miamaxent-ms.bib

@article{aarts2012,
  author = {Aarts, Geert and Fieberg, John and Matthiopoulos, Jason},
  doi = {10.1111/j.2041-210X.2011.00141.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Aarts et al_2012_Comparative interpretation of count, presence-absence and point methods for.pdf},
  journal = {Methods in Ecology and Evolution},
  number = {1},
  pages = {177-187},
  title = {Comparative Interpretation of Count, Presence-Absence and Point Methods for Species Distribution Models},
  volume = {3},
  year = {2012}
}

@article{ahmed2015,
  author = {Ahmed, Sadia E and McInerny, Greg and O'Hara, Kenton and Harper, Richard and Salido, Lara and Emmott, Stephen and Joppa, Lucas N},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Ahmed et al_2015_Scientists and software–surveying the species distribution modelling community.pdf},
  journal = {Diversity and Distributions},
  number = {3},
  pages = {258-267},
  title = {Scientists and Software \textendash{} Surveying the Species Distribution Modelling Community},
  volume = {21},
  year = {2015}
}

@article{aiello-lammens2015,
  author = {{Aiello-Lammens}, Matthew E and Boria, Robert A and Radosavljevic, Aleksandar and Vilela, Bruno and Anderson, Robert P},
  doi = {10.1111/ecog.01132},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Aiello-Lammens et al_2015_spThin.pdf},
  journal = {Ecography},
  keywords = {R package,software note},
  number = {5},
  pages = {541-545},
  title = {{{spThin}}: An {{R}} Package for Spatial Thinning of Species Occurrence Records for Use in Ecological Niche Models},
  volume = {38},
  year = {2015}
}

@article{araujo2005,
  abstract = {Abstract Increasing concern over the implications of climate change for biodiversity has led to the use of species?climate envelope models to project species extinction risk under climate-change scenarios. However, recent studies have demonstrated significant variability in model predictions and there remains a pressing need to validate models and to reduce uncertainties. Model validation is problematic as predictions are made for events that have not yet occurred. Resubstituition and data partitioning of present-day data sets are, therefore, commonly used to test the predictive performance of models. However, these approaches suffer from the problems of spatial and temporal autocorrelation in the calibration and validation sets. Using observed distribution shifts among 116 British breeding-bird species over the past ?20 years, we are able to provide a first independent validation of four envelope modelling techniques under climate change. Results showed good to fair predictive performance on independent validation, although rules used to assess model performance are difficult to interpret in a decision-planning context. We also showed that measures of performance on nonindependent data provided optimistic estimates of models' predictive ability on independent data. Artificial neural networks and generalized additive models provided generally more accurate predictions of species range shifts than generalized linear models or classification tree analysis. Data for independent model validation and replication of this study are rare and we argue that perfect validation may not in fact be conceptually possible. We also note that usefulness of models is contingent on both the questions being asked and the techniques used. Implementations of species?climate envelope models for testing hypotheses and predicting future events may prove wrong, while being potentially useful if put into appropriate context.},
  author = {Ara{\'u}jo, Miguel B. and Pearson, Richard G. and Thuiller, Wilfried and Erhard, Markus},
  doi = {10.1111/j.1365-2486.2005.01000.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Araujo et al_2005_Validation of species–climate impact models under climate change.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Araújo et al_2005_Validation of species–climate impact models under climate change.pdf;C:\\Users\\julienv\\Zotero\\storage\\97IQKGHF\\j.1365-2486.2005.01000.html},
  issn = {1354-1013},
  journal = {Global Change Biology},
  keywords = {bioclimatic-envelope models,breeding birds,climate change,model accuracy,uncertainty,validation},
  month = sep,
  number = {9},
  pages = {1504-1513},
  title = {Validation of Species\textendash{}Climate Impact Models under Climate Change},
  volume = {11},
  year = {2005}
}

@article{araujo2006,
  author = {Ara{\'u}jo, Miguel B and Guisan, Antoine},
  doi = {10.1111/j.1365-2699.2006.01584.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Araújo_Guisan_2006_Five (or so) challenges for species distribution modelling.pdf},
  journal = {Journal of Biogeography},
  number = {10},
  pages = {1677-1688},
  title = {Five (or so) Challenges for Species Distribution Modelling},
  volume = {33},
  year = {2006}
}

@article{araujo2019,
  abstract = {Demand for models in biodiversity assessments is rising, but which models are adequate for the task? We propose a set of best-practice standards and detailed guidelines enabling scoring of studies based on species distribution models for use in biodiversity assessments. We reviewed and scored 400 modeling studies over the past 20 years using the proposed standards and guidelines. We detected low model adequacy overall, but with a marked tendency of improvement over time in model building and, to a lesser degree, in biological data and model evaluation. We argue that implementation of agreed-upon standards for models in biodiversity assessments would promote transparency and repeatability, eventually leading to higher quality of the models and the inferences used in assessments. We encourage broad community participation toward the expansion and ongoing development of the proposed standards and guidelines.
Biodiversity assessments use a variety of data and models. We propose best-practice standards for studies in these assessments.
Biodiversity assessments use a variety of data and models. We propose best-practice standards for studies in these assessments.},
  author = {Ara{\'u}jo, Miguel B. and Anderson, Robert P. and Barbosa, A. M{\'a}rcia and Beale, Colin M. and Dormann, Carsten F. and Early, Regan and Garcia, Raquel A. and Guisan, Antoine and Maiorano, Luigi and Naimi, Babak and O'Hara, Robert B. and Zimmermann, Niklaus E. and Rahbek, Carsten},
  copyright = {Copyright \textcopyright{} 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial License 4.0 (CC BY-NC).. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.},
  doi = {10.1126/sciadv.aat4858},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Araújo et al. - 2019 - Standards for distribution models in biodiversity .pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Araújo et al. - 2019 -supp .pdf;C:\\Users\\julienv\\Zotero\\storage\\VH3L3TXS\\eaat4858.html},
  issn = {2375-2548},
  journal = {Science Advances},
  language = {en},
  month = jan,
  number = {1},
  pages = {eaat4858},
  title = {Standards for Distribution Models in Biodiversity Assessments},
  volume = {5},
  year = {2019}
}

@article{austin2002,
  author = {Austin, M P},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Austin_2002_Spatial prediction of species distribution.pdf},
  journal = {Ecological Modelling},
  pages = {101-118},
  title = {Spatial Prediction of Species Distribution: An Interface between Ecological Theory and Statistical Modelling},
  volume = {157},
  year = {2002}
}

@article{austin2007,
  author = {Austin, Mike},
  doi = {10.1016/j.ecolmodel.2006.07.005},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Austin_2007_Species distribution models and ecological theory.pdf},
  journal = {Ecological Modelling},
  number = {1-2},
  pages = {1-19},
  title = {Species Distribution Models and Ecological Theory: {{A}} Critical Assessment and Some Possible New Approaches},
  volume = {200},
  year = {2007}
}

@article{bahn2013,
  author = {Bahn, Volker and McGill, Brian J.},
  doi = {10.1111/j.1600-0706.2012.00299.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Bahn and McGill - 2013 - Testing the predictive performance of distribution.pdf},
  issn = {00301299},
  journal = {Oikos},
  language = {en},
  month = mar,
  number = {3},
  pages = {321-331},
  title = {Testing the Predictive Performance of Distribution Models},
  volume = {122},
  year = {2013}
}

@article{bell2016,
  author = {Bell, David M and Schlaepfer, Daniel R},
  doi = {10.1016/j.ecolmodel.2016.03.012},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Bell_Schlaepfer_2016_On the dangers of model complexity without ecological justification in species.pdf},
  journal = {Ecological Modelling},
  keywords = {extrapolation,Favorite,SDM,transferability},
  pages = {50-59},
  title = {On the Dangers of Model Complexity without Ecological Justification in Species Distribution Modeling},
  volume = {330},
  year = {2016}
}

@article{bendiksby2014,
  author = {Bendiksby, Mika and Mazzoni, Sabrina and J{\o}rgensen, Marte H and Halvorsen, Rune and Holien, H{\aa}kon},
  doi = {10.1111/jbi.12347},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Bendiksby et al_2014_Combining genetic analyses of archived specimens with distribution modelling to.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Bendiksby et al_2014_zapp.pdf},
  journal = {Journal of Biogeography},
  number = {11},
  pages = {2020-2031},
  title = {Combining Genetic Analyses of Archived Specimens with Distribution Modelling to Explain the Anomalous Distribution of the Rare Lichen {{Staurolemma}} Omphalarioides: Long-Distance Dispersal or Vicariance?},
  volume = {41},
  year = {2014}
}

@article{berger1996,
  author = {Berger, Adam L and Pietra, Vincent J Della and Pietra, Stephen A Della},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Berger et al_1996_A maximum entropy approach to natural language processing.pdf},
  journal = {Computational linguistics},
  keywords = {maximum entropy},
  number = {1},
  pages = {39-71},
  title = {A Maximum Entropy Approach to Natural Language Processing},
  volume = {22},
  year = {1996}
}

@article{bolduc2018,
  abstract = {We present R2MCDS, an R package that provides tools to call the multiple-covariate distance sampling (MCDS) engine from the Distance 6.2 software in the R environment. We built R2MCDS for the analysis of line transects data collected using distance intervals. Based on the user inputs, the package writes a command file, runs the MCDS engine and re-imports the resulting statistics file as an R object. The package is particularly useful when the user wants to repeat the same distance analysis, as in multi-species surveys. We demonstrate the applicability of the package for multi-species surveys using data from the Eastern Canada Seabirds at Sea (ECSAS) database.},
  author = {Bolduc, Fran{\c c}ois and Roy, Christian and Rousseu, Fran{\c c}ois},
  doi = {10.1016/j.ecoinf.2017.10.003},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Bolduc et al_2018_R2MCDS.pdf},
  issn = {15749541},
  journal = {Ecological Informatics},
  keywords = {R package,software note},
  month = sep,
  pages = {23-25},
  title = {{{R2MCDS}}: {{An R}} Package for the Analysis of Multi-Species Datasets Collected Using Distance Sampling},
  volume = {47},
  year = {2018}
}

@article{breiman2001,
  author = {Breiman, Leo},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Breiman and others - 2001 - Statistical modeling The two cultures (with comme.pdf},
  journal = {Statistical science},
  number = {3},
  pages = {199-215},
  publisher = {{Institute of Mathematical Statistics}},
  title = {Statistical Modeling: {{The}} Two Cultures},
  volume = {16},
  year = {2001}
}

@article{dahlgren2010,
  abstract = {Murtaugh (2009) recently illustrated that all subsets variable selection is very similar to stepwise regression. This, however, does not necessarily mean both methods are useful. On the contrary, the same problems with overfitting should apply. Ecologists should, if model building is indeed necessary, consider more reliable regression methods now available.},
  author = {Dahlgren, Johan P},
  doi = {10.1111/j.1461-0248.2010.01460.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Dahlgren_2010_Alternative regression methods are not considered in Murtaugh (2009) or by.pdf},
  issn = {1461023X},
  journal = {Ecology Letters},
  month = apr,
  number = {5},
  pages = {E7-E9},
  title = {Alternative Regression Methods Are Not Considered in {{Murtaugh}} (2009) or by Ecologists in General},
  volume = {13},
  year = {2010}
}

