Novel methods improve prediction of species' distributions from occurrence data

By Jane Elith, Catherine H. Graham, R. P. Anderson, Miroslav Dud'ik, Simon Ferrier, Antoine Guisan, Robert J. Hijmans, Falk Huettmann, J. Leathwick, Anthony Lehmann and 17 more. 2006

7,428 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/3aa1b70fdc97ae96091c5fb39cd911015ac5253e

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

Modelling geographic distribution and detecting conservation gaps in Italy for the threatened beetle Rosalia alpina (2013, positive): “This approach is especially valuable for species whose available presence data are limited and false absence in surveys is a significant risk (Elith et al. 2006; Phillips & Dudik 2008).”
Here be dragons: a tool for quantifying novelty due to covariate range and correlation change when projecting species distribution models (2014, neutral): “For example, correlation coefficients as high as Pearson’s r = 0.8 are common in the literature (Elith et al., 2006, 2010), and indeed, (multi)collinearity is considered as a serious challenge to most of the correlative modelling methods (Graham, 2003; Freckleton, 2011; Dormann et al., 2013).”
Is there a correlation between abundance and environmental suitability derived from ecological niche modelling? A meta‐analysis (2017, positive): “In general, this method performs better than statistical and similarity methods in predicting species distributions (Elith et al. 2006).”
Can species distribution modelling provide estimates of population densities? A case study with jaguars in the Neotropics (2012, positive): “Taking into consideration the variation among SDM methods now available and the recent discussions about this topic (e.g. Guisan & Zimmermann, 2000; Elith et al., 2006; Austin, 2007; Elith & Graham, 2009; Elith & Leathwick, 2009), it is recommended to apply more than one of these methods and…”
Environmental influences on the occurrence of coastal sharks in estuarine waters (2010, positive): “…relationships and often has superior predictive performance to other techniques such as generalized linear and additive models that are often used to model species-habitat relationships (Elith et al. 2006, Leathwick et al. 2006, Elith et al. 2008, Leathwick et al. 2008, Parisien & Moritz 2009).”
Use of Maximum Entropy Modeling in Wildlife Research (2009, neutral): “Kappa [3,26,27] and jackknife statistics [16,20,27] are also used to test models.”
Insights into the area under the receiver operating characteristic curve (AUC) as a discrimination measure in species distribution modelling (2012, neutral): “In the same way, background data have also been used with classical presence–absence techniques such as generalized linear models (e.g. Elith et al., 2006).”
Selecting thresholds for the prediction of species occurrence with presence‐only data (2013, neutral): “Numerous modelling techniques have been used to predict species distributions (e.g. Segurado & Ara ujo, 2004; Elith et al., 2006; Thuiller et al., 2009).”
Hotspots of plant invasion predicted by propagule pressure and ecosystem characteristics (2011, positive): “BRTs have been shown to perform well for both explanatory and predictive models (Elith et al., 2006) and are now being applied to a variety of ecological problems (Leathwick et al., 2008; Vesk et al., 2010).”
Mapping the Global Distribution of Livestock (2014, positive): “New machine-learning methods such as Boosted Regression Trees, or Random Forests have recently demonstrated their increased predictive capacities over more standard statistical methods for species distribution modelling [66].”
Feeding habitat of the whale shark Rhincodon typus in the northern Gulf of Mexico determined using species distribution modelling (2012, positive): “MaxEnt is a machine-learning technique that attempts to fit a probability distribution of species occurrence over the entire study area using a complex suite of transformations on the environmental variables believed to be important to the target species (Phillips et al. 2006, Elith et al. 2006).”
A statistical explanation of MaxEnt for ecologists (2011, positive): “MaxEnt’s predictive performance is consistently competitive with the highest performing methods (Elith et al., 2006).”