Effects of sample size on the performance of species distribution models

By M. Wisz, Robert J. Hijmans, Jin Li, A. Townsend Peterson, Catherine H. Graham, Antoine Guisan. 2008

2,393 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/b1239b970b264dc3dabe9c9d7b9f6b6601c2830a

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

The strong influence of collection bias on biodiversity knowledge shortfalls of Brazilian terrestrial biodiversity (2016, background): “Wallacean shortfall and sampling bias The SDM algorithms showed significant differences between the median AUC’s, except Domain and Maxent, which showed no significant differences between their medians (Fig.”
Mapping the Global Potential Geographical Distribution of Black Locust (Robinia Pseudoacacia L.) Using Herbarium Data and a Maximum Entropy Model (2014, background): “[24] found that MaxEnt was one of the best predictors among 12 different models tested.”
Pollination services at risk: Bee habitats will decrease owing to climate change in Brazil (2012, background): “The Maxent approach is especially useful because it can be applied to analyse small and presence-only data sets (Wisz et al., 2008), a form of data that is frequent on tropical regions (Feeley and Siman, 2011).”
Keep collecting: accurate species distribution modelling requires more collections than previously thought (2011, methodology): “maxent has consistently been shown to be one of the most robust SDMs and to perform very well using presence-only data even with limited sample sizes (Elith et al., 2006; Hernandez et al., 2006; Wisz et al., 2008; Elith & Graham, 2009).”
The effect of prevalence and its interaction with sample size on the reliability of species distribution models (2009, background): “It is well-known that sample size influences the results of species distribution models (Wisz et al. 2008) and different minimum sample sizes have been suggested in the species distribution modeling literature (Pearce and Ferrier 2000, Stockwell and Peterson 2002, McPherson et al.”
The Effects of Sampling Bias and Model Complexity on the Predictive Performance of MaxEnt Species Distribution Models (2013, background): “Species location data are increasingly made available by museums, herbaria, and other scientific institutions [6,7,8] via open-access data portals such as the Global Biodiversity Information Facility (GBIF; www.”
On the brink of extinction? How climate change may affect global chelonian species richness and distribution (2012, background): “This was necessary as SDMs may not be reliable when the sample size is too low (Hernandez et al., 2006; Wisz et al., 2008).”
Applications and future challenges in marine species distribution modeling (2011, background): “, 2008) and availability of species records (Hernandez et al., 2006; Wisz et al., 2008) and selection of appropriate predictors (Peterson and Nakazawa, 2008),”
Overcoming limitations of modelling rare species by using ensembles of small models (2015, methodology): “Maxent has typically been chosen for modelling rare species (Parolo, Rossi & Ferrarini 2008; Gogol-Prokurat 2011; Niamir et al. 2011; Bean, Stafford & Brashares 2012) because this technique performs comparatively well with low sample sizes (Hernandez et al. 2006; Wisz et al. 2008).”
A framework for species distribution modelling with improved pseudo-absence generation (2015, background): “Sample size effect on results362 As sample sizes are heterogeneous across Oak groups, this allowed us363 to indirectly evaluate the influence of the sample size in the performance.364 Caution has to be given to interpreting inflated AUC values due to small365 number of records (Wisz et al., 2008).”
A strict maximum likelihood explanation of MaxEnt, and some implications for distribution modelling (2013, methodology): “…et al. (2006), ranking MaxEnt as the best method or among the best (Hernandez et al. 2006, Guisan et al. 2007, Sérgio et al. 2007, Wang et al. 2007, Wisz et al. 2008, Roura-Pascual et al. 2009, Tognelli et al. 2009, Václávik & Meentemeyer 2009, Veloz 2009, Williams et al. 2009, Mateo et al. 2010,…”