Spatial factor analysis: a new tool for estimating joint species distributions and correlations in species range

By J. Thorson, M. Scheuerell, A. Shelton, Kevin E. See, Hans Julius Skaug, Kasper Kristensen. 2015

156 citations recorded by the source corpus.

Predicting and explaining the distribution and density of species is one of the oldest concerns in ecology. Species distributions can be estimated using geostatistical methods, which estimate a latent spatial variable explaining observed variation in densities, but geostatistical methods may be imprecise for species with low densities or few observations. Additionally, simple geostatistical methods fail to account for correlations in distribution among species and generally estimate such cross‐correlations as a post hoc exercise. We therefore present spatial factor analysis (SFA), a spatial model for estimating a low‐rank approximation to multivariate data, and use it to jointly estimate the distribution of multiple species simultaneously. We also derive an analytic estimate of cross‐correlations among species from SFA parameters. As a first example, we show that distributions for 10 bird species in the breeding bird survey in 2012 can be parsimoniously represented using only five spatial factors. As a second case study, we show that forward prediction of catches for 20 rockfishes (Sebastes spp.) off the U.S. West Coast is more accurate using SFA than analysing each species individually. Finally, we show that single‐species models give a different picture of cross‐correlations than joint estimation using SFA. Spatial factor analysis complements a growing list of tools for jointly modelling the distribution of multiple species and provides a parsimonious summary of cross‐correlation without requiring explicit declaration of habitat variables. We conclude by proposing future research that would model species cross‐correlations using dissimilarity of species’ traits, and the development of spatial dynamic factor analysis for a low‐rank approximation to spatial time‐series data.

Subfields: Ecological Modeling, Ecology, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/e65a53999b48af3ffa92d3443dfb24e9d0568038

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

How to make more out of community data? A conceptual framework and its implementation as models and software. (2017, background): “…joint species distribution models e.g. to presence–absence data or abundance data (Hui et al. 2015; Clark et al. 2017), as well as to study designs of spatial (Thorson et al. 2015; Ovaskainen et al. 2016b), temporal (Sebasti an-Gonz alez et al. 2010) or spatio-temporal (Thorson et al. 2016) nature.”
Biotic interactions in species distribution modelling: 10 questions to guide interpretation and avoid false conclusions (2018, background): “Spatial factor analysis (Thorson et al., 2015) describes species associations by regressing occurrence (or abundance) of multiple species against common but unknown (latent) covariates, which are modelled as random spatial fields.”
So Many Variables: Joint Modeling in Community Ecology. (2015, background): “This analysis assumed that abundances were independent across samples, but future analyses could look at whether there is spatial structuring in abundances beyond that explained by predictors in the model by incorporating spatially autocorrelated latent variables, along the lines of [77].”
Joint dynamic species distribution models: a tool for community ordination and spatio‐temporal monitoring (2016, methodology): “We therefore impose constraints on the form of the loadings matrix L , and specify that the marginal variance of each factor Var( w j ) is one (Harvey, 1990; Zuur et al ., 2003; Thorson et al ., 2015a), as explained in Appendix S2.”
Joint Species Distribution Modeling: Dimension Reduction Using Dirichlet Processes (2017, background): “, 2010; Ovaskainen and Soininen, 2011), continuous or discrete abundance (Latimer et al., 2009; Thorson et al., 2015), abundance with large number of zeros (Clark et al.”
Using joint species distribution models for evaluating how species‐to‐species associations depend on the environmental context (2017, background): “2016a) and spatially explicit data (Thorson et al. 2015; Ovaskainen et al. 2016b).”
A comprehensive evaluation of predictive performance of 33 species distribution models at species and community levels (2019, background): “…the occurrences of all species in a community simultaneously and include a covariance structure to capture species-to-species associations, without necessarily assuming rigid species-by-species relationships (Clark et al. 2014, Pollock et al. 2014, Thorson et al. 2015, Ovaskainen et al. 2017).”