Generalized models vs. classification tree analysis: Predicting spatial distributions of plant species at different scales

By Wilfried Thuiller, Miguel B. Araújo, S. Lavorel. 2003

354 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/bd7230d9ded81a34517c30b91b63edfd46820163

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

Effect of species rarity on the accuracy of species distribution models for reptiles and amphibians in southern California (2009, result): “Although there was a trend that has also been noted in other studies, with GAMs performing best, followed by GLMs and then CTs (e.g. Thuiller et al., 2003), the magnitude of the difference was small.”
Evaluation of consensus methods in predictive species distribution modelling (2009, methodology): “However, models based on continuous curves (e.g. GLM and GAM) may be more reliable than those based on classification trees (Thuiller et al., 2004) in order to transfer the calibrated models into new areas or time periods via extrapolation modelling (where inputs may be outside the range within…”
Predicting species distributions: a critical comparison of the most common statistical models using artificial species (2007, result): “As with other comparative studies (Franklin, 1998; Elith & Burgman, 2002; Thuiller et al., 2003; Segurado & Araújo, 2004), these two modelling strategies seem to represent a good trade-off between model complexity and performance for a diverse set of species.”
Potential effects of climate change on ecosystem and tree species distribution in British Columbia. (2006, methodology): “…employed for bioclimate envelope modeling include generalized linear models (Bakkenes et al. 2002), generalized additive models (Frescino et al. 2001), regression tree analysis (Iverson and Prasad 1998, Thuiller et al. 2003b), and artificial neural networks (Berry et al. 2002, Pearson et al. 2002).”
Patterns and uncertainties of species' range shifts under climate change (2004, methodology): “%) and tested on the remaining data sets with both the receiver–operating characteristic (ROC) curve and the k statistic (Pearce & Ferrier, 2000; Thuiller et al., 2003).”
Dispersal will limit ability of mammals to track climate change in the Western Hemisphere (2012, background): “Incorporating vegetation or soil types as predictor variables or using alternate bioclimatic models would likely result in different projections of species distributions and possibly in greater or lesser velocities of climate change (24, 25).”
Effects of restricting environmental range of data to project current and future species distributions (2004, methodology): “This is mostly due to restrictions in data accessibility and sampling strategies, to geographic constraints (species at the edge of ranges) or both, preventing the correct capture of the entire environmental range of species (Thuiller et al. 2003b).”
Soil nutritional factors improve models of plant species distribution: an illustration with Acer campestre (L.) in France (2006, background): “Indeed, with the development of geographical information systems (GIS), access to climatic data layers interpolated from meteorological station measurements is facilitated (McKenzie & Halpern, 1999; McKenzie et al., 2003a,b; Thuiller et al., 2003b, 2004).”
Response of British lichens to climate change scenarios: Trends and uncertainties in the projected impact for contrasting biogeographic groups (2007, background): “…the bioclimatic envelope approach have been largely methodological, and have tested variations in statistical design and modelling (Thuiller, 2003; Thuiller et al., 2003; Segurado and Araújo, 2004; Luoto et al., 2005; Thuiller et al., 2004; Araújo et al., 2005a; Araújo and Guisan, 2006;…”
The performance of state-of-the-art modelling techniques depends on geographical distribution of species. (2009, methodology): “…variables, and may then be used to project spatial shifts in species climatic envelopes according to selected climate change scenarios (Bakkenes et al., 2002; Beaumont and Hughes, 2002; Berry et al., 2002; Pearson and Dawson, 2003; Thuiller, 2003; Huntley et al., 2004; Thuiller et al., 2004a,b).”
Can niche-based distribution models outperform spatial interpolation? (2007, background): “At the coarser scale at which our study was set such direct requirements give way to more indirect, large-scale variables such as climate (Thuiller et al., 2003).”
Vulnerability of African mammals to anthropogenic climate change under conservative land transformation assumptions (2006, methodology): “%) and a stepwise selection methodology, with the most parsimonious model being selected using the Akaike information criterion (AIC) (e.g. Thuiller et al., 2003, 2004a).”