2007 GSA Denver Annual Meeting (28–31 October 2007)

Paper No. 19
Presentation Time: 8:00 AM-12:00 PM

SEISMIC VELOCITY VARIATIONS IN CALIFORNIA MIDDLE AND LOWER CONTINENTAL CRUST - OBSERVATIONS FROM THE SAN GABRIEL MOUNTAINS


DEWALT, Heather A., Akron, OH 44312 and BARTH, Andy, Indianapolis, IN 46202, hd173106@ohio.edu

An improved understanding of modern seismicity and crustal evolution requires knowledge of wave velocity variations as a function of rock composition and mineral assemblage at elevated pressure and temperature. Here we test two computational approaches (Christensen and Mooney, 1995; Hacker and Abers, 2004) and compare the results to limited direct measurements (McCaffree and Christensen, 1998) of Vp and Vs in an exposed section of middle to lower continental crust. In this crustal section, amphibolite and granulite facies Paleoproterozoic gneisses are intruded by an anorthosite-gabbro-syenite suite and a variety of Mesozoic plutons. Calculated Vp based on measured rock density of framework gneisses ranges from 5.8 – 7.0 km/sec, and velocities are bimodal, with mafic gneisses ranging from 6.3 – 7.0 km/sec and felsic gneisses from 5.8 – 6.2 km/sec. Calculated Vp based on a mineralogical model results in a smaller range and higher calculated velocities, from 6.4 to 7.1 km/sec for mafic and 6.4 to 6.5 km/sec for felsic gneisses. Based on measured rock densities, Vp of rocks in the anorthosite complex ranges from 5.9 to 7.75 km sec. Using the mineralogical model and published geochemical data (Ekstrom et al., 1994), anorthosite and gabbro range from 6.34 – 6.92 and jotunite-syenite 6.32 – 7.6 km/sec. Hydrous Mesozoic plutons intruding the gneiss and anorthosite complex are also bimodal, but consistently yield lower calculated velocities; both calculation techniques yield Vp between 5.8 and 6.3 km/sec. As expected, calculated values based on measured densities are in good agreement with limited direct Vp measurements on these samples. The mineralogical model yields calculated Vp values that are as much as 6% higher. These computational approaches allow us to rapidly estimate Vp for a wide variety of rock compositions, giving a better view of the range and average Vp of each rock unit and variations within the middle to lower crust.