The performance of state-of-the-art modelling techniques depends on geographical distribution of species

We explored the effects of prevalence, latitudinal range and clumping (spatial autocorrelation) of species distribution patterns on the predictive accuracy of eight state-of-the-art modelling techniques: Generalized Linear Models (GLMs), Generalized Boosting Method (GBM), Generalized Additive Models (GAMs), Classification Tree Analysis (CTA), Artificial Neural Network (ANN), Multivariate Adaptive Regression Splines (MARS), Mixture Discriminant Analysis (MDA) and Random Forest (RF). One hundred species of Lepidoptera, selected from the Distribution Atlas of European Butterflies, and three climate variables were used to determine the bioclimatic envelope for each butterfly species. The data set consisting of 2620 grid squares 30′ × 60′ in size all over Europe was randomly split into the calibration and the evaluation data sets. The performance of different models was assessed using the area under the curve (AUC) of a receiver operating characteristic (ROC) plot. Observed differences in modelling accuracy among species were then related to the geographical attributes of the species using GAM. The modelling performance was negatively related to the latitudinal range and prevalence, whereas the effect of spatial autocorrelation on prediction accuracy depended on the modelling technique. These three geographical attributes accounted for 19–61% of the variation in the modelling accuracy. Predictive accuracy of GAM, GLM and MDA was highly influenced by the three geographical attributes, whereas RF, ANN and GBM were moderately, and MARS and CTA only slightly affected. The contrasting effects of geographical distribution of species on predictive performance of different modelling techniques represent one source of uncertainty in species spatial distribution models. This should be taken into account in biogeographical modelling studies and assessments of climate change impacts.     

The evaluation strip: A new and robust method for plotting predicted responses from species distribution models

Increasing use is being made in conservation management of statistical models that couple extensive collections of species and environmental data to make predictions of the geographic distributions of species. While the relationships fitted between a species and its environment are relatively transparent for many of these modeling techniques, others are more ‘black box’ in character, only producing geographic predictions and providing minimal or untraditional summaries of the fitted relationships on which these predictions are based. This in turn prevents robust evaluation of the ecological sensibility of such models, a necessary process if model predictions are to be treated with confidence. Here we propose a new but simple method for visualizing modeled responses that can be implemented with any modeling method, and demonstrate its application using five common methods applied to the prediction of an Australian tree species. This is achieved by insetting an “evaluation strip” into the spatial data layers, which, after predictions have been made, can be clipped out and used for creating plots of the modelled responses. We present findings of the application strip for algorithms GLMs, GAMs, CLIM, DOMAIN and MARS. Evaluation strips can be constructed to investigate either uni-variate responses, or the simultaneous variation in predicted values in relation to two variables. The latter option is particularly useful for evaluating responses in models that allow the fitting of complex interaction terms.     

Spatial residual analysis of six modeling techniques

In recent years alternative modeling techniques have been used to account for spatial autocorrelations among data observations. They include linear mixed model (LMM), generalized additive model (GAM), multi-layer perceptron (MLP) neural network, radial basis function (RBF) neural network, and geographically weighted regression (GWR). Previous studies show these models are robust to the violation of model assumptions and flexible to nonlinear relationships among variables. However, many of them are non-spatial in nature. In this study, we utilize a local spatial analysis method (i.e., local Moran coefficient) to investigate spatial distribution and heterogeneity in model residuals from those modeling techniques with ordinary least-squares (OLS) as the benchmark. The regression model used in this study has tree crown area as the response variable, and tree diameter and the coordinates of tree locations as the predictor variables. The results indicate that LMM, GAM, MLP and RBF may improve model fitting to the data and provide better predictions for the response variable, but they generate spatial patterns for model residuals similar to OLS. The OLS, LMM, GAM, MLP and RBF models yield more residual clusters of similar values, indicating that trees in some sub-areas are either all underestimated or all overestimated for the response variable. In contrast, GWR estimates model coefficients at each location in the study area, and produces more accurate predictions for the response variable. Furthermore, the residuals of the GWR model have more desirable spatial distributions than the ones derived from the OLS, LMM, GAM, MLP and RBF models.     

