Species-specific tuning increases robustness to sampling bias in models of species distributions: An implementation with Maxent

By R. P. Anderson, Israel Gonzalez. 2011

563 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Genetics

View Paper: https://www.semanticscholar.org/paper/902dc65fd432d3eb6e3fecaed5dcfa4cb1355db3

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

In Genetics

Citations in the corpus

SDMtoolbox 2.0: the next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses (2017, unknown): “This regularization imposes a penalty for each term included in the model and for higher 190 weights given to a term (Phillips et al., 2006; Anderson & Gonzalez, 2011).”
A strict maximum likelihood explanation of MaxEnt, and some implications for distribution modelling (2013, methodology): “Model selection by ℓ1-regularisation (the lasso) to prevent overfitting This standard MaxEnt practice is followed by a large majority of users of Maxent software and has only occasionally been questioned [but see Anderson & Gonzalez (2011) and Warren & Seifert (2011)].”
A review of evidence about use and performance of species distribution modelling ensembles like BIOMOD (2019, background): “…would expect model tuning to be import‐ ant because: (a) there is strong evidence that model tuning affects performance of individual models (e.g., Anderson & Gonzalez, 2011); and (b) it has been repeatedly suggested that ensemble model per‐ formance is dependent on the quality of individual…”
Spatial filtering to reduce sampling bias can improve the performance of ecological niche models (2014, background): “Maxent has performed well in comparison with other techiques and is commonly used (Elith et al., 2006; Wisz et al., 2008) ut sensitive to sampling biases (Anderson and Gonzalez, 2011; hillips et al., 2009).”
Protected areas network is not adequate to protect a critically endangered East Africa Chelonian: Modelling distribution of pancake tortoise, Malacochersus tornieri under current and future climates (2020, background): “…class (linear, quadratic, etc.; Phillips and Dudik 2008) but can be tuned by multiplying it by a user-specified constant to amplify or dampen its effect to reflect the desired level of confidence and produce more or less complex models, respectively (Anderson and Gonzalez 2011, Elith et al. 2011).”
The challenge of modeling niches and distributions for data‐poor species: a comprehensive approach to model complexity (2018, methodology): “…Elith et al. 2010) has been shown to result in simpler and substantially more realistic Maxent models than those built using default settings (Anderson and Gonzalez 2011, Warren and Seifert 2011, Jueterbock et al. 2013, Muscarella et al. 2014, Radosavljevic and Anderson 2014, Warren et al.…”
Estimating optimal complexity for ecological niche models: A jackknife approach for species with small sample sizes (2013, methodology): “In this aspect, we followed Anderson and Gonzalez 2011) (see Elith et al. (2010), Warren and Seifert (2011), adosavljevic and Anderson (in press) for tuning Maxent models by arying only regularization and Syfert et al. (2013) for tuning Maxnt models by varying only feature class).”