@article{dellapietra1997,
  author = {Della Pietra, S. and Della Pietra, V. and Lafferty, J.},
  doi = {10.1109/34.588021},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Della Pietra et al_1997_Inducing features of random fields.pdf},
  issn = {01628828},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  keywords = {maxent,maximum entropy},
  month = apr,
  number = {4},
  pages = {380-393},
  title = {Inducing Features of Random Fields},
  volume = {19},
  year = {1997}
}

@article{dicola2017,
  abstract = {The aim of the ecospat package is to make available novel tools and methods to support spatial analyses and modeling of species niches and distributions in a coherent workflow. The package is written in the R language (R Development Core Team) and contains several features, unique in their implementation, that are complementary to other existing R packages. Pre-modeling analyses include species niche quantifications and comparisons between distinct ranges or time periods, measures of phylogenetic diversity, and other data exploration functionalities (e.g. extrapolation detection, ExDet). Core modeling brings together the new approach of ensemble of small models (ESM) and various implementations of the spatially-explicit modeling of species assemblages (SESAM) framework. Post-modeling analyses include evaluation of species predictions based on presence-only data (Boyce index) and of community predictions, phylogenetic diversity and environmentally-constrained species co-occurrences analyses. The ecospat package also provides some functions to supplement the `biomod2' package (e.g. data preparation, permutation tests and cross-validation of model predictive power). With this novel package, we intend to stimulate the use of comprehensive approaches in spatial modelling of species and community distributions.},
  author = {Di Cola, Valeria and Broennimann, Olivier and Petitpierre, Blaise and Breiner, Frank T. and D'Amen, Manuela and Randin, Christophe and Engler, Robin and Pottier, Julien and Pio, Dorothea and Dubuis, Anne and Pellissier, Loic and Mateo, Rub{\'e}n G. and Hordijk, Wim and Salamin, Nicolas and Guisan, Antoine},
  doi = {10.1111/ecog.02671},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Di Cola et al_2017_ecospat.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Di Cola et al_2017_zapp.pdf},
  issn = {16000587},
  journal = {Ecography},
  number = {6},
  pages = {774--787},
  title = {Ecospat: An {{R}} Package to Support Spatial Analyses and Modeling of Species Niches and Distributions},
  volume = {40},
  year = {2017}
}

@incollection{dormann2011,
  author = {Dormann, Carsten F},
  booktitle = {Modelling Complex Ecological Dynamics},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Dormann_2011_Modelling species’ distributions.pdf},
  pages = {179-196},
  publisher = {{Springer}},
  title = {Modelling Species' Distributions},
  year = {2011}
}

@article{dormann2013,
  author = {Dormann, Carsten F and Elith, Jane and Bacher, Sven and Buchmann, Carsten and Carl, Gudrun and Carr{\'e}, Gabriel and Marqu{\'e}z, Jaime R Garc{\'i}a and Gruber, Bernd and Lafourcade, Bruno and Leit{\~a}o, Pedro J and M{\"u}nkem{\"u}ller, Tamara and McClean, Colin and Osborne, Patrick E and Reineking, Bj{\"o}rn and Schr{\"o}der, Boris and Skidmore, Andrew K and Zurell, Damaris and Lautenbach, Sven},
  doi = {10.1111/j.1600-0587.2012.07348.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Dormann et al_2013_Collinearity.pdf},
  journal = {Ecography},
  keywords = {Favorite},
  number = {1},
  pages = {27-46},
  title = {Collinearity: A Review of Methods to Deal with It and a Simulation Study Evaluating Their Performance},
  volume = {36},
  year = {2013}
}

@article{edvardsen2011,
  author = {Edvardsen, Anette and Bakkestuen, Vegar and Halvorsen, Rune},
  doi = {10.1111/j.1756-1051.2010.00984.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Edvardsen et al_2011_A fine-grained spatial prediction model for the red-listed vascular plant.pdf},
  journal = {Nordic Journal of Botany},
  keywords = {calibration,independent evaluation,model validation},
  number = {4},
  pages = {495-504},
  title = {A Fine-Grained Spatial Prediction Model for the Red-Listed Vascular Plant {{{\emph{Scorzonera}}}}{\emph{ Humilis}}},
  volume = {29},
  year = {2011}
}

@article{efron2001,
  author = {Efron, Brad},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Breiman and others - 2001 - Statistical modeling The two cultures (with comme.pdf},
  journal = {Statistical Science},
  number = {3},
  pages = {218-219},
  title = {[{{Statistical Modeling}}: {{The Two Cultures}}]: {{Comment}}},
  volume = {16},
  year = {2001}
}

@article{elith2006,
  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.},
  author = {Elith, Jane and Graham, Catherine H and Anderson, Robert P and Dud{\'i}k, Miroslav and Simon, Ferrier and Guisan, Antoine and Hijmans, Robert J and Huettmann, Falk and Leathwick, John R and Lehmann, Anthony and Li, Jin and Lohmann, Lucia G and Loiselle, Bette A and Manion, Glenn and Moritz, Craig and Nakamura, Miguel and Nakazawa, Yoshinori and Overton, Jacob McC and Peterson, A Townsend and Phillips, Steven J and Richardson, Karen and {Scachetti-Pereira}, Ricardo and Schapire, Robert E and Sober{\'o}n, Jorge and Williams, Stephen and Wisz, Mary S and Zimmermann, Niklaus E and Araujo, Miguel},
  doi = {10.2307/3683475},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith et al_2006_Novel Methods Improve Prediction of Species' Distributions from Occurrence Data.pdf},
  journal = {Ecography},
  number = {2},
  pages = {129-151},
  title = {Novel Methods Improve Prediction of Species' Distributions from Occurrence Data},
  volume = {29},
  year = {2006}
}

@article{elith2009,
  author = {Elith, Jane and Graham, Catherine H},
  doi = {10.1111/j.1600-0587.2008.05505.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith and Graham - 2009 - Do they How do they WHY do they differ On findi.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith and Graham - 2009 - zappendix.pdf},
  journal = {Ecography},
  number = {1},
  pages = {66-77},
  title = {Do They? {{How}} Do They? {{WHY}} Do They Differ? {{On}} Finding Reasons for Differing Performances of Species Distribution Models},
  volume = {32},
  year = {2009}
}

@article{elith2009a,
  author = {Elith, Jane and Leathwick, John R},
  doi = {10.1146/annurev.ecolsys.110308.120159},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith_Leathwick_2009_Species Distribution Models.pdf},
  journal = {Annual Review of Ecology, Evolution, and Systematics},
  number = {1},
  pages = {677-697},
  title = {Species {{Distribution Models}}: {{Ecological Explanation}} and {{Prediction Across Space}} and {{Time}}},
  volume = {40},
  year = {2009}
}

@article{elith2010,
  author = {Elith, Jane and Kearney, Michael and Phillips, Steven},
  doi = {10.1111/j.2041-210X.2010.00036.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith et al_2010_The art of modelling range-shifting species.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith et al_2010_zapp.pdf},
  journal = {Methods in Ecology and Evolution},
  number = {4},
  pages = {330-342},
  title = {The Art of Modelling Range-Shifting Species},
  volume = {1},
  year = {2010}
}

@article{elith2011,
  author = {Elith, Jane and Phillips, Steven J and Hastie, Trevor and Dud{\'i}k, Miroslav and Chee, Yung En and Yates, Colin J},
  doi = {10.1111/j.1472-4642.2010.00725.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith et al_2011_A statistical explanation of MaxEnt for ecologists.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Elith et al_2011_appendices.pdf},
  issn = {13669516},
  journal = {Diversity and Distributions},
  keywords = {maxent,tuning},
  month = jan,
  number = {1},
  pages = {43-57},
  title = {A Statistical Explanation of {{MaxEnt}} for Ecologists},
  volume = {17},
  year = {2011}
}

@article{escobar2018,
  abstract = {Many previous studies have attempted to assess ecological niche modeling performance using receiver operating characteristic (ROC) approaches, even though diverse problems with this metric have been pointed out in the literature. We explored different evaluation metrics based on independent testing data using the Darwin's Fox (Lycalopex fulvipes) as a detailed case in point. Six ecological niche models (ENMs; generalized linear models, boosted regression trees, Maxent, GARP, multivariable kernel density estimation, and NicheA) were explored and tested using six evaluation metrics (partial ROC, Akaike information criterion, omission rate, cumulative binomial probability), including two novel metrics to quantify model extrapolation versus interpolation (E-space index I) and extent of extrapolation versus Jaccard similarity (E-space index II). Different ENMs showed diverse and mixed performance, depending on the evaluation metric used. Because ENMs performed differently according to the evaluation metric employed, model selection should be based on the data available, assumptions necessary, and the particular research question. The typical ROC AUC evaluation approach should be discontinued when only presence data are available, and evaluations in environmental dimensions should be adopted as part of the toolkit of ENM researchers. Our results suggest that selecting Maxent ENM based solely on previous reports of its performance is a questionable practice. Instead, model comparisons, including diverse algorithms and parameterizations, should be the sine qua non for every study using ecological niche modeling. ENM evaluations should be developed using metrics that assess desired model characteristics instead of single measurement of fit between model and data. The metrics proposed herein that assess model performance in environmental space (i.e., E-space indices I and II) may complement current methods for ENM evaluation.},
  author = {Escobar, Luis E. and Qiao, Huijie and Cabello, Javier and Peterson, A. Townsend},
  copyright = {\textcopyright{} 2018 The Authors. Ecology and Evolution published by John Wiley \& Sons Ltd.},
  doi = {10.1002/ece3.4014},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Escobar et al_2018_Ecological niche modeling re-examined.pdf;C:\\Users\\julienv\\Zotero\\storage\\H3JCN8HS\\ece3.html},
  issn = {2045-7758},
  journal = {Ecology and Evolution},
  language = {en},
  number = {10},
  pages = {4757-4770},
  shorttitle = {Ecological Niche Modeling Re-Examined},
  title = {Ecological Niche Modeling Re-Examined: {{A}} Case Study with the {{Darwin}}'s Fox},
  volume = {8},
  year = {2018}
}

@article{fielding1997,
  author = {Fielding, Alan H and Bell, John F},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Fielding_Bell_1997_A review of methods for the assessment of prediction errors in conservation.pdf},
  journal = {Environmental conservation},
  number = {1},
  pages = {38-49},
  title = {A Review of Methods for the Assessment of Prediction Errors in Conservation Presence/Absence Models},
  volume = {24},
  year = {1997}
}

@article{fithian2013,
  abstract = {Statistical modeling of presence-only data has attracted much recent attention in the ecological literature, leading to a proliferation of methods, including the inhomogeneous Poisson process (IPP) model, maximum entropy (Maxent) modeling of species distributions and logistic regression models. Several recent articles have shown the close relationships between these methods. We explain why the IPP intensity function is a more natural object of inference in presence-only studies than occurrence probability (which is only defined with reference to quadrat size), and why presence-only data only allows estimation of relative, and not absolute intensity of species occurrence. All three of the above techniques amount to parametric density estimation under the same exponential family model (in the case of the IPP, the fitted density is multiplied by the number of presence records to obtain a fitted intensity). We show that IPP and Maxent give the exact same estimate for this density, but logistic regression in general yields a different estimate in finite samples. When the model is misspecified-as it practically always is-logistic regression and the IPP may have substantially different asymptotic limits with large data sets. We propose "infinitely weighted logistic regression," which is exactly equivalent to the IPP in finite samples. Consequently, many already-implemented methods extending logistic regression can also extend the Maxent and IPP models in directly analogous ways using this technique.},
  author = {Fithian, William and Hastie, Trevor},
  doi = {10.1214/13-AOAS667},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Fithian and Hastie - 2013 - Finite-sample equivalence in statistical models fo.pdf},
  issn = {1932-6157},
  journal = {The Annals of Applied Statistics},
  month = dec,
  number = {4},
  pages = {1917-1939},
  pmid = {25493106},
  title = {Finite-Sample Equivalence in Statistical Models for Presence-Only Data},
  volume = {7},
  year = {2013}
}

@incollection{fletcher2018,
  abstract = {Predictive models of species distributions are increasingly used in both basic and applied ecology to map species distributions and predict the effects of environmental change. Here, we describe the key concepts relevant to predicting species distributions (focusing on the use of niche theory), the types of data typically used, some common modeling algorithms, and illustrate how models are frequently evaluated. Our general goal is to illustrate how concepts, data and models are used to interpret species\textendash{}environment relationships and create maps of species distributions for addressing ecological questions and conservation problems. To do so, we model the distribution of the varied thrush (Ixoreus naevius) in the western USA using several model algorithms, such as climate envelopes, generalized linear and additive models, Random Forests, and Maxent. This example illustrates the different assumptions of modeling algorithms and how understanding the utility of models can vary based on how models are evaluated. Finally, we link these diverse approaches by emphasizing how many of these approaches can be cast as approximations of inhomogeneous point process models, which can help guide modeling decisions. We end by discussing further advanced in applied modeling of species distributions that aim to improve predictions, account for measurement error, and incorporate dynamics into the modeling process.},
  address = {Cham},
  author = {Fletcher, Robert and Fortin, Marie-Jos{\'e}e},
  booktitle = {Spatial Ecology and Conservation Modeling: Applications with {{R}}},
  doi = {10.1007/978-3-030-01989-1_7},
  editor = {Fletcher, Robert and Fortin, Marie-Jos{\'e}e},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Fletcher_Fortin_2018_Species Distributions.pdf},
  isbn = {978-3-030-01989-1},
  language = {en},
  pages = {213-269},
  publisher = {{Springer International Publishing}},
  title = {Species Distributions},
  year = {2018}
}

@article{galante2018,
  author = {Galante, Peter J. and Alade, Babatunde and Muscarella, Robert and Jansa, Sharon A. and Goodman, Steven M. and Anderson, Robert P.},
  doi = {10.1111/ecog.02909},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Galante et al_2018_The challenge of modeling niches and distributions for data-poor species.pdf},
  issn = {09067590},
  journal = {Ecography},
  keywords = {AIC,Favorite,maxent,model selection,niche,overfitting,sample bias,SDM},
  month = may,
  number = {5},
  pages = {726-736},
  title = {The Challenge of Modeling Niches and Distributions for Data-Poor Species: A Comprehensive Approach to Model Complexity},
  volume = {41},
  year = {2018}
}

@article{gaston2011,
  abstract = {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\textendash{}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.},
  author = {Gast{\'o}n, Aitor and {Garc{\'i}a-Vi{\~n}as}, Juan I.},
  doi = {10.1016/j.ecolmodel.2011.04.015},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Gastón_García-Viñas_2011_Modelling species distributions with penalised logistic regressions.pdf;C:\\Users\\julienv\\Zotero\\storage\\9LX424KA\\S0304380011002237.html},
  issn = {0304-3800},
  journal = {Ecological Modelling},
  keywords = {Calibration,Generalised linear models,Regularisation,Species distribution models},
  month = jul,
  number = {13},
  pages = {2037-2041},
  shorttitle = {Modelling Species Distributions with Penalised Logistic Regressions},
  title = {Modelling Species Distributions with Penalised Logistic Regressions: {{A}} Comparison with Maximum Entropy Models},
  volume = {222},
  year = {2011}
}

@book{gelman2006,
  abstract = {Data Analysis Using Regression and Multilevel/Hierarchical Models, first published in 2007, is a comprehensive manual for the applied researcher who wants to perform data analysis using linear and nonlinear regression and multilevel models. The book introduces a wide variety of models, whilst at the same time instructing the reader in how to fit these models using available software packages. The book illustrates the concepts by working through scores of real data examples that have arisen from the authors' own applied research, with programming codes provided for each one. Topics covered include causal inference, including regression, poststratification, matching, regression discontinuity, and instrumental variables, as well as multilevel logistic regression and missing-data imputation. Practical tips regarding building, fitting, and understanding are provided throughout.},
  author = {Gelman, Andrew and Hill, Jennifer},
  isbn = {978-1-139-46093-4},
  keywords = {Mathematics / Probability \& Statistics / General,Political Science / General,Psychology / Assessment; Testing \& Measurement,Social Science / Research},
  language = {en},
  month = dec,
  publisher = {{Cambridge University Press}},
  title = {Data Analysis Using Regression and Multilevel/Hierarchical Models},
  year = {2006}
}

@article{golding2017,
  author = {Golding, Nick and August, Tom A and Lucas, Tim C D and Gavaghan, David J and {van Loon}, E Emiel and McInerny, Greg},
  doi = {10.1111/2041-210x.12858},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Golding et al_2017_The zoon R package for reproducible and shareable species distribution modelling.pdf},
  journal = {Methods in Ecology and Evolution},
  keywords = {Favorite,R package,SDM,software note},
  number = {2},
  pages = {260-268},
  title = {The Zoon {{R}} Package for Reproducible and Shareable Species Distribution Modelling},
  volume = {9},
  year = {2017}
}

@article{guevara2018,
  author = {Guevara, L{\'a}zaro and Gerstner, Beth E. and Kass, Jamie M. and Anderson, Robert P.},
  doi = {10.1111/gcb.13992},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Guevara et al_2018_Toward ecologically realistic predictions of species distributions.pdf},
  issn = {13541013},
  journal = {Global Change Biology},
  keywords = {climate change,Favorite,maxent,niche,paleoecology,SDM,transferability},
  month = apr,
  number = {4},
  pages = {1511-1522},
  title = {Toward Ecologically Realistic Predictions of Species Distributions: {{A}} Cross-Time Example from Tropical Montane Cloud Forests},
  volume = {24},
  year = {2018}
}

@article{guillera-arroita2014,
  abstract = {1.Thibaud et al. (2014) present a framework for simulating species and evaluating the relative effects of factors affecting the predictions from species distribution models (SDMs). They demonstrate their approach by generating presence-absence datasets for different simulated species and analysing them using four modelling methods: three presence-absence methods and Maxent, which is a presence-background modelling tool. One of their results is striking: that their use of Maxent performs well in estimating occupancy probabilities and even outperforms the other methods on small sample sizes. This result is of concern to us, because it suggests that Maxent directly offers a useful alternative for modelling presence-absence data, which may prompt widespread adoption of this use of Maxent. In this paper we explore why this would be a mistake. 2.We draw on the theory underlying how the Maxent model operates and on simulations to discover: i) why Maxent appears to fare as well as it does in their evaluation and ii) why the best suited presence-absence method for data analysis (the generating model; a GLM) does not perform as well as we would expect. 3.We demonstrate that: i) the good performance observed for Maxent is largely a coincidence; the simulated species match well the arbitrary default parameter that Maxent applies to map its relative output into a 0-1 scale, but errors are much larger for other species we simulate; ii) the performance of the GLM is poorer than expected because Thibaud et al. do not use model selection and fit a model that is too complex for the amount of data available. 4.Maxent is a presence-background method and only provides estimates of relative suitability regardless of how the background sample is specified. When presence-absence data are available, one can transform Maxent's relative estimates into estimates of occupancy probability, and we provide methods to do so. However this requires the user to post-process Maxent's output. Proper PA methods such as GLMs can perform well under small sample sizes, provided care is taken during modelling to avoid overfitting. We demonstrate an effective method using regularisation with the R package glmnet. This article is protected by copyright. All rights reserved.},
  author = {{Guillera-Arroita}, Gurutzeta and {Lahoz-Monfort}, Jos{\'e} J. and Elith, Jane},
  doi = {10.1111/2041-210X.12252},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Guillera-Arroita et al_2014_Maxent is not a presence-absence method.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Guillera-Arroita et al. - 2014 zapp1-4.pdf},
  issn = {2041210X},
  journal = {Methods in Ecology and Evolution},
  number = {11},
  pages = {1192-1197},
  pmid = {99709634},
  title = {Maxent Is Not a Presence-Absence Method: A Comment on {{Thibaud}} et Al.},
  volume = {5},
  year = {2014}
}

@article{guisan2000,
  author = {Guisan, Antoine and Zimmermann, Niklaus E},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Guisan_Zimmermann_2000_Predictive habitat distribution models in ecology.pdf},
  journal = {Ecological Modelling},
  pages = {147-186},
  title = {Predictive Habitat Distribution Models in Ecology},
  volume = {135},
  year = {2000}
}

@article{guisan2013,
  author = {Guisan, Antoine and Tingley, Reid and Baumgartner, John B. and {Naujokaitis-Lewis}, Ilona and Sutcliffe, Patricia R. and Tulloch, Ayesha I. T. and Regan, Tracey J. and Brotons, Lluis and {McDonald-Madden}, Eve and {Mantyka-Pringle}, Chrystal and Martin, Tara G. and Rhodes, Jonathan R. and Maggini, Ramona and Setterfield, Samantha A. and Elith, Jane and Schwartz, Mark W. and Wintle, Brendan A. and Broennimann, Olivier and Austin, Mike and Ferrier, Simon and Kearney, Michael R. and Possingham, Hugh P. and Buckley, Yvonne M.},
  doi = {10.1111/ele.12189},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Guisan et al_2013_Predicting species distributions for conservation decisions.pdf},
  issn = {1461023X},
  journal = {Ecology Letters},
  month = dec,
  number = {12},
  pages = {1424-1435},
  title = {Predicting Species Distributions for Conservation Decisions},
  volume = {16},
  year = {2013}
}

@article{halvorsen2012,
  author = {Halvorsen, Rune},
  doi = {10.2478/v10208-011-0015-3},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen_2012_A gradient analytic perspective on distribution modelling.pdf},
  journal = {Sommerfeltia},
  keywords = {naturtyper},
  pages = {1-165},
  title = {A Gradient Analytic Perspective on Distribution Modelling},
  volume = {35},
  year = {2012}
}

@article{halvorsen2013,
  author = {Halvorsen, Rune},
  doi = {10.2478/v10208-011-0016-2},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen_2013_A strict maximum likelihood explanation of MaxEnt, and some implications for.pdf},
  journal = {Sommerfeltia},
  number = {1},
  pages = {1-132},
  title = {A Strict Maximum Likelihood Explanation of {{MaxEnt}}, and Some Implications for Distribution Modelling},
  volume = {36},
  year = {2013}
}

@article{halvorsen2015,
  author = {Halvorsen, Rune and Mazzoni, Sabrina and Bryn, Anders and Bakkestuen, Vegar},
  doi = {10.1111/ecog.00565},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2015_Opportunities for improved distribution modelling practice via a strict maximum.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2015_zapp1_2.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2015_zapp3.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2015_zapp4.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2015_zapp5.pdf},
  journal = {Ecography},
  number = {2},
  pages = {172-183},
  title = {Opportunities for Improved Distribution Modelling Practice via a Strict Maximum Likelihood Interpretation of {{MaxEnt}}},
  volume = {38},
  year = {2015}
}

@article{halvorsen2016,
  author = {Halvorsen, Rune and Mazzoni, Sabrina and Dirksen, John Wirkola and N{\ae}sset, Erik and Gobakken, Terje and Ohlson, Mikael},
  doi = {10.1016/j.ecolmodel.2016.02.021},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen et al_2016_How important are choice of model selection method and spatial autocorrelation.pdf},
  journal = {Ecological Modelling},
  pages = {108-118},
  title = {How Important Are Choice of Model Selection Method and Spatial Autocorrelation of Presence Data for Distribution Modelling by {{MaxEnt}}?},
  volume = {328},
  year = {2016}
}

@book{hastie2009,
  author = {Hastie, T. and Tibshirani, R. and Friedman, J.},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Hastie et al. - 2009 - The Elements of Statistical Learning Data Mining,.pdf},
  isbn = {978-0-387-21606-5},
  lccn = {2001031433},
  publisher = {{Springer New York}},
  series = {Springer {{Series}} in {{Statistics}}},
  title = {The Elements of Statistical Learning: Data Mining, Inference, and Prediction},
  year = {2009}
}

@article{hastie2013,
  abstract = {Presence-only data abounds in ecology, often accompanied by a background sample. Although many interesting aspects of the species' distribution can be learned from such data, one cannot learn the overall species occurrence probability, or prevalence, without making unjustified simplifying assumptions. In this forum article we question the approach of Royle et al. (2012) that claims to be able to do this.},
  author = {Hastie, T and Fithian, W},
  doi = {10.1111/j.1600-0587.2013.00321.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Hastie_Fithian_2013_Inference from presence-only data\; the ongoing controversy.pdf},
  journal = {Ecography},
  number = {8},
  pages = {864-867},
  pmid = {25492992},
  title = {Inference from Presence-Only Data; the Ongoing Controversy},
  volume = {36},
  year = {2013}
}

@article{heikkinen2012,
  author = {Heikkinen, Risto K and Marmion, Mathieu and Luoto, Miska},
  doi = {10.1111/j.1600-0587.2011.06999.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Heikkinen et al_2012_Does the interpolation accuracy of species distribution models come at the.pdf},
  journal = {Ecography},
  number = {3},
  pages = {276-288},
  title = {Does the Interpolation Accuracy of Species Distribution Models Come at the Expense of Transferability?},
  volume = {35},
  year = {2012}
}

@article{huntley1995,
  abstract = {[It is hypothesized that the principal features of higher plant distributions at continental scales are determined by the macroclimate. Bioclimate data have been computed on a 50 km grid across Europe. Along with published maps of higher plant distributions based upon the same grid, these data have been used to derive climate response surfaces that model the relationship between a species' distribution and the present climate. Eight species representative of a variety of phytogeographic patterns have been investigated. The results support the hypothesis that the European distributions of all eight species are principally determined by macroclimate and illustrate the nature of the climatic constraints upon each species. Simulated future distributions in equilibrium with 2\texttimes{} CO2 climate scenarios derived from two alternative GCMs show that all of the species are likely to experience major shifts in their potential range if such climatic changes take place. Some species may suffer substantial range and population reductions and others may face the threat of extinction. The rate of the forecast climate changes is such that few, if any, species may be able to maintain their ranges in equilibrium with the changing climate. In consequence, the transient impacts upon ecosystems will be varied but often may lead to a period of dominance by opportunist, early-successional species. Our simulations of potential ranges take no account of such factors as photoperiod or the direct effects of CO2, both of which may substantially alter the realized future equilibrium.]},
  author = {Huntley, Brian and Berry, Pamela M. and Cramer, Wolfgang and McDonald, Alison P.},
  doi = {10.2307/2845830},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Huntley et al_1995_Special Paper.pdf},
  issn = {0305-0270},
  journal = {Journal of Biogeography},
  number = {6},
  pages = {967-1001},
  shorttitle = {Special {{Paper}}},
  title = {Special {{Paper}}: {{Modelling Present}} and {{Potential Future Ranges}} of {{Some European Higher Plants Using Climate Response Surfaces}}},
  volume = {22},
  year = {1995}
}

@article{jarnevich2015,
  author = {Jarnevich, Catherine S and Stohlgren, Thomas J and Kumar, Sunil and Morisette, Jeffery T and Holcombe, Tracy R},
  doi = {10.1016/j.ecoinf.2015.06.007},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Jarnevich et al_2015_Caveats for correlative species distribution modeling.pdf},
  journal = {Ecological Informatics},
  pages = {6-15},
  title = {Caveats for Correlative Species Distribution Modeling},
  volume = {29},
  year = {2015}
}

@article{jaynes1957,
  author = {Jaynes, Edwin T},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Jaynes_1957_Information theory and statistical mechanics.pdf},
  journal = {Physical review},
  keywords = {maxent,maximum entropy},
  number = {4},
  pages = {620},
  title = {Information Theory and Statistical Mechanics},
  volume = {106},
  year = {1957}
}

@article{jaynes1957a,
  author = {Jaynes, Edwin T},
  journal = {Physical review},
  keywords = {maxent,maximum entropy},
  number = {2},
  pages = {171},
  title = {Information Theory and Statistical Mechanics. {{II}}},
  volume = {108},
  year = {1957}
}

@article{kass2018,
  author = {Kass, Jamie M and Vilela, Bruno and {Aiello-Lammens}, Matthew E and Muscarella, Robert and Merow, Cory and Anderson, Robert P},
  doi = {10.1111/2041-210x.12945},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Kass et al_2018_Wallace.pdf},
  journal = {Methods in Ecology and Evolution},
  number = {4},
  pages = {1151-1156},
  title = {Wallace: {{A}} Flexible Platform for Reproducible Modeling of Species Niches and Distributions Built for Community Expansion},
  volume = {9},
  year = {2018}
}

@article{leroy2016,
  author = {Leroy, Boris and Meynard, Christine N. and Bellard, C{\'e}line and Courchamp, Franck},
  doi = {10.1111/ecog.01388},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Leroy et al_2015_virtualspecies, an R package to generate virtual species distributions.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Leroy et al_2016_virtualspecies, an R package to generate virtual species distributions.pdf},
  issn = {09067590},
  journal = {Ecography},
  language = {en},
  month = jun,
  number = {6},
  pages = {599-607},
  title = {Virtualspecies, an {{R}} Package to Generate Virtual Species Distributions},
  volume = {39},
  year = {2016}
}

@article{mazzoni2015,
  author = {Mazzoni, Sabrina and Halvorsen, Rune and Bakkestuen, Vegar},
  doi = {10.1016/j.ecoinf.2015.07.001},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Mazzoni et al_2015_MIAT.pdf},
  issn = {15749541},
  journal = {Ecological Informatics},
  month = nov,
  pages = {215-221},
  title = {{{MIAT}}: {{Modular R}}-Wrappers for Flexible Implementation of {{MaxEnt}} Distribution Modelling},
  volume = {30},
  year = {2015}
}

@phdthesis{mazzoni2016,
  address = {Oslo, Norway},
  author = {Mazzoni, Sabrina},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Mazzoni_2016_Distribution modelling by Maxent.pdf},
  note = {Publication Title: Natural History Museum},
  school = {Universiy of Oslo},
  title = {Distribution Modelling by {{Maxent}}: From Black Box to Flexible Toolbox},
  year = {2016}
}

@article{merow2013,
  abstract = {The MaxEnt software package is one of the most popular tools for species distribution and environmental niche modeling, with over 1000 published applications since 2006. Its popularity is likely for two reasons: 1) MaxEnt typically outperforms other methods based on predictive accuracy and 2) the software is particularly easy to use. MaxEnt users must make a number of decisions about how they should select their input data and choose from a wide variety of settings in the software package to build models from these data. The underlying basis for making these decisions is unclear in many studies, and default settings are apparently chosen, even though alternative settings are often more appropriate. In this paper, we provide a detailed explanation of how MaxEnt works and a prospectus on modeling options to enable users to make informed decisions when preparing data, choosing settings and interpreting output. We explain how the choice of background samples reflects prior assumptions, how nonlinear functions of environmental variables (features) are created and selected, how to account for environmentally biased sampling, the interpretation of the various types of model output and the challenges for model evaluation. We demonstrate MaxEnt's calculations using both simplified simulated data and occurrence data from South Africa on species of the flowering plant family Proteaceae. Throughout, we show how MaxEnt's outputs vary in response to different settings to highlight the need for making biologically motivated modeling decisions.},
  author = {Merow, Cory and Smith, Matthew J. and Silander, John A.},
  copyright = {\textcopyright{} 2013 The Authors},
  doi = {10.1111/j.1600-0587.2013.07872.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Merow et al_2013_A practical guide to MaxEnt for modeling species’ distributions.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Merow et al_2013_app.pdf;C:\\Users\\julienv\\Zotero\\storage\\BED2U66P\\j.1600-0587.2013.07872.html},
  issn = {1600-0587},
  journal = {Ecography},
  language = {en},
  month = oct,
  number = {10},
  pages = {1058-1069},
  shorttitle = {A Practical Guide to {{MaxEnt}} for Modeling Species' Distributions},
  title = {A Practical Guide to {{MaxEnt}} for Modeling Species' Distributions: What It Does, and Why Inputs and Settings Matter},
  volume = {36},
  year = {2013}
}

@article{merow2014,
  author = {Merow, Cory and Smith, Mathew J and Edwards, Thomas C and Guisan, Antoine and McMahon, Sean M and Normand, Signe and Thuiller, Wilfried and W{\"u}est, Rafael O and Zimmermann, Niklaus E and Elith, Jane},
  doi = {10.1111/ecog.00845},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Merow et al_2014_What do we gain from simplicity versus complexity in species distribution models.pdf},
  journal = {Ecography},
  keywords = {IPM,overfitting,SDM},
  number = {12},
  pages = {1267-1281},
  title = {What Do We Gain from Simplicity versus Complexity in Species Distribution Models?},
  volume = {37},
  year = {2014}
}

