Fauna habitat modelling and mapping: A review and case study in the Lower Hunter Central Coast region of NSW

Abstract Habitat models are now broadly used in conservation planning on public lands. If implemented correctly, habitat modelling is a transparent and repeatable technique for describing and mapping biodiversity values, and its application in peri‐urban and agricultural landscape planning is likely to expand rapidly. Conservation planning in such landscapes must be robust to the scrutiny that arises when biodiversity constraints are placed on developers and private landholders. A standardized modelling and model evaluation method based on widely accepted techniques will improve the robustness of conservation plans. We review current habitat modelling and model evaluation methods and provide a habitat modelling case study in the New South Wales central coast region that we hope will serve as a methodological template for conservation planners. We make recommendations on modelling methods that are appropriate when presence‐absence and presence‐only survey data are available and provide methodological details and a website with data and training material for modellers. Our aim is to provide practical guidelines that preserve methodological rigour and result in defendable habitat models and maps. The case study was undertaken in a rapidly developing area with substantial biodiversity values under urbanization pressure. Habitat maps for seven priority fauna species were developed using logistic regression models of species‐habitat relationships and a bootstrapping methodology was used to evaluate model predictions. The modelled species were the koala, tiger quoll, squirrel glider, yellow‐bellied glider, masked owl, powerful owl and sooty owl. Models ranked sites adequately in terms of habitat suitability and provided predictions of sufficient reliability for the purpose of identifying preliminary conservation priority areas. However, they are subject to multiple uncertainties and should not be viewed as a completely accurate representation of the distribution of species habitat. We recommend the use of model prediction in an adaptive framework whereby models are iteratively updated and refined as new data become available.     

Evaluation of consensus methods in predictive species distribution 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.     

BIOMOD – optimizing predictions of species distributions and projecting potential future shifts under global change

A new computation framework (BIOMOD: BIOdiversity MODelling) is presented, which aims to maximize the predictive accuracy of current species distributions and the reliability of future potential distributions using different types of statistical modelling methods. BIOMOD capitalizes on the different techniques used in static modelling to provide spatial predictions. It computes, for each species and in the same package, the four most widely used modelling techniques in species predictions, namely Generalized Linear Models (GLM), Generalized Additive Models (GAM), Classification and Regression Tree analysis (CART) and Artificial Neural Networks (ANN). BIOMOD was applied to 61 species of trees in Europe using climatic quantities as explanatory variables of current distributions. On average, all the different modelling methods yielded very good agreement between observed and predicted distributions. However, the relative performance of different techniques was idiosyncratic across species, suggesting that the most accurate model varies between species. The results of this evaluation also highlight that slight differences between current predictions from different modelling techniques are exacerbated in future projections. Therefore, it is difficult to assess the reliability of alternative projections without validation techniques or expert opinion. It is concluded that rather than using a single modelling technique to predict the distribution of several species, it would be more reliable to use a framework assessing different models for each species and selecting the most accurate one using both evaluation methods and expert knowledge.     

An evaluation of methods for modelling species distributions

Aim Various statistical techniques have been used to model species probabilities of occurrence in response to environmental conditions. This paper provides a comprehensive assessment of methods and investigates whether errors in model predictions are associated to specific kinds of geographical and environmental distributions of species. Location Portugal, Western Europe. Methods Probabilities of occurrence for 44 species of amphibians and reptiles in Portugal were modelled using seven modelling techniques: Gower metric, Ecological Niche Factor Analysis, classification trees, neural networks, generalized linear models, generalized additive models and spatial interpolators. Generalized linear and additive models were constructed with and without a term accounting for spatial autocorrelation. Model performance was measured using two methods: sensitivity and Kappa index. Species were grouped according to their spatial (area of occupancy and extent of occurrence) and environmental (marginality and tolerance) distributions. Two‐way comparison tests were performed to detect significant interactions between models and species groups. Results Interaction between model and species groups was significant for both sensitivity and Kappa index. This indicates that model performance varied for species with different geographical and environmental distributions. Artificial neural networks performed generally better, immediately followed by generalized additive models including a covariate term for spatial autocorrelation. Non‐parametric methods were preferred to parametric approaches, especially when modelling distributions of species with a greater area of occupancy, a larger extent of occurrence, lower marginality and higher tolerance. Main conclusions This is a first attempt to relate performance of modelling techniques with species spatial and environmental distributions. Results indicate a strong relationship between model performance and the kinds of species distributions being modelled. Some methods performed generally better, but no method was superior in all circumstances. A suggestion is made that choice of the appropriate method should be contingent on the goals and kinds of distributions being modelled.     

