REVISITING THE PALEOMAGNETISM OF A PALEOCENE TRENCH-RIDGE-TRENCH TRIPLE JUNCTION; KODIAK ISLANDS, ALASKA

The Ghost Rocks Formation of the Kodiak Islands, AK is part of an accretionary complex which contains turbidites interlayered with pillow lavas and intruded by gabbro and mafic sills. It is generally accepted that this anomalous near-trench magmatism is the result of a Paleocene spreading center subducted beneath the Kodiak Islands. However, it is unclear where this event occurred in relation to the North American margin. Paleomagnetic directions could reveal this location, and a study by Plumley et al. (1983) concluded that the Ghost Rocks Fm. was formed at 40°N latitude. However, discordant site mean directions between the two major bays studied (Alitak and Kiliuda) as well as a 2-stage structural correction in Alitak Bay complicate the results. This collaborative structural and paleomagnetic study seeks to resolve the location of the Paleocene triple junction by providing modern methods, and more detailed sampling and structural analysis. Results from some of our sites at Alitak Bay are reported.

Oriented samples were taken at 67 sites to include a baked contact test, conglomerate test, and fold test to ensure reliable data. 13 sites from Alitak Bay have been analyzed, and 95 of 114 specimens display one to two component vectors during thermal demagnetization. Second components were averaged at each site, and 6 of the 13 sites showed α95 values of less than 20. In-situ inclination-only mean direction for these 6 sites (I= -34.6, k = 25, α95 = 14.9) is improved when corrected for local stratigraphy (I = -61.8, k = 54, α95 = 10.2). This indicates a positive fold test, and suggests that the paleomagnetic signature was acquired before deformation in the accretionary prism. These results are in partial agreement with the original data, showing weakly magnetized samples with unblocking temperatures from 400 to 500°C. New data reveal very similar inclination directions, but unlike the original study show both normal and reverse polarities. Field evidence has not been found for regional folding used by Plumley et al. (1983); structural corrections were instead made on a site-by-site basis. Structural measurements show regional block rotations between Alitak and Kiliuda Bays, which may explain the discrepancy between declination directions in the original study. Continued study of remaining 54 sites will further constrain the data.