Presence‐only modelling using MAXENT: when can we trust the inferences?

By C. Yackulic, R. Chandler, Elise F. Zipkin, J. Andrew Royle, James D. Nichols, E. Grant, S. Véran. 2013

762 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Evolution, Behavior and Systematics, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/618c29d685b65b51231f3b6a534014005fff60de

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecological Modeling

In Ecology, Evolution, Behavior and Systematics

In Nature and Landscape Conservation

Citations in the corpus

The importance of correcting for sampling bias in MaxEnt species distribution models (2013, background): “A recent article by Yackulic et al. (2013) revealed that 87% of MaxEnt models were based on data that were likely to suffer from sample selection bias.”
Climate change, genetic markers and species distribution modelling (2015, neutral): “Phillips et al. (2006) were careful to highlight this distinction, but unfortunately, the scaled index that MaxEnt creates ('logistic output') has almost always been interpreted by the user community as the probability of occurrence (Yackulic et al., 2013).”
Paintings predict the distribution of species, or the challenge of selecting environmental predictors and evaluation statistics (2018, background): “However, owing to the many potential issues that may arise when settings and inputs are not adequately chosen (Merow, Smith, & Silander, 2013; Yackulic et al., 2013), such a simplistic approach can be considered as bad practice.”
MaxEnt versus MaxLike: empirical comparisons with ant species distributions (2013, background): “More importantly, for most species, the relative suitability index estimated by MaxEnt often was poorly correlated with the probability of occurrence estimated by MaxLike, suggesting that the two methods are estimating different quantities.”
The area under the precision‐recall curve as a performance metric for rare binary events (2019, background): “All performance metric values are contingent on the dataset with which they were computed, and comparison of the value of performance metrics across datasets that diverge in attributes – especially species prevalence – is generally subjective (Yackulic et al., 2012; Lawson et al., 2014).”
A Systematic Review of Marine-Based Species Distribution Models (SDMs) with Recommendations for Best Practice (2017, background): “While in many studies, this is a pragmatic solution, buildingmodels using these data can violate assumptions of some of the models (Yackulic et al., 2013).”
A comparison between Ensemble and MaxEnt species distribution modelling approaches for conservation: A case study with Egyptian medicinal plants (2020, methodology): “MaxEnt models have been critised for being incorrectly applied and oversimplified (Morales et al., 2017; Yackulic et al., 2013), and to have produced overfitting models (Halvorsen, 2013; Merckx et al., 2011).”
A comparison of Maxlike and Maxent for modelling species distributions (2014, background): “Substantial challenges exist in estimating sampling effort and detection probability for PO data (Dorazio 2012; Yackulic et al. 2012).”