Predictive model of San Andreas fault system paleogeography, Late Cretaceous to early Miocene, derived from detailed multidisciplinary conglomerate correlations

摘要

Paleogeographic reconstruction of the region of the San Andreas fault system in western California, USA, was hampered for more than two decades by the apparent incompatibility of authoritative lithologic correlations. These led to disparate estimates of dextral strike-slip offsets across the San Andreas fault, notably 315 km between Pinnacles and Neenach Volcanics, versus 563 km offset between Anchor Bay and Eagle Rest peak. Furthermore, one section of the San Andreas fault between Pinnacles and Point Reyes had been reported to have six pairs of features showing only similar to 30 km offset, while several younger features in that same area were reported consistent with similar to 315 km offset. Estimates of total dextral slip on the adjoining San Gregorio fault have ranged from 5 km to 185 km. Sixteen Upper Cretaceous and Paleogene conglomerates of the California Coast Ranges, from Anchor Bay to Simi Valley, were included in a multidisciplinary study centered on identification of matching unique clast varieties, rather than on simply counting general clast types. Detailed analysis verified the prior correlation of the Upper Cretaceous strata of Anchor Bay at Anchor Bay with a then-unnamed conglomerate at Highway 92 and Skyline Road (south of San Francisco); anti verified that the Paleocene or Eocene Point Reyes Conglomerate at Point Reyes is a tectonically displaced segment of the Carmelo Formation of Point Lobos (near Monterey). The work also led to three new correlations: Point Reyes Conglomerate with granitic source rock at Point Lobos; a magnetic anomaly at Black Point (near Sea Ranch) with a magnetic anomaly near San Gregorio; and strata of Anchor Bay with previously established source rock, the potassium-poor Logan Gabbro of Eagle Rest peak, at a more recently recognized subsurface location just east of the San Gregorio fault, south of San Gregorio. From these correlations, a Late Cretaceous to early Oligocene paleogeography was constructed which was unique in utilizing modern shapes of tectonic blocks in a palcogeographic resolution synthesizing both the proposed 315 and similar to 30 km San Andreas fault offsets, as well as honoring the lithologic correlation of Anchor Bay with Eagle Res: peak. The model has proved predictive. Since its first introduction, in April and June 1998, other authors have reported seven subsequently identified correlative pairs of geological and geophysical features consistent with it. These required both lateral and temporal expansion of the model: The paleogeography now incorporates 58 pairs of correlative features, covers the period from 70 to 21.3 Ma, and extends from Pelona and Orocopia to the Mendocino triple junction. The model supports the view that fault is not the boundary between the Pacific and North American plates is a wide zone encompassing the San Andreas fault system. However, the model suggests the San Andreas fault is a temporary assemblage of separate segments having different motions, and is neither the primary plate boundary, nor the dominant fault of the San Andreas fault system. By improving resolution of complex spatial-temporal distribution of slip along this evolving tectonic margin, the model provides a firmer foundation for resolution of geophysical issues such as slab window and Pacific Plate subduction models.