How to best threshold and validate stacked species assemblages? Community optimisation might hold the answer

Scherrer, Daniel; D’Amen, Manuela; Fernandes, Rui F.; Mateo, Rubén G.; Guisan, Antoine

doi:10.1111/2041-210x.13041

Cited by 37 publications

(42 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To do so, they need to define an adequate threshold to transform the data. Several different thresholds have been proposed depending on whether the presence-absence or presenceonly data are being used for modelling (Liu et al 2005(Liu et al , 2013 or when modelling communities (Scherrer et al 2018). Here, authors need to specify which threshold is used and explain why thresholding is deemed necessary.…”

Section: Threshold Selectionmentioning

confidence: 99%

A standard protocol for reporting species distribution models

et al. 2020

Self Cite

View full text Add to dashboard Cite

Species distribution models (SDMs) constitute the most common class of models across ecology, evolution and conservation. The advent of ready‐to‐use software packages and increasing availability of digital geoinformation have considerably assisted the application of SDMs in the past decade, greatly enabling their broader use for informing conservation and management, and for quantifying impacts from global change. However, models must be fit for purpose, with all important aspects of their development and applications properly considered. Despite the widespread use of SDMs, standardisation and documentation of modelling protocols remain limited, which makes it hard to assess whether development steps are appropriate for end use. To address these issues, we propose a standard protocol for reporting SDMs, with an emphasis on describing how a study's objective is achieved through a series of modeling decisions. We call this the ODMAP (Overview, Data, Model, Assessment and Prediction) protocol, as its components reflect the main steps involved in building SDMs and other empirically‐based biodiversity models. The ODMAP protocol serves two main purposes. First, it provides a checklist for authors, detailing key steps for model building and analyses, and thus represents a quick guide and generic workflow for modern SDMs. Second, it introduces a structured format for documenting and communicating the models, ensuring transparency and reproducibility, facilitating peer review and expert evaluation of model quality, as well as meta‐analyses. We detail all elements of ODMAP, and explain how it can be used for different model objectives and applications, and how it complements efforts to store associated metadata and define modelling standards. We illustrate its utility by revisiting nine previously published case studies, and provide an interactive web‐based application to facilitate its use. We plan to advance ODMAP by encouraging its further refinement and adoption by the scientific community.

show abstract

Section: Threshold Selectionmentioning

confidence: 99%

A standard protocol for reporting species distribution models

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…When dealing with species richness (SR), the simplest and most common method consists in modelling the distribution of all individual species in a pool and then summing their predictions to obtain assemblages (stacked-SDM, S-SDM; Ferrier & Guisan, 2006;Dubuis et al, 2011). However, this method has some limitations, such as over-predicting species richness per site (Guisan & Rahbek, 2011) or being sensitive to methodological biases (Calabrese et al, 2014;Scherrer et al, 2018a). Additionally, while single species models are useful, numerous factors (e.g.…”

Section: Introductionmentioning

confidence: 99%

How much should one sample to accurately predict the distribution of species assemblages? A virtual community approach

Fernandes

Scherrer

Guisan

2018

Ecological Informatics

Self Cite

View full text Add to dashboard Cite

Correlative species distribution models (SDMs) are widely used to predict species distributions and assemblages, with many fundamental and applied uses. Different factors were shown to affect SDM prediction accuracy. However, real data cannot give unambiguous answers on these issues, and for this reason, artificial data have been increasingly used in recent years. Here, we move one step further by assessing how different factors can affect the prediction accuracy of virtual assemblages obtained by stacking individual SDM predictions (stacked SDMs, S-SDM). We modelled 100 virtual species in a real study area, testing five different factors: sample size (200-800-3200), sampling method (nested, non-nested), sampling prevalence (25%, 50%, 75% and species true prevalence), modelling technique (GAM, GLM, BRT and RF) and thresholding method (ROC, MaxTSS, and MaxKappa). We showed that the accuracy of S-SDM predictions is mostly affected by modelling technique followed by sample size. Models fitted by GAM/GLM had a higher accuracy and lower variance than BRT/RF. Model accuracy increased with sample size and a sampling strategy reflecting the true prevalence of the species was most successful. However, even with sample sizes as high as >3000 sites, residual uncertainty remained in the predictions, potentially reflecting a bias introduced by creating and/or resampling the virtual species. Therefore, when evaluating the accuracy of predictions from S-SDMs fitted with real field data, one can hardly expect reaching perfect accuracy, and reasonably high values of similarity or predictive success can already be seen as valuable predictions. We recommend the use of a 'plot-like' sampling method (best approximation of the species' true prevalence) and not simply increasing the number of presences-absences of species. As presented here, virtual simulations might be used more systematically in future studies to inform about the best accuracy level that one could expect given the characteristics of the data and the methods used to fit and stack SDMs.

