety of disturbances in the southeastern U.S. and other regions. Our approach builds on other harmonic approaches that use the Landsat archive to detect forest change, such as the Continuous Change Detection and Classification (CCDC) algorithm, and provides a tool to describe post-disturbance forest change.