Impact of Dust Radiative Forcing upon Climate - Mineral Dust: A Key Player in the Earth System - page 341

extends over the Sahara Desert, upwind of the precipitation anomaly; in other
seasons, the dust concentration and forcing are largest outside the region of
convection. Thus, the overturning anomaly ıM is not solely related to the TOA
forcing within the convecting region, although this case can be accommodated by
including forcing within the subsidence region when calculating the anomalous
energy transport and overturning. Second, Fig. 13.8 assumes that anomalous energy
transport occurs through adjustment of the overturning, but changes in the contrast
of moist static energy between the convecting and subsiding regions can also
contribute. In fact, the anomalous overturning is more closely related to atmospheric
forcing F T

F S within the subsiding region, so long as changes to the stratification
by F T are of secondary importance, as discussed in Sect. 13.3.1.2 (cf. Miller and
Tegen 1998 , 1999 ). (In this case, the regional contrast of moist static energy
adjusts to allow the necessary anomaly of energy transport.) Third, the sensitivity
of moisture import to the overturning anomaly requires that the inflow is humid and
that perturbations to humidity by dust are small, allowing the neglect of the last
term in ( 13.7 ). Where the inflow is from arid regions that are dust sources (so that
q S;d is small), the moisture import will be relatively insensitive to changes in the
circulation strength. Finally, where the dust layer extends offshore and evaporation
is reduced to compensate the surface forcing, moisture export from the subsiding
region may be sensitive to the anomalous humidity in addition to the anomalous
circulation strength, so that the last term in Eq. 13.7 is not negligible. In summary,
the mechanism illustrated by Fig. 13.8 is consistent with the relation between TOA
forcing and modeled Sahel precipitation in Table 13.2 , but several assumptions need
further consideration.
The coupling of TOA forcing and precipitation illustrated by Fig. 13.8 has three
consequences. First, ascent and precipitation can increase where F T >0within
a convecting region (Fig. 13.7 b, c), despite strong negative forcing at the surface
and intense heating within the aerosol layer. These latter conditions occasionally
lead to claims of column “stabilization” by dust, but precipitation in fact increases;
heating of the aerosol layer is nullified by adiabatic cooling associated with ascent
related to the increased export of energy. This again illustrates that aerosol forcing is
not automatically balanced by a local temperature anomaly because the circulation
allows the forcing to be balanced nonlocally. The second consequence, as shown
by Chou et al. ( 2005 ), is that the anomalous export of energy from the region of
forcing is driven by F T and not the aerosol forcing within the column F T F S .

That is, the “elevated heat pump” described by Lau and Kim ( 2006 ) drives ascent
only in the initial period, before there is significant compensation of the surface
forcing. Full compensation occurs within a few weeks over land, but takes several
months over the ocean, where heat is stored over a deeper layer. Thus, the “heat
pump” mechanism is not relevant for precipitation anomalies on interannual and
longer time scales. Again, the exception is where ocean heat transport adjusts in
response to aerosol forcing, allowing the net surface energy flux to remain nonzero
in equilibrium. The third consequence is that negative forcing at TOA by reflective
dust particles is equally effective at driving precipitation anomalies, compared to