Updated on June 3, 2003

CEA-DSM work examples

1. Our main objectives

1.1 Development and calibration of 1-D biogeochemical model components: early diagenesis sediment module.
1.2 Implementation of a module describing particle dynamics (coagulation, disaggregation and settling of biogenic particles) in the IPSL biogeochemical ocean general circulation model (OPA-PISCES).
1.3 Laboratory experiments on calcium carbonate reactivity (solubility and dissolution kinetics).

2. Examples of our work

1.1 1-D sediment early diagenesis model: Implementation of the sedimentary Si-cycle within the FEMME modelling environment. The model distinguishes between two classes of biogenic Si with differing dissolution characteristics : a fast and a slow dissolving phase. Dissolution of either fraction follows first order kinetics. The respective dissolution rate parameters are constant with depth (no aging). The Si early diagenesis model has been compared to observed dissolved Si and BSi distributions (Rabouille et al. 1996). First results are encouraging (Figures 1 and 2).

Figure 1: Comparison between observed data of dissolved silica from Antarctic and fitted data by SIDIA Model including 2 types of biogenic silica.

Figure 2: Comparison between observed data of solid silica from Antarctic and fitted data by SIDIA Model including 2 types of biogenic silica.

1.2 Implementation of a module describing particle dynamics: Sensitivity of the ocean-atmosphere CO2 partitioning to the representation of vertical organic matter fluxes. We have run 3 different versions of the PISCES model, but forced with the same dynamical forcing fields (obtained with the Océan Parallélisé (OPA) model).

CASE 1.
One particle size fraction sinking with a constant speed (10 m/day) - control simulation.

CASE 2.
Two particle size fractions. Sinking speeds are constant over time but differ for the 2 fractions. For the small particles, the sinking speed is set to 3 m/day in the euphotic layer and increase to 10 m/day at depth. For the large particles, the sinking speed is set to 10 m/day in the photic layer and increases to 200 m/day at depth - standard PISCES simulation.

CASE 3.
We have introduced a more advanced particles dynamic module, based on the work of Kriest and Evans (1999,2000). We explicitely represent the sinking of particles below the photic zone and incorporate a mechanistic representation of aggregation/disaggregation of particles.

Figure 3 compares computed settling velocities of particles (preliminary results) to estimates derived from sediment trap data. Settling velocity of particles: deduced from sediment trap data for the Arabian Sea (black boxes) and the Equatorial Pacific (yellow boxes) from Berelson (2002) ; imposed in the Case 1 simulation (black line) and in the Case 2 simulation for small and large particles (green lines) ; in the Case 3 simulation, the sinking speed is prognostically computed from the mean aggregate size and averaged over the globe (red line).

1.3 Parameterisation of carbonate dissolution.

The quantification of in situ carbonate dissolution rates relies on an accurate parameterisation of carbonate dissolution kinetics, which implies an improved knowledge of the rate constant and the solubility product.

Experimental approach:
Free-drift experiments are run under controlled T and pCO2. The chemical composition of the experimental solution is allowed to drift in response to the ongoing reaction. Reaction progress is followed by monitoring solution pH. The speciation of the carbonate system at any time is obtained by combining pCO2 (constant) and observed pH. This approach allows the simultaneous acquisition of kinetic and solubility data.

Figure 4

Figure 5

Kinetic experiments were run using core top sediments sampled along a bathymetric transect across Sierra Leone Rise (tropical Atlantic). Figure 6 exemplifies our experimental results.

Figure 6