Numerical Prediction of Dilution Process and Biological Impacts

in CO2 Ocean Sequestration


Toru Sato and Kei Sato


Department of Environmental and Ocean Engineering, University of Tokyo

7-3-1 Hongo, Bunkyo-ku, Tokyo 133-8656, Japan

Tel: 03-5841-6523, Fax: 03-5802-3374



In order to investigate biological impacts caused by the ocean sequestration of CO2 (carbon dioxide), the dilution processes of CO2 should be elucidated near injection points in the deep ocean. From a combination of fluid-dynamical, chemical, and biological points of view, a two-phase CFD (Computational Fluid Dynamics) method with mass transfer was developed in order to predict droplet plume flow, the dissolution of CO2 from droplets into seawater, and the advection-diffusion of DCO2 (dissolved CO2) in the deep ocean. Change of pH due to the concentration of DCO2 is also calculated. In addition, the iso-mortality concept is incorporated. The simulation results suggest that the biological impacts of the CO2 sequestration can be insignificant in terms of individual mortality both in small-amount field experiment and business cases we proposed.


Key Words: CO2 ocean sequestration, two-phase flow, mass transfer, CFD, droplet, biological impact


Contour Maps of Dissolved CO2 in Small-Amount Field Experiment Case


Volume Fraction of LCO2       DCO2                 Salinity              Temperature

Contour Maps in Business Case


As is well known, the ocean is chemically able to dissolve 1800GtC (gigatonne of carbon), while it is said that the ocean only absorbs 2GtC/yr of the CO2 originated from man-utilised fossil fuel, which is about 6GtC/yr. The dissolution of CO2 into seawater depends not only on the temperature and pressure of surface water, the thickness of which is the order of magnitude of 102m, but also on pCO2 (partial pressure of CO2) in both the atmosphere and the surface water. However, there is a big delay until the equilibrium between the atmosphere and ocean, because a thermocline barrier separates the deep water from the surface water and the global ocean circulation is the almost only mechanism to exchange water between the surface and the deep sea. Therefore, even though the surface water reaches the equilibrium, most oceans, the average depth of which is about 4000m, are hardly affected within the time range of hundreds years.

Hoffert et al.1) estimated by their box diffusion model that the 90% of fossil fuels in the earth would be depleted by 2100 and that the pCO2 in the atmosphere settles at the equilibrium of 1150ppmv(part per million by volume) between the atmosphere and ocean in about 3000 years time. The current pCO2 is 367ppmv in 20002). According to them, before the equilibrium, we would face the maximum pCO2 of 2800ppmv temporarily at the end of the 21st century, if we do not pay attention to the reduction of CO2 emission (so-called the business-as-usual case) and continue to inject CO2 into the atmosphere. This will be disastrous in many aspects, such as deserting fertile lands, rising the sea surface level, and imposing impacts on marine lives in the surface water. Direct injection of CO2 in the deep ocean is one of the ways to avoid this catastrophic overrun of pCO2 up to 2800ppmv. Therefore, it seems to be widely agreed that this option is not regarded as the permanent nor definite mitigation of greenhouse-gas effects, but as temporary evacuation from such disasters.

A very recent IPCC (Intergovernmental Panel of Climate Change) report2) says that pCO2 in the atmosphere predicted in the SRES (Special Report on Emission Scenarios) ranges from 550ppmv in the B1 case to 1000ppmv in the A1FI case at the end of the 21st century. The B1 scenario describes a convergent world with global population that peaks in mid-century and decline thereafter, with rapid change economic structures towards a service and information economy, with reduction in material intensity and the introduction of clean and resource-efficient technologies. On the other hand, the A1FI scenario implements the same global population growth as the B1, the rapid introduction of new and more efficient technologies, but more fossil-intensive energy systems. The scenario that changes the last part of the A1FI to gnon-fossil energy systemsh is called the A1T and predicts about 600ppmv in 2100.

Recently, Thornton and Shirayama3), however, pointed out that even the pCO2 of 550ppmv causes nontrivial damages on benthic lives such as sea urchins in the surface water. It is known that most marine lives are in the surface water because there is the sunshine, which give rise to the ocean food chain including phytoplankton. As was mentioned previously in this section, it takes thousands years to reach the equilibrium of pCO2 between the surface and deep waters and this fact delays the reduction of pCO2 in the atmosphere. The experimental data obtained by Thornton and Shirayama3) might force us to make a difficult choice, i.e. which do we prefer, 550ppmv or more with non-ignorable damages on marine lives in the surface water or less pCO2 by sequestrating CO2 in the deep ocean by directly injecting it?

At the moment, the CO2 ocean sequestration can be categorised into three methods, namely, shallow dissolution4, 5), middle-depth dissolution6, 7, 8), and deep storing9, 10). In this study, we focus on the middle-depth dissolution. One of uncertainties in this method is its impact on marine organisms around injection points before CO2 is diluted widely in the ocean. In the ocean, especially in the deep ocean, where the average current speed is said to be less than a couple of centimetres per second, we cannot expect as quick dilution as that in the air. Accordingly, it is significant to investigate the dilution process of DCO2 on the scale of O(102-103)m in space.

Since field experiments cost enormously, numerical simulations are expected to show detailed information on the dilution process near injection points from a fluid-dynamical point of view. This paper presents the numerical simulations of CO2 behaviour injected in the forms of droplets and solute in the deep ocean. We consider two cases, i.e. small-amount field experiment and business cases, the latter of which copes with CO2 emitted from a middle-sized thermal power plant. Our target scale covers hundreds meters in space and hours in time, roughly. Here focuses are concentrated on the movement of LCO2 droplet plume and the intrusion behaviour of DCO2. We also try predicting the impacts of CO2-rich water on marine organisms by incorporating the so-called MIT mortality curve11, 12).