Skip to content

Multivariate Spatiotemporal Regionalization Bottleneck

Home / Open Questions / Multivariate Spatiotemporal Regionalization Bottleneck

Background: Spatial regionalization methods often treat time series as univariate, ignoring the interdependent multidimensional nature of many real-world spatiotemporal processes.

Question / Future Work: The methodology needs to be extended to multivariate time series to allow regionalization to be informed by the joint dynamics of multiple indicators simultaneously, which requires new joint MDL objectives that account for inter-variable dependencies across space and time.

Why It Matters: Environmental and social phenomena are inherently multivariate; restricting analysis to a single time series can obscure important cross-variable spatial regularities.

Evidence: First, the framework could be extended to incorporate multivariate time series, allowing multiple environmental or socio-economic variables to jointly inform regionalization.

Metadata & Links