Dynamic spatio-temporal sequential ordinal models: Application to invasive weeds
Stochastic Environmental Research and Risk Assessment · 2026-12 · 40 (12)
DOI: 10.1007/s00477-026-03389-7 ↗Kurz gefasst
Fachartikel in der Zeitschrift Stochastic Environmental Research and Risk Assessment, online erschienen am 14.11.2026. Verfasst von Geoffrey R. Hosack und 9 weiteren Autor:innen. Die Arbeit nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The multivariate sequential ordinal model is investigated for use in the Bayesian analysis of spatio-temporal ordinal data. The sequential ordinal model likelihood is equivalent to a binary model conditional on unknown regression coefficients and spatio-temporal random effects. Therefore estimation and prediction in the space-time context can proceed using the well-established dynamic generalised linear model framework. Moreover, the sequential ordinal model avoids the ordering constraints on the threshold parameters that determine the category break points required by cumulative ordinal models, and so simplifies the estimation procedure for high-dimensional space-time applications using Bayesian inference. The dynamic spatio-temporal sequential ordinal model is applied to estimate the cover-abundance of four invasive alien weed species. These weeds are observed by means of a modified Braun-Blanquet score that is commonly used in vegetation studies and constitutes ordinal data. As all four weeds are actively managed, the multivariate ordinal data are sparsely distributed in space and time with few observations recorded in high cover-abundance categories. A separable model for space-time dependence that maintains parameter interpretability in the presence of aggregated ordinal categories is therefore developed and applied to each species of interest. The dynamic spatio-temporal sequential ordinal model is shown to be particularly useful for biological populations. Estimation and prediction is demonstrated using integrated nested Laplace approximation (INLA) methods developed for univariate spatio-temporal models. Bayesian estimation and prediction shows that the four invasive weed species differentially respond to habitat type, control effort and accessibility, and share similar magnitudes of dependence with short effective spatial ranges and strong temporal autocorrelations.
Zitieren
APA
Hosack, G. R., Yang, W., Pike, K. N., O’Loughlin, L. S., Cook, E., Howland, B., Sutcliffe, E., Milner, R. N. C., Gooden, B., & Froese, J. G. (2026). Dynamic spatio-temporal sequential ordinal models: Application to invasive weeds. Stochastic Environmental Research and Risk Assessment, 40(12). https://doi.org/10.1007/s00477-026-03389-7
Vancouver
Hosack GR, Yang W, Pike KN, O’Loughlin LS, Cook E, Howland B, et al. Dynamic spatio-temporal sequential ordinal models: Application to invasive weeds. Stochastic Environmental Research and Risk Assessment. 2026;40(12). doi:10.1007/s00477-026-03389-7
BibTeX
@article{Hosack2026,
author = {Hosack, Geoffrey R. and Yang, Wen-Hsi and Pike, Kyana N. and O’Loughlin, Luke S. and Cook, Emma and Howland, Brett and Sutcliffe, Emily and Milner, Richard N. C. and Gooden, Ben and Froese, Jens G.},
title = {Dynamic spatio-temporal sequential ordinal models: Application to invasive weeds},
journal = {Stochastic Environmental Research and Risk Assessment},
year = {2026},
volume = {40},
number = {12},
doi = {10.1007/s00477-026-03389-7}
}
EUVIMED