Brief outline of the ASGAMAGE model simulations

Gas exchange simulations for CO2, helium (He) and sulphur hexafluoride (SF6) were performed using complementary one and two-dimensional models (1-D model, 2-D model, respectively) of the Oceanic Boundary Layer.

Model phenomenology
Simulations
    Forcing; initial and boundary conditions
    dt-simulations
    ec-simulations
References

The Models

The 1-D model is suitable for studying diurnal variations. It allows the penetration of light into the water, and it parameterizes the bottom boundary layer at the seabed, including mixing by tidal currents. Enhancement of turbulence by wave breaking is parameterized as well. The 2-D model is suited to studying the effect of horizontal inhomogeneities on fluxes and concentrations, and it allows detailed resolution of the Atmospheric Surface Layer. But it runs for strictly neutral stability conditions in water and air, and ignores the influence of solar radiation. The main model characteristics are compared in the table below.

Brief characterization and comparison of the models used in this study
 
      Model 1: one-dimensional Model 2: two-dimensional
  • designed to model Oceanic Boundary Layer structure and fluxes; adjusted and extended from Large et al [1994]
  • diurnal evolution: dynamic simulation of water speed, temperature, salinity and passive tracer (inert or CO2)
  • CO2 chemistry: slow hydration and chemical equilibrium of bicarbonate and carbonate [Emerson, 1995] 
  • non local, 1st order turbulence closure
  • skin layer and kw: surface renewal theory
  • wave breaking enhances turbulence; it increases diffusivity and decreases surface renewal time scale; parameterization uses significant wave height and wave age [Terray et al, 1996]
  • buoyancy effects and light penetration included
  • Bottom Boundary Layer instead of deep-sea formulation of Large et al [1994]
  • tidal current + wind induced current
  • flux-profile relationships for atmospheric surface layer (reference level at 10 m)
  • staggered grid with spacing 0.005-0.2 m in the water, 110 levels
  • designed to study effects of inhomogeneities on gas flux (Kjeld, 1999)
  • spatial distribution: dynamic simulation of passive tracer (inert or CO2)
  • CO2 chemistry: chemical equilibrium 
  • local, 1st order turbulence closure
  • kw = 12.4 Sc-1/2 u*2 [Coantic, 1986]
  • roughness length in water adjusted to wave breaking [Zillitinkevitch and Kreiman, 1994]
  • strictly neutral stratification
  • conditions at 5 m (water) fixed
  • fixed water current + wind induced current
  • ASL concentrations and fluxes resolved with high vertical resolution (logarithmic grid with 102 levels)
  • logarithmic grid between z0,w and reference depth, 102 levels

[top]

Simulations

Forcing; initial and boundary conditions

In both models the momentum flux across the air-sea interface and the reference level wind are related by a logarithmic wind profile, using roughness length z0=0.015u*2/g, where u* is friction velocity. Initial conditions were chosen to resemble the environmental conditions during the ASGAMAGE field experiment in the fall of 1996 (ASGAMAGE-B): water temperature=12oC, salinity=31‰.

In the 1-D model, the amplitude of tidal current was set to 0.5 m/s (two cycles per day). Significant wave height and wave age were set to conform with the ASGAMAGE-B data (range 0.32-4.94m and 2-72, respectively). The net heat flux was driven by conditions at a reference level in air (10 m), taken as constant during a run (temperature 12oC, relative humidity 85%). Bulk formulations were used to compute the sensible and latent heat flux (DeCosmo et al, 1996). For the amount of solar radiation absorbed by the water values were used corresponding to fair weather conditions in De Bilt (Netherlands) (Holtslag and van Ulden, 1983), and assuming an albedo of 0.1.

