Effects of sample size on accuracy of species distribution models

By David R. B. Stockwell, A. Townsend Peterson. 2002

1,177 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/404dbbe1227e028b778335996e92ba74112425c2

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

Using species distribution models in paleobiogeography: A matter of data, predictors and concepts (2011, methodology): “The accuracy of SDM predictions also relies on the explanatory variables used to build the models (Stockwell and Peterson, 2002).”
Predicting species distribution at range margins: testing the effects of study area extent, resolution and threshold selection in the Sahara–Sahel transition zone (2014, background): “These results also support that widespread species may show regional ecological adaptations, resulting in different habitat preferences in discrete parts of the species’ range (Stockwell & Peterson, 2002).”
Selecting thresholds of occurrence in the prediction of species distributions (2005, background): “A representative in the first category is taking 0.5 as the threshold, which is widely used in ecology (Manel et al. 1999, 2001, Luck 2002, Stockwell and Peterson 2002, Bailey et al. 2002, Woolf et al. 2002).”
Sample size matters: investigating the effect of sample size on a logistic regression susceptibility model for debris flows (2014, positive): “Such inclusion of more information does not necessarily lead to better model performance;Stockwell and Townsend Peterson(2002) describes “plateaus” wherein new data add little to model performance.”
The biogeography of prediction error: why does the introduced range of the fire ant over-predict its native range? (2006, methodology): “The first group ( t 1 ) encompassed the first 20 years of fire ant invasion (i.e. 1930–52) and included 30 US counties, a minimum number of locations required for robust predictions in GARP (Stockwell & Peterson, 2002).”
A null‐model for significance testing of presence‐only species distribution models (2007, background): “The number of records by which a species is represented in herbaria and natural history museums range from 1 to 150200 records ( Stockwell and Peterson 2002 ).”
Distribution modelling and statistical phylogeography: an integrative framework for generating and testing alternative biogeographical hypotheses (2007, background): “For example, generating an accurate projection of a species’ distribution typically requires samples from at least 20 localities (Stockwell & Peterson, 2002; but see Pearson et al., 2007).”
Rethinking receiver operating characteristic analysis applications in ecological niche modeling (2008, methodology): “Many facets of these tools and heir application have been examined in detailed analyses Stockwell and Peterson, 2002a,b, 2003; Anderson et al., 2003; earson and Dawson, 2003; Araújo et al., 2005a,b; Guisan and huiller, 2005; Guisan et al., 2006; Pearson et al., 2007) that have reatly clarified the…”
Minimum required number of specimen records to develop accurate species distribution models (2016, background): “Model performance is known to rapidly decrease for sample sizes smaller than 20 (Stockwell and Peterson 2002) or 15 (Papeş and Gaubert 2007), and is dramatically poor for samples sizes smaller than 5 records (Pearson et al.”
Risk Analysis and Bioeconomics of Invasive Species to Inform Policy and Management (2016, background): “In some cases, reliable models may be fit to as few as 20 observations, if those points are sufficiently representative of the niche range (123).”
Improving generalized regression analysis for the spatial prediction of forest communities (2006, background): “Sample size (Pearce & Ferrier, 2000; Stockwell & Peterson, 2002) and species prevalence (frequency of species’ occurrences in the data set: Fielding & Bell, 1997; Manel et al., 1999b, 2001) are known to have a major impact on model performance.”
Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS) (2006, methodology): “An alternative approach to obtaining a validation set with effective prevalence of 50% is by random resampling from the data available ( Stockwell & Peterson 2002 ).”