气候变化对马尾松潜在分布影响预估的多模型比较

物种分布模型被广泛应用于评估气候变化对物种分布的影响。随着计算机和统计学的发展, 模拟物种分布的模型层出不穷, 但对这些模型的相对表现知之甚少, 因此需要对其进行对比分析, 以便更可靠地评估气候变化的影响。该文采用3个比较新颖的组合集成学习(ensemble learning)模型(随机森林(random forest, RF)、广义助推法和NeuralEnsembles)、3个常规模型(广义线性模型、广义加法模型和分类回归树)、3个大气环流模型(global circulation model, GCM) (MIROC32_medres, JP; CCCMA_CGCM3, CA; BCCR-BCM2.0, NW)和一个气体排放情景(SRES_A2), 模拟分析了马尾松(Pinus massoniana)历史基准气候(1961-1990)和未来3个不同时期(2010-2039, 2020s; 2040-2069, 2050s; 2070-2099, 2080s)的潜在分布。基于环境阈值方法选择物种不发生区, 依据ClimateChina软件进行当前和未来气候数据的降尺度处理, 采用接收机工作特征曲线(receiver operator characteristic, ROC)下的面积(area under the curve, AUC)、Kappa值和真实技巧统计法(true skill statistic, TSS)以及马尾松种子区划范围来评价模型的预测精度。结果表明: 6个物种分布模型都具有较高的预测精度, 但组合集成学习模型的预测精度稍高于其他常规模型, 其中RF的预测精度最高。3个GCM和6个模型模拟条件下, 马尾松对气候变化的响应格局既有一致性也有异同性。一致性表现在: 随着时间的推移, 马尾松分布区将逐渐向北迁移, 未来潜在分布区的面积将逐渐增加; 异同性表现在: 在不同模型和不同气候情景下, 马尾松潜在分布区的迁移距离和面积变化幅度不同, 其中NW模式下预测的变化幅度小于CA和JP模式; RF模型预测的分布区迁移距离和面积变化幅度最大。随着时间的推移, 未来马尾松的18个潜在分布空间预测图(6个模型 × 3 GCM)之间的差异也逐渐增大, 其中空间不一致性地区主要集中发生在马尾松潜在分布区的北部和西部边缘地带。模型本身不同的构建原理以及GCM之间的差异是导致预测结果存在差异的主要原因。     

Does the interpolation accuracy of species distribution models come at the expense of transferability?

Model transferability (extrapolative accuracy) is one important feature in species distribution models, required in several ecological and conservation biological applications. This study uses 10 modelling techniques and nationwide data on both (1) species distribution of birds, butterflies, and plants and (2) climate and land cover in Finland to investigate whether good interpolative prediction accuracy for models comes at the expense of transferability – i.e. markedly worse performance in new areas. Models’ interpolation and extrapolation performance was primarily assessed using AUC (the area under the curve of a receiver characteristic plot) and Kappa statistics, with supplementary comparisons examining model sensitivity and specificity values. Our AUC and Kappa results show that extrapolation to new areas is a greater challenge for all included modelling techniques than simple filling of gaps in a well‐sampled area, but there are also differences among the techniques in the degree of transferability. Among the machine‐learning modelling techniques, MAXENT, generalized boosting methods (GBM), and artificial neural networks (ANN) showed good transferability while the performance of GARP and random forest (RF) decreased notably in extrapolation. Among the regression‐based methods, generalized additive models (GAM) and generalized linear models (GLM) showed good transferability. A desirable combination of good prediction accuracy and good transferability was evident for three modelling techniques: MAXENT, GBM, and GAM. However, examination of model sensitivity and specificity revealed that model types may differ in their tendencies to either increased over‐prediction of presences or absences in extrapolation, and some of the methods show contrasting changes in sensitivity vs specificity (e.g. ANN and GARP). Among the three species groups, the best transferability was seen with birds, followed closely by butterflies, whereas reliable extrapolation for plant species distribution models appears to be a major challenge at least at this scale. Overall, detailed knowledge of the behaviour of different techniques in various study settings and with different species groups is of utmost importance in predictive modelling.     

