Here be dragons: a tool for quantifying novelty due to covariate range and correlation change when projecting species distribution models

By M. Mesgaran, R. Cousens, B. L. Webber. 2014

228 citations recorded by the source corpus.

Correlative species distribution models (SDMs) often involve some degree of projection into novel covariate space (i.e. extrapolation), because calibration data may not encompass the entire space of interest. Most methods for identifying extrapolation focus on the range of each model covariate individually. However, extrapolation can occur that is well within the range of univariate variation, but which exhibits novel combinations between covariates. Our objective was to develop a tool that can detect, distinguish and quantify these two types of novelties: novel univariate range and novel combinations of covariates.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/1b99829f1df39f7d472f4ce44634c713513b996e

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 Nature and Landscape Conservation

Citations in the corpus

Bunching up the background betters bias in species distribution models (2019, methodology): “The comparison was performed multivariately using Mahalanobis distance, which takes into account the correlation structure between variables (Mesgaran et al. 2014).”
The potential global distribution of the brown marmorated stink bug, Halyomorpha halys, a critical threat to plant biosecurity (2017, unknown): “To deal with this, software tools have been developed to identify where correlative models are being projected into novel environments (Elith et al. 2011; Mesgaran et al. 2014).”
Predicting into unknown space? Estimating the area of applicability of spatial prediction models (2020, background): “…al. 2016, Toplak40 et al. 2014, Gadaleta et al. 2016, where the concept is usually referred to as "domain of applicability") and have been41 addressed in the field of species distribution modelling (Elith et al. 2010, Mesgaran et al. 2014, Zurell et al. 2012) and42 soil mapping (Zhu et al. 2015).”
Outstanding Challenges in the Transferability of Ecological Models. (2018, background): “, [42,48]), and these can help assess the potential impacts of non-analog conditions on predictive performance.”
Predicting distributions of invasive species (2013, background): “Mapping novel environments (Elith et al., 2010; Mesgaran et al., 2014; Williams et al., 2007) helps interpretation of model output and guides users as to where predictions may be highly unreliable.”
ntbox: An r package with graphical user interface for modelling and evaluating multidimensional ecological niches (2020, unknown): “…distribu - tion modelling: NicheA (Qiao et al., 2016), kuenm (Cobos, Peterson, Barve, & Osorio-Olvera, 2019), Wallace (Kass et al., 2018), ENMeval (Muscarella et al., 2014), biomod2 (Thuiller, Lafourcade, Engler, & Araújo, 2009), dismo (Hijmans, Phillips, Leathwick, & Elith, 2019), sdm (Naimi &…”
A standard protocol for reporting species distribution models (2020, background): “Authors should specify exactly how novelty was defined and quantified; for example novel environments along single environmental gradients (Elith et al. 2010) or novel combinations of environments (Zurell et al. 2012, Mesgaran et al. 2014).”