Comparison of six correlative models in predictive vegetation mapping on a local scale

A variety of statistical techniques has been used in predictive vegetation modelling (PVM) that attempt to predict occurrence of a given community or species in respect to environmental conditions. We compared the performance of three profile models, BIOCLIM, GARP and MAXENT with three nonparametric models of group discrimination techniques, MARS, NPMR and LRT. The two latter models are relatively new statistical techniques that have just entered the field of PVM. We ran all models on a local scale for a given grassland community (Teucrio-Seslerietum) using the same input data to examine their performance. Model accuracy was evaluated both by Cohen’s kappa statistics (κ) and by area under receiver operating characteristics curve based both on resubstitution of training data and on an independent test data set. MAXENT of profile models and MARS of group discrimination techniques achieved the best prediction.     

The effect of species response form on species distribution model prediction and inference

Ecological theory and current evidence support the validity of various species response curves according to a variety of environmental gradients. Various methods have been developed for building species distribution models but it is not well known how these methods perform under various assumptions about the form of the underlying species response. It is also not well known how spatial correlation in species occurrence affects model performance. These effects were investigated by applying an environmental envelope method (BIOCLIM) and three regression-based methods: logistic regression (LR), generalized additive modelling (GAM), and classification and regression tree (CART) to simulated species occurrence data. Each simulated species was constructed as a sum of responses with varying weights. Three basic species response curves were assumed: Gaussian (bell-shaped), Beta (skew) and linear. The two non-linear responses conform to standard ecological niche theory. All three responses were applied in turn to three simulated environmental variables, each with varying degrees of spatial autocorrelation. GAM produced the most consistent model performance over all forms of simulated species response. BIOCLIM and CART were inclined to underrate the performance of variables with a linear response. BIOCLIM was less sensitive to data density. LR was susceptible to model misspecification. The use of a linear function in LR underestimated the performance of variables with non-linear species response and contributed to increased spatial autocorrelation in model residuals. Omission of important environmental variables with non-linear species response also contributed to increased spatial autocorrelation in model residuals. Adding a spatial autocovariate term to the LR model (autologistic model) reduced the spatial autocorrelation and improved model performance, but did not correct the misidentification of the dominant environmental determinant. This is to be expected since the autologistic approach was designed primarily for prediction and not for inference. Given that various forms of species response to environmental determinants arise commonly in nature: (1) higher order functions should always be tested when applying LR in modelling species distribution; (2) spatial autocorrelation in species distribution model residuals can indicate that environmental determinants with non-linear response are missing from the model; and (3) deficiencies in LR model performance due to model misspecification can be addressed by adding a spatial autocovariate to the model, but care should be taken when interpreting the coefficients of the model parameters.     

Species distribution models and ecological theory: A critical assessment and some possible new approaches

Given the importance of knowledge of species distribution for conservation and climate change management, continuous and progressive evaluation of the statistical models predicting species distributions is necessary. Current models are evaluated in terms of ecological theory used, the data model accepted and the statistical methods applied. Focus is restricted to Generalised Linear Models (GLM) and Generalised Additive Models (GAM). Certain currently unused regression methods are reviewed for their possible application to species modelling.A review of recent papers suggests that ecological theory is rarely explicitly considered. Current theory and results support species responses to environmental variables to be unimodal and often skewed though process-based theory is often lacking. Many studies fail to test for unimodal or skewed responses and straight-line relationships are often fitted without justification.Data resolution (size of sampling unit) determines the nature of the environmental niche models that can be fitted. A synthesis of differing ecophysiological ideas and the use of biophysical processes models could improve the selection of predictor variables. A better conceptual framework is needed for selecting variables.Comparison of statistical methods is difficult. Predictive success is insufficient and a test of ecological realism is also needed. Evaluation of methods needs artificial data, as there is no knowledge about the true relationships between variables for field data. However, use of artificial data is limited by lack of comprehensive theory.Three potentially new methods are reviewed. Quantile regression (QR) has potential and a strong theoretical justification in Liebig's law of the minimum. Structural equation modelling (SEM) has an appealing conceptual framework for testing causality but has problems with curvilinear relationships. Geographically weighted regression (GWR) intended to examine spatial non-stationarity of ecological processes requires further evaluation before being used.Synthesis and applications: explicit theory needs to be incorporated into species response models used in conservation. For example, testing for unimodal skewed responses should be a routine procedure. Clear statements of the ecological theory used, the nature of the data model and sufficient details of the statistical method are needed for current models to be evaluated. New statistical methods need to be evaluated for compatibility with ecological theory before use in applied ecology. Some recent work with artificial data suggests the combination of ecological knowledge and statistical skill is more important than the precise statistical method used. The potential exists for a synthesis of current species modelling approaches based on their differing ecological insights not their methodology.     

