Novel methods improve prediction of species' distributions from occurrence data
University of Melbourne1, Stony Brook University2, City University of New York3, Princeton University4, University of Lausanne5, University of California, Berkeley6, University of Alaska Fairbanks7, National Institute of Water and Atmospheric Research8, Commonwealth Scientific and Industrial Research Organisation9, University of São Paulo10, University of Missouri11, Consejo Nacional de Ciencia y Tecnología12, University of Kansas13, Landcare Research14, AT&T15, McGill University16, James Cook University17, Swiss Federal Institute for Forest, Snow and Landscape Research18
TL;DR: This work compared 16 modelling methods over 226 species from 6 regions of the world, creating the most comprehensive set of model comparisons to date and found that presence-only data were effective for modelling species' distributions for many species and regions.
Abstract: Prediction of species' distributions is central to diverse applications in ecology, evolution and conservation science. There is increasing electronic access to vast sets of occurrence records in museums and herbaria, yet little effective guidance on how best to use this information in the context of numerous approaches for modelling distributions. To meet this need, we compared 16 modelling methods over 226 species from 6 regions of the world, creating the most comprehensive set of model comparisons to date. We used presence-only data to fit models, and independent presence-absence data to evaluate the predictions. Along with well-established modelling methods such as generalised additive models and GARP and BIOCLIM, we explored methods that either have been developed recently or have rarely been applied to modelling species' distributions. These include machine-learning methods and community models, both of which have features that may make them particularly well suited to noisy or sparse information, as is typical of species' occurrence data. Presence-only data were effective for modelling species' distributions for many species and regions. The novel methods consistently outperformed more established methods. The results of our analysis are promising for the use of data from museums and herbaria, especially as methods suited to the noise inherent in such data improve.
Citations
More filters
TL;DR: These new risk maps and infection estimates provide novel insights into the global, regional and national public health burden imposed by dengue and will help to guide improvements in disease control strategies using vaccine, drug and vector control methods, and in their economic evaluation.
Abstract: Dengue is a systemic viral infection transmitted between humans by Aedes mosquitoes. For some patients, dengue is a life-threatening illness. There are currently no licensed vaccines or specific therapeutics, and substantial vector control efforts have not stopped its rapid emergence and global spread. The contemporary worldwide distribution of the risk of dengue virus infection and its public health burden are poorly known. Here we undertake an exhaustive assembly of known records of dengue occurrence worldwide, and use a formal modelling framework to map the global distribution of dengue risk. We then pair the resulting risk map with detailed longitudinal information from dengue cohort studies and population surfaces to infer the public health burden of dengue in 2010. We predict dengue to be ubiquitous throughout the tropics, with local spatial variations in risk influenced strongly by rainfall, temperature and the degree of urbanization. Using cartographic approaches, we estimate there to be 390 million (95% credible interval 284-528) dengue infections per year, of which 96 million (67-136) manifest apparently (any level of disease severity). This infection total is more than three times the dengue burden estimate of the World Health Organization. Stratification of our estimates by country allows comparison with national dengue reporting, after taking into account the probability of an apparent infection being formally reported. The most notable differences are discussed. These new risk maps and infection estimates provide novel insights into the global, regional and national public health burden imposed by dengue. We anticipate that they will provide a starting point for a wider discussion about the global impact of this disease and will help to guide improvements in disease control strategies using vaccine, drug and vector control methods, and in their economic evaluation.
7,238 citations
Cites methods from "Novel methods improve prediction of..."
...The optimal number of trees was found with 10 fold cross-validation using the methods of Elith et al.(2008) 106 ....
[...]
...For parameters 2-5, we follow Elith et al. (2008) 106 in setting the interaction depth K equal to 4, the stochastic subsampling proportion π equal to 0....
[...]
TL;DR: It was found that methods specifically designed for collinearity, such as latent variable methods and tree based models, did not outperform the traditional GLM and threshold-based pre-selection and the value of GLM in combination with penalised methods and thresholds when omitted variables are considered in the final interpretation.
