GEOCHEMISTRY OF CAMBRIAN RHYOLITES IN THE WICHITA MOUNTAINS, SOUTHWESTERN OKLAHOMA: A-TYPE FELSIC VOLCANISM WITHIN THE SOUTHERN OKLAHOMA AULACOGEN

The Carlton Rhyolite Group forms the most extensive unit of a bimodal igneous assemblage emplaced during Cambrian rifting within the Southern Oklahoma aulacogen. The rhyolites occupy an area of ~ 40,000 km² in the subsurface but are exposed only in the Arbuckle and Wichita Mountains, with the latter area providing the most complete outcrops. Here we report geochemical data for ~ 50 samples from stratigraphic sequences of lava flows (and possibly some rheoignimbrites) from the main rhyolite outcrop areas in the Wichita Mountains. These outcrops are up to 45 km apart and include Bally Mountain (9 flows), Zodletone Mountain (at least 2 flows), Blue Creek Canyon (6 flows) and Fort Sill (2 flows). Individual flows are tabular and range from 80 to 400 m thick. The most complete sequence of flows, at Bally Mountain, has a stratigraphic thickness of > 2 km.

Carlton rhyolites and associated subvolcanic granites (Wichita Granite Group) show typical features of A-type felsic rocks (e.g., low CaO, high FeO/MgO, elevated contents of high-field-strength trace elements), consistent with emplacement in a rift setting. Rhyolites and granites show nearly identical patterns on MORB-normalized multi-element diagrams, with strong depletions in Sr, P and Ti and less extreme depletion in Ba, consistent with fractionation of plagioclase, apatite, titanomagnetite and K-feldspar.

In Harker variation diagrams for the rhyolites, many of the major oxides show considerable scatter, implying that the more mobile elements have been disturbed by secondary alteration. In contrast, TiO₂ increases regularly with P₂O₅ along a single well defined trend for the entire rhyolite suite. This trend, which is based on elements that are less susceptible to alteration, suggests that rhyolite eruptions may have tapped a single, progressively differentiating magma chamber, or may have tapped separate smaller chambers evolving along similar petrogenetic pathways. Alternatively, the trend could record generation of individual magma batches during progressive melting of a common source. However, immobile trace elements such as Th, Zr, Nb and Y generally do not show coherent variations when plotted versus each other or TiO₂, suggesting that more complex petrogenetic models need to be considered.