MaxEnt is a program for modelling species distributions from presence-only species records. This paper is written for ecologists and describes the MaxEnt model from a statistical perspective, making explicit links between the structure of the model, decisions required in producing a modelled distribution, and knowledge about the species and the data that might affect those decisions. To begin we discuss the characteristics of presence-only data, highlighting implications for modelling distributions. We particularly focus on the problems of sample bias and lack of information on species prevalence. The keystone of the paper is a new statistical explanation of MaxEnt which shows that the model minimizes the relative entropy between two probability densities (one estimated from the presence data and one, from the landscape) defined in covariate space. For many users, this viewpoint is likely to be a more accessible way to understand the model than previous ones that rely on machine learning concepts. We then step through a detailed explanation of MaxEnt describing key components (e.g. covariates and features, and definition of the landscape extent), the mechanics of model fitting (e.g. feature selection, constraints and regularization) and outputs. Using case studies for a Banksia species native to south-west Australia and a riverine fish, we fit models and interpret them, exploring why certain choices affect the result and what this means. The fish example illustrates use of the model with vector data for linear river segments rather than raster (gridded) data. Appropriate treatments for survey bias, unprojected data, locally restricted species, and predicting to environments outside the range of the training data are demonstrated, and new capabilities discussed. Online appendices include additional details of the model and the mathematical links between previous explanations and this one, example code and data, and further information on the case studies.
Aim Advancement in ecological methods predicting species distributions is a crucial precondition for deriving sound management actions. Maximum entropy (MaxEnt) models are a popular tool to predict species distributions, as they are considered able to cope well with sparse, irregularly sampled data and minor location errors. Although a fundamental assumption of MaxEnt is that the entire area of interest has been systematically sampled, in practice, MaxEnt models are usually built from occurrence records that are spatially biased towards better-surveyed areas. Two common, yet not compared, strategies to cope with uneven sampling effort are spatial filtering of occurrence data and background manipulation using environmental data with the same spatial bias as occurrence data. We tested these strategies using simulated data and a recently collated dataset on Malay civet Viverra tangalunga in Borneo. Location Borneo, Southeast Asia. Methods We collated 504 occurrence records of Malay civets from Borneo of which 291 records were from 2001 to 2011 and used them in the MaxEnt analysis (baseline scenario) together with 25 environmental input variables. We simulated datasets for two virtual species (similar to a range-restricted highland and a lowland species) using the same number of records for model building. As occurrence records were biased towards north-eastern Borneo, we investigated the efficacy of spatial filtering versus background manipulation to reduce overprediction or underprediction in specific areas. Results Spatial filtering minimized omission errors (false negatives) and commission errors (false positives). We recommend that when sample size is insufficient to allow spatial filtering, manipulation of the background dataset is preferable to not correcting for sampling bias, although predictions were comparatively weak and commission errors increased. Main Conclusions We conclude that a substantial improvement in the quality of model predictions can be achieved if uneven sampling effort is taken into account, thereby improving the efficacy of species conservation planning.
Aim Woody plants were not widely considered to be important invasive alien species until fairly recently. Thousands of species of trees and shrubs have, however, been moved around the world. Many species have spread from planting sites, and some are now among the most widespread and damaging of invasive organisms. This article presents a global list of invasive alien trees and shrubs. It discusses taxonomic biases, geographical patterns, modes of dispersal, reasons for introductions and key issues regarding invasions of non-native woody plants around the world. Location Global. Methods An exhaustive survey was made of regional and national databases and the literature. Correspondence with botanists and ecologists and our own observations in many parts of the world expanded the list. Presence of invasive species was determined for each of 15 broad geographical regions. The main reasons for introduction and dissemination were determined for each species. Results The list comprises 622 species (357 trees, 265 shrubs in 29 plant orders, 78 families, 286 genera). Regions with the largest number of woody invasive alien species are: Australia (183); southern Africa (170); North America (163); Pacific Islands (147); and New Zealand (107). Species introduced for horticulture dominated the list (62% of species: 196 trees and 187 shrubs). The next most important reasons for introduction and dissemination were forestry (13%), food (10%) and agroforestry (7%). Three hundred and twenty-three species (52%) are currently known to be invasive in only one region, and another 126 (20%) occur in only two regions. Only 38 species (6%) are very widespread (invasive in six or more regions). Over 40% of invasive tree species and over 60% of invasive shrub species are bird dispersed. Main conclusions Only between 0.5% and 0.7% of the world's tree and shrub species are currently invasive outside their natural range, but woody plant invasions are rapidly increasing in importance around the world. The objectively compiled list of invasive species presented here provides a snapshot of the current dimensions of the phenomenon and will be useful for screening new introductions for invasive potential.
