Karen M. Shell     


Direct radiative effect of dust on the climate system

Relevant Papers:

Shell, K. M., and R. C. J. Somerville (2007), Sensitivity of climate forcing and response to dust optical properties in an idealized model, J. Geophys. Res., 112, D03206, doi:10.1029/2006JD007198. (**)

Shell, K. M., and R. C. J. Somerville (2007), Direct radiative effect of mineral dust and volcanic aerosols in a simple aerosol climate model, J. Geophys. Res., 112, D03205, doi:10.1029/2006JD007197. Online supplement (**)

Shell, Karen M., and Richard C. J. Somerville, 2005: A generalized energy balance climate model with parameterized dynamics and diabatic heating, J. Clim., 18: 1753-1772. (*)

Figure from Shell and Somerville (2007b). Global average steady state (top) atmospheric temperature, (middle) surface temperature, and (bottom) latent heat change as a function of (left) Top-of-the-Atmosphere and (right) surface shortwave dust forcing for each set of optical parameters in the shortwave sensitivity experiment. Also shown are results from versions of the model which omit the latent heat and sensible heat, emissivity, or lapse rate feedback.

Airborne mineral dust can influence the climate by altering the radiative properties of the atmosphere. However, dust optical properties and concentrations are uncertain, and thus the signs and magnitudes of climate forcing and response are not well known. For my dissertation research, with Richard Somerville, I developed a new zonally-averaged energy-balance climate model to study the effect of dust. The model is similar to earlier energy-balance models but includes unique features, such as separate surface and atmospheric layers, an interactive atmospheric lapse rate, and individual calculations of the meridional transport of heat due to the three major processes: the Hadley circulation, mid-latitude baroclinic eddies, and ocean currents and eddies. My simple model has two main benefits. First, because it is much less complicated than general circulation models (GCMs), it is easier to gain a conceptual understanding of how dust and the climate system interact. GCMs are, of course, necessary to obtain detailed estimates of climate change; however, they are often difficult to interpret. My model can be a useful tool for understanding both the actual climate system and GCM results. The other main advantage to my model is its computational speed. In one day, I can perform over a thousand experiments on a workstation. Similar experiments with a GCM would take years of supercomputer time. Thus, with my model, I can test many different configurations of dust distributions and optical properties. The ability to explore large regions of parameter space in detail is especially useful for the dust-climate problem, since there are still large uncertainties in the amount and optical properties of dust.

After developing my model, I compared it with observational results to evaluate its realism. The model reproduces both the observed mean variation of temperature with latitude and the global average heat budget. The model also produces realistic radiative forcing values for the present day dust distribution. Finally, modeled radiative, temperature, and hydrological changes caused by the Mount Pinatubo eruption compare favorably to observations and general circulation model results.

Recent observations suggest that dust absorbs less solar radiation than previously thought. Studies which use a shortwave single scattering albedo corresponding to World Climate Program dust optical properties may underestimate the effect of dust. Using a solar single scattering albedo value of 0.97, corresponding to recent measurements, the modeled global average top-of-atmosphere (TOA) shortwave forcing is -0.73 W/m2 for the present day dust distribution. The TOA longwave forcing is 0.23 W/m2, the surface shortwave forcing is -1.3 W/m2, and the surface longwave forcing is 0.37 W/m2. For plausible ranges in dust optical properties, the net TOA forcing varies from -2.0 W/m2 to +3.1 W/m2.

Dust decreases the surface and atmospheric temperatures by about 0.1 K and reduces latent and sensible heat fluxes from the surface to the atmosphere by 1%, moderating the surface cooling. These results suggest that changes in the hydrological cycle may be a significant effect of dust. Based on the range of optical property values used, the surface temperature change ranges from -0.3 K to 0.6 K. The latent heat change ranges from -1.9 W/m2 to 0.5 W/m2 . In addition, temporal variations in dust concentration contribute to interannual climate variability. While changes in the temperature can be estimated from the TOA forcing, the hydrological cycle changes require knowledge of the vertical distribution of dust forcing. Thus, focusing on the TOA forcing alone, rather than the vertical distribution of forcing, neglects important climatic changes.

One key concept that emerges from this work is the importance of vertical resolution in a climate model, especially when considering the radiative effects of aerosols. Even the rough vertical resolution of this model, which consists of one surface layer, a single atmospheric layer, and a climate-dependent lapse rate, allows for exploration of processes not normally considered by simple models. In fact, models without vertical resolution are of limited use in determining the effect of dust, because the climate response depends not only on the TOA forcing but also on the surface forcing, and the dust forcing is itself sensitive to the height of the dust. In addition, changes in latent and sensible heat flux from the surface to the atmosphere are a significant component of the climate response, modifying the way the temperatures adjust to achieve a new equilibrium after a climate forcing is applied. Other feedbacks, such as the lapse rate and emissivity feedbacks, also rely on some amount of vertical differentiation. Most energy balance models consist of a single column-average layer, omitting vertical resolution. Thus, this model fills a particular need for a model which includes some vertical variation while remaining simple, easy-to-understand, and fast.

Last modified: Thu Nov 1 11:44:11 PDT 2012