Abstract: Collinearity refers to the non independence of predictor variables, usually in a regression-type analysis. It is a common feature of any descriptive ecological data set and can be a problem for parameter estimation because it inflates the variance of regression parameters and hence potentially leads to the wrong identification of relevant predictors in a statistical model. Collinearity is a severe problem when a model is trained on data from one region or time, and predicted to another with a different or unknown structure of collinearity. To demonstrate the reach of the problem of collinearity in ecology, we show how relationships among predictors differ between biomes, change over spatial scales and through time. Across disciplines, different approaches to addressing collinearity problems have been developed, ranging from clustering of predictors, threshold-based pre-selection, through latent variable methods, to shrinkage and regularisation. Using simulated data with five predictor-response relationships of increasing complexity and eight levels of collinearity we compared ways to address collinearity with standard multiple regression and machine-learning approaches. We assessed the performance of each approach by testing its impact on prediction to new data. In the extreme, we tested whether the methods were able to identify the true underlying relationship in a training dataset with strong collinearity by evaluating its performance on a test dataset without any collinearity. We found that methods specifically designed for collinearity, such as latent variable methods and tree based models, did not outperform the traditional GLM and threshold-based pre-selection. Our results highlight the value of GLM in combination with penalised methods (particularly ridge) and threshold-based pre-selection when omitted variables are considered in the final interpretation. However, all approaches tested yielded degraded predictions under change in collinearity structure and the ‘folk lore’-thresholds of correlation coefficients between predictor variables of |r| >0.7 was an appropriate indicator for when collinearity begins to severely distort model estimation and subsequent prediction. The use of ecological understanding of the system in pre-analysis variable selection and the choice of the least sensitive statistical approaches reduce the problems of collinearity, but cannot ultimately solve them.
6,199 citations
AT&T1
TL;DR: This paper presents a tuning method that uses presence-only data for parameter tuning, and introduces several concepts that improve the predictive accuracy and running time of Maxent and describes a new logistic output format that gives an estimate of probability of presence.
Abstract: Accurate modeling of geographic distributions of species is crucial to various applications in ecology and conservation. The best performing techniques often require some parameter tuning, which may be prohibitively time-consuming to do separately for each species, or unreliable for small or biased datasets. Additionally, even with the abundance of good quality data, users interested in the application of species models need not have the statistical knowledge required for detailed tuning. In such cases, it is desirable to use "default settings", tuned and validated on diverse datasets. Maxent is a recently introduced modeling technique, achieving high predictive accuracy and enjoying several additional attractive properties. The performance of Maxent is influenced by a moderate number of parameters. The first contribution of this paper is the empirical tuning of these parameters. Since many datasets lack information about species absence, we present a tuning method that uses presence-only data. We evaluate our method on independently collected high-quality presence-absence data. In addition to tuning, we introduce several concepts that improve the predictive accuracy and running time of Maxent. We introduce "hinge features" that model more complex relationships in the training data; we describe a new logistic output format that gives an estimate of probability of presence; finally we explore "background sampling" strategies that cope with sample selection bias and decrease model-building time. Our evaluation, based on a diverse dataset of 226 species from 6 regions, shows: 1) default settings tuned on presence-only data achieve performance which is almost as good as if they had been tuned on the evaluation data itself; 2) hinge features substantially improve model performance; 3) logistic output improves model calibration, so that large differences in output values correspond better to large differences in suitability; 4) "target-group" background sampling can give much better predictive performance than random background sampling; 5) random background sampling results in a dramatic decrease in running time, with no decrease in model performance.
5,314 citations
Cites background or methods or result from "Novel methods improve prediction of..."
...A recent comprehensive comparison of presence-only modeling techniques (Elith et al. 2006) found that some new methods have better predictive accuracy than the established methods....
[...]
...Species occurrence data and environmental predictors We used a comprehensive collection of data developed by a working group at the National Center for Ecological Analysis and Synthesis (NCEAS) as part of a large-scale comparison of species distribution modeling methods (Elith et al. 2006)....
[...]
...For more details, see Elith et al. (2006)....
[...]
...…parameter settings developed in this study were used (without hinge features or logistic output, which came later) in the modeling comparison of Elith et al. (2006), and we feel that our careful parameter tuning, especially for small sample sizes, contributed to Maxent’s good showing in that…...
[...]
...First, we explore tuning of Maxent settings with the comprehensive dataset used by Elith et al. (2006)....
[...]
TL;DR: Species distribution models (SDMs) as mentioned in this paper are numerical tools that combine observations of species occurrence or abundance with environmental estimates, and are used to gain ecological and evolutionary insights and to predict distributions across landscapes, sometimes requiring extrapolation in space and time.
