I have known some indicators, like negative Zr anomalies, low Th, Zr and LREE contents, low La/Sm, (Th/Ta)N, Th/La and high