Effects of sample survey design on the accuracy of classification tree models in species distribution models

By THOMAS C. EDWARDS JR., D. R. Cutler, N. E. Zimmermann, L. Geiser, Gretchen G. Moisen. 2006

150 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/b915b994d90122c6919fdc102a9a2ecb84ba5ff1

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

Habitat suitability modelling of an invasive plant with advanced remote sensing data (2009, background): “Habitat suitability predictor variables derived from remote sensing data include land-use/land-cover/habitat classifications (e.g. Higgins et al., 1999; Newbold & Eadie, 2004; Gutierréz et al., 2005; Edwards et al., 2006; Thompson et al., 2006; Davis et al., 2007; Foxcroft et al., 2007), canopy cove”
A strict maximum likelihood explanation of MaxEnt, and some implications for distribution modelling (2013, methodology): “responses to the environment and distributions in geographical and environmental variables spaces (Elith et al. 2011). Accordingly, choosing the optimal combination of MaxEnt options and settings when the distribution of a specific modelled target is to be modelled for a specific purpose, often by u”
Effects of the number of presences on reliability and stability of MARS species distribution models: the importance of regional niche variation and ecological heterogeneity (2010, background): “We consider that this unexpected result is mostly due to errors that occurred when geolocating specimens in natural history collections without full locality information (Margules & Pressey 2000; Soberon & Peterson 2004; Rowe 2005; Edwards et al. 2006; Papeş & Gaubert 2007), and also perhaps to unce”
Comparing sample bias correction methods for species distribution modeling using virtual species (2021, background): “Random or stratified-random sampling designs can be used to collect observations in a manner free of sampling bias (Edwards et al. 2006), but the high cost of conducting these surveys often precludes their use and has resulted in the reliance on widely available biodiversity observations from…”
A comparison between Ensemble and MaxEnt species distribution modelling approaches for conservation: A case study with Egyptian medicinal plants (2020, methodology): “Suggested sources of uncertainty in predictions include the SDM method, choice of predictor variables, spatial distribution and sample size, all of which can impact model accuracy and predictive power (Austin and Niel, 2011; Buisson et al., 2010; Edwards et al., 2006; Segurado and Araújo, 2004).”
Sampling in ecology and evolution – bridging the gap between theory and practice (2010, background): “How the sampling design may influence the patterns revealed from the sampled data is rarely discussed in ecological studies, where the focus is generally on the estimation of simple parameters such as population averages, size, or measures of diversity (Yoccoz et al. 2001, Edwards et al. 2006).”
Modelling and predicting fungal distribution patterns using herbarium data (2008, background): “If recorded presences do not adequately represent the actual distribution of a species, it is not surprising that the quality of distribution models is affected (Edwards et al., 2006).”
Effect of species rarity on the accuracy of species distribution models for reptiles and amphibians in southern California (2009, methodology): “Although these data were not collected using a spatial sampling scheme that would be considered ideal for predictive modelling (Fortin et al., 1989; Edwards et al., 2006), often data collected for another purpose are used for SDM because they are the best data available on regional species distribut”
Evaluation of consensus methods in predictive species distribution modelling (2009, background): “Such differences in final model variables may be a function of underlying design and model form (Edwards et al., 2006), because different methods combine responses to individual predictors in different ways (Elith et al., 2005).”
Random forests as a tool for ecohydrological distribution modelling (2007, background): “Furthermore, model generality should be tested on a spatially independent data set since the use of accuracy estimates based on two-fold cross-validation data and on spatially independent evaluation data tend to differ (Edwards et al., 2006).”
Impact of training and validation sample selection on classification accuracy and accuracy assessment when using reference polygons in object-based classification (2013, methodology): “The protocol used to select the reference polygons determines whether the advantage of rigorous statistical inference based on having implemented a probability sampling design is available (Edwards et al. 2006; Stehman and Foody 2009).”
Mapping Species Distributions with MAXENT Using a Geographically Biased Sample of Presence Data: A Performance Assessment of Methods for Correcting Sampling Bias (2014, background): “Several studies questioned the effect of sampling design [38], or the biased nature of museum and herbarium datasets [39] on the predictive performance of SDMs.”
Uncertainty associated with survey design in Species Distribution Models (2014, background): “In fact, the design of the surveys can have a large impact on SDM performance (Hirzel & Guisan, 2002; Edwards et al., 2006; Albert et al., 2010; Braunisch & Suchant, 2010).”