A wide range of modelling algorithms is used by ecologists, conservation practitioners, and others to predict species ranges from point locality data. Unfortunately, the amount of data available is limited for many taxa and regions, making it essential to quantify the sensitivity of these algorithms to sample size. This is the first study to address this need by rigorously evaluating a broad suite of algorithms with independent presence-absence data from multiple species and regions. We evaluated predictions from 12 algorithms for 46 species (from six different regions of the world) at three sample sizes (100, 30, and 10 records). We used data from natural history collections to run the models, and evaluated the quality of model predictions with area under the receiver operating characteristic curve (AUC). With decreasing sample size, model accuracy decreased and variability increased across species and between models. Novel modelling methods that incorporate both interactions between predictor variables and complex response shapes (i.e. GBM, MARS-INT, BRUTO) performed better than most methods at large sample sizes but not at the smallest sample sizes. Other algorithms were much less sensitive to sample size, including an algorithm based on maximum entropy (MAXENT) that had among the best predictive power across all sample sizes. Relative to other algorithms, a distance metric algorithm (DOMAIN) and a genetic algorithm (OM-GARP) had intermediate performance at the largest sample size and among the best performance at the lowest sample size. No algorithm predicted consistently well with small sample size (n < 30) and this should encourage highly conservative use of predictions based on small sample size and restrict their use to exploratory modelling.
Aim Spatial modelling techniques are increasingly used in species distribution modelling. However, the implemented techniques differ in their modelling performance, and some consensus methods are needed to reduce the uncertainty of predictions. In this study, we tested the predictive accuracies of five consensus methods, namely Weighted Average (WA), Mean(All), Median(All), Median(PCA), and Best, for 28 threatened plant species. Location North-eastern Finland, Europe. Methods The spatial distributions of the plant species were forecasted using eight state-of-the-art single-modelling techniques providing an ensemble of predictions. The probability values of occurrence were then combined using five consensus algorithms. The predictive accuracies of the single-model and consensus methods were assessed by computing the area under the curve (AUC) of the receiver-operating characteristic plot. Results The mean AUC values varied between 0.697 (classification tree analysis) and 0.813 (random forest) for the single-models, and from 0.757 to 0.850 for the consensus methods. WA and Mean(All) consensus methods provided significantly more robust predictions than all the single-models and the other consensus methods. Main conclusions Consensus methods based on average function algorithms may increase significantly the accuracy of species distribution forecasts, and thus they show considerable promise for different conservation biological and biogeographical applications.
Aim The global sprawl of marine hard infrastructure (e.g. breakwaters, sea walls and jetties) can extensively modify coastal seascapes, but the knowledge of such impacts remains limited to local scales. We examined the regional-scale effects of marine artificial habitats on the distribution and abundance of assemblages of ascidians, a key group of ecosystem engineer species in benthic fouling systems. Location Five hundred kilometers of coastline in the North Adriatic Sea. Methods We sampled a variety of natural reefs, marine infrastructures and marinas, and tested hypotheses about the role of habitat type and location in influencing the relative distribution and abundance of both native and non-indigenous species. Results Assemblages differed significantly between natural and artificial habitats and among different types of artificial habitats. Non-indigenous species were 2–3 times more abundant on infrastructures built along sedimentary coastlines than on natural rocky reefs or infrastructures built close to rocky coastlines. Conversely, native species were twice as abundant on natural reefs than on nearby infrastructures and were scarce to virtually absent on infrastructures built along sedimentary coasts. The species composition of assemblages in artificial habitats was more similar to that of marinas than of natural reefs, independently of their location. Main conclusions Our results show that marine infrastructures along sandy shores disproportionally favour non-indigenous over native hard bottom species, affecting their spread at regional scales. This is particularly concerning for coastal areas that have low natural densities of rocky reef habitats. We discuss design and management options to improve the quality as habitat of marine infrastructures and to favour their preferential use by native species over non-indigenous ones.
