Large silicic calderas pose a major global hazard, but reconstructing the long-term assembly of their magma reservoirs remains challenging because climactic eruptions often destroy pre-caldera records. Here we use zircon petrochronology on 18 samples, including caldera-forming units and pre-caldera lithics entombed within the 161 ka Kos Plateau Tuff (Aegean Sea, Greece), as well as post-caldera data from the subsequent Nisyros and Yali eruptions, to reconstruct >1.5 million years of magmatic evolution at the Kos-Nisyros volcanic field. Analysis of >680 zircon crystals reveals progressive reservoir assembly from andesite–dacite to late-stage, pre-caldera crystal-poor rhyolite, reflecting coalescence of smaller magma bodies into a large upper crustal mush system capable of producing high-SiO₂ rhyolite lenses by melt extraction prior to the cataclysmic eruption. After caldera collapse, the system transitioned to a rapid reconstructive phase, with erupted compositions evolving from basalt to rhyolite in ~40 ky. Notably, we find no zircon recycling between andesite/dacite and rhyolitic phases, nor between the caldera-forming, and pre- or post-caldera magmas. We attribute this to processes that limit antecryst entrainment: (1) spatial migration of reservoir foci, (2) slow melt extraction from a persistent mush and absence of self-mixing following rejuvenation, (3) solidification of residual mush after large eruptions, and (4) high-temperature resorption of older zircons during recharge. These results illuminate cyclic behaviour in cataclysmic-eruption-capable systems and provide a framework for predicting zircon recycling patterns during the prolonged evolution of long-lived silicic volcanoes.
No takes yet. Share an insight, caveat, or question.
Mikosch et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: