THE MAGMATIC AND TECTONIC EVOLUTION OF ATTU ISLAND IN THE NEAR ISLANDS OF THE ALEUTIAN ARC

The massive Aleutian Ridge has long been argued to have been rapidly built along a fracture zone after volcanism ceased to the north at ~55-51 Ma. An enigmatic area is the western Attu Island region, south of the junction of the Bowers and Shirshov Ridges. The oldest ages on Attu Island are a K/Ar age of 42.5 Ma on an amphibolite (Vallier et al., 1994) and a new plagioclase Ar/Ar age of 40.28±0.12 on the volcanic host rocks of 35.6-29.6 Ma gabbroic plutons in the eastern part of the Island. All of the rocks have island arc tholeiite-like trace element patterns with depleted LREE patterns and HFSE anomalies that resemble Marianas arc rocks. They form a NE-striking band, which is flanked to the west by a parallel band of pillow lavas, breccias and dikes with Ar/Ar ages from 35.6-29.3 Ma and more MORB-like compositions. All are chemically distinct from similar age Aleutian rocks to the east. To the west lies a similar trending band of rhyolite-albite granite with Ar/Ar ages of ~19 Ma and MORB-like characteristics. The youngest rocks on Attu are 8-6 Ma calc-alkaline volcanic rocks that outcrop along a distinct E-W trend. All Attu magmatic rocks have high ¹⁴³Nd/¹⁴⁴Nd ratios with the ~ 40 Ma host rocks and Miocene hornblende-bearing calc-alkaline rocks having lower eNd (~ +7.5-9.0) than the backarc-like lavas (~ +9-10.8). The older rocks have high ⁸⁷Sr/⁸⁶Sr ratios suggesting hydrothermal interaction with seawater. Tectonic models to explain the setting of Attu Island have ranged from formation to the east followed by westward translation and clockwise rotation to formation in the west behind an extensional basin related to the opening Komandorsky basin (Yogodzinski et al., 1993). The new ages and chemical data require the oldest units to form in a Marianas-like arc in the late Eocene with extension continuing into the Miocene as evidenced by the bimodal early Miocene mafic and silicic volcanic rocks. The subsequent eruption of the late Miocene-early Pliocene volcanic rocks occurred at a time of calc-alkaline volcanism along much of the arc. The lack of adakites and high Mg andesites on the relatively intact Attu block is consistent with the increasingly commonly recognized western Aleutian adakites in more disrupted parts of the ridge being related to forearc subduction eroded material melted in the mantle wedge rather than slab melting.