Secular Control of Saturation State on Carbonate Precipitation Rates in The Oceans: New Insights from Ancient Carbonate Platforms

Carbonate precipitation in the oceans is a crucial part of the global carbon cycle and plays a significant role in the formation and evolution of carbonate platforms. Carbonate precipitation rates (G) are defined as the volume of carbonate formed per unit area, in the unit time. Factors such as temperature, alkalinity, salinity, hydrodynamics, and nutrient availability play significant roles in regulating carbonate precipitation and dissolution in these systems. A critical parameter controlling carbonate precipitation is the ocean saturation state of calcium carbonate (Ω), which is a measure of the stability of CaCO₃ in seawater. Laboratory experiments have demonstrated a clear correlation between Ω and the rate of carbonate precipitation in abiotic conditions (Zhong and Mucci, 1989). However, field and mesocosm data from modern reef ecosystems demonstrate a positive correlation between the Ω and G, with higher saturation levels corresponding to elevated G values. This is particularly true when looking at the geological past, where calcifying organisms that once dominated carbonate platforms got extinct while new ones emerged and the modes of carbonate precipitation were not always the same. This PhD thesis aims to explore the potential link between Ω and G in the geological past by estimating G in a number of case studies, selected among well exposed and studied carbonate platforms of different age, and comparing to modeled secular variations of Ω. A broad correlation was observed between Ω reconstructions and accumulation rates, where periods of high saturation coincided with increased rates of CaCO₃ accumulation. However, accumulation rates are calculated as thickness per unit time and are influenced by accommodation space, varying significantly in space at a given time. Quantifying G in fossil carbonate systems is challenging, as methods such as hydrochemistry, census-based estimates, sediment accumulation analysis cannot be applied to ancient carbonate platforms. To estimate G in fossil carbonate platforms, a 3D modeling approach was applied, based on stratigraphic sections and geological maps to reconstruct volumes within selected stratigraphic intervals. Productive surfaces were then estimated based on available depositional models. Estimated Gs display a negative correlation with the duration of the stratigraphic interval that was considered, with longer time intervals corresponding to lower G values. This trend is consistent with previous observations that highlighted a negative correlation between accumulation rates and the length of the time of observation (Schlager, 2003). Therefore, Gs estimates for each case study were normalized based on the linear regression between G and duration of the duration of the examined time interval. A clear linear relationship between G and Ω emerges when plotting G/Ω in a log-log relationship, using Ω values from the Ridgwell (2005) model. To account for the effects of temperature, a Generalized Linear Model (GLM) approach was used to fit a linear regression model by considering the variance distribution of multiple parameters. The resulting prediction planes were statistically indistinguishable between geological and modern data, while a notable offset indicated faster precipitation rates under biotic versus abiotic conditions, consistent with previous observations. The GLM confirmed that the same G/Ω relationship holds across all datasets, accurately predicting both class distinctions and trends. This suggests that Ω steered carbonate precipitation in modern and ancient shallow water systems in the same way, which is surprising as dominant carbonate precipitation mode and calcifying organisms of ancient and modern platforms differ and significant differences also exist within the examined fossil case studies.