A model for marine sedimentary carbonate diagenesis and paleoclimate proxy signal tracking: IMP v1.0

Abstract. The preservation of calcium carbonate in marine sediments is central to controlling the alkalinity balance of the ocean and, hence, the ocean–atmosphere partitioning of CO2 . To successfully address carbon cycle–climate dynamics on geologic ( ≫1  kyr) timescales, Earth system models then require an appropriate representation of the primary controls on CaCO3 preservation. At the same time, marine sedimentary carbonates represent a major archive of Earth history, as they have the potential to preserve how seawater chemistry, isotopic composition, and even properties of planktic and benthic ecosystems, change with time. However, changes in preservation and even chemical erosion of previously deposited CaCO3 , along with the biogenic reworking of upper portions of sediments, whereby sediment particles are translocated both locally and nonlocally between different depths in the sediments, all act to distort the recorded signal. Numerical models can aid in recovering what the “true” environmental changes might have been, but only if they appropriately account for these processes. Building on a classical 1-D reaction-transport framework, we present a new diagenetic model – IMP (Implicit model of Multiple Particles (and diagenesis)) – that simulates biogeochemical transformations in carbonate-hosted proxy signals by allowing for populations of solid carbonate particles to possess different physicochemical characteristics such as isotopic value, solubility and particle size. The model also utilizes a variable transition matrix to implement different styles of bioturbation. We illustrate the utility of the model for deciphering past environmental changes using several hypothesized transitions of seawater proxies obscured by sediment mixing and chemical erosion. To facilitate the use of IMP, we provide the model in Fortran, MATLAB and Python versions. We described IMP with integration into Earth system models in mind, and we present the description of this coupling of IMP with the “cGENIE.muffin” model in a subsequent paper.