6/6 After the 2021 eruption, magma accumulated and fractionated at near-Moho depths. In 2022–2023, a looser mush network enabled faster magma assembly. Diffusion modelling, barometry, and geophysical data show dikes started propagating ~1 month before eruption, then breached the upper crust in days.

5/6 Plg timescales in 2022–23 record mush erosion over months to years, progressively shortening with time, suggesting a loosened mush and faster response to deep melt input. Ol timescales (days-weeks) reflect timing and duration of dike ascent from near-Moho depths, correlating with pre-seismicity

4/6 The 2021–2023 Fagradalsfjall eruptions were sourced from a persistent near-Moho magma domain. However, cpx–melt barometry suggests that in 2022–2023, magmas stalled at mid-crustal levels or ascended more slowly, unlike in the 2021 eruption.

3/6 Unlike 2021, the 2022–2023 Fagradalsfjall lavas show limited geochemical variability, likely sourced from a single, well-mixed reservoir. Trace element ratios and isotopes lie along 2021 mixing trends, suggesting little changes in melt compositions over three years.

2/6 In 2022–2023, olivine (ol) and clinopyroxene (cpx) are phenocrysts, while plagioclase (plg) crystals are antecrysts eroded from crystal mushes. This contrasts with 2021, when all crystals had antecrystic origins, suggesting two distinct crystal populations in the 2022-2023 eruptions.

1/6 We investigate the 2022–2023 Fagradalsfjall eruptions to track short-term changes in magmatic processes and timescales beneath the volcano. Our petrological and geochemical data are integrated with published data from the 2021 eruption (Halldórsson et al. 2022; Kahl 2023; Marshall 2024).

@albertocaracciolo.bsky.social talks about their Rannís (Icelandic Research Fund) project on hazard implications of the ongoing eruptive cycle at the Reykjanes Peninsula, Iceland rdcu.be/eqQAx

Thank you! I am actually not familiar at all. Hopefully I can edit it now