Cooling and crystallization of tholeiitic basalt, 1965 Makaopuhi Lava Lake, Hawaii
Thomas L. Wright, Reginald T. Okamura · USGS professional paper · 1977
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
USGS professional paper · United States Government Publishing Office · report · 1977
Abstract
A pond of lava 84 m deep and 800 m wide formed in Makaopuhi crater during an eruption of Kilauea volcano March 5-15, 1965.The tholeiitic lava (MgO about 8 percent, SiCh about 50 percent) was erupted at about 1,190°C.Foundering of the lake-surface crust fol lowing the eruption reduced the temperature of the upper 10m of the lake by as much as 40°C.The following studies were conducted be tween March 1965 and February 1969, when the lake was covered by lava from a subsequent eruption of Kilauea.1.
Twenty four small-diameter (1.5 cm) and three large-diameter (6 cm) holes were drilled to trace the growth of the upper crust of the lake.The 1,070°C isothermal surface separates rigid, partly molten crust (solidus=980°C) from melt into which probes may be pushed by hand.2. Open holes and holes cased with stainless steel were used sub sequently for measurement of temperature, oxygen fugacity, a single measurement of melt viscosity, and collection of samples of melt and gases exsolving from the melt.3. Elevations of a grid of nails set in the lake surface crust were periodically determined.4.
The specific gravity of drill core was measured, and the core was used in a variety of petrographic and petrochemical studies.This report summarizes all of the quantitative studies relevant to the cooling and crystallization of ponded basaltic lava.Principal re sults and interpretations of these aspects are as follows:1.
Thermal history: Isotherms in the upper crust migrated down ward in the lake, first, as a linear function of square root of elapsed time (Vt); then, they were irregularly depressed to greater depths than predicted by this function, the initial change in slope being triggered by a period of heavy rainfall.Isotherms in the melt (1,070-1,130°C) were likewise initially linear with Vt, but their slopes flattened and began to vary erratically in the period March-December 1966.This behavior is interpreted as being caused by the initiation of convection in the ponded basaltic melt.By late 1968 all isothermal surfaces were at greater depths than predicted from the Vt function.2.
Oxygen fugacity (fOz): Drill holes showed buffered/XIh-T pro files over much of the period, with/X)2 values slightly higher than the quartz-fayalite-magnetite buffer.Superimposed on normal profiles were transient high values offCh.as much as 10-2 atmospheres (atm) between 400-800°C.Core became oxidized soon after exposure to high fOz, evidenced by hematitic alteration of mafic minerals and increased FteOs/FeO ratios in analyzed core.The high oxygen fugacities are tentatively ascribed to deep circulation of oxygensaturated rain water.3.
Chemistry and petrography: Samples collected during the erup tion and to a depth of 7.9 m have MgO contents of 7.5-8.5 percent and show olivine-controlled chemical variation.Large olivine crystals are inferred to have settled toward the bottom of the lake.Below 7.9 m, MgO decreases to 6.1 percent at 16.5 m, and samples are dif ferentiated by removal of olivine, augite, and plagioclase.This proc ess is interpreted as flow differentiation promoted by convection of the melt.Segregation veins have compositions explainable by lowtemperature (1,030°-1,070°C) filtration of liquid from partially mol ten crust into open-gash fractures.Grain size increases markedly to a maximum at a depth of 14 m, where median-grain values exceed 0.001 mm3 (diam=0.2mm).Residual glass composition is that of a calc-alkaline rhyolite, and the content of residual glass increases slightly with depth as an apparent function of the amount of differ entiation.4.
Core density: Core density reaches a maximum at 6.1 m depth (sp gr=2.7 g/cc) and then decreases to 2.5 at 15.2 m.Melt density is low (<2.5 g/cc) at a depth of 4.6-7.6 m (numerous vesicles) but high (2.8 g/cc) at a depth of 16.5-18.3m.Evidently the temperature of vesiculation decreases with increasing pressure, and the resultant difference in the amount of an exsolved gas phase in the melt causes density differences that are considered to be the driving force for convection.5.
Surface altitude changes: The surface of the central part of the lake subsided at a decreasing rate (relative to Vt) until the middle of 1967 and was subsequently uplifted through the last measurement date in 1968.This variation in altitude is the resultant of thermal contraction and density change on solidification, the latter critically controlled by the temperature at which vesiculation takes place.
Open the resolved locator · 10.3133/pp1004
Metadata licence: OpenAlex metadata (CC0)