Document Type


Date of Degree Completion

Winter 2006

Degree Name

Master of Science (MS)


Geological Sciences

Committee Chair

Wendy A. Bohrson

Second Committee Member

Paul W. O. Hoskin

Third Committee Member

Michael Clynne


Mount St. Helens (MSH) volcano in southwestern Washington has intermittently erupted dacitic products for the last 40,000 years. On limited occasions, the volcano has produced andesite lava flows, and during one short-lived period, basaltic lava flows. This time interval has been termed the Castle Creek eruptive period and occurred between approximately 2500 and 1700 years B.P. The Castle Creek period erupted dacite, andesite and basalt within this short span of time. Andesite and dacite eruptions dominate the first approximately 700 years of the period, and all basaltic units were erupted in approximately the last 100 years of the period. This is the only known occurrence of basaltic eruptive products at MSH, and yet these lava flows were a major contributor to the buildup of the modern stratocone associated with MSH. Three litho- stratigraphic units exist within the basalts of Castle Creek; from youngest to oldest they are Cave basalt, Precave basalt and North Flank basalt. Petrogenetic relations among these units provide insight into the nature of the subjacent magma chamber during Castle Creek time. Presented here are whole-rock major and trace element data compiled from both published and unpublished works, along with 17 new whole rock trace element analyses and 87Sr/86Sr isotopic ratios, plus textural and compositional imaging integrated with feldspar crystal chemistry data collected by electron microprobe. Petrographic and quantitative petrographic analyses were performed on 5 samples of Castle Creek basalt. MELTS closed-system fractional crystallization simulations were also executed to model major element evolution during the Castle Creek eruptive period.

MELTS simulations reveal primitive Cave basalt can be related to most Precave basalts, andesite and dacite compositions of Castle Creek age through isobaric fractional crystallization. Simulations of North Flank basalt evolution reveal fractional crystallization is potentially a contributing process within the basalts but does not likely control North Flank composition. High initial abundances of K2O and TiO2 within the North Flank and data trends of those oxides prohibit a strict fractional crystallization relationship. Strontium isotopic compositions reveal fractional crystallization alone is not responsible for variation within the three basalt units, as evidenced by the relatively large range of 87Sr/86Sr ratios (approximately 0.7030-0.7034) and increasing 87Sr/86Sr with decreasing whole rock MgO contents. Feldspar analyses by electron microprobe reveal maximum plagioclase feldspar core compositions remain at near constant levels (approximately An80) throughout an approximately 2 wt% change in whole rock MgO content. Best-fit MELTS simulations predict equilibrium plagioclase crystallization that occurs over this same interval should produce crystals with lower anorthite content (down to An52). Crystal size distributions (CSD) suggest plagioclase feldspar from basalt samples had maximum average residence times between 30 and 60 years, well within the approximately 100-year timescale of basaltic magmatism at MSH. Low numbers of large phenocrysts among all samples, similar shapes of CSD plots and similar feldspar compositions all suggest the basalts of Castle Creek time were exposed to a similar thermal regime while plagioclase was a stable phase.

Magmas of the Castle Creek eruptive period of Mount St. Helens were affected by a complex interplay of open-system magmatic processes. Preservation of distinct compositional characteristics between more primitive North Flank and Cave samples (e.g., K2O) suggests that the magma reservoir system at MSH during Castle Creek time was sufficiently geometrically complex to effectively isolate discreet magma batches. Petrologic evidence, however, suggests interaction among more differentiated samples of all units (e.g., magma mixing) may have contributed to magmatic evolution throughout Castle Creek time, and may be responsible for creating some of the compositional diversity during that episode.