Interspecific competition constrains local abundance in highly suitable areas

Sites with high environmental suitability for species’ occurrence, in terms of abiotic conditions, may hold populations with higher local abundances by increasing reproductive and survival rates and decreasing extinction rate. Interspecific competition, however, may affect this relationship. Here we tested the hypothesis that local abundance of the gray slender opossum Marmosops incanus is affected by the local richness of potential competitors and environmental suitability derived from ecological niche models (ENMs). We also discuss the ability of distinct modelling methods to predict species’ abundance. We compiled occurrence records and information about M. incanus’ relative abundance from museums and published articles. Environmental suitability was derived from five algorithms using seven environmental predictors. To assess our hypothesis, we chose the best statistical models among generalised linear models and quantile regressions, and then tested whether the effects of richness of competitors on local abundance are stronger under highly suitable conditions. We found that environmental suitability given by presence-only methods are positively related to the maximum abundance of M. incanus. That is, species’ local abundance is low when suitability is low but can be either low or high when suitability is high. The richness of competitors, in turn, explains the abundance variation within sites with high environmental suitability. We strongly recommend that the relationship between abundance and suitability must be carefully interpreted when using ENMs to predict species’ distribution because biotic interactions can be the main driver of local abundance within highly suitable environments.     

Testing methods in species distribution modelling using virtual species: what have we learnt and what are we missing?

Species distribution models (SDMs) have become one of the major predictive tools in ecology. However, multiple methodological choices are required during the modelling process, some of which may have a large impact on forecasting results. In this context, virtual species, i.e. the use of simulations involving a fictitious species for which we have perfect knowledge of its occurrence–environment relationships and other relevant characteristics, have become increasingly popular to test SDMs. This approach provides for a simple virtual ecologist framework under which to test model properties, as well as the effects of the different methodological choices, and allows teasing out the effects of targeted factors with great certainty. This simplification is therefore very useful in setting up modelling standards and best practice principles. As a result, numerous virtual species studies have been published over the last decade. The topics covered include differences in performance between statistical models, effects of sample size, choice of threshold values, methods to generate pseudo‐absences for presence‐only data, among many others. These simulations have therefore already made a great contribution to setting best modelling practices in SDMs. Recent software developments have greatly facilitated the simulation of virtual species, with at least three different packages published to that effect. However, the simulation procedure has not been homogeneous, which introduces some subtleties in the interpretation of results, as well as differences across simulation packages. Here we 1) review the main contributions of the virtual species approach in the SDM literature; 2) compare the major virtual species simulation approaches and software packages; and 3) propose a set of recommendations for best simulation practices in future virtual species studies in the context of SDMs.     

A critical assessment of two species distribution models: a case study of the vervet monkey (Cercopithecus aethiops)

