Geoscience Reference
In-Depth Information
care should be taken in assigning far-travelled dusts (e.g., paleo dusts in ice cores)
to “continental source areas” without other evidence. Finally, it should be borne in
mind that especially 87 Sr/ 86 Sr analyses are grain-size sensitive and that only similar
(carbonate-free) size fractions can be compared in a strict sense (Meyer et al. 2011 ).
2.2.3
Elemental Data
The third important bulk analytical technique to characterize mineral dust samples
is the chemical analysis of major and/or trace elements. However, the compilation
and evaluation of this data set is hampered, for example, by the different types of
data obtained (mass fraction [wt%] versus aerosol mass concentration [g/m 3 ]),
the different analytical methods used (atomic absorption spectroscopy (AAS),
inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled
plasma optical (atomic) emission spectroscopy (ICP-OES or ICP-AES), X-ray
fluorescence spectrometry (XRF), particle-induced X-ray emission (PIXE), or
instrumental neutron activation analysis (INAA)), and the grain-size dependence
of the elemental composition. Nevertheless, the elemental analyses of northern
African dust samples were compiled and discussed in detail by Scheuvens et al.
( 2013 ) and will be briefly reviewed in the following (starting with the mass fraction
data set). As expected, in many northern African dust samples, Si is the most
abundant element and is in average slightly enriched compared to the composition
of the upper continental crust. Si/Al ratios fall mainly into a range between 2 and 7,
pointing to mixtures between quartz and aluminosilicates in agreement with the
mineralogical data. Regional trends in the Si/Al ratio are not observed. Scheuvens
et al. ( 2013 ) could show that especially the ratio (Ca
Mg)/Fe has a high potential
for the discrimination between the different potential source areas in northern Africa
(see their Fig. 2). Mineral dusts from the Atlas region, central Algeria, Libya,
and Egypt are generally characterized by (Ca
C
Mg)/Fe ratios >1, whereas dust
samples collected in the sub-Saharan belt (including the Bodélé depression in
Chad) mainly exhibit (Ca
C
Mg)/Fe ratios <1 in agreement with the mineralogical
data presented above (carbonate-rich mineral dusts in more northern latitudes). All
other major elements (e.g., K, Na, Fe, Mn, Ti, P) and elemental ratios (with the
“crustal” element Al as the denominator) do not show any distinct regional trends
and should not be used for any source apportionment in northern Africa. In average,
northern African dusts are slightly depleted in K and Na and slightly enriched
in Ti and P when compared to the average composition of the upper continental
crust. The abundance of Fe in mineral dusts arouses interest because of its possible
capability to serve as nutrient in downwind marine (Atlantic ocean) and terrestrial
ecosystems (Amazon forest) and its significant influence on the optical properties
of mineral dust in the visible range of the spectrum (e.g., Arimoto et al. 2002 ).
Meanwhile, a series of studies focused on the partitioning of Fe between iron oxides
(e.g., hematite) and iron hydroxides (e.g., goethite), on one hand, and crystal-bound
Fe (mainly in clay minerals), on the other hand, mainly using the CBD extraction
C
Search WWH ::




Custom Search