Abstract: Species distribution models (SDMs) are numerical tools that combine observations of species occurrence or abundance with environmental estimates. They are used to gain ecological and evolutionary insights and to predict distributions across landscapes, sometimes requiring extrapolation in space and time. SDMs are now widely used across terrestrial, freshwater, and marine realms. Differences in methods between disciplines reflect both differences in species mobility and in “established use.” Model realism and robustness is influenced by selection of relevant predictors and modeling method, consideration of scale, how the interplay between environmental and geographic factors is handled, and the extent of extrapolation. Current linkages between SDM practice and ecological theory are often weak, hindering progress. Remaining challenges include: improvement of methods for modeling presence-only data and for model selection and evaluation; accounting for biotic interactions; and assessing model uncertainty.
5,076 citations
Cites background from "Novel methods improve prediction of..."
...Reviews and comparisons include Franklin (2009) and Elith et al. (2006)....
[...]
...As a consequence their predictive performance may exceed that of more conventional techniques (Elith et al. 2006)....
[...]
TL;DR: This study provides a working guide to boosted regression trees (BRT), an ensemble method for fitting statistical models that differs fundamentally from conventional techniques that aim to fit a single parsimonious model.
Abstract: Summary 1 Ecologists use statistical models for both explanation and prediction, and need techniques that are flexible enough to express typical features of their data, such as nonlinearities and interactions 2 This study provides a working guide to boosted regression trees (BRT), an ensemble method for fitting statistical models that differs fundamentally from conventional techniques that aim to fit a single parsimonious model Boosted regression trees combine the strengths of two algorithms: regression trees (models that relate a response to their predictors by recursive binary splits) and boosting (an adaptive method for combining many simple models to give improved predictive performance) The final BRT model can be understood as an additive regression model in which individual terms are simple trees, fitted in a forward, stagewise fashion 3 Boosted regression trees incorporate important advantages of tree-based methods, handling different types of predictor variables and accommodating missing data They have no need for prior data transformation or elimination of outliers, can fit complex nonlinear relationships, and automatically handle interaction effects between predictors Fitting multiple trees in BRT overcomes the biggest drawback of single tree models: their relatively poor predictive performance Although BRT models are complex, they can be summarized in ways that give powerful ecological insight, and their predictive performance is superior to most traditional modelling methods 4 The unique features of BRT raise a number of practical issues in model fitting We demonstrate the practicalities and advantages of using BRT through a distributional analysis of the short-finned eel ( Anguilla australis Richardson), a native freshwater fish of New Zealand We use a data set of over 13 000 sites to illustrate effects of several settings, and then fit and interpret a model using a subset of the data We provide code and a tutorial to enable the wider use of BRT by ecologists
4,787 citations
Cites background or methods from "Novel methods improve prediction of..."
...…as a problem in statistical modelling, our experience with BRT is that prediction to independent data is not compromised – indeed, it is generally superior to other methods (see e.g. comparisons with GLM, GAM and multivariate adaptive regression splines, Elith et al. 2006; Leathwick et al. 2006)....
[...]
...A growing body of literature quantifies this difference in performance (Elith et al. 2006; Leathwick et al. 2006; Moisen et al. 2006)....
[...]
References
More filters
TL;DR: A new method for estimation in linear models called the lasso, which minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant, is proposed.
Abstract: SUMMARY We propose a new method for estimation in linear models. The 'lasso' minimizes the residual sum of squares subject to the sum of the absolute value of the coefficients being less than a constant. Because of the nature of this constraint it tends to produce some coefficients that are exactly 0 and hence gives interpretable models. Our simulation studies suggest that the lasso enjoys some of the favourable properties of both subset selection and ridge regression. It produces interpretable models like subset selection and exhibits the stability of ridge regression. There is also an interesting relationship with recent work in adaptive function estimation by Donoho and Johnstone. The lasso idea is quite general and can be applied in a variety of statistical models: extensions to generalized regression models and tree-based models are briefly described.
40,785 citations
"Novel methods improve prediction of..." refers methods in this paper
...determining model complexity (Tibshirani 1996)....
[...]
...For example, the ‘‘lasso,’’ used for regularization in MAXENT, can be applied to variable selection and coefficient estimation in regression (Hastie et al. 2001), and has been shown to perform better than stepwise selection for determining model complexity (Tibshirani 1996)....
[...]
Book•
19 Jun 2013
TL;DR: The second edition of this book is unique in that it focuses on methods for making formal statistical inference from all the models in an a priori set (Multi-Model Inference).