The 2-D model was driven by concentration changes at the reference levels in air and water (10 and 5m respectively). Such conditions can represent horizontal homogeneity as well as advection. In the examples presented here, u* was fixed at 0.4 m/s. The velocity of the water was computed assuming a logarithmic profile, with water roughness length 5.5u*2/g [Zilitinkevitch and Kreimann, 1994].
[top]

Simulation of a dual tracer experiment in the North Sea

With the 1-D model a dt-experiment in the North Sea can be simulated as follows. He and SF6 are “released” into the water in initially equal concentrations with instantaneous and complete mixing. Because tracer concentrations in the air are maintained at zero,  the tracer concentrations in the water will gradually decrease by gas exchange and mixing. Then, a “measured” transfer velocity can be determined from the change in the concentration ratio of He and SF6  at some depth below the sea surface (5 m) just like in a field experiment (Nightingale et al, 2000). These results are then compared to the "true" transfer velocity, that is determined by the model equations.
The 2-D model delivers a stationary solution. Thus, the concentrations in the water cannot be followed in time. However, 2-D concentration distribution of a tracer patch can be simulated. To achieve this, a Gaussian distribution of the concentration is applied at the reference level in the water. From the 2-D concentration field, the errors in the "measured" transfer velocity due to deviations from the assumption of perfect mixing can be estimated (Kjeld, 1999).

[top]

Simulation of eddy correlation measurements

EC-measurements are simulated using CO2 as the passive tracer. We take the initial fugacity in the water 46 Pa (well mixed); the fugacity in the air is 36 Pa throughout. This  resembles the conditions during ASGAMAGE-B. The 1-D model can be forced to quasi-stationarity, that is, diurnal averages do not vary significantly anymore. This is done using a source term for dissolved inorganic carbon chosen so as to balance exactly the rate of CO2 efflux. In another set of simulations, the CO2 concentration in the water was allowed to increase with a tidal modulation, just like observed during ASGAMAGE-B. In this case, the source strength was chosen increase the CO2 fugacity in the water by 1 Pa/day. Further, we took it inversely proportional to the square of the distance from the air-sea interface.
Similar simulations were performed with the 2-D model, for a horizontally homogeneous concentration field and for Gaussian distributions of the concentration at the reference level. It contains no sources or sinks for inorganic carbon.
In all cases, the simulated transfer velocity can be computed using the simulated CO2 flux and concentration difference between air and water. This "observed" value can then be compared to the true value, given by the model equations.

[top]

References

Coantic, M., A model for gas transfer across air-water interfaces with capillary waves, J. Geophys. Res., 91, 3925-3943, 1986.

Decosmo, J., K.B. Katsaros, S.D. Smith, R.J. Anderson, W.A. Oost, K. Bumke and H. Chadwick, Air-sea exchange of water vapor and sensible heat: the Humidity Exchange of the Sea (HEXOS) results, J. Geophys. Res., 101, 12001-12016, 1996.

Emerson, S., Enhanced transport of Carbon Dioxide during gas exchange, in Air-water gas transfer. Selected papers from the Third International Symposium on air-water gas transfer, edited by B. Jähne and E.C. Monahan, AEON Verlag & Studio, Hanau, pp. 23-35, 1995.

Holtslag, A.A.M. and A.P. van Ulden, 1983: A simple scheme for daytime estimates of the surface fluxes from routine weather data. J. Climate Appl. Meteor., 22, 517-529.

Kjeld, J.F., A model study of air-sea gas exchange of trace gases and wind flow in complex terrain, Ph.D. Thesis, Risø, 1999.

Large, W.G., J.C. McWilliams and S.C. Doney, 1994: Oceanic vertical mixing: a review and a model with a nonlocal bound-ary layer parameterization. Rev. Geophys., 32, 364-403.

Nightingale, P.D., G. Malin, C.S. Law, A.J. Watson, P.S. Liss, M.I. Liddicoat, J. Boutin, and R.C. Upstill-Goddard, In situ evaluation of air-sea gas exchange parameterizations using novel conservative and volatile tracers, Global Biogeochem. Cycles, 14, 373-387, 2000.

Terray, E.A., M.A. Donelan, Y.C. Agrawal, W.M. Drennan, K.K. Kahma, A.J. Williams III, P.A. Hwang and S.A. Kitaigorodskii, Estimates of kinetic energy dissipation under breaking waves, J. Phys. Oceanogr., 26, 792-807,1996.

Zilitinkevitch, S.S. and K.D. Kreimann, Wind-induced drift of surface films, in Modeling air-lake interaction, edited by S.S. Zilitinkevitch, pp. 63-73, Springer Verlag, Heidelberg, 1991.

[top]