Aim Interest in species distribution models (SDMs) and related niche studies has increased dramatically in recent years, with several books and reviews being prepared since 2000. The earliest SDM studies are dealt with only briefly even in the books. Consequently, many researchers are unaware of when the first SDM software package (bioclim) was developed and how a broad range of applications using the package was explored within the first 8 years following its release. The purpose of this study is to clarify these early developments and initial applications, as well as to highlight bioclim's continuing relevance to current studies. Location Mainly Australia and New Zealand, but also some global applications. Methods We outline the development of the bioclim package, early applications (1984–1991) and its current relevance. Results bioclim was the first SDM package to be widely used. Early applications explored many of the possible uses of SDMs in conservation biogeography, such as quantifying the environmental niche of species, identifying areas where a species might be invasive, assisting conservation planning and assessing the likely impacts of climate change on species distributions. Main conclusions Understanding this pioneering work is worthwhile as bioclim was for many years one of the leading SDM packages and remains widely used. Climate interpolation methods developed for bioclim were used to create the WorldClim database, the most common source of climate data for SDM studies, and bioclim variables are used in about 76% of recent published MaxEnt analyses of terrestrial ecosystems. Also, some of the bioclim studies from the late 1980s, such as measuring niche (both realized and fundamental) and assessing possible impacts of climate change, are still highly relevant to key conservation biogeography issues.
Biodiversity hotspots are conservation priorities. We identify the North American Coastal Plain (NACP) as a global hotspot based on the classic definition, a region with > 1500 endemic plant species and > 70% habitat loss. This region has been bypassed in prior designations due to misconceptions and myths about its ecology and history. These fallacies include: (1) young age of the NACP, climatic instability over time and submergence during high sea-level stands; (2) climatic and environmental homogeneity; (3) closed forest as the climax vegetation; and (4) fire regimes that are mostly anthropogenic. We show that the NACP is older and more climatically stable than usually assumed, spatially heterogeneous and extremely rich in species and endemics for its range of latitude, especially within pine savannas and other mostly herbaceous and fire-dependent communities. We suspect systematic biases and misconceptions, in addition to missing information, obscure the existence of similarly biologically significant regions world-wide. Potential solutions to this problem include (1) increased field biological surveys and taxonomic determinations, especially within grassy biomes and regions with low soil fertility, which tend to have much overlooked biodiversity; (2) more research on the climatic refugium role of hotspots, given that regions of high endemism often coincide with regions with low velocity of climate change; (3) in low-lying coastal regions, consideration of the heterogeneity in land area generated by historically fluctuating sea levels, which likely enhanced opportunities for evolution of endemic species; and (4) immediate actions to establish new protected areas and implement science-based management to restore evolutionary environmental conditions in newly recognized hotspots.
Aim Invasion ecology includes many hypotheses. Empirical evidence suggests that most of these can explain the success of some invaders to some degree in some circumstances. If they all are correct, what does this tell us about invasion? We illustrate the major themes in invasion ecology, and provide an overarching framework that helps organize research and foster links among subfields of invasion ecology and ecology more generally. Location Global. Methods We review and synthesize 29 leading hypotheses in plant invasion ecology. Structured around propagule pressure (P), abiotic characteristics (A) and biotic characteristics (B), with the additional influence of humans (H) on P, A and B (hereon PAB), we show how these hypotheses fit into one paradigm. P is based on the size and frequency of introductions, A incorporates ecosystem invasibility based on physical conditions, and B includes the characteristics of invading species (invasiveness), the recipient community and their interactions. Having justified the PAB framework, we propose a way in which invasion research could progress. Results By highlighting the common ground among hypotheses, we show that invasion ecology is encumbered by theoretical redundancy that can be removed through integration. Using both holistic and incremental approaches, we show how the PAB framework can guide research and quantify the relative importance of different invasion mechanisms. Main conclusions If the prime aim is to identify the main cause of invasion success, we contend that a top-down approach that focuses on PAB maximizes research efficiency. This approach identifies the most influential factors first, and subsequently narrows the number of potential causal mechanisms. By viewing invasion as a multifaceted process that can be partitioned into major drivers and broken down into a series of sequential steps, invasion theory can be rigorously tested, understanding improved and effective weed management techniques identified.
