Working Group I: The Scientific Basis

Other reports in this collection

5.2.2 Primary and Secondary Sources of Aerosols Soil dust

Soil dust is a major contributor to aerosol loading and optical thickness, especially in sub-tropical and tropical regions. Estimates of its global source strength range from 1,000 to 5,000 Mt/yr (Duce, 1995; see Table 5.3), with very high spatial and temporal variability. Dust source regions are mainly deserts, dry lake beds, and semi-arid desert fringes, but also areas in drier regions where vegetation has been reduced or soil surfaces have been disturbed by human activities. Major dust sources are found in the desert regions of the Northern Hemisphere, while dust emissions in the Southern Hemisphere are relatively small. Unfortunately, this is not reflected in the source distribution shown in Figure 5.2(f), and represents a probable shortcoming of the dust mobilisation model used. Dust deflation occurs in a source region when the surface wind speed exceeds a threshold velocity, which is a function of surface roughness elements, grain size, and soil moisture. Crusting of soil surfaces and limitation of particle availability can reduce the dust release from a source region (Gillette, 1978). On the other hand, the disturbance of such surfaces by human activities can strongly enhance dust mobilisation. It has been estimated that up to 50% of the current atmospheric dust load originates from disturbed soil surfaces, and should therefore be considered anthropogenic in origin (Tegen and Fung, 1995), but this estimate must be considered highly uncertain. Furthermore, dust deflation can change in response to naturally occurring climate modes. For example, Saharan dust transport to Barbados increases during El Niño years (Prospero and Nees, 1986), and dust export to the Mediterranean and the North Atlantic is correlated with the North Atlantic Oscillation (Moulin et al., 1997). Analysis of dust storm records shows regions with both increases and decreases in dust storm frequency over the last several decades (Goudie and Middleton, 1992).

Figure 5.2: Annual average source strength in kg km-2 hr-1 for each of the aerosol types considered here (a to g) with total aerosol optical depth (h). Shown are (a) the column average H2 SO4 production rate from anthropogenic sources, (b) the column average H2 SO4 production rate from natural sources (DMS and SO2 from volcanoes), (c) anthropogenic sources of organic matter, (d) natural sources of organic matter, (e) anthro-pogenic sources of black carbon, (f) dust ources for dust with diameters less than 2 µm, (g) sea salt sources for sea salt with diameters less than 2 µm, and (h) total optical depth for the sensitivity case CHAM/GRANTOUR model (see Section

Table 5.3: Primary particle emissions for the year 2000 (Tg/yr)a.
Global Low High
Carbonaceous aerosols          

Organic Matter (0-2 Ám)


Biomass burning

28 26 54 45 80 Liousse et al. (1996),
Scholes and Andreae (2000)
    Fossil fuel 28 0.4 28 10 30 Cook et al. (1999),
Penner et al. (1993)
    Biogenic (>1 Ám) - - 56 0 90 Penner (1995)
  Black Carbon (0-2 µm)          
    Biomass burning 2.9 2.7 5.7 5 9 Liousse et al. (1996);
Scholes and Andreae (2000)
    Fossil fuel 6.5 0.1 6.6 6 8 Cooke et al. (1999);
Penner et al. (1993)
    Aircraft 0.005 0.0004 0.006  
Industrial Dust, etc. (>1 Ám)   100 40 130 Wolf and Hidy (1997);
Andreae (1995)
Gong et al. (1998)
Sea Salt            
     d< 1 µm 23 31 54 18 100  
    d=1-16 µm 1,420 1,870 3,290 1,000 6,000  
    Total 1,440 1,900 3,340 1,000 6,000  
Mineral (Soil) Dustb            
    d< 1 µm 90 17 110 - -  
    d=1-2 µm 240 50 290 - -  
    d=2-20 µm 1,470 282 1,750 - -  
    Total 1,800 349 2,150 1,000 3,000  
a Range reflects estimates reported in the literature. The actual range of uncertainty may encompass values larger and smaller than those reported here.
b Source inventory prepared by P. Ginoux for the IPCC Model Intercomparison Workshop.

Continues on next page

Other reports in this collection

IPCC Homepage