Vertical Decoupling and Network Connectivity Shifts in Permian Radiolarian Faunas During Greenhouse Events

The Permian Period experienced three warming events, during which the expansion of the Neo-Tethys Ocean and the gradual amalgamation of Gondwana drove the reorganization of the Global Meridional Overturning Circulation (GMOC), accompanied by pronounced shifts in vertical and latitudinal biodiversity gradients. Modern biological studies indicate a positive correlation between ocean currents and population connectivity, however, strong physical connectivity does not equate to strong ecological connectivity. Radiolaria, first appearing in the Cambrian Period, are widely distributed across shallow to deep-water environments. Different taxonomic units exhibit distinct depth-related ecological preferences, making them ideal proxies for reconstructing vertical paleobiogeographic dynamics. This study integrated the PBDB and published literature to construct a comprehensive dataset of genus-level occurrence records of Permian radiolarians. Sampling localities were assigned to major paleo-oceanic domains based on paleo-plate reconstructions, and categorized into shallow-water and deep-water assemblages. Network analysis was employed to quantify inter-domain connectivity and modularity across geological stages. The Sørensen similarity coefficient was used to assess taxonomic similarity of faunas among localities, while paleo-latitude and paleo-longitude data facilitated reconstruction of the latitudinal diversity gradient (LDG). Our results reveal that deep-water connectivity exceeded shallow-water connectivity, and the biogeographic network exhibited a significantly vertically decoupled response to climate forcing. Shallow-water networks were more sensitive to temperature fluctuations, showing increased connectivity and reduced modularity—reflecting intense environmental filtering and habitat fragmentation. In contrast, deep-water assemblages maintained higher connectivity through sustained inter-regional linkages that offset local extinctions. LDG analysis further uncovered alternating bimodal and unimodal distribution patterns, along with systematic latitudinal migration of diversity peaks associated with greenhouse intervals. This study elucidates the differential response mechanisms of Permian marine vertical biogeographic patterns to the coupled evolution of climate and ocean circulation. It confirms that deep-water ecosystems can serve as early-warning indicators of network collapse, providing critical paleontological evidence for predicting shifts in connectivity and vulnerability of marine ecosystems under contemporary rapid warming scenarios.