Overcoming limitations of modelling rare species by using ensembles of small models

By F. Breiner, Antoine Guisan, A. Bergamini, M. Nobis. 2015

465 citations recorded by the source corpus.

Subfields: Ecological Modeling, Genetics, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/c88011913d81600d58029e79e57f12588ef63bab

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 Genetics

In Nature and Landscape Conservation

Citations in the corpus

A standard protocol for reporting species distribution models (2020, methodology): “…the range of predictions given the different uncertainties (initial conditions, model classes, model parameters, boundary conditions, Araújo and New 2007, Thuiller et al. 2019), but it is also increasingly used for mapping, e.g. to model rare species (Breiner et al. 2015, 2018, Hao et al. 2019).”
Testing the ability of species distribution models to infer variable importance (2019, background): “Recent work demonstrates that very small sample sizes may be adequate for generating accurate models in some situations (Soultan & Safi 2017), especially when coupled with novel techniques (Lomba et al. 2010; Breiner et al. 2015).”
A comparison between Ensemble and MaxEnt species distribution modelling approaches for conservation: A case study with Egyptian medicinal plants (2020, background): “…to improve model predictions (Grenouillet et al., 2011; Oppel et al., 2012; Stohlgren et al., 2010), reduce overfitting when modelling rare species (Breiner et al., 2016), and has been advocated as a better alternative to single models for future climate projection modelling with large numbers of…”
Optimizing ensembles of small models for predicting the distribution of species with few occurrences (2018, methodology): “For the second subset, we used the five most common species (sample sizes: 140, 128, 123, 115 and 112) in the data of Breiner et al. (2015).”
Testing species assemblage predictions from stacked and joint species distribution models (2019, methodology): “To avoid overfitting for rare species (Breiner et al., 2015), we only considered species with at least 50 presences resulting in Local species pool”
Predictor complexity and feature selection affect Maxent model transferability: Evidence from global freshwater invasive species (2020, methodology): “…invasions (e.g., Rodda et al., 2011; Rödder et al., 2009), the identification of priority conservation areas for rare or threatened species (e.g., Breiner et al., 2015), quantification of climate change risks (e.g., Warren et al., 2014), and elucidating ecological processes driving species…”