show abstract

“…These procedures return the predicted species composition and richness within each assemblage (i.e., grid cell) across a geographical domain (Ferrier and Guisan, 2006;Guisan and Rahbek, 2011). More specifically, binary S-SDM (bS-SDM) predictions were obtained by converting individual species ensembles into binary predictions or occurrence probabilities by using its corresponding TSS-maximization threshold (Scherrer et al, 2018;Zurell et al, 2020).…”

Section: 4modeling Frameworkmentioning

confidence: 99%

“…PRR emulates ecological assembly rules by ranking the species in each assemblage based on the occurrence probability obtained from each species and the number of species per assemblage. The species with the highest probabilities in an assemblage is selected until the number of species in an assemblage, based on observed data, is reached (D'Amen et al, 2015;Scherrer et al, 2018). We used the maximum number of species per assemblage from the FIA dataset as assemblage-level constraint for cS-SDM estimations.…”

Section: 4modeling Frameworkmentioning

confidence: 99%

“…To evaluate the performance of species assemblage predictions, we used four different metrics that are commonly used for this purpose (Pottier et al, 2013;Scherrer et al, 2018;Zurell et al, 2020). The metrics used were: a) the deviation of the predicted species richness to the observed (SR deviation); b) the proportion of correctly predicted as present or absent (Prediction success); c) true skill statistic (TSS) and d) similarity between the observed and predicted community composition (Sørensen index).…”

Section: 5evaluating Species and Assemblage Predictionsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting species distributions and community composition using remote sensing products

Pinto‐Ledezma

Cavender‐Bares

2020

Preprint

View full text Add to dashboard Cite

AbstractAccurate predictions of species composition and diversity are critical to the development of conservation actions and management strategies. In this paper using oak assemblages distributed across the conterminous United States as study model, we assessed the performance of stacked species distribution models (S-SDMs) and remote sensing products in building the next-generation of biodiversity models. This study represents the first attempt to evaluate the integrated predictions of biodiversity models—including assemblage diversity and composition—obtained by stacking next-generation SDMs. We found three main results. First, environmental predictors derived entirely from remote sensing products represent adequate covariates for biodiversity modeling. Second, applying constraints to assemblage predictions, such as imposing the probability ranking rule, results in more accurate species diversity predictions. Third, independent of the stacking procedure (bS-SDM versus cS-SDM), biodiversity models do not recover the observed species composition with high spatial resolution, i.e., correct species identities at the scale of individual plots. However, they do return reasonable predictions at macroecological scales (1 km). Our results provide insights for the prediction of assemblage diversity and composition at different spatial scales. An important task for future studies is to evaluate the reliability of combining S-SDMs with direct detection of species using image spectroscopy to build a new generation of biodiversity models to accurately predict and monitor ecological assemblages through time and space.

show abstract

How to best threshold and validate stacked species assemblages? Community optimisation might hold the answer

Cited by 37 publications

References 75 publications

A standard protocol for reporting species distribution models

A standard protocol for reporting species distribution models

How much should one sample to accurately predict the distribution of species assemblages? A virtual community approach

Predicting species distributions and community composition using remote sensing products

Contact Info

Product

Resources

About