Abstract: Introduction * Information and Likelihood Theory: A Basis for Model Selection and Inference * Basic Use of the Information-Theoretic Approach * Formal Inference From More Than One Model: Multi-Model Inference (MMI) * Monte Carlo Insights and Extended Examples * Statistical Theory and Numerical Results * Summary
36,993 citations
TL;DR: In this article, the authors present a procedure for having two or more judges independently categorize a sample of units and determine the degree, significance, and significance of the units. But they do not discuss the extent to which these judgments are reproducible, i.e., reliable.
Abstract: CONSIDER Table 1. It represents in its formal characteristics a situation which arises in the clinical-social-personality areas of psychology, where it frequently occurs that the only useful level of measurement obtainable is nominal scaling (Stevens, 1951, pp. 2526), i.e. placement in a set of k unordered categories. Because the categorizing of the units is a consequence of some complex judgment process performed by a &dquo;two-legged meter&dquo; (Stevens, 1958), it becomes important to determine the extent to which these judgments are reproducible, i.e., reliable. The procedure which suggests itself is that of having two (or more) judges independently categorize a sample of units and determine the degree, significance, and
34,965 citations
"Novel methods improve prediction of..." refers methods in this paper
...Kappa (Cohen 1960), which is a chance-corrected measure of agreement, is commonly used in ecological studies with presence-absence data....
[...]
TL;DR: A representation and interpretation of the area under a receiver operating characteristic (ROC) curve obtained by the "rating" method, or by mathematical predictions based on patient characteristics, is presented and it is shown that in such a setting the area represents the probability that a randomly chosen diseased subject is (correctly) rated or ranked with greater suspicion than a random chosen non-diseased subject.
Abstract: A representation and interpretation of the area under a receiver operating characteristic (ROC) curve obtained by the "rating" method, or by mathematical predictions based on patient characteristics, is presented. It is shown that in such a setting the area represents the probability that a randomly chosen diseased subject is (correctly) rated or ranked with greater suspicion than a randomly chosen non-diseased subject. Moreover, this probability of a correct ranking is the same quantity that is estimated by the already well-studied nonparametric Wilcoxon statistic. These two relationships are exploited to (a) provide rapid closed-form expressions for the approximate magnitude of the sampling variability, i.e., standard error that one uses to accompany the area under a smoothed ROC curve, (b) guide in determining the size of the sample required to provide a sufficiently reliable estimate of this area, and (c) determine how large sample sizes should be to ensure that one can statistically detect difference...
19,398 citations
"Novel methods improve prediction of..." refers methods in this paper
...AUC has been used extensively in the species’ distribution modelling literature, and measures the ability of a model to discriminate between sites where a species is present, versus those where it is absent (Hanley and McNeil 1982)....
[...]
...Standards errors were calculated with the methods of Hanley and McNeil (1982)....
[...]
Book•
28 Jul 2013TL;DR: In this paper, the authors describe the important ideas in these areas in a common conceptual framework, and the emphasis is on concepts rather than mathematics, with a liberal use of color graphics.
Abstract: During the past decade there has been an explosion in computation and information technology. With it have come vast amounts of data in a variety of fields such as medicine, biology, finance, and marketing. The challenge of understanding these data has led to the development of new tools in the field of statistics, and spawned new areas such as data mining, machine learning, and bioinformatics. Many of these tools have common underpinnings but are often expressed with different terminology. This book describes the important ideas in these areas in a common conceptual framework. While the approach is statistical, the emphasis is on concepts rather than mathematics. Many examples are given, with a liberal use of color graphics. It is a valuable resource for statisticians and anyone interested in data mining in science or industry. The book's coverage is broad, from supervised learning (prediction) to unsupervised learning. The many topics include neural networks, support vector machines, classification trees and boosting---the first comprehensive treatment of this topic in any book.
This major new edition features many topics not covered in the original, including graphical models, random forests, ensemble methods, least angle regression and path algorithms for the lasso, non-negative matrix factorization, and spectral clustering. There is also a chapter on methods for ``wide'' data (p bigger than n), including multiple testing and false discovery rates.
Trevor Hastie, Robert Tibshirani, and Jerome Friedman are professors of statistics at Stanford University. They are prominent researchers in this area: Hastie and Tibshirani developed generalized additive models and wrote a popular book of that title. Hastie co-developed much of the statistical modeling software and environment in R/S-PLUS and invented principal curves and surfaces. Tibshirani proposed the lasso and is co-author of the very successful An Introduction to the Bootstrap. Friedman is the co-inventor of many data-mining tools including CART, MARS, projection pursuit and gradient boosting.
19,261 citations