Predicting tree species presence and basal area in Utah: A comparison of stochastic gradient boosting,generalized additive models,and tree-based methods

Many efforts are underway to produce broad-scale forest attribute maps by modelling forest class and structure variables collected in forest inventories as functions of satellite-based and biophysical information. Typically, variants of classification and regression trees implemented in Rulequest's^© See5 and Cubist (for binary and continuous responses, respectively) are the tools of choice in many of these applications. These tools are widely used in large remote sensing applications, but are not easily interpretable, do not have ties with survey estimation methods, and use proprietary unpublished algorithms. Consequently, three alternative modelling techniques were compared for mapping presence and basal area of 13 species located in the mountain ranges of Utah, USA. The modelling techniques compared included the widely used See5/Cubist, generalized additive models (GAMs), and stochastic gradient boosting (SGB). Model performance was evaluated using independent test data sets. Evaluation criteria for mapping species presence included specificity, sensitivity, Kappa, and area under the curve (AUC). Evaluation criteria for the continuous basal area variables included correlation and relative mean squared error. For predicting species presence (setting thresholds to maximize Kappa), SGB had higher values for the majority of the species for specificity and Kappa, while GAMs had higher values for the majority of the species for sensitivity. In evaluating resultant AUC values, GAM and/or SGB models had significantly better results than the See5 models where significant differences could be detected between models. For nine out of 13 species, basal area prediction results for all modelling techniques were poor (correlations less than 0.5 and relative mean squared errors greater than 0.8), but SGB provided the most stable predictions in these instances. SGB and Cubist performed equally well for modelling basal area for three species with moderate prediction success, while all three modelling tools produced comparably good predictions (correlation of 0.68 and relative mean squared error of 0.56) for one species.     

Comparison and ranking of different modelling techniques for prediction of site index in Mediterranean mountain forests

Forestry science has a long tradition of studying the relationship between stand productivity and abiotic and biotic site characteristics, such as climate, topography, soil and vegetation. Many of the early site quality modelling studies related site index to environmental variables using basic statistical methods such as linear regression. Because most ecological variables show a typical non-linear course and a non-constant variance distribution, a large fraction of the variation remained unexplained by these linear models. More recently, the development of more advanced non-parametric and machine learning methods provided opportunities to overcome these limitations. Nevertheless, these methods also have drawbacks. Due to their increasing complexity they are not only more difficult to implement and interpret, but also more vulnerable to overfitting. Especially in a context of regionalisation, this may prove to be problematic. Although many non-parametric and machine learning methods are increasingly used in applications related to forest site quality assessment, their predictive performance has only been assessed for a limited number of methods and ecosystems.In this study, five different modelling techniques are compared and evaluated, i.e. multiple linear regression (MLR), classification and regression trees (CART), boosted regression trees (BRT), generalized additive models (GAM), and artificial neural networks (ANN). Each method is used to model site index of homogeneous stands of three important tree species of the Taurus Mountains (Turkey): Pinus brutia, Pinus nigra and Cedrus libani. Site index is related to soil, vegetation and topographical variables, which are available for 167 sample plots covering all important environmental gradients in the research area. The five techniques are compared in a multi-criteria decision analysis in which different model performance measures, ecological interpretability and user-friendliness are considered as criteria.When combining these criteria, in most cases GAM is found to outperform all other techniques for modelling site index for the three species. BRT is a good alternative in case the ecological interpretability of the technique is of higher importance. When user-friendliness is more important MLR and CART are the preferred alternatives. Despite its good predictive performance, ANN is penalized for its complex, non-transparent models and big training effort.     

