Application of sequence stratigraphic concepts to the cretaceous Urgonian carbonate platform, southeast France

Abstract

Carbonate platforms are increasingly being studied using sequence stratigraphic concepts and models borrowed from the study of siliciclastic shelves in passive margin settings. The direct transposition of the stratigraphic model for a siliciclastic shelf to its carbonate counterpart, the carbonate shelf, assumes that the two systems respond in a very similar way to changes of relative sea-level, the interpreted major control upon depositional stacking patterns. Current models depicting the sequence stratigraphic evolution of carbonate shelves are and have been frequently applied without regard for the differences between the siliciclastic and carbonate shelf depositional systems. It is the purpose of this study to test the current sequence stratigraphic model and its assumptions for a carbonate shelf. Carbonate shelves do differ quite fundamentally from their siliciclastic equivalents. The carbonate shelf has the capacity to respond in quite different ways to changes in relative sea-level, compared to siliciclastic systems, as a result of the strong physio-chemical control upon carbonate sedimentation and the potential high rates of carbonate production at the shelf margin in comparison to rates of relative sea-level rise. Carbonate sedimentation rates are also differential across a shelf and highly sensitive to slight environmental' changes such as nutrient upwelling and temperature increases or decreases. This can lead to abrupt changes of sedimentation rate not necessarily related to changes of relative sea-level. Because of these differences carbonate shelves can develop stratal patterns similar to siliciclastic settings, but in the majority of cases they are very different. In direct contrast to siliciclastic systems the lowstand systems tract is normally impoverished on the flanks of carbonate shelves. Two different end-members of lowstand sedimentation are distinguished for carbonate shelves and these reflect the inherited morphology of the slope: low angle, mud-dominated slopes are characterized by basin-floor slides and debrites during times of falling relative sea-level and by a relatively large volume autochthonous slope wedge. In direct contrast, high angle slopes are characterized by basin-floor megabreccias and volumetrically very small or even absent autochthonous slope wedges. The carbonate transgressive systems tract can also develop a wide variety of stratal patterns, a reflection of the often complex interplay of variable sedimentation rates and rates of relative sea-level rise. Two different types of geometric stacking pattern are distinguished: type 1 geometries, developed when sedimentation rates are less than rates of relative sea-level rise, and type 2 geometries formed when sedimentation rates are equal to or greater than rates of relative sea-level rise. The highstand systems tract is the time of maximum carbonate production potential and is normally associated with rapid basinwards progradation. For the highstand systems tract two different types of foreslope progradation are distinguished, slope aprons and toe-of-slope aprons. These differences between carbonate and siliciclastic depositional models suggest that simple application of the previously published models can lead to incorrect interpretation of systems tracts, sequences and therefore relative sea-level curves. Sequence stratigraphic models and concepts are tested by application to the spectacular seismic scale exposures of the mid-Cretaceous Urgonian platform, SE France. The platform is divided into a lower regressive' part, the Glandasse Formation and an upper 'transgressive' part the Urgonian Limestone Formation. These are dominated by progradational outer-shelf grainstone facies and aggradational shelf-lagoonal facies respectively. Criteria are developed to identify key surfaces and stratal packages upon the Urgonian platform. On the shelf sequence boundaries are readily defined and are marked by sub-aerial exposure surfaces associated with meteoric diagenesis. Lowstand sedimentation is generally absent, but can be represented by lacustrine facies. Strong erosional truncation is only developed on the shelf if siliciclastics are introduced during lowstand of sea-level. Thus, the transgressive and highstand systems tracts dominate shelf sedimentation but can only be distinguished if a clear flooding surface is developed, and this is not always the case. On the slope large-scale erosional surfaces developed by sedimentary bypass and/or slope collapse can develop at any stage of a sequence and make identification of the sequence boundary more difficult Similarly, on the basin-floor allochthonous debris derived from slope collapse and/or bypassing is not restricted to times of falling relative sea-level. From the criteria developed for identification of key stral surfaces and packages a sequence stratigraphy for the Urgonian platform is built This is placed within the time scale of Haq et al. (1987), and relative sea-level curves for the platform are constructed. These are compared to the eustatic sea-level charts from which they differ significantly. Minimum aggradation rates are also compared to other well known ancient carbonate platforms, from which the Urgonian is shown to have very high sedimentation rates.