Aim Species distribution models are invaluable tools in biogeographical, ecological and applied biological research, but specific concerns have been raised in relation to different modelling techniques in terms of their validity. Here we compare two fundamentally different approaches to species distribution modelling, one based on simple occurrence data where the lack of an ecological framework has been criticized, and the other firmly based in socio‐ecological theory but requiring highly detailed behavioural information that is often limited in availability. Location (Sub‐Saharan) Africa. Methods We used two distinct techniques to predict the realized distribution of a model species, the vervet monkey (Cercopithecus aethiops Linnaeus, 1758). A maximum entropy model was produced taking 13 environmental variables and presence‐only data from 174 sites throughout Africa as input, with an additional 58 sites retained to test the model. A time‐budget model considering the same environmental variables was constructed from detailed behavioural data on 20 groups representing 14 populations, with presence‐only data from the remaining 218 sites reserved to test model predictions on vervet monkey occurrence. Both models were further validated against a reference species distribution map as drawn up by the African Mammals Databank. Results Both models performed well, with the time budget and maximum entropy algorithms correctly predicting vervet monkey presence at 78.4% and 91.4% of their respective test sites. Similarly, the time‐budget model correctly predicted presence and absence at 87.4% of map pixels against the reference distribution map, and the maximum entropy model achieved a success rate of 81.8%. Finally, there was a high level of agreement (81.6%) between the presence–absence maps produced by the two models, and the environmental variables identified as most strongly driving vervet monkey distribution were the same in both models. Main conclusions The time‐budget and maximum entropy models produced accurate and remarkably similar species distribution maps, despite fundamental differences in their conceptual and methodological approaches. Such strong convergence not only provides support for the credibility of current results, but also relieves concerns about the validity of the two modelling approaches.     

Towards novel approaches to modelling biotic interactions in multispecies assemblages at large spatial extents

Aim Biotic interactions – within guilds or across trophic levels – have widely been ignored in species distribution models (SDMs). This synthesis outlines the development of ‘species interaction distribution models’ (SIDMs), which aim to incorporate multispecies interactions at large spatial extents using interaction matrices. Location Local to global. Methods We review recent approaches for extending classical SDMs to incorporate biotic interactions, and identify some methodological and conceptual limitations. To illustrate possible directions for conceptual advancement we explore three principal ways of modelling multispecies interactions using interaction matrices: simple qualitative linkages between species, quantitative interaction coefficients reflecting interaction strengths, and interactions mediated by interaction currencies. We explain methodological advancements for static interaction data and multispecies time series, and outline methods to reduce complexity when modelling multispecies interactions. Results Classical SDMs ignore biotic interactions and recent SDM extensions only include the unidirectional influence of one or a few species. However, novel methods using error matrices in multivariate regression models allow interactions between multiple species to be modelled explicitly with spatial co‐occurrence data. If time series are available, multivariate versions of population dynamic models can be applied that account for the effects and relative importance of species interactions and environmental drivers. These methods need to be extended by incorporating the non‐stationarity in interaction coefficients across space and time, and are challenged by the limited empirical knowledge on spatio‐temporal variation in the existence and strength of species interactions. Model complexity may be reduced by: (1) using prior ecological knowledge to set a subset of interaction coefficients to zero, (2) modelling guilds and functional groups rather than individual species, and (3) modelling interaction currencies and species’ effect and response traits. Main conclusions There is great potential for developing novel approaches that incorporate multispecies interactions into the projection of species distributions and community structure at large spatial extents. Progress can be made by: (1) developing statistical models with interaction matrices for multispecies co‐occurrence datasets across large‐scale environmental gradients, (2) testing the potential and limitations of methods for complexity reduction, and (3) sampling and monitoring comprehensive spatio‐temporal data on biotic interactions in multispecies communities.     

物种分布模型理论研究进展

利用物种分布模型估计物种的真实和潜在分布区,已成为区域生态学与生物地理学中非常活跃的研究领域。然而,到目前为止,这项技术的理论基础仍然存在不足之处,一些关键的生态过程未能被有效纳入到物种分布模型的理论框架中,从而为解释物种分布模型预测的结果带来了诸多困惑。鉴于此,总结了物种分布模型的理论基础;系统探讨了物种分布模型与物种分布区的关系;特别指出了物种分布模型研究中存在的理论问题;重点阐述了物种分布模型未来的发展方向。研究认为,物种分布模型与生态位理论、源-库理论、种群动态理论、集合种群理论、进化理论等具有重要的联系;正确理解物种分布模型的预测结果与物种分布区的关系,有赖于对影响物种分布的3个主要因素(环境条件、物种相互作用与物种迁移能力)做出定量的分离;目前物种分布模型主要存在的问题是未能将物种的相互作用和物种的迁移能力有效纳入到模型的构建过程中;未来物种分布模型的发展应该加强模型背后理论框架的研究,并进一步加强整合物种相互作用过程、种群动态过程、迁移过程和物种进化过程等内容。研究还认为,从更高的理论层次模拟功能群和群落结构将是未来物种分布模型的重要发展方向。     