@article{merow2016,
  author = {Merow, Cory and Allen, Jenica M and {Aiello-Lammens}, Matthew and Silander, John A},
  doi = {10.1111/geb.12453},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Merow et al_2016_Improving niche and range estimates with Maxent and point process models by.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Merow et al_2016_zapp.pdf},
  journal = {Global Ecology and Biogeography},
  pages = {1022-1036},
  title = {Improving Niche and Range Estimates with {{Maxent}} and Point Process Models by Integrating Spatially Explicit Information},
  volume = {25},
  year = {2016}
}

@article{morales2017,
  abstract = {Environmental niche modeling (ENM) is commonly used to develop probabilistic maps of species distribution. Among available ENM techniques, MaxEnt has become one of the most popular tools for modeling species distribution, with hundreds of peer-reviewed articles published each year. MaxEnt's popularity is mainly due to the use of a graphical interface and automatic parameter configuration capabilities. However, recent studies have shown that using the default automatic configuration may not be always appropriate because it can produce non-optimal models; particularly when dealing with a small number of species presence points. Thus, the recommendation is to evaluate the best potential combination of parameters (feature classes and regularization multiplier) to select the most appropriate model. In this work we reviewed 244 articles published between 2013 and 2015 to assess whether researchers are following recommendations to avoid using the default parameter configuration when dealing with small sample sizes, or if they are using MaxEnt as a ``black box tool.'' Our results show that in only 16\% of analyzed articles authors evaluated best feature classes, in 6.9\% evaluated best regularization multipliers, and in a meager 3.7\% evaluated simultaneously both parameters before producing the definitive distribution model. We analyzed 20 articles to quantify the potential differences in resulting outputs when using software default parameters instead of the alternative best model. Results from our analysis reveal important differences between the use of default parameters and the best model approach, especially in the total area identified as suitable for the assessed species and the specific areas that are identified as suitable by both modelling approaches. These results are worrying, because publications are potentially reporting over-complex or over-simplistic models that can undermine the applicability of their results. Of particular importance are studies used to inform policy making. Therefore, researchers, practitioners, reviewers and editors need to be very judicious when dealing with MaxEnt, particularly when the modelling process is based on small sample sizes.},
  author = {Morales, Narkis S. and Fern{\'a}ndez, Ignacio C. and {Baca-Gonz{\'a}lez}, Victoria},
  doi = {10.7717/peerj.3093},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Morales et al_2017_MaxEnt’s parameter configuration and small samples.pdf},
  issn = {2167-8359},
  journal = {PeerJ},
  pages = {e3093},
  pmid = {28316894},
  title = {{{MaxEnt}}'s Parameter Configuration and Small Samples: Are We Paying Attention to Recommendations? {{A}} Systematic Review},
  volume = {5},
  year = {2017}
}

@article{moreno-amat2015,
  author = {{Moreno-Amat}, Elena and Mateo, Rub{\'e}n G and {Nieto-Lugilde}, Diego and {Morueta-Holme}, Naia and Svenning, Jens-Christian and {Garc{\'i}a-Amorena}, Ignacio},
  doi = {10.1016/j.ecolmodel.2015.05.035},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Moreno-Amat et al_2015_Impact of model complexity on cross-temporal transferability in Maxent species.pdf},
  journal = {Ecological Modelling},
  pages = {308-317},
  title = {Impact of Model Complexity on Cross-Temporal Transferability in {{Maxent}} Species Distribution Models: {{An}} Assessment Using Paleobotanical Data},
  volume = {312},
  year = {2015}
}

@article{murtaugh2009,
  author = {Murtaugh, Paul A.},
  doi = {10.1111/j.1461-0248.2009.01361.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Murtaugh - 2009 - Performance of several variable-selection methods .pdf},
  issn = {1461023X, 14610248},
  journal = {Ecology Letters},
  language = {en},
  month = oct,
  number = {10},
  pages = {1061-1068},
  title = {Performance of Several Variable-Selection Methods Applied to Real Ecological Data},
  volume = {12},
  year = {2009}
}

@article{muscarella2014,
  author = {Muscarella, Robert and Galante, Peter J. and {Soley-Guardia}, Mariano and Boria, Robert A. and Kass, Jamie M. and Uriarte, Mar{\'i}a and Anderson, Robert P.},
  doi = {10.1111/2041-210X.12261},
  editor = {McPherson, Jana},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Muscarella et al_2014_ENMeval.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Muscarella et al_2014_erratum.pdf},
  issn = {2041210X},
  journal = {Methods in Ecology and Evolution},
  keywords = {erratum},
  month = nov,
  number = {11},
  pages = {1198-1205},
  title = {{{ENMeval}}: {{An R}} Package for Conducting Spatially Independent Evaluations and Estimating Optimal Model Complexity for {{Maxent}} Ecological Niche Models},
  volume = {5},
  year = {2014}
}

@article{naimi2016,
  author = {Naimi, Babak and Ara{\'u}jo, Miguel B},
  doi = {10.1111/ecog.01881},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Naimi_Araújo_2016_sdm.pdf},
  journal = {Ecography},
  keywords = {R package},
  number = {4},
  pages = {368-375},
  title = {SDM: A Reproducible and Extensible {{R}} Platform for Species Distribution Modelling},
  volume = {39},
  year = {2016}
}

@article{okland2001,
  author = {{\O}kland, Rune Halvorsen and {\O}kland, Tonje and Rydgren, Knut},
  journal = {Sommerfeltia},
  pages = {1-190},
  title = {Vegetation-Environment Relationships of Boreal Spruce Swamp Forests in {{{\O}stmarka Nature Reserve}}, {{SE Norway}}},
  volume = {29},
  year = {2001}
}

@article{okland2003,
  abstract = {Previous studies point to biogeographic (i.e., evolutionary and demographic) and ecological (i.e., habitat differentiation and disturbance) processes as the most important causes of spatial variation in species richness and species composition (occurrence and abundance). We examined patterns of variation in vascular plant and bryophyte species composition among 150 1-m2 plots distributed semi-randomly over 11 Norwegian boreal swamp-forest localities. Swamp forests are species-rich islands in an otherwise species-poor forest landscape. For each plot, 53 environmental variables were recorded. By using Canonical Correspondence Analysis (CCA), we found that {$\sim$}20\% of the explainable variation in species composition was due to swamp-forest affiliation, in addition to the {$\sim$}35\% that was due to environmental differences between swamp-forest localities. The uniqueness of the species composition of each swamp forest was also emphasized by analyses of compositional dissimilarity. Plots were significantly more dissimilar if situated in different swamp forests than if situated in the same swamp forest, after environmental differences had been corrected for. The lack of any significant relationship between compositional dissimilarity and geographical distance or swamp-forest area indicated that this pattern was not mainly due to recent successful dispersal and establishment events. We argue that the distinctness of swamp forests, in particular, those richer in species and soil nutrients, is due to a combination of factors among which randomness in establishment in gaps (``windows of opportunity'') and persistence of established clonal species are important. Furthermore, we argue that the probability for successful recruitment may have been higher in previous time periods than it is today. The unique combination of important determinants of the species composition in boreal swamp forests supports the view that there exists a diversity of explanations for diversity, and that these, to a large extent, are system and/or area specific.},
  author = {{\O}kland, Rune Halvorsen and Rydgren, Knut and {\O}kland, Tonje},
  copyright = {\textcopyright{} 2003 by the Ecological Society of America},
  doi = {10.1890/0012-9658(2003)084[1909:PSCOBS]2.0.CO;2},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Halvorsen Økland et al. - 2003 - Plant species composition of boreal spruce swamp f.pdf;C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Økland et al_2003_Plant Species Composition of Boreal Spruce Swamp Forests.pdf;C:\\Users\\julienv\\Zotero\\storage\\PETLLK2Y\\0012-9658(2003)084[1909PSCOBS]2.0.html},
  issn = {1939-9170},
  journal = {Ecology},
  keywords = {bryophytes,ecological similarity,habitat islands,immigration history,spatial variation,swamp forest,variation partitioning,vascular plants},
  language = {en},
  number = {7},
  pages = {1909-1919},
  shorttitle = {Plant {{Species Composition}} of {{Boreal Spruce Swamp Forests}}},
  title = {Plant Species Composition of Boreal Spruce Swamp Forests: Closed Doors and Windows of Opportunity},
  volume = {84},
  year = {2003}
}

@article{owens2013,
  abstract = {Correlational models of species' ecological niches are commonly used to transfer model rules onto other sets of conditions to evaluate species' distri\ldots{}},
  author = {Owens, Hannah L. and Campbell, Lindsay P. and Dornak, L. Lynnette and Saupe, Erin E. and Barve, Narayani and Sober{\'o}n, Jorge and Ingenloff, Kate and {Lira-Noriega}, Andr{\'e}s and Hensz, Christopher M. and Myers, Corinne E. and Peterson, A. Townsend},
  doi = {10.1016/j.ecolmodel.2013.04.011},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\2013 - Constraints on interpretation of ecological niche .pdf;C:\\Users\\julienv\\Zotero\\storage\\UMMSUW8D\\S0304380013002159.html},
  issn = {03043800},
  journal = {Ecological Modelling},
  language = {en},
  month = aug,
  pages = {10-18},
  title = {Constraints on Interpretation of Ecological Niche Models by Limited Environmental Ranges on Calibration Areas},
  volume = {263},
  year = {2013}
}

@article{pearce2006,
  author = {Pearce, Jennie L. and Boyce, Mark S.},
  doi = {10.1111/j.1365-2664.2005.01112.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Pearce_Boyce_2006_Modelling distribution and abundance with presence-only data.pdf},
  issn = {0021-8901},
  journal = {Journal of Applied Ecology},
  month = jun,
  number = {3},
  pages = {405-412},
  title = {Modelling Distribution and Abundance with Presence-Only Data},
  volume = {43},
  year = {2006}
}

@inproceedings{phillips2004,
  address = {Banff, Canada},
  author = {Phillips, Steven J and Dud{\'i}k, Miroslav and Schapire, Robert E},
  booktitle = {Proceedings of the Twenty-First International Conference on Machine Learning},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Phillips et al_2004_A maximum entropy approach to species distribution modeling.pdf},
  pages = {655-662},
  publisher = {{ACM}},
  title = {A Maximum Entropy Approach to Species Distribution Modeling},
  year = {2004}
}

@article{phillips2006,
  author = {Phillips, Steven J and Anderson, Robert P and Schapire, Robert E},
  doi = {10.1016/j.ecolmodel.2005.03.026},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Phillips et al_2006_Maximum entropy modeling of species geographic distributions.pdf},
  journal = {Ecological Modelling},
  number = {3-4},
  pages = {231-259},
  title = {Maximum Entropy Modeling of Species Geographic Distributions},
  volume = {190},
  year = {2006}
}

@article{phillips2008,
  author = {Phillips, Steven J and Dud{\'i}k, Miroslav},
  doi = {10.1111/j.2007.0906-7590.05203.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Phillips_Dudík_2008_Modeling of species distributions with Maxent.pdf},
  journal = {Ecography},
  pages = {161-175},
  title = {Modeling of Species Distributions with {{Maxent}}: New Extensions and a Comprehensive Evaluation},
  volume = {31},
  year = {2008}
}

