What drives the structure of the stratigraphic record? A registered report study using forward modelling

The structure of the sedimentary record comprises the information which parts of the geological history are preserved and what physical record represents this history. Different depositional environments and sedimentary basins have different types of time completeness and resolution, which affects what environmental and evolutionary information can be gleaned from them. How time is preserved in rocks is reconstructed under different paradigms, which can yield different – sometimes radically – results for the same geological sections. Two widely contrasted paradigms around the formation of stratigraphic patterns are:

• external forcing through changes in the orientation and tilt of the Earth’s axis and shape of the Earth’s orbit around the Sun (Milanković cycles), which – through the insolation of Earth’s surface – affect production and erosion of sediments;
• autocyclicity, which can arise through the alternation of environments forming a mosaic, which is particularly prominent in carbonate settings, where it is reinforced by the dispersal and competition between sediment-producing biota.

We use carbonate forward models to create virtual stratigraphic architectures representing these two possibilities or their mixtures. As a “null model” we use architectures driven by a stochastic sea-level change, in this case an Ornstein-Uhlenbeck process, to account for the possibility that neither of these two paradigms allows us to obtain realistic stratigraphic patterns or that their stratigraphic expression is not implemented realistically.

Our study consists of two parts:

• a registered report which describes how the stratigraphic architectures are produced, processed using the admtools package for the R software and then classified without supervision with respect to their driving mechanism,
• an implementation using a newly developed Open Source model of carbonate platform growth, CarboKitten.jl, inspired by previous models, CarboCAT and CARBONATE 3D.

Such design allows reproducing the study using other forward models, such as SedFlux, either representing different depositional environments or using different algorithms to construct the same environment. It does not exhaust possible drivers of time-preservation in rocks. For example, it does not yet allow to account for diagenetic self-organization, i.e. stratigraphic patterns emerging post-deposition, nor lags of the climate system associated with processes such as ice dynamics. However, it allows to explore how time-completeness and distribution of gaps affects reconstructions of environmental and evolutionary processes. The forward modelling approach allows for testing hypotheses on the mechanisms of the formation of stratigraphic record in a reproducible way.