Assessing the accuracy of species distribution models to predict amphibian species richness patterns

1. Evaluating the distribution of species richness where biodiversity is high but has been insufficiently sampled is not an easy task. Species distribution modelling has become a useful approach for predicting their ranges, based on the relationships between species records and environmental variables. Overlapping predictions of individual distributions could be a useful strategy for obtaining estimates of species richness and composition in a region, but these estimates should be evaluated using a proper validation process, which compares the predicted richness values and composition with accurate data from independent sources. 2. In this study, we propose a simple approach to estimate model performance for several distributional predictions generated simultaneously. This approach is particularly suitable when species distribution modelling techniques that require only presence data are used. 3. The individual distributions for the 370 known amphibian species of Mexico were predicted using maxent to model data on their known presence (66,113 presence-only records). Distributions were subsequently overlapped to obtain a prediction of species richness. Accuracy was assessed by comparing the overall species richness values predicted for the region with observed and predicted values from 118 well-surveyed sites, each with an area of c. 100 km(2), which were identified using species accumulation curves and nonparametric estimators. 4. The derived models revealed a remarkable heterogeneity of species richness across the country, provided information about species composition per site and allowed us to obtain a measure of the spatial distribution of prediction errors. Examining the magnitude and location of model inaccuracies, as well as separately assessing errors of both commission and omission, highlights the inaccuracy of the predictions of species distribution models and the need to provide measures of uncertainty along with the model results. 5. The combination of a species distribution modelling method like maxent and species richness estimators offers a useful tool for identifying when the overall pattern provided by all model predictions might be representing the geographical patterns of species richness and composition, regardless of the particular quality or accuracy of the predictions for each individual species.     

LIVES: a new habitat modelling technique for predicting the distribution of species’ occurrences using presence-only data based on limiting factor theory

Predictive modelling techniques using presence-only data have attracted increasing attention because they can provide information on species distributions and their potential habitat for conservation and ecosystem management. However, the existing predictive modelling techniques have several limitations. Here, we propose a novel predictive modelling technique, Limiting Variable and Environmental Suitability (LIVES), for predicting the distributions and potential habitats of species using presence-only data. It is based on limiting factor theory, which postulates that the occurrence of a species is only determined by the factor that most limits its distribution. LIVES predicts the suitability of a candidate grid cell for a species in terms of limiting environmental factor. It also predicts the most limiting factor or the potential limiting factor at the grid cell. The environmental factors can be climatic, geological, biological and any other relevant environmental factors, whether quantitative or qualitative. The predicted habitats consist of the current distribution of the species and the potentially suitable areas for the species where there is currently no record of occurrence. We also compare several properties of LIVES and other predictive modelling techniques. On the basis of 1,000 simulations, the average predictions of LIVES are more accurate than the two other commonly used modelling techniques (BIOCLIM and DOMAIN) for presence-only data.     

A fine‐grained spatial prediction model for the red‐listed vascular plant Scorzonera humilis

We explored the applied use of distribution modelling as a tool for making spatial predictions of occurrences of the red‐listed vascular plant species Scorzonera humilis in a study area in southeast Norway. Scorzonera is typical of extensively managed semi‐natural grasslands. A Maxent model was trained on all known records of the species, accurately georeferenced and gridded to fine resolution (grid cells of 25×25 m). Model performance was assessed on the training data by data‐splitting (by which some records were set off for evaluation) and on independent evaluation data collected in the field. Of the eight predictor variables used in the modelling, distance to roads and to arable land were most important followed by land‐cover class and altitude. Judged from the area under curve (AUC), the model was good to excellent and a significant, positive relationship was found between relative probabilities of occurrence predicted by the model and true probability of presence provided by the independently collected evaluation data. The model was used together with the evaluation data to estimate presence of Scorzonera humilis in 0.7% of the grid cells in the study area. The grid cells in which the model predicted highest probability for Scorzonera to be present had a true probability of presence of ca 12%, i.e. 17×higher than in an average cell. The present study demonstrates that, even when only simple predictor variables are available, spatial prediction modelling contributes important knowledge about rare species such as prevalence estimates, spatial prediction maps and insights into the species’ autecology. Spatial prediction modelling also makes cost‐efficient monitoring of rare species possible. However, it is pointed out that these benefits require evaluation of the model on independently sampled evaluation data.     