@article{phillips2009,
  author = {Phillips, Steven J and Dud{\'i}k, Miroslav and Elith, Jane and Graham, Catherine H and Lehmann, Anthony and Leathwick, John and Ferrier, Simon},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Phillips et al_2009_Sample selection bias and presence-only distribution models.pdf},
  journal = {Ecological Applications},
  number = {1},
  pages = {181-197},
  title = {Sample Selection Bias and Presence-Only Distribution Models: Implications for Background and Pseudo-Absence Data},
  volume = {19},
  year = {2009}
}

@article{phillips2017,
  author = {Phillips, Steven J and Anderson, Robert P and Dud{\'i}k, Miroslav and Schapire, Robert E and Blair, Mary E},
  doi = {10.1111/ecog.03049},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Phillips et al_2017_Opening the black box.pdf},
  journal = {Ecography},
  number = {7},
  pages = {887-893},
  title = {Opening the Black Box: An Open-Source Release of {{Maxent}}},
  volume = {40},
  year = {2017}
}

@article{pinto2016,
  author = {Pinto, Cecilia and Thorburn, James A and Neat, Francis and Wright, Peter J and Wright, Serena and Scott, Beth E and Cornulier, Thomas and Travis, Justin M J},
  doi = {10.1111/ddi.12437},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Pinto et al_2016_Using individual tracking data to validate the predictions of species.pdf},
  journal = {Diversity and Distributions},
  number = {6},
  pages = {682-693},
  title = {Using Individual Tracking Data to Validate the Predictions of Species Distribution Models},
  volume = {22},
  year = {2016}
}

@article{ponder2001,
  author = {Ponder, Winston F and Carter, G A and Flemons, P and Chapman, R R},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Ponder et al_2001_Evaluation of museum collection data for use in biodiversity assessment.pdf},
  journal = {Conservation biology},
  number = {3},
  pages = {648-657},
  title = {Evaluation of Museum Collection Data for Use in Biodiversity Assessment},
  volume = {15},
  year = {2001}
}

@article{qiao2015,
  abstract = {The field of ecological niche modeling or species distribution modeling has seen enormous activity and attention in recent years, in light of exciting biological inferences that can be drawn from correlational models of species' environmental requirements (i.e., ecological niches) and inferences of potential geographic distributions. Among the many methods used in the field, one or two are in practice assumed to be `best' and are used commonly, often without explicit testing. We explore herein implications of the ``No Free Lunch'' theorem, which suggests that no single optimization approach will prove to be best under all circumstances: we developed diverse virtual species with known niche and dispersal properties to test a suite of niche modeling algorithms designed to estimate potential areas of distribution. The result was that (1) indeed, no single `best' algorithm was found, and (2) different algorithms perform in very different manners depending on the particularities of the virtual species. The conclusion is that niche or distribution modeling studies should begin by testing a suite of algorithms for predictive ability under the particular circumstances of the study, and choose an algorithm for a particular challenge based on the results of those tests. Studies that do not take this step may use algorithms that are not optimal for that particular challenge. This article is protected by copyright. All rights reserved.},
  author = {Qiao, Huijie and Sober{\'o}n, Jorge and Peterson, Andrew Townsend},
  doi = {10.1111/2041-210X.12397},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Qiao et al_2015_No silver bullets in correlative ecological niche modelling.pdf},
  issn = {2041210X},
  journal = {Methods in Ecology and Evolution},
  pages = {1126-1136},
  title = {No Silver Bullets in Correlative Ecological Niche Modelling: Insights from Testing among Many Potential Algorithms for Niche Estimation},
  volume = {6},
  year = {2015}
}

@article{radosavljevic2014,
  abstract = {Aim Models of species niches and distributions have become invaluable to biogeographers over the past decade, yet several outstanding methodological issues remain. Here we address three critical ones: selecting appropriate evaluation data, detecting overfitting, and tuning program settings to approximate optimal model complexity. We integrate solutions to these issues for Maxent models, using the Caribbean spiny pocket mouse, Heteromys anomalus, as an example. Location: North-western South America. Methods We partitioned data into calibration and evaluation datasets via three variations of k-fold cross-validation: randomly partitioned, geographically structured and masked geographically structured (which restricts background data to regions corresponding to calibration localities). Then, we carried out tuning experiments by varying the level of regularization, which controls model complexity. Finally, we gauged performance by quantifying discriminatory ability and overfitting, as well as via visual inspections of maps of the predictions in geography. Results Performance varied among data-partitioning approaches and among regularization multipliers. The randomly partitioned approach inflated estimates of model performance and the geographically structured approach showed high overfitting. In contrast, the masked geographically structured approach allowed selection of high-performing models based on all criteria. Discriminatory ability showed a slight peak in performance around the default regularization multiplier. However, regularization levels two to four times higher than the default yielded substantially lower overfitting. Visual inspection of maps of model predictions coincided with the quantitative evaluations. Main conclusions Species-specific tuning of model parameters can improve the performance of Maxent models. Further, accurate estimates of model performance and overfitting depend on using independent evaluation data. These strategies for model evaluation may be useful for other modelling methods as well.},
  author = {Radosavljevic, Aleksandar and Anderson, Robert P.},
  doi = {10.1111/jbi.12227},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Radosavljevic_Anderson_2014_Making better Maxent models of species distributions.pdf},
  issn = {13652699},
  journal = {Journal of Biogeography},
  keywords = {cross validation,maxent,niche,overfitting},
  pages = {629-643},
  title = {Making Better {{Maxent}} Models of Species Distributions: Complexity, Overfitting and Evaluation},
  volume = {41},
  year = {2014}
}

@article{raffalovich2008,
  abstract = {Model selection strategies play an important, if not explicit, role in quantitative research. The inferential properties of these strategies are largely unknown, therefore, there is little basis for recommending (or avoiding) any particular set of strategies. In this paper, we evaluate several commonly used model selection procedures [Bayesian information criterion (BIC), adjusted R 2, Mallows' C p, Akaike information criteria (AIC), AICc, and stepwise regression] using Monte-Carlo simulation of model selection when the true data generating processes (DGP) are known. We find that the ability of these selection procedures to include important variables and exclude irrelevant variables increases with the size of the sample and decreases with the amount of noise in the model. None of the model selection procedures do well in small samples, even when the true DGP is largely deterministic; thus, data mining in small samples should be avoided entirely. Instead, the implicit uncertainty in model specification should be explicitly discussed. In large samples, BIC is better than the other procedures at correctly identifying most of the generating processes we simulated, and stepwise does almost as well. In the absence of strong theory, both BIC and stepwise appear to be reasonable model selection strategies in large samples. Under the conditions simulated, adjusted R 2, Mallows' C p AIC, and AICc are clearly inferior and should be avoided.},
  author = {Raffalovich, Lawrence E. and Deane, Glenn D. and Armstrong, David and Tsao, Hui-Shien},
  doi = {10.1080/03081070802203959},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Raffalovich et al_2008_Model selection procedures in social research.pdf;C:\\Users\\julienv\\Zotero\\storage\\CTWWYYHX\\03081070802203959.html},
  issn = {0266-4763},
  journal = {Journal of Applied Statistics},
  month = oct,
  number = {10},
  pages = {1093-1114},
  shorttitle = {Model Selection Procedures in Social Research},
  title = {Model Selection Procedures in Social Research: {{Monte}}-{{Carlo}} Simulation Results},
  volume = {35},
  year = {2008}
}

@article{randin2006,
  abstract = {Aim To assess the geographical transferability of niche-based species distribution models fitted with two modelling techniques. Location Two distinct geographical study areas in Switzerland and Austria, in the subalpine and alpine belts. Methods Generalized linear and generalized additive models (GLM and GAM) with a binomial probability distribution and a logit link were fitted for 54 plant species, based on topoclimatic predictor variables. These models were then evaluated quantitatively and used for spatially explicit predictions within (internal evaluation and prediction) and between (external evaluation and prediction) the two regions. Comparisons of evaluations and spatial predictions between regions and models were conducted in order to test if species and methods meet the criteria of full transferability. By full transferability, we mean that: (1) the internal evaluation of models fitted in region A and B must be similar; (2) a model fitted in region A must at least retain a comparable external evaluation when projected into region B, and vice-versa; and (3) internal and external spatial predictions have to match within both regions. Results The measures of model fit are, on average, 24\% higher for GAMs than for GLMs in both regions. However, the differences between internal and external evaluations (AUC coefficient) are also higher for GAMs than for GLMs (a difference of 30\% for models fitted in Switzerland and 54\% for models fitted in Austria). Transferability, as measured with the AUC evaluation, fails for 68\% of the species in Switzerland and 55\% in Austria for GLMs (respectively for 67\% and 53\% of the species for GAMs). For both GAMs and GLMs, the agreement between internal and external predictions is rather weak on average (Kulczynski's coefficient in the range 0.3\textendash{}0.4), but varies widely among individual species. The dominant pattern is an asymmetrical transferability between the two study regions (a mean decrease of 20\% for the AUC coefficient when the models are transferred from Switzerland and 13\% when they are transferred from Austria). Main conclusions The large inter-specific variability observed among the 54 study species underlines the need to consider more than a few species to test properly the transferability of species distribution models. The pronounced asymmetry in transferability between the two study regions may be due to peculiarities of these regions, such as differences in the ranges of environmental predictors or the varied impact of land-use history, or to species-specific reasons like differential phenotypic plasticity, existence of ecotypes or varied dependence on biotic interactions that are not properly incorporated into niche-based models. The lower variation between internal and external evaluation of GLMs compared to GAMs further suggests that overfitting may reduce transferability. Overall, a limited geographical transferability calls for caution when projecting niche-based models for assessing the fate of species in future environments.},
  author = {Randin, Christophe F. and Dirnb{\"o}ck, Thomas and Dullinger, Stefan and Zimmermann, Niklaus E. and Zappa, Massimiliano and Guisan, Antoine},
  doi = {10.1111/j.1365-2699.2006.01466.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Randin et al_2006_Are niche-based species distribution models transferable in space.pdf;C:\\Users\\julienv\\Zotero\\storage\\VCMGBLSU\\j.1365-2699.2006.01466.html},
  issn = {1365-2699},
  journal = {Journal of Biogeography},
  language = {en},
  number = {10},
  pages = {1689-1703},
  title = {Are Niche-Based Species Distribution Models Transferable in Space?},
  volume = {33},
  year = {2006}
}

@misc{rcoreteam2018,
  address = {Vienna, Austria},
  author = {{R Core Team}},
  howpublished = {R Foundation for Statistical Computing},
  keywords = {R package},
  title = {R: A Language and Environment for Statistical Computing},
  year = {2018}
}

@article{reineking2006,
  abstract = {Predictive habitat models are an important tool for ecological research and conservation. A major cause of unreliable models is excessive model complexity, and regularization methods aim to improve the predictive performance by adequately constraining model complexity. We compare three regularization methods for logistic regression: variable selection, lasso, and ridge. They differ in the way model complexity is measured: variable selection uses the number of estimated parameters, the lasso uses the sum of the absolute values of the parameter estimates, and the ridge uses the sum of the squared values of the parameter estimates. We performed a simulation study with environmental data of a real landscape and artificial species occupancy data. We investigated the effect of three factors on relative model performance: (1) the number of parameters (16, 10, 6, 2) in the 'true' model that determined the distribution of the artificial species, (2) the prevalence, i.e. the proportion of sites occupied by the species, and (3) the sample size (measured in events per variable, EPV). Regularization improved model discrimination and calibration. However, no regularization method performed best under all circumstances: the ridge generally performed best in the 16-parameter scenario. The lasso generally performed best in the 10-parameter scenario. Variable selection with AIC was best at large sample sizes (EPV {$\geq$} 10) when less than half of the variables influenced the species distribution. However, at low sample sizes (EPV {$<$} 10), ridge and lasso always performed best, regardless of the parameter scenario or prevalence. Overall, calibration was best in ridge models. Other methods showed overconfidence, particularly at low sample sizes. The percentage of correctly identified models was low for both lasso and variable selection. Variable selection should be used with caution. Although it can produce the best performing models under certain conditions, these situations are difficult to infer from the data. Ridge and lasso are risk-averse model strategies that can be expected to perform well under a wide range of underlying species\textendash{}habitat relationships, particularly at small sample sizes.},
  author = {Reineking, Bj{\"o}rn and Schr{\"o}der, Boris},
  doi = {10.1016/j.ecolmodel.2005.10.003},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Reineking_Schröder_2006_Constrain to perform.pdf},
  issn = {03043800},
  journal = {Ecological Modelling},
  month = mar,
  number = {3-4},
  pages = {675-690},
  title = {Constrain to Perform: {{Regularization}} of Habitat Models},
  volume = {193},
  year = {2006}
}

@article{renner2013,
  abstract = {Modeling the spatial distribution of a species is a fundamental problem in ecology. A number of modeling methods have been developed, an extremely popular one being MAXENT, a maximum entropy modeling approach. In this article, we show that MAXENT is equivalent to a Poisson regression model and hence is related to a Poisson point process model, differing only in the intercept term, which is scale-dependent in MAXENT. We illustrate a number of improvements to MAXENT that follow from these relations. In particular, a point process model approach facilitates methods for choosing the appropriate spatial resolution, assessing model adequacy, and choosing the LASSO penalty parameter, all currently unavailable to MAXENT. The equivalence result represents a significant step in the unification of the species distribution modeling literature.},
  author = {Renner, I W and Warton, D I},
  doi = {10.1111/j.1541-0420.2012.01824.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Renner_Warton_2013_Equivalence of MAXENT and Poisson point process models for species distribution.pdf},
  journal = {Biometrics},
  keywords = {independence,spatial autocorrelation},
  number = {1},
  pages = {274-281},
  pmid = {23379623},
  title = {Equivalence of {{MAXENT}} and {{Poisson}} Point Process Models for Species Distribution Modeling in Ecology},
  volume = {69},
  year = {2013}
}

@article{renner2015,
  abstract = {* Presence-only data are widely used for species distribution modelling, and point process regression models are a flexible tool that has considerable potential for this problem, when data arise as point events. * In this paper, we review point process models, some of their advantages and some common methods of fitting them to presence-only data. * Advantages include (and are not limited to) clarification of what the response variable is that is modelled; a framework for choosing the number and location of quadrature points (commonly referred to as pseudo-absences or `background points') objectively; clarity of model assumptions and tools for checking them; models to handle spatial dependence between points when it is present; and ways forward regarding difficult issues such as accounting for sampling bias. * Point process models are related to some common approaches to presence-only species distribution modelling, which means that a variety of different software tools can be used to fit these models, including maxent or generalised linear modelling software.},
  author = {Renner, Ian W. and Elith, Jane and Baddeley, Adrian and Fithian, William and Hastie, Trevor and Phillips, Steven J. and Popovic, Gordana and Warton, David I.},
  doi = {10.1111/2041-210X.12352},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Renner et al_2015_Point process models for presence-only analysis.pdf},
  issn = {2041210X},
  journal = {Methods in Ecology and Evolution},
  keywords = {maxent,SDM},
  number = {4},
  pages = {366-379},
  title = {Point Process Models for Presence-Only Analysis},
  volume = {6},
  year = {2015}
}

@article{rydgren2003,
  author = {Rydgren, Knut and {\O}kland, Rune Halvorsen and {\O}kland, Tonje},
  doi = {10.1111/j.1654-1103.2003.tb02220.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Rydgren et al. - 2003 - Species response curves along environmental gradie.pdf;C:\\Users\\julienv\\Zotero\\storage\\FJVMEGFY\\j.1654-1103.2003.tb02220.html},
  issn = {1654-1103},
  journal = {Journal of Vegetation Science},
  language = {en},
  month = dec,
  number = {6},
  pages = {869-880},
  title = {Species Response Curves along Environmental Gradients. {{A}} Case Study from {{SE Norwegian}} Swamp Forests},
  volume = {14},
  year = {2003}
}

@article{searcy2014,
  author = {Searcy, Christopher A and Shaffer, H Bradley},
  doi = {10.1111/ecog.00733},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Searcy_Shaffer_2014_Field validation supports novel niche modeling strategies in a cryptic.pdf},
  journal = {Ecography},
  number = {10},
  pages = {983-992},
  title = {Field Validation Supports Novel Niche Modeling Strategies in a Cryptic Endangered Amphibian},
  volume = {37},
  year = {2014}
}

@article{stoa2018,
  author = {St{\o}a, Bente and Halvorsen, Rune and Mazzoni, Sabrina and Gusarov, Vladimir I.},
  doi = {10.2478/som-2018-0001},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Støa et al_2018_Sampling bias in presence-only data used for species distribution modelling.pdf},
  journal = {Sommerfeltia},
  language = {en},
  month = oct,
  number = {1},
  pages = {1-53},
  shorttitle = {Sampling Bias in Presence-Only Data Used for Species Distribution Modelling},
  title = {Sampling Bias in Presence-Only Data Used for Species Distribution Modelling: Theory and Methods for Detecting Sample Bias and Its Effects on Models},
  volume = {38},
  year = {2018}
}

@article{syfert2013,
  abstract = {Species distribution models (SDMs) trained on presence-only data are frequently used in ecological research and conservation planning. However, users of SDM software are faced with a variety of options, and it is not always obvious how selecting one option over another will affect model performance. Working with MaxEnt software and with tree fern presence data from New Zealand, we assessed whether (a) choosing to correct for geographical sampling bias and (b) using complex environmental response curves have strong effects on goodness of fit. SDMs were trained on tree fern data, obtained from an online biodiversity data portal, with two sources that differed in size and geographical sampling bias: a small, widely-distributed set of herbarium specimens and a large, spatially clustered set of ecological survey records. We attempted to correct for geographical sampling bias by incorporating sampling bias grids in the SDMs, created from all georeferenced vascular plants in the datasets, and explored model complexity issues by fitting a wide variety of environmental response curves (known as ``feature types'' in MaxEnt). In each case, goodness of fit was assessed by comparing predicted range maps with tree fern presences and absences using an independent national dataset to validate the SDMs. We found that correcting for geographical sampling bias led to major improvements in goodness of fit, but did not entirely resolve the problem: predictions made with clustered ecological data were inferior to those made with the herbarium dataset, even after sampling bias correction. We also found that the choice of feature type had negligible effects on predictive performance, indicating that simple feature types may be sufficient once sampling bias is accounted for. Our study emphasizes the importance of reducing geographical sampling bias, where possible, in datasets used to train SDMs, and the effectiveness and essentialness of sampling bias correction within MaxEnt.},
  author = {Syfert, Mindy M. and Smith, Matthew J. and Coomes, David A.},
  doi = {10.1371/journal.pone.0055158},
  editor = {Roberts, David L.},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Syfert et al_2013_The Effects of Sampling Bias and Model Complexity on the Predictive Performance.pdf},
  issn = {1932-6203},
  journal = {PLoS ONE},
  month = feb,
  number = {2},
  pages = {e55158},
  title = {The {{Effects}} of {{Sampling Bias}} and {{Model Complexity}} on the {{Predictive Performance}} of {{MaxEnt Species Distribution Models}}},
  volume = {8},
  year = {2013}
}

@article{tibshirani1996,
  author = {Tibshirani, Robert},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Tibshirani_1996_Regression shrinkage and selection via the lasso.pdf},
  journal = {Journal of the Royal Statistical Society. Series B (Methodological)},
  number = {1},
  pages = {267-288},
  title = {Regression Shrinkage and Selection via the Lasso},
  volume = {58},
  year = {1996}
}

@article{varela2014a,
  abstract = {Citizen science projects store an enormous amount of information about species distribution, diversity and characteristics. Researchers are now beginning to make use of this rich collection of data. However, access to these databases is not always straightforward. Apart from the largest and international projects, citizen science repositories often lack specific Application Programming Interfaces (APIs) to connect them to the scientific environments. Thus, it is necessary to develop simple routines to allow researchers to take advantage of the information collected by smaller citizen science projects, for instance, programming specific packages to connect them to popular scientific environments (like R). Here, we present rAvis, an R-package to connect R-users with Proyecto AVIS (http://proyectoavis.com), a Spanish citizen science project with more than 82,000 bird observation records. We develop several functions to explore the database, to plot the geographic distribution of the species occurrences, and to generate personal queries to the database about species occurrences (number of individuals, distribution, etc.) and birdwatcher observations (number of species recorded by each collaborator, UTMs visited, etc.). This new R-package will allow scientists to access this database and to exploit the information generated by Spanish birdwatchers over the last 40 years.},
  author = {Varela, Sara and {Gonz{\'a}lez-Hern{\'a}ndez}, Javier and Casabella, Eduardo and Barrientos, Rafael},
  doi = {10.1371/journal.pone.0091650},
  editor = {White, Stephanie Ann},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Varela et al_2014_rAvis.pdf},
  issn = {1932-6203},
  journal = {PLoS ONE},
  keywords = {R package,software note},
  month = mar,
  number = {3},
  pages = {e91650},
  title = {{{rAvis}}: {{An R}}-{{Package}} for {{Downloading Information Stored}} in {{Proyecto AVIS}}, a {{Citizen Science Bird Project}}},
  volume = {9},
  year = {2014}
}

@article{veloz2009,
  author = {Veloz, Samuel D},
  doi = {10.1111/j.1365-2699.2009.02174.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Veloz_2009_Spatially autocorrelated sampling falsely inflates measures of accuracy for.pdf},
  journal = {Journal of Biogeography},
  number = {12},
  pages = {2290-2299},
  title = {Spatially Autocorrelated Sampling Falsely Inflates Measures of Accuracy for Presence-Only Niche Models},
  volume = {36},
  year = {2009}
}

