TY - JOUR
T1 - Spatiotemporal joint species distribution modelling
T2 - A basis function approach
AU - Hui, Francis K.C.
AU - Warton, David I.
AU - Foster, Scott D.
AU - Haak, Christopher R.
N1 - Publisher Copyright:
© 2023 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society.
PY - 2023
Y1 - 2023
N2 - We introduce community-level basis function models (CBFMs) as an approach for spatiotemporal joint distribution modelling. CBFMs can be viewed as related to spatiotemporal latent variable models, where the latent variables are replaced by a set of pre-specified spatiotemporal basis functions which are common across species. In a CBFM, the coefficients that link the basis functions to each species are treated as random slopes. As such, the CBFM can be formulated to have a similar structure to a generalised additive model. This allows us to adapt existing techniques to fit CBFMs efficiently. CBFMs can be used for a variety of reasons, such as inferring patterns of habitat use in space and time, understanding how residual covariation between species varies spatially and/or temporally, and spatiotemporal predictions of species- and community-level quantities. A simulation study and an application to data from a bottom trawl survey conducted across the U.S. Northeast shelf show that CBFMs can achieve similar and sometimes better predictive performance compared to existing approaches for spatiotemporal joint species distribution modelling, while being computationally more scalable.
AB - We introduce community-level basis function models (CBFMs) as an approach for spatiotemporal joint distribution modelling. CBFMs can be viewed as related to spatiotemporal latent variable models, where the latent variables are replaced by a set of pre-specified spatiotemporal basis functions which are common across species. In a CBFM, the coefficients that link the basis functions to each species are treated as random slopes. As such, the CBFM can be formulated to have a similar structure to a generalised additive model. This allows us to adapt existing techniques to fit CBFMs efficiently. CBFMs can be used for a variety of reasons, such as inferring patterns of habitat use in space and time, understanding how residual covariation between species varies spatially and/or temporally, and spatiotemporal predictions of species- and community-level quantities. A simulation study and an application to data from a bottom trawl survey conducted across the U.S. Northeast shelf show that CBFMs can achieve similar and sometimes better predictive performance compared to existing approaches for spatiotemporal joint species distribution modelling, while being computationally more scalable.
KW - community-level modelling
KW - environmental filtering
KW - fixed rank kriging
KW - latent variable model
KW - random effects
KW - spatial statistics
KW - time series
UR - http://www.scopus.com/inward/record.url?scp=85164495628&partnerID=8YFLogxK
U2 - 10.1111/2041-210X.14184
DO - 10.1111/2041-210X.14184
M3 - Article
SN - 2041-210X
VL - 14
SP - 2150
EP - 2164
JO - Methods in Ecology and Evolution
JF - Methods in Ecology and Evolution
IS - 8
ER -