Vol. 12, issue 03, article # 6

Barker H. W., Fu Qiang. Modelling Domain-averaged Solar Fluxes for an Evolving Tropical Cloud System. // Atmospheric and oceanic optics. 1999. V. 12. No. 03. P. 211-217.    PDF
Copy the reference to clipboard

Abstract:

Domain-averaged, broadband solar radiative budgets for an evolving tropical mesoscale convective cloud system are computed by two approximate 1D models which make different assumptions about the structure of unresolved clouds. One model is the standard plane-parallel, homogeneous (PPH) two-stream approximation. The other assumes that fluctuations in cloud extinction b can be described by a gamma distribution pG(β) and so weights the two-stream equations by pG(β) and integrates over all β. A 3D Monte Carlo (MC) algorithm provides reference calculations. The cloud system was simulated by a 2D cloud-resolving model and the domain measures 514 km horizontally and ~18 km vertically. Horizontal grid-spacing is 1 km while the 35 layers vary in thickness. The hydrometeors accounted for are liquid droplets, ice crystals, rain, graupel, and snow. Snapshots of the domain were saved every 5 model-minutes for 10 hours thus spanning the life-cycle of the system.
It is shown that the conventional PPH two-stream is thoroughly inappropriate as it yields 10 hour mean TOA albedo atoa and surface absorptance asfc of 0.56 and 0.20 but the corresponding 3D MC values are 0.32 and 0.47. For the gamma-weighted two-stream approximation (GWTSA), however, αtoa and αsfc are 0.32 and 0.49. Moreover, while heating rate errors for the PPH model are about –0.5 K/day near the surface and almost +2 K/day at 10 km, they are diminished at both altitudes to 0.25 K/day for the GWTSA. For reference, it is shown that the independent column approximation is almost identical to the 3D MC and that while the best possible PPH model (i.e., perfect account of cloud overlap) is vastly superior to the regular PPH model, it is significantly inferior to the GWTSA.