الفهرس 
Emission of mineral dustOnly 14 pages are availabe for public view
 |  | from | 136 |  |  |
| | | from | 136 | | |
Abstract
In this study, we assessed the role played by mineral dust in three major environmental impacts as followed i) modifying the optical properties of aerosols due to anthropization, ii) modifying the composition of atmosphere by heterogeneous reactions with gaseous pollutants, iii) tracing the sources of mineral dust by elemental ratios. First, we use the level-2 Aeronet inversion products of 19 stations of North Africa, the Middle East, or downwind of them, to characterize the degree of anthropization of the corresponding sub-regions. For this, we define a new dust-index (Idust) based on the statistical analysis of the frequency distributions of the aerosol optical depth at 440nm and its Fine Mode Fraction (FMF440). Idust is found to be more than one order of magnitude larger at the remote Saharan site of Tamanrasset (Idust=6.3) and at the 4 Sahelian stations (2.1<Idust<5.8) than along the Mediterranean coast (Idust=0.06, 0.08, and 0.09, in Blida, Tunis, and Lampedusa, respectively) or in the Middle East (Idust=0.11 in Sede Boker and 0.05 in Cairo). With Idust varying from 0.16 to 0.81, the stations of Morrocco (Oujda, Saada), south Tunisia (Medenine), the Egyptian western desert (El Farafra), and the Arabian Peninsula (Kaust Campus, Mezaira, and Solar Village) occupy an intermediate position. In spite of the different degrees of anthropization of the 19 stations, the Aeronet inversions do not reveal any differentiation of the real (n) and imaginary (k) parts of the refractive index, asymmetry parameter (g), and single scattering albedo (SSA) of the coarse mineral dust component of the aerosol. As a result of these commonalities, the inter-site variability of the instantaneous dust radiative forcing efficiencies at TOA and BOA (relative to the AOD440, and for solar zenith angles in the range 60±5°) is found to be entirely explainable by the differences of albedo () of the underlying surfaces: FEBOA=161-145 (R2=0.91, n=19), and FETOA=192-103 W m-2 AOD440-1 (R2=0.93, n=19). Conversely, the characteristics of the fine anthropogenic aerosols are found to be more variable. In particular, the large k (0.010, 0.012, 0.013, and 0.017 at Banizoumbou, Dakar, Cinzana, and Cabo Verde, respectively) and low SSA of the pollution aerosols of the Sahelian region denote a strong absorbing character to be linked to their most probable bio-mass burning origin. Generally, under pollution episodes, the forcing efficiency is found to be more negative (cooling) at BOA than during dust episodes, but less negative at TOA, or even close to 0 over the brightest surfaces. These results stress the importance of accounting for the surface albedo in the quantification of the climate impact of the aerosols.
Second, the data of 17 air quality monitoring stations of Greater Cairo are used to perform a statistical analysis aiming to detect any heterogeneous surface effects of mineral dust on the distribution of reactive trace gases (SO2 NO2, and ozone) in. After a thorough quality check, the methodology consisted of i) selecting representative stations by agglomerative hierarchical clustering, ii) identifying dust events based on PM10 measurements, remote sensing observations, and meteorology, and iii) applying the non-parametric Kruskal Wallis (KW) hypothesis test to compare (at the 95% confidence level) trace gas concentrations during dust and non-dust events. The representative stations display either a background-like or a bimodal variability with concentrations (even that of the secondary product NO2) peaking at traffic rush hours but during dust storms all stations capture the signal of mineral dust advection. Eight wintertime and springtime dust cases are retained for the study. After the role of the confounding factors (i.e., ventilation index, relative humidity, and photolysis) has been carefully discussed and taken into account, the KW test shows that there is no significant reduction of the SO2, NO2 and ozone concentrations attributable to dust during 7 of the 8 events. The DROP of the concentrations coinciding with the advection of dry dust-laden Saharan air masses is rather an effect of the dilution resulting from the combination of large wind speed and mixing layer height than of the heterogeneous uptake of these gases on the mineral dust surface.
The third objective of this study is to confirm the possibility of tracing sources of airborne mineral dust using elemental ratios. The region of study is the north-eastern part of the Sahara where dust compositional data were lacking. 272 mineral dust samples collected at 3 experimental stations of Egypt were analyzed by X-Ray fluorescence. Based on the mineral dust concentration, 65% of the samples correspond to moderate (3-10 μgm−3), 22% to medium (10- 25 μgm−3), and 13% to intense (> 25 μgm−3) dust-events. The Ca/Al and (Mg+Ca)/Fe ratios were found to be strongly correlated and their ranges of variation to cover more than one order of magnitude, which confirms their interest for the tracing of the dust sources. Using a combination of MODIS satellite observations, HYSPLIT back-trajectory analysis, and simulations with dust emission models, the large (>6) and intermediate (2-6) Ca/Al ratios were shown to correspond to dust originating from saline playas (chotts) and calcareous surfaces, respectively. Conversely, the lowest (<2) ratios correspond to dust emitted by siliceous sand seas (ergs). Therefore, a classification of the dust in 3 categories (erg-like, calcareous, and chott-like) based on Ca/Al is proposed. The events of moderate to medium intensity are dominated by chott-like and calcareous dust, which suggests that these sources are activated during erosion events of limited (local) extension. Conversely, the fact that the dust composition is more likely to be of the erg-like type during intense events underlines the dominant role played by sand dune fields during major erosion events.