Aim A major issue in ecology, biogeography, conservation biology and invasion biology is the extent to which climate, and hence climate change, contributes to the positions of species' range boundaries. Thirty years of rapid climate warming provides an excellent opportunity to test the hypothesis that climate acts as a major constraint on range boundaries, treating anthropogenic climate change as a large-scale experiment. Location UK and global data, and literature. Methods This article analyses the frequencies with which species have responded to climate change by shifting their range boundaries. It does not consider abundance or other changes. Results For the majority of species, boundaries shifted in a direction that is concordant with being a response to climate change; 84% of all species have expanded in a polewards direction as the climate has warmed (for the best data available), which represents an excess of 68% of species after taking account of the fact that some species may shift in this direction for non-climatic reasons. Other data sets also show an excess of animal range boundaries expanding in the expected direction. Main conclusions Climate is likely to contribute to the majority of terrestrial and freshwater range boundaries. This generalization excludes species that are endemic to specific islands, lakes, rivers and geological outcrops, although these local endemics are not immune from the effects of climate change. The observed shifts associated with recent climate change are likely to have been brought about through both direct and indirect (changes to species' interactions) effects of climate; indirect effects are discussed in relation to laboratory experiments and invasive species. Recent observations of range boundary shifts are consistent with the hypothesis that climate contributes to, but is not the sole determinant of, the position of the range boundaries of the majority of terrestrial animal species.
The global database of invasive trees and shrubs (Richardson & Rejmánek, 2011; Diversity Distrib. 17, 788-809) has been updated, resulting in a total of 751 species (434 trees and 317 shrubs) from 90 families. Ten originally listed species were deleted (synonyms, inconclusive identification, etc.) and 139 additional invasive species (86 trees and 53 shrubs) are now included in the database. For many species, new records on their adventive distributions are added. The updated database also includes the native ranges for all listed species.
Aim Climate change poses significant threats to biodiversity, including impacts on species distributions, abundance and ecological interactions. At a landscape scale, these impacts, and biotic responses such as adaptation and migration, will be mediated by spatial heterogeneity in climate and climate change. We examine several aspects of the geography of climate change and their significance for biodiversity conservation. Location California and Nevada, USA. Methods Using current climate surfaces (PRISM) and two scenarios of future climate (A1b, 2070-2099, warmer-drier and warmer-wetter), we mapped disappearing, declining, expanding and novel climates, and the velocity and direction of climate change in California and Nevada. We also examined fine-scale spatial heterogeneity in protected areas of the San Francisco Bay Area in relation to reserve size, topographic complexity and distance from the ocean. Results Under the two climate change scenarios, current climates across most of California and Nevada will shrink greatly in extent, and the climates of the highest peaks will disappear from this region. Expanding and novel climates are projected for the Central Valley. Current temperature isoclines are projected to move up to 4.9 km year−1 in flatter regions, but substantially slower in mountainous areas because of steep local topoclimate gradients. In the San Francisco Bay Area, climate diversity within currently protected areas increases with reserve size and proximity to the ocean (the latter because of strong coastal climate gradients). However, by 2100 of almost 500 protected areas (>100 ha), only eight of the largest are projected to experience temperatures within their currently observed range. Topoclimate variability will further increase the range of conditions experienced and needs to be incorporated in future analyses. Main Conclusions Spatial heterogeneity in climate, from mesoclimate to topoclimate scales, represents an important spatial buffer in response to climate change, and merits increased attention in conservation planning.
In the past decades, Brazil made important progress in the conservation of forest ecosystems. Non-forest ecosystems (NFE), in contrast, have been neglected, even though they cover large parts of the country and have biodiversity levels comparable to forests. To avoid losing much of its biodiversity and ecosystem services, conservation and sustainable land use policies in Brazil need to be extended to NFE. A strategy for conservation of Brazil's NFE should encompass the following elements: (1) creation of new large protected areas in NFE; (2) enforcement of legal restrictions of land use; (3) extension of subsidy programs and governance commitments to NFE; (4) improvement of ecosystem management and sustainable use in NFE; and (5) improvement of monitoring of land use change in NFE. If Brazil managed to extend its conservation successes to NFE, it not only would contribute significantly to conservation of its biodiversity, but also could take the lead in conservation of NFE world-wide.