Species interactions weakly modify climate‐induced tree co‐occurrence patterns

Aims

Species distributions are hypothesized to be underlain by a complex association of processes that span multiple spatial scales including biotic interactions, dispersal limitation, fine‐scale resource gradients and climate. Species disequilibrium with climate may reflect the effects of non‐climatic processes on species distributions, yet distribution models have rarely directly considered non‐climatic processes. Here, we use a Joint Species Distribution Model (JSDM) to investigate the influence of non‐climatic factors on species co‐occurrence patterns and to directly quantify the relative influences of climate and alternative processes that may generate correlated responses in species distributions, such as species interactions, on tree co‐occurrence patterns.

Location

US Rocky Mountains.

Methods

We apply a Bayesian JSDM to simultaneously model the co‐occurrence patterns of ten dominant tree species across the Rocky Mountains, and evaluate climatic and residual correlations from the fitted model to determine the relative contribution of each component to observed co‐occurrence patterns. We also evaluate predictions generated from the fitted model relative to a single‐species modelling approach.

Results

For most species, correlation due to climate covariates exceeded residual correlation, indicating an overriding influence of broad‐scale climate on co‐occurrence patterns. Accounting for covariance among species did not significantly improve predictions relative to a single‐species approach, providing limited evidence for a strong independent influence of species interactions on distribution patterns.

Conclusions

Overall, our findings indicate that climate is an important driver of regional biodiversity patterns and that interactions between dominant tree species contribute little to explain species co‐occurrence patterns among Rocky Mountain trees.     

Intra‐specific variability and plasticity influence potential tree species distributions under climate change

Aim To assess the effect of local adaptation and phenotypic plasticity on the potential distribution of species under future climate changes. Trees may be adapted to specific climatic conditions; however, species range predictions have classically been assessed by species distribution models (SDMs) that do not account for intra‐specific genetic variability and phenotypic plasticity, because SDMs rely on the assumption that species respond homogeneously to climate change across their range, i.e. a species is equally adapted throughout its range, and all species are equally plastic. These assumptions could cause SDMs to exaggerate or underestimate species at risk under future climate change. Location The Iberian Peninsula. Methods Species distributions are predicted by integrating experimental data and modelling techniques. We incorporate plasticity and local adaptation into a SDM by calibrating models of tree survivorship with adaptive traits in provenance trials. Phenotypic plasticity was incorporated by calibrating our model with a climatic index that provides a measure of the differences between sites and provenances. Results We present a new modelling approach that is easy to implement and makes use of existing tree provenance trials to predict species distribution models under global warming. Our results indicate that the incorporation of intra‐population genetic diversity and phenotypic plasticity in SDMs significantly altered their outcome. In comparing species range predictions, the decrease in area occupancy under global warming conditions is smaller when considering our survival–adaptation model than that predicted by a ‘classical SDM’ calibrated with presence–absence data. These differences in survivorship are due to both local adaptation and plasticity. Differences due to the use of experimental data in the model calibration are also expressed in our results: we incorporate a null model that uses survival data from all provenances together. This model always predicts less reduction in area occupancy for both species than the SDM calibrated with presence–absence. Main conclusions We reaffirm the importance of considering adaptive traits when predicting species distributions and avoiding the use of occurrence data as a predictive variable. In light of these recommendations, we advise that existing predictions of future species distributions and their component populations must be reconsidered.     

Modelling the occurrence of threatened plant species in taiga landscapes: methodological and ecological perspectives

Aim Understanding the spatial patterns of species distribution and predicting the occurrence of high biological diversity and rare species are central themes in biogeography and environmental conservation. The aim of this study was to model and scrutinize the relative contributions of climate, topography, geology and land‐cover factors to the distributions of threatened vascular plant species in taiga landscapes in northern Finland. Location North‐east Finland, northern Europe. Methods The study was performed using a data set of 28 plant species and environmental variables at a 25‐ha resolution. Four different stepwise selection algorithms [Akaike information criterion (AIC), Bayesian information criterion (BIC), adaptive backfitting, cross selection] with generalized additive models (GAMs) were fitted to identify the main environmental correlates for species occurrences. The accuracies of the distribution models were evaluated using fourfold cross‐validation based on the area under the curve (AUC) derived from receiver operating characteristic plots. The GAMs were tentatively extrapolated to the whole study area and species occurrence probability maps were produced using GIS techniques. The effect of spatial autocorrelation on the modelling results was also tested by including autocovariate terms in the GAMs. Results According to the AUC values, the model performance varied from fair to excellent. The AIC algorithm provided the highest mean performance (mean AUC = 0.889), whereas the lowest mean AUC (0.851) was obtained from BIC. Most of the variation in the distribution of threatened plant species was related to growing degree days, temperature of the coldest month, water balance, cover of mire and mean elevation. In general, climate was the most powerful explanatory variable group, followed by land cover, topography and geology. Inclusion of the autocovariate only slightly improved the performance of the models and had a minor effect on the importance of the environmental variables. Main conclusions The results confirm that the landscape‐scale distribution patterns of plant species can be modelled well on the basis of environmental parameters. A spatial grid system with several environmental variables derived from remote sensing and GIS data was found to produce useful data sets, which can be employed when predicting species distribution patterns over extensive areas. Landscape‐scale maps showing the predicted occurrences of individual or multiple threatened plant species may provide a useful basis for focusing field surveys and allocating conservation efforts.     

Insights into the area under the receiver operating characteristic curve (AUC) as a discrimination measure in species distribution modelling

Aim The area under the receiver operating characteristic (ROC) curve (AUC) is a widely used statistic for assessing the discriminatory capacity of species distribution models. Here, I used simulated data to examine the interdependence of the AUC and classical discrimination measures (sensitivity and specificity) derived for the application of a threshold. I shall further exemplify with simulated data the implications of using the AUC to evaluate potential versus realized distribution models. Innovation After applying the threshold that makes sensitivity and specificity equal, a strong relationship between the AUC and these two measures was found. This result is corroborated with real data. On the other hand, the AUC penalizes the models that estimate potential distributions (the regions where the species could survive and reproduce due to the existence of suitable environmental conditions), and favours those that estimate realized distributions (the regions where the species actually lives). Main conclusions Firstly, the independence of the AUC from the threshold selection may be irrelevant in practice. This result also emphasizes the fact that the AUC assumes nothing about the relative costs of errors of omission and commission. However, in most real situations this premise may not be optimal. Measures derived from a contingency table for different cost ratio scenarios, together with the ROC curve, may be more informative than reporting just a single AUC value. Secondly, the AUC is only truly informative when there are true instances of absence available and the objective is the estimation of the realized distribution. When the potential distribution is the goal of the research, the AUC is not an appropriate performance measure because the weight of commission errors is much lower than that of omission errors.     

物种分布模型的发展及评价方法

物种分布模型已被广泛地应用于以保护区规划、气候变化对物种分布的影响等为目的的研究。回顾了已经得到广泛应用的多种物种分布模型,总结了评价模型性能的方法。基于物种分布模型的发展和应用以及性能评价中尚存在的问题,本文认为:在物种分布模型中集成样本选择模块能够避免模型预测过程中的过度拟合及欠拟合,增加变量选择模块可评估和降低变量之间自相关性的影响,增加生物因子以及将物种对环境的适应性机制(及扩散行为特征)和潜在分布模型进行结合,是提高模型预测性能的可行方案;在模型性能的评价方面,采用赤池信息量可对模型的预测性能进行客观评价。相关建议可为物种分布建模提供参考。