Crystallization of Mount St. Helens 1980–1986 dacite: A quantitative textural approach

Abstract

Quantitative measurements of crystal size distributions (CSDs) have been used to obtain kinetic information on crystallization of industrial compounds (Randolph and Larson 1971) and more recently on Hawaiian basalts (Cashman and Marsh 1988). The technique is based on a population balance resulting in a differential equation relating the population densityn of crystals to crystal sizeL, i.e., at steady staten =n ^o exp(−L/itGτ), wheren ^o is nucleation density,G is the average crystal growth rate,τ is the average growth time, and the nucleation rateJ =n ^o G. CSD (Inn vsL) plots of plagioclase phenocrysts in 12 samples of Mount St. Helens “blast” dacite and 14 samples of dacite from the 1980–1986 Mount St. Helens dome are similar and averageGτ = 9.6 (± 1.1) × 10⁻³ cm andn ^o = 1−2 × 10⁶ cm⁻⁴. Reproducibility of the measurements was tested by measuring CSDs of 12 sections cut from a single sample in three mutually perpendicular directions; precision of the size distributions is good in terms of relative, but not necessarily absolute values (± 10%). Growth and nucleation rates for plagioclase have been calculated from these measurements using time brackets ofτ = 30–150 years; growth ratesG are 3−10 × 10⁻¹²cm/s, and nucleation ratesJ are 5−21 × 10⁻⁶/cm³ s.G andJ for Fe-Ti oxides calculated from CSD data areG = 2−13 ± 10⁻¹³ cm/sec andJ = 7−33 × 10⁻⁵/cm³ s, respectively. The higher nucleation rate and lower growth rate of oxides resulted in a smaller average crystal size than for plagioclase. Sizes of plagioclase microlites (<0.01 mm) in the blast dacite groundmass have been measured from backscatter SEM photographs. Nucleation of these microlites was probably triggered by intrusion of material into the cone of Mount St. Helens in spring 1980. This residence time of 52 days gives minimum crystallization estimates ofG = 1−3 × 10⁻¹¹ cm/s andJ = 9−16 × 1O³/cm³ s. The skeletal form of the microlites provides evidence for nucleation and growth at high values of undercooling (ΔT) relative to the phenocrysts. A comparison of nucleation and growth rates for the two crystal populations (phenocrysts vs microlites) suggests that while growth rate seems to be only slightly affected by changes inΔT, nucleation rate is a very strong function of undercooling. A comparison of plagioclase nucleation and growth rates measured in the Mount St. Helens dacite and in basalt from Makaopuhi lava lake in Hawaii suggests that plagioclase nucleation rates are not as dependent on composition. Groundmass textures suggest that plagioclase microphenocrysts crystallized at depth rather than in the conduit, in the dome, or after extrusion onto the surface. Most of this crystallization appears to be in the form of crystal growth (coarsening) of groundmass microphenocrysts at small degrees of undercooling rather than extensive nucleation of new crystals. This continuous crystallization in a shallow magmatic reservoir may provide the “overpressurization” needed to drive the continuing periodic domebuilding extrusions, which have been the pattern of activity at Mount St. Helens since December 1980.