Aim Nowadays, large amounts of species distribution data and software for implementing different species distribution modelling methods are freely available through the internet. As a result, methodological works that analyse the relative performance of modelling techniques, as well as those that study which species characteristics affect their performance, are necessary. We discuss three important topics that must be kept in mind when modelling species distributions, namely (i) the distinction between potential and realized distribution, (ii) the effect of the relative occurrence area of the species on the results of the evaluation of model performance, and (iii) the general inaccuracy of the predictions of the realized distribution provided by species distribution modelling methods. Location Unspecific. Methods Using some recent papers as a basis, we illustrate the three issues mentioned above and discuss the negative implications of neglecting them. Results Considering a potential-realized distribution gradient, different modelling methods may be arranged along this gradient according to their ability to model any concept. Complex techniques may be more suitable to model the realized distribution than simple ones, which may be more appropriate to estimate the potential distribution. Comparisons among techniques must consider this scenario. The relative occurrence area of the species conditions the results of the evaluation scores, implying that models of rare species will unavoidably yield higher discrimination values. Moreover, discrimination values that are usually reported in the literature may imply considerable over or underestimations of the distribution of the species. Main conclusions It is extremely important to establish a solid conceptual and methodological framework on which the emergent field of species distribution modelling can stand and develop.
Aim Invasive alien species (IAS) pose a significant threat to biodiversity. The Convention on Biological Diversity's 2010 Biodiversity Target, and the associated indicator for IAS, has stimulated globally coordinated efforts to quantify patterns in the extent of biological invasion, its impact on biodiversity and policy responses. Here, we report on the outcome of indicators of alien invasion at a global scale. Location Global. Methods We developed four indicators in a pressure-state-response framework, i.e. number of documented IAS (pressure), trends in the impact of IAS on biodiversity (state) and trends in international agreements and national policy adoption relevant to reducing IAS threats to biodiversity (response). These measures were considered best suited to providing globally representative, standardized and sustainable indicators by 2010. Results We show that the number of documented IAS is a significant underestimate, because its value is negatively affected by country development status and positively by research effort and information availability. The Red List Index demonstrates that IAS pressure is driving declines in species diversity, with the overall impact apparently increasing. The policy response trend has nonetheless been positive for the last several decades, although only half of countries that are signatory to the Convention on Biological Diversity (CBD) have IAS-relevant national legislation. Although IAS pressure has apparently driven the policy response, this has clearly not been sufficient and/or adequately implemented to reduce biodiversity impact. Main conclusions For this indicator of threat to biodiversity, the 2010 Biodiversity Target has thus not been achieved. The results nonetheless provide clear direction for bridging the current divide between information available on IAS and that needed for policy and management for the prevention and control of IAS. It further highlights the need for measures to ensure that policy is effectively implemented, such that it translates into reduced IAS pressure and impact on biodiversity beyond 2010.
Aim We argue that `propagule pressure', a key term in invasion biology, has been attributed at least three distinct definitions (with usage of a related term causing additional confusion). All of the definitions refer to fundamental concepts within the invasion process, with the result that the distinct importance of these different concepts has been at best diluted, and at worst lost. Location Global. Methods We reviewed pertinent literature on propagule pressure to resolve confusion about different uses of the term `propagule pressure' and we introduced a new term for one variant, colonization pressure. We conducted a computer simulation whereby the introduction of species is represented as a simple sampling process to elucidate the relationship between propagule and colonization pressure. Results We defined colonization pressure as the number of species introduced or released to a single location, some of which will go on to establish a self-sustaining population and some of which will not. We subsequently argued that colonization pressure should serve as a null hypothesis for understanding temporal or spatial differences in exotic species richness, as the more species that are introduced, the more we should expect to establish. Finally, using a simple simulation, we showed that propagule pressure is related to colonization pressure, but in a non-linear manner. Main conclusion We suggest that the nature of the relationship between propagule pressure and colonization pressure, as well as the efficacy of various proxy measures of each, require more detailed exploration if invasion ecology is to continue to develop into a more predictive science.
Aim: Invasive alien species (IAS) are recognized as major drivers of biodiversity loss, but few causal relationships between IAS and species declines have been documented. In this study, we compare the distribution (Belgium and Britain) and abundance (Belgium, Britain and Switzerland) of formerly common and widespread native ladybirds before and after the arrival of Harmonia axyridis, a globally rapidly expanding IAS. Location: Europe Methods: We used generalized linear mixed-effects models (GLMMs) to assess the distribution trends of eight conspicuous and historically widespread and common species of ladybird within Belgium and Britain before and after the arrival of H. axyridis. The distribution data were collated largely through public participatory surveys but verified by a recognized expert. We also used GLMMs to model trends in the abundance of ladybirds using data collated through systematic surveys of deciduous trees in Belgium, Britain and Switzerland. Results: Five (Belgium) and seven (Britain) of eight species studied show substantial declines attributable to the arrival of H. axyridis. Indeed, the two-spot ladybird, Adalia bipunctata, declined by 30% (Belgium) and 44% (Britain) over 5 years after the arrival of H. axyridis. Trends in ladybird abundance revealed similar patterns of declines across three countries. Main conclusion: Together, these analyses show H. axyridis to be displacing native ladybirds with high niche overlap, probably through predation and competition. This finding provides strong evidence of a causal link between the arrival of an IAS and decline in native biodiversity. Rapid biotic homogenization at the continental scale could impact on the resilience of ecosystems and severely diminish the services they deliver.
A large proportion of European biodiversity today depends on habitat provided by low-intensity farming practices, yet this resource is declining as European agriculture intensifies. Within the European Union, particularly the central and eastern new member states have retained relatively large areas of species-rich farmland, but despite increased investment in nature conservation here in recent years, farmland biodiversity trends appear to be worsening. Although the high biodiversity value of Central and Eastern European farmland has long been reported, the amount of research in the international literature focused on farmland biodiversity in this region remains comparatively tiny, and measures within the EU Common Agricultural Policy are relatively poorly adapted to support it. In this opinion study, we argue that, 10 years after the accession of the first eastern EU new member states, the continued under-representation of the low-intensity farmland in Central and Eastern Europe in the international literature and EU policy is impeding the development of sound, evidence-based conservation interventions. The biodiversity benefits for Europe of existing low-intensity farmland, particularly in the central and eastern states, should be harnessed before they are lost. Instead of waiting for species-rich farmland to further decline, targeted research and monitoring to create locally appropriate conservation strategies for these habitats is needed now.
Aim A major challenge for invasion ecology is to identify high-impact invaders to guide prioritization of management interventions. We argue that species with the potential to cause regime shifts (altered states of ecosystem structure and function that are difficult or impossible to reverse) should be prioritized. These are species that modify ecosystems in ways that enhance their own persistence and suppress that of native species through reinforcing feedback processes. Methods Using both systems analysis and meta-analysis approaches, we synthesized changes to ecosystems caused by 173 invasive plant species. For the systems analysis, we examined published studies of impacts of invasive plants to determine which presented evidence consistent with a reinforcement of feedback processes. For the meta-analysis, we calculated the effect size ratio between standardized changes in recipient ecosystem and in the status of introduced species as an indication of a reinforcing feedback in particular species-environment combinations. The systems analysis approach allowed us to conceptualize regime shifts in invader-dominated landscapes and to estimate the likelihood of such changes occurring. The meta-analysis allowed us to quantitatively verify the conceptual model and the key invader-context feedbacks and to detect the strength and direction of feedbacks. Results Most reinforcing feedbacks involve impacts on soil-nutrient cycling by shrub and tree invaders in forests and herbaceous invaders in wetlands. Feedbacks resulting in regime shifts were most likely related to processes associated with seed banks, fire and nutrient cycling. Results were used to derive a key for identifying high-impact invaders. Main conclusions Identifying combinations of plant life-forms and ecosystems most likely to result in regime shifts is a robust approach for predicting highimpact invasions and therefore for prioritizing management interventions. The meta-analysis revealed the need for more quantitative studies, including manipulative experiments, on ecosystem feedbacks.
Aim To demonstrate that multi-modelling methods have effectively been used to combine static species distribution models (SDM), predicting the geographical pattern of suitable habitat, with dynamic landscape and population models to forecast the impacts of environmental change on species' status, an important goal of conservation biogeography. Methods Three approaches were considered: (1) incorporating models of species migration to understand the ability of a species to occupy suitable habitat in new locations; (2) linking models of landscape disturbance and succession to models of habitat suitability; and (3) fully linking models of habitat suitability, habitat dynamics and spatially explicit population dynamics. Results Linking species-environment relationships, landscape dynamics and population dynamics in a multi-modelling framework allows the combined impacts of climate change (affecting species distribution and vital rates) and land cover dynamics (land use change, altered disturbance regimes) on species to be predicted. This approach is only feasible if the life history parameters and habitat requirements of the species are well understood. Main conclusions Forecasts of the impacts of global change on species may be improved by considering multiple causes. A range of methods are available to address the interactions of changing habitat suitability, habitat dynamics and population response that vary in their complexity, realism and data requirements.