@misc{vollering2018,
  author = {Vollering, Julien and Mazzoni, Sabrina and Halvorsen, Rune},
  copyright = {All rights reserved},
  keywords = {maxent,R package},
  title = {{{MIAmaxent}}: A Modular, Integrated Approach to Maximum Entropy Distribution Modeling},
  year = {2018}
}

@article{warren2011,
  author = {Warren, Dan L and Seifert, Stephanie N},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Warren_Seifert_2011_Ecological niche modeling in Maxent.pdf},
  journal = {Ecological Applications},
  number = {2},
  pages = {335-342},
  title = {Ecological Niche Modeling in {{Maxent}}: The Importance of Model Complexity and the Performance of Model Selection Criteria},
  volume = {21},
  year = {2011}
}

@article{warren2012,
  abstract = {There is a growing awareness of problems with the estimation of the ecological tolerances of species through correlative modeling approaches. These problems have led some investigators to argue for abandoning terms such as `ecological niche model' and `environmental niche model' in favor of the ostensibly more value-neutral `species distribution model', as the models are thought to frequently be poor estimators of the niche. Here, I argue that most applications to which these models are put require the assumption that they do estimate the niche, however imperfectly, and that obscuring this inescapable and potentially flawed assumption in the terminology may only serve to hinder the development of the field.},
  author = {Warren, Dan L.},
  doi = {10.1016/J.TREE.2012.03.010},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Warren_2012_In defense of ‘niche modeling’.pdf},
  issn = {0169-5347},
  journal = {Trends in Ecology \& Evolution},
  month = sep,
  number = {9},
  pages = {497-500},
  title = {In Defense of `Niche Modeling'},
  volume = {27},
  year = {2012}
}

@article{warton2010,
  author = {Warton, David I. and Shepherd, Leah C.},
  doi = {10.1214/10-AOAS331},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Warton_Shepherd_2010_Poisson point process models solve the “pseudo-absence problem” for.pdf},
  issn = {1932-6157},
  journal = {The Annals of Applied Statistics},
  language = {en},
  month = sep,
  number = {3},
  pages = {1383-1402},
  title = {Poisson Point Process Models Solve the ``Pseudo-Absence Problem'' for Presence-Only Data in Ecology},
  volume = {4},
  year = {2010}
}

@article{wenger2012,
  abstract = {1. Ecologists have long sought to distinguish relationships that are general from those that are idiosyncratic to a narrow range of conditions. Conventional methods of model validation and selection assess in- or out-of-sample prediction accuracy but do not assess model generality or transferability, which can lead to overestimates of performance when predicting in other locations, time periods or data sets. 2. We propose an intuitive method for evaluating transferability based on techniques currently in use in the area of species distribution modelling. The method involves cross-validation in which data are assigned non-randomly to groups that are spatially, temporally or otherwise distinct, thus using heterogeneity in the data set as a surrogate for heterogeneity among data sets. 3. We illustrate the method by applying it to distribution modelling of brook trout (Salvelinus fontinalis Mitchill) and brown trout (Salmo trutta Linnaeus) in western United States. We show that machine-learning techniques such as random forests and artificial neural networks can produce models with excellent in-sample performance but poor transferability, unless complexity is constrained. In our example, traditional linear models have greater transferability. 4. We recommend the use of a transferability assessment whenever there is interest in making inferences beyond the data set used for model fitting. Such an assessment can be used both for validation and for model selection and provides important information beyond what can be learned from conventional validation and selection techniques.},
  author = {Wenger, Seth J. and Olden, Julian D.},
  copyright = {\textcopyright{} 2012 The Authors. Methods in Ecology and Evolution \textcopyright{} 2012 British Ecological Society},
  doi = {10.1111/j.2041-210X.2011.00170.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Wenger_Olden_2012_Assessing transferability of ecological models.pdf;C:\\Users\\julienv\\Zotero\\storage\\6WF2A76X\\j.2041-210X.2011.00170.html},
  issn = {2041-210X},
  journal = {Methods in Ecology and Evolution},
  keywords = {cross-validation,Favorite,generality,niche model,performance,species distribution model,statistical},
  language = {en},
  number = {2},
  pages = {260-267},
  shorttitle = {Assessing Transferability of Ecological Models},
  title = {Assessing Transferability of Ecological Models: An Underappreciated Aspect of Statistical Validation},
  volume = {3},
  year = {2012}
}

@article{west2016,
  abstract = {Accurate and reliable predictions of invasive species distributions are urgently needed by land managers for developing management plans and monitoring new potential areas of establishment. Presence-only species distribution models are commonly used in these evaluations, however they are rarely tested with independent data over time or compared with presence-absence models fit with the same presence data. Using Maxent, we developed a presence-only model of invasive cheatgrass (Bromus tectorum L.) distribution in Rocky Mountain National Park, Colorado, USA in 2007 fit with limited data, and then tested the model with independent presence and absence data collected between 2008 and 2013. This model was verified using threshold dependent and threshold independent evaluation metrics. Next, we developed a Maxent model with cheatgrass presence data from 2007 through 2013 (i.e. Maxent 2013), and compared this model to a presence-absence method (i.e., generalized linear model; GLM 2013) using the same data. Threshold dependent and threshold independent evaluation metrics suggested Maxent 2013 outperformed GLM 2013, and a two-tailed Wilcoxon signed rank test indicated relative probability outputs were not significantly different between the models in geographic space. Based on known presences and absences of cheatgrass collected in the field, the Maxent 2013 and GLM 2013 relative probability outputs were highly correlated at absence locations but less correlated at presence locations. A Kappa comparison of Maxent 2007 and Maxent 2013 binary output provides evidence that Maxent is robust when fit with limited data. Our results indicate Maxent is an appropriate model for use when land management objectives are supported by limited resources and thus require a conservative, but highly accurate estimate of habitat suitability for invasive species on the landscape.},
  author = {West, Amanda M. and Kumar, Sunil and Brown, Cynthia S and Stohlgren, Thomas J. and Bromberg, Jim},
  doi = {10.1016/J.ECOINF.2016.11.001},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\West et al_2016_Field validation of an invasive species Maxent model.pdf},
  issn = {1574-9541},
  journal = {Ecological Informatics},
  month = nov,
  pages = {126-134},
  title = {Field Validation of an Invasive Species {{Maxent}} Model},
  volume = {36},
  year = {2016}
}

@article{whittingham2006,
  abstract = {1 The biases and shortcomings of stepwise multiple regression are well established within the statistical literature. However, an examination of papers published in 2004 by three leading ecological and behavioural journals suggested that the use of this technique remains widespread: of 65 papers in which a multiple regression approach was used, 57\% of studies used a stepwise procedure. 2 The principal drawbacks of stepwise multiple regression include bias in parameter estimation, inconsistencies among model selection algorithms, an inherent (but often overlooked) problem of multiple hypothesis testing, and an inappropriate focus or reliance on a single best model. We discuss each of these issues with examples. 3 We use a worked example of data on yellowhammer distribution collected over 4 years to highlight the pitfalls of stepwise regression. We show that stepwise regression allows models containing significant predictors to be obtained from each year's data. In spite of the significance of the selected models, they vary substantially between years and suggest patterns that are at odds with those determined by analysing the full, 4-year data set. 4 An information theoretic (IT) analysis of the yellowhammer data set illustrates why the varying outcomes of stepwise analyses arise. In particular, the IT approach identifies large numbers of competing models that could describe the data equally well, showing that no one model should be relied upon for inference.},
  author = {Whittingham, Mark J. and Stephens, Philip A. and Bradbury, Richard B. and Freckleton, Robert P.},
  doi = {10.1111/j.1365-2656.2006.01141.x},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Whittingham et al_2006_Why do we still use stepwise modelling in ecology and behaviour.pdf;C:\\Users\\julienv\\Zotero\\storage\\GKTZKIWE\\j.1365-2656.2006.01141.html},
  issn = {1365-2656},
  journal = {Journal of Animal Ecology},
  language = {en},
  number = {5},
  pages = {1182-1189},
  title = {Why Do We Still Use Stepwise Modelling in Ecology and Behaviour?},
  volume = {75},
  year = {2006}
}

@article{wiegand2010,
  abstract = {Some research studies in the medical literature use multiple stepwise variable selection (SVS) algorithms to build multivariable models. The purpose of this study is to determine whether the use of multiple SVS algorithms in tandem (stepwise agreement) is a valid variable selection procedure. Computer simulations were developed to address stepwise agreement. Three popular SVS algorithms were tested (backward elimination, forward selection, and stepwise) on three statistical methods (linear, logistic, and Cox proportional hazards regression). Other simulation parameters explored were the sample size, number of predictors considered, degree of correlation between pairs of predictors, p-value-based entrance and exit criteria, predictor type (normally distributed or binary), and differences between stepwise agreement between any two or all three algorithms. Among stepwise methods, the rate of agreement, agreement on a model including only those predictors truly associated with the outcome, and agreement on a model containing the predictors truly associated with the outcome were measured. These rates were dependent on all simulation parameters. Mostly, the SVS algorithms agreed on a final model, but rarely on a model with only the true predictors. Sample size and candidate predictor pool size are the most influential simulation conditions. To conclude, stepwise agreement is often a poor strategy that gives misleading results and researchers should avoid using multiple SVS algorithms to build multivariable models. More research on the relationship between sample size and variable selection is needed. Published in 2010 by John Wiley \& Sons, Ltd.},
  author = {Wiegand, Ryan E.},
  copyright = {This article is a U.S. Government work and is in the public domain in the U.S.A. Published in 2010 by John Wiley \& Sons, Ltd.},
  doi = {10.1002/sim.3943},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Wiegand_2010_Performance of using multiple stepwise algorithms for variable selection.pdf;C:\\Users\\julienv\\Zotero\\storage\\S4S9GRQ5\\sim.html},
  issn = {1097-0258},
  journal = {Statistics in Medicine},
  language = {en},
  number = {15},
  pages = {1647-1659},
  title = {Performance of Using Multiple Stepwise Algorithms for Variable Selection},
  volume = {29},
  year = {2010}
}

@article{yackulic2013,
  author = {Yackulic, Charles B and Chandler, Richard and Zipkin, Elise F and Royle, J Andrew and Nichols, James D and Campbell Grant, Evan H and Veran, Sophie and O'Hara, Robert B},
  doi = {10.1111/2041-210x.12004},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Yackulic et al_2013_Presence-only modelling using MAXENT.pdf},
  journal = {Methods in Ecology and Evolution},
  number = {3},
  pages = {236-243},
  title = {Presence-Only Modelling Using {{MAXENT}}: When Can We Trust the Inferences?},
  volume = {4},
  year = {2013}
}

@book{zuur2007,
  author = {Zuur, A F and Ieno, E N and Smith, G M},
  file = {C:\\Users\\julienv\\Google Drive\\ZoteroLibrary\\Zuur et al_2007_Analysing Ecological Data.pdf},
  isbn = {10: 0-387-45967-7},
  keywords = {Favorite},
  publisher = {{Springer Science \& Business Media}},
  series = {Statistics for Biology and Health},
  title = {Analysing {{Ecological Data}}},
  year = {2007}
}