Radiative forcing of a tropical direct circulation by soil dust aerosols

The effect of soil dust aerosols upon the tropical climate is estimated by forcing a simple model of a tropical direct circulation. The model consists of a region vertically mixed by deep convection and a nonconvecting region , for which budgets of dry static energy and moisture are constructed. Dyna mical effects are included implicitly, by prohibiting horizontal temperatur e contrasts above the boundary layer. Dust aerosols absorb sunlight to a greater extent than industrial sulfate a nd sea-salt aerosols. In a companion study, where the climate response to d ust is calculated using an atmospheric general circulation model, the globa l-average dust radiative forcing is negligible at the top of the dust layer ,in comparison to the large reduction of the net flux at the surface. Thus, dust aerosols redistribute radiative heating from the surface into the dus t layer, unlike industrial sulfates and sea salt, which through reflection reduce the total radiative energy gained by the column. The simple model is perturbed by a reduction in the net radiative flux at t he surface. Forcing at the top of the dust layer is idealized to be zero. C ooling occurs at the surface of the nonconvecting region, but surface tempe rature within the convecting region is only slightly perturbed. It is shown that the disproportionately small response within the convecting region is a consequence of the trivial radiative forcing at the top of the dust laye r, and the occurrence of deep convection, which prevents the surface temper ature from changing without a corresponding change of the emitting temperat ure in the upper troposphere. Additional experiments, where the absorptivity of the dust particles is var ied, indicate that the anomalous surface temperature is most sensitive to t he radiative forcing at the top of the dust layer. The reduction of the sur face net radiation is less important per se but introduces an asymmetry in the response between the convecting and nonconvecting regions through the r adiative forcing within the dust layer, which is the difference between the forcing at the surface and the layer top. This heating can offset radiativ e cooling above the boundary layer, reducing the strength of the circulatio n that links the nonconvecting and convecting regions. The weakened circula tion requires cooling of the nonconvecting region relative to the convectin g region in order to maintain the export of energy from the latter to the f ormer. It is suggested that the "semi-indirect" effect of aerosols, wherein cloud cover is changed in response to aerosol heating, is sensitive to the vertic al extent and magnitude of the aerosol forcing. The experiments suggest that dust optical properties (to which the top of t he atmosphere forcing is sensitive) should be allowed to vary with the mine ral composition of the source region in a computation of the climate respon se. More extensive measurements of the dust optical properties, along with the vertical distribution of the dust layer, are needed to reduce the uncer tainty of the climate response to dust aerosols.