Modelling species distributions with penalised logistic regressions: A comparison with maximum entropy models

An important aspect of species distribution modelling is the choice of the modelling method because a suboptimal method may have poor predictive performance. Previous comparisons have found that novel methods, such as Maxent models, outperform well-established modelling methods, such as the standard logistic regression. These comparisons used training samples with small numbers of occurrences per estimated model parameter, and this limited sample size may have caused poorer predictive performance due to overfitting. Our hypothesis is that Maxent models would outperform a standard logistic regression because Maxent models avoid overfitting by using regularisation techniques and a standard logistic regression does not. Regularisation can be applied to logistic regression models using penalised maximum likelihood estimation. This estimation procedure shrinks the regression coefficients towards zero, causing biased predictions if applied to the training sample but improving the accuracy of new predictions. We used Maxent and logistic regression (standard and penalised) to analyse presence/pseudo-absence data for 13 tree species and evaluated the predictive performance (discrimination) using presence-absence data. The penalised logistic regression outperformed standard logistic regression and equalled the performance of Maxent. The penalised logistic regression may be considered one of the best methods to develop species distribution models trained with presence/pseudo-absence data, as it is comparable to Maxent. Our results encourage further use of the penalised logistic regression for species distribution modelling, especially in those cases in which a complex model must be fitted to a sample with a limited size.     

Using and comparing two nonparametric methods (CART and MARS) to model the potential distribution of gullies

Gully erosion represents an important soil degradation process in rangelands. In order to take preventive or control measures and to reduce its environmental damages and economical costs it is useful to localize the points in the landscape where gullying takes place and to determine the importance of the different factors involved. The study is carried out in Extremadura, southwest Spain. The main objectives of this work are: (a) comparing two nonparametric schemes to model the potential distribution of gullies, (b) evaluating the importance of the different factors involved in gullying processes, (c) analyzing the role of prevalence in the success of the model and finally, (d) implementing and mapping the results with the help of a Geographical Information System (GIS). Two methods were used to model the response of a dependent variable (gullying) from a set of independent variables: Classification And Regression Trees (CART) and Multivariate Adaptive Regression Splines (MARS). Three different datasets were used; the first one for constructing the model (training dataset) and the others for validating the model (external datasets). These datasets are formed by a target variable (presence or absence of gullies) and a set of independent variables. The dependent variable was obtained by mapping the locations of gullies with the help of a GPS and high resolution aerial ortophotographs. A set of 32 independent variables reflecting topography, lithology, soil type, climate, land use and vegetation cover of each area were used. The performance of the models was evaluated using a non-dependent threshold method: the Receiver Operating Characteristic (ROC) curve. The results showed a better performance of MARS for predicting gullying with areas under the ROC curve of 0.98 and 0.97 for the validation datasets, while CART presented values of 0.96 and 0.66.     

New approaches to modelling fish-habitat relationships

Ecologists often develop models that describe the relationship between faunal communities and their habitat. Coral reef fishes have been the focus of numerous such studies, which have used a wide range of statistical tools to answer an equally wide range of questions. Here, we apply a series of both conventional statistical techniques (linear and generalized additive regression models) and novel machine-learning techniques (the support vector machine and three ensemble techniques used with regression trees) to predict fish species richness, biomass, and diversity from a range of habitat variables. We compare the techniques in terms of their predictive performance, and we compare a subset of the models in terms of the influence each habitat variable has for the predictions. Prediction errors are estimated by cross-validation, and variable importance is assessed using permutations of individual variable values. For predictions of species richness and diversity the tree-based models generally and the random forest model specifically are superior (produce the lowest errors). These model types are all able to model both nonlinear and interaction effects. The linear model, unable to model either effect type, performs the worst (produces the highest errors). For predictions of biomass, the generalized additive model is superior, and the support vector machine performs the worst. Depth range, the difference between maximum and minimum water depth at a given site, is identified as the most important variable in the majority of models predicting the three fish community variables. However, variable importance is highly dependent upon model type, which leads to questions regarding the interpretation of variable importance and its proper use as an indicator of causality. The representation of ecological relationships by tree-based ensemble learners will improve predictive performance, and provide a new avenue for exploring ecological relationships, both statistical and causal.     

Predicting species distributions from samples collected along roadsides

Predictive models of species distributions are typically developed with data collected along roads. Roadside sampling may provide a biased (nonrandom) sample; however, it is currently unknown whether roadside sampling limits the accuracy of predictions generated by species distribution models. We tested whether roadside sampling affects the accuracy of predictions generated by species distribution models by using a prospective sampling strategy designed specifically to address this issue. We built models from roadside data and validated model predictions at paired locations on unpaved roads and 200 m away from roads (off road), spatially and temporally independent from the data used for model building. We predicted species distributions of 15 bird species on the basis of point-count data from a landbird monitoring program in Montana and Idaho (U.S.A.). We used hierarchical occupancy models to account for imperfect detection. We expected predictions of species distributions derived from roadside-sampling data would be less accurate when validated with data from off-road sampling than when it was validated with data from roadside sampling and that model accuracy would be differentially affected by whether species were generalists, associated with edges, or associated with interior forest. Model performance measures (kappa, area under the curve of a receiver operating characteristic plot, and true skill statistic) did not differ between model predictions of roadside and off-road distributions of species. Furthermore, performance measures did not differ among edge, generalist, and interior species, despite a difference in vegetation structure along roadsides and off road and that 2 of the 15 species were more likely to occur along roadsides. If the range of environmental gradients is surveyed in roadside-sampling efforts, our results suggest that surveys along unpaved roads can be a valuable, unbiased source of information for species distribution models.     

Modeling and Predicting Species Occurrence Using Broad-Scale Environmental Variables: an Example with Butterflies of the Great Basin

Abstract: If occurrence of individual species can be modeled as a function of easily quantified environmental variables (e.g., derived from a geographic information system [GIS]) and the predictions of these models are demonstrably successful, then the scientific foundation for management planning will be strengthened. We used Bayesian logistic regression to develop predictive models for resident butterflies in the central Great Basin of western North America. Species inventory data and values for 14 environmental variables from 49 locations (segments of canyons) in the Toquima Range ( Nevada, U.S.A.) were used to build the models. Squares of the environmental variables were also used to accommodate possibly nonmonotonic responses. We obtained statistically significant models for 36 of 56 (64%) resident species of butterflies. The models explained 8–72% of the deviance in occurrence of those species. Each of the independent variables was significant in at least one model, and squared versions of five variables contributed to models. Elevation was included in more than half of the models. Models included one to four variables; only one variable was significant in about half the models. We conducted preliminary tests of two of our models by using an existing set of data on the occurrence of butterflies in the neighboring Toiyabe Range. We compared conventional logistic classification with posterior probability distributions derived from Bayesian modeling. For the latter, we restricted our predictions to locations with a high ( 70%) probability of predicted presence or absence. We will perform further tests after conducting inventories at new locations in the Toquima Range and nearby Shoshone Mountains, for which we have computed environmental variables by using remotely acquired topographic data, digital-terrain and microclimatic models, and GIS computation.     

An evaluation of environmental factors affecting species distributions

Many different models can be built to explain the distributions of species. Often there is no single model that is clearly better than the alternatives, and this leads to uncertainty over which environmental factors are limiting species’ distributions. We investigated the support for different environmental factors by determining the drop in model performance when selected predictors were excluded from the model building process. We used a paired t-test over 37 plant species so that an environmental factor was only deemed significant if it consistently improved the results for multiple species. Geology and winter minimum temperatures were found to be the environmental factors with the most support, with a significant drop in model performance when either of these factors was excluded. However, there was less support for summer maximum temperature, as other environmental factors could combine to produce similar model performance. Our method of evaluating environmental factors using multiple species will not be capable of detecting predictors that are only important for one or two species, but it is difficult to distinguish these from spurious correlations. The strength of the method is that it increases inference for factors that consistently affect the distributions of many species. We discourage the assessment of models against predefined benchmarks, such as an area under the curve (AUC) of more than 0.7, as many alternative models for the same species produce similar results. Therefore, the benchmarks do not provide any indication of how the performance of the selected model compares to alternative models, and they provide weak inference to accept any selected model.     

Species distribution modelling: Does one size fit all? A phytogeographic analysis of Salix in Ontario

Empirical models for predicting the distribution of organisms from environmental data have often focused on principles of ecological niche theory. However, even at large scales, there is little agreement over how to represent the dimensions of a species’ niche. The performance of such models is greatly affected by the nature of species distributional and environmental data. Regional scale distribution models were developed for 30 willow species in Ontario to examine (i) the predictive ability of logistic regression analysis, and (ii) the effects of using different distributional and environmental data sets. Two original measures of model accuracy and over-prediction were employed and evaluated using independent data. Models based on unique combinations of monthly climate data predicted distributions most accurately for all species. Models based on a fixed set of variables, while generating the highest average probabilities of occurrence for certain species with limited ranges, resulted in the greatest under- and over-estimates of willow distributions. Comparisons of models demonstrated climatic patterns among willows of differing habit and habitat. The distribution of dwarf willow species, present only in the Ontario arctic, followed gradients of summer maximum temperatures. The distribution of the tree species in the southerly portions of the province followed gradients of fall and winter minimum temperatures. Regardless of distributional and environmental data input, no algorithm maximized model performance for all species. Individual species models require individual approaches; i.e., the variable selection technique, the set of environmental factors used as predictors, and the nature of species distributional data must be carefully matched to the intended application. An understanding of evolutionary processes enhances the meaningful interpretation of individual species models. Unless sampling bias and species prevalence can be accounted for, models based on collection point data are best used to guide field surveys. While inferred range data may be better suited to determine potential ecological niches, overestimation of species prevalence and environmental tolerance must be recognized. A combination of available distributional data types is recommended to best determine species niches, an important step in developing conservation strategies.     

Diameter growth models for mixed-species stands of Coastal British Columbia including thinning and fertilization effects

In this study, diameter growth models for three species growing in mixed-stands of Coastal British Columbia (BC), Canada, under a variety of silvicultural treatments were developed. The three species were: Douglas-fir (Pseudotsuga menziesii var. menziesii (Mirb.) Franco), western hemlock (Tsuga heterophylla (Raf.) Sarg.), and western redcedar (Thuja plicata Donn). A Box and Lucas model (1959) was initially fitted to the diameter growth series for each tree, as this model is very flexible and was based on processes reflective of the metabolic processes governing tree growth. Next, a random coefficients modelling approach (i.e., parameter prediction approach) was used to modify the estimated parameters for each species using functions of tree size and stage of development, site productivity, and inter-tree competition variables, while accounting for temporal correlation within trees. Impacts of fertilization on diameter growth were estimated by including the time since fertilization as an additional variable. Since state variables that are changed as a result of thinning were already in the model, accurate results post-thinning were obtained with no changes to the model. For the combined effects of thinning and fertilization, a two-step additive approach was used, where the state variables were changed following thinning and the diameter increment was modified for fertilization using the time since fertilization variable. Results indicated that multiple treatments sustain a change in growth for a longer time period following treatment than thinning or fertilization alone.     

A heuristic approach to predicting water beetle diversity in temporary and fluctuating waters

An understanding of the causal mechanisms and processes that shape macroinvertebrate communities at a local scale has important implications for the management and conservation of freshwater biodiversity. Here we compare the performance of linear and non-linear statistics to explore diversity-environment relationships using data from 76 temporary and fluctuating ponds in two regions of southern England. We focus on aquatic beetle assemblages, which have been shown to be excellent surrogates of wider freshwater macroinvertebrate diversity. Ponds in the region contained a rich coleopteran fauna, totaling 68 species, which provided an excellent model system with which to compare the performance of two non-linear procedures (artificial neural networks—ANNs and generalised additive models—GAMs) and one more traditional linear approach (Multiple linear regression—MLR) to modelling diversity-environment relationships. Of all approaches employed, the best fit was obtained using an ANN model with only four input variables (conductivity, turbidity, magnesium concentration and depth). This model accounted for 82% of the observed variability in Shannon diversity index across ponds. In contrast, the best GAM and MLR models only explained 50% and 14% of this variation, respectively. Contribution profile analysis of conductivity, turbidity, magnesium concentration and depth, obtained from the best fit ANN through a hierarchical cluster analysis, allowed the identification of direct and proxy effects in relation to the environmental variables measured in this study. In each case, distinct clusters of ponds were identified in contribution profile analysis, suggesting that ponds across the two regions fall into a number of discrete groups, whose beetle faunas respond in subtly yet significantly different ways to key environmental variables. Aquatic coleopteran diversity in ponds in the two regions appears to be driven at a local scale by changes in relatively few physicochemical gradients, which are related to diversity in a clearly non-linear manner.     

The importance of spatial autocorrelation,extent and resolution in predicting forest bird occurrence

Concerns about declines in forest biodiversity underscore the need for accurate estimates of the distribution and abundance of organisms at large scales and at resolutions that are fine enough to be appropriate for management. This paper addresses three major objectives: (i) to determine whether the resolution of typical air photo-derived forest inventory is sufficient for the accurate prediction of site occupancy by forest birds. We compared prediction success of habitat models using air photo variables to models with variables derived from finer resolution, ground-sampled vegetation plots. (ii) To test whether incorporating spatial autocorrelation into habitat models via autologistic regression increases prediction success. (iii) To determine whether landscape structure is an important factor in predicting bird distribution in forest-dominated landscapes. Models were tested locally (Greater Fundy Ecosystem [GFE]) using cross-validation, and regionally using an independent data set from an area located ca. 250 km to the northwest (Riley Brook [RB]). We found significant positive spatial autocorrelation in the residuals of at least one habitat model for 76% (16/21) of species examined. In these cases, the logistic regression assumption of spatially independent errors was violated. Logistic models that ignored spatial autocorrelation tended to overestimate habitat effects. Though overall prediction success was higher for autologistic models than logistic models in the GFE, the difference was only significantly improved for one species. Further, the inclusion of spatial covariates did little to improve model performance in the geographically discrete study area. For 62% (13/21) of species examined, landscape variables were significant predictors of forest bird occurrence even after statistically controlling for stand-level variability. However, broad spatial extents explained less variation than local factors. In the GFE, 76% (16/21) of air photo and 81% (17/21) of ground plot models were accurate enough to be of practical utility (AUC > 0.7). When applied to RB, both model types performed effectively for 55% (11/20) of the species examined. We did not detect an overall difference in prediction success between air photo and ground plot models in either study area. We conclude that air photo data are as effective as fine resolution vegetation data for predicting site occupancy for the majority of species in this study. These models will be of use to forest managers who are interested in mapping species distributions under various timber harvest scenarios, and to protected areas planners attempting to optimize reserve function.     

Invasive species distribution modeling (iSDM): Are absence data and dispersal constraints needed to predict actual distributions?

Species distribution models (SDMs) based on statistical relationships between occurrence data and underlying environmental conditions are increasingly used to predict spatial patterns of biological invasions and prioritize locations for early detection and control of invasion outbreaks. However, invasive species distribution models (iSDMs) face special challenges because (i) they typically violate SDM's assumption that the organism is in equilibrium with its environment, and (ii) species absence data are often unavailable or believed to be too difficult to interpret. This often leads researchers to generate pseudo-absences for model training or utilize presence-only methods, and to confuse the distinction between predictions of potential vs. actual distribution. We examined the hypothesis that true-absence data, when accompanied by dispersal constraints, improve prediction accuracy and ecological understanding of iSDMs that aim to predict the actual distribution of biological invasions. We evaluated the impact of presence-only, true-absence and pseudo-absence data on model accuracy using an extensive dataset on the distribution of the invasive forest pathogen Phytophthora ramorum in California. Two traditional presence/absence models (generalized linear model and classification trees) and two alternative presence-only models (ecological niche factor analysis and maximum entropy) were developed based on 890 field plots of pathogen occurrence and several climatic, topographic, host vegetation and dispersal variables. The effects of all three possible types of occurrence data on model performance were evaluated with receiver operating characteristic (ROC) and omission/commission error rates. Results show that prediction of actual distribution was less accurate when we ignored true-absences and dispersal constraints. Presence-only models and models without dispersal information tended to over-predict the actual range of invasions. Models based on pseudo-absence data exhibited similar accuracies as presence-only models but produced spatially less feasible predictions. We suggest that true-absence data are a critical ingredient not only for accurate calibration but also for ecologically meaningful assessment of iSDMs that focus on predictions of actual distributions.