ABSTRACT

The passive-margin succession of the Diablo Platform (Lower Ordovician, west Texas) is represented by the second-order Sauk C supersequence set, consisting of a basal transgressive clastic unit (The Bliss Sandstone) above the breakup unconformity, marking the second-order basal lowstand transgressive phase), overlain by 750 m of drift-related, shallow-marine platform carbonates (the El Paso Group) recording the second-order highstand. Due to late Paleozoic structuring of the Gondwanan passive margin, present exposures in Texas are in an updip shelf position and lack internal stratal geometries across depositional strike, so sequences and systems tracts are identified solely by the vertical stacking patterns of depositional subfacies and higher-frequency, fifth-order cycles.

Accommodation plots (Fischer plots) of high-frequency cycles gauge systematic shifts in third-order accommodation of two complete third-order sequences, each about 2 m.y. long and 60-140 m thick, within the El Paso Group. This is expressed in the vertical succession of cycle types, systematic changes in cycle thickness, and variations in subfacies as revealed by histograms of subfacies types tied to Fischer plots. A complete El Paso shelf sequence contains a thin lowstand systems tract of quartz arenite, a thick transgressive systems tract dominated by upward-thickening, thrombolitic subtidal cycles, and a highstand systems tract marked by dolomitic, thinning-upward peritidal cycles containing quartz sand. The El Paso third-order depositional sequences have been correlated by biostrat graphy and cycle stacking pattern from the Franklin Mountains southeast to the Diablo Arch, northeast into the Ardmore Basin, and thence to the Appalachian Basin. Hence, the third-order sequences appear to be eustatic, based on biostratigraphic correlation of accommodation plots around the periphery of Gondwana.

Time-series analyses suggest that a strict allocyclic, Milankovitch-driven, glacio-eustatic mechanism alone is insufficient to account for the higher-frequency fifth-order and lower-frequency third-order accommodation cycles. We used a two-dimensional forward computer simulation to compare allocyclic and autocyclic models for high-frequency cycles. The allocyclic simulation superimposes a deterministic, Milankovitch glacio-eustatic driver on a lower-frequency, third-order accommodation cycle and generates a two-dimensional synthetic cross section composed of high-frequency peritidal and subtidal cycles stacked into third-order sequences. The autocyclic simulation is generated by a Ginsburg autocyclic mechanism for high-frequency cyclicity triggered by third-order accommodation changes The autocyclic simulation quantitatively replicates an autocyclic mechanism for fifth-order cycle development linked to onshore sediment transport and seaward progradation of tidal flats. The two-dimensional internal facies architecture of a third-order sequence and the high-frequency cycle stacking patterns are simulated in both scenarios. Important differences in patterns of cycles and subfacies stacking in the simulations shed light on the interpretation of the Diablo platform cycles and sequences. As an alternative or a supplement to autocyclicity or Milankovitch-forced glacio-eustasy, we have studied stochastic models for driving high-frequency changes in relative sea level. These intriguing results maintain essentially all the features observed in the rock record. Our smoothed sto hastic simulations suggest that the "cycle problem" is much more complex than presently realized.