Insights into magma storage depths and eruption controls at Kīlauea Volcano during explosive and effusive periods of the past 500 years based on melt and fluid inclusions

K & imacr;lauea Volcano experiences centuries-long cycles of explosive and effusive eruptive behavior, but the relation, if any, between these eruptive styles and changing conditions in the magma plumbing system remains poorly known. We analyze olivine-hosted melt and fluid inclusions to determine magma storage depths during the explosive-era Keanak & amacr;ko'i Tephra eruptions (similar to 1500-1840 CE) and compare these results to modern effusive-era K & imacr;lauea eruptions (1959 K & imacr;lauea Iki, 1960 Kapoho, 2018 lower East Rift Zone). We find that shallow (1-3 km) magma storage has persisted for centuries at K & imacr;lauea, spanning both explosive and effusive periods. In contrast, mid-crustal zones of magma storage shallowed over time, from 5 to 8 km during the Keanak & amacr;ko'i sequence to 3-5 km during the modern effusive period. Melt and fluid inclusions in high-forsterite olivine (Fo(86-89)) trapped at shallow depths indicate that high-temperature magmas (1200 to similar to 1300 degrees C) commonly reach depths of <= 3 km. CO2-rich fluid inclusions are present in olivine from all investigated K & imacr;lauea eruptions but are larger and much more abundant in Keanak & amacr;ko'i units, which we interpret as indicating that a greater volume fraction of exsolved CO2-rich fluid was present in pre-eruptive Keanak & amacr;ko'i melts. Increased amounts of CO2-rich fluids in the Keanak & amacr;ko'i-era magmas would have increased magma buoyancy and driven rapid magma ascent, thereby increasing eruption energy and enhancing near-surface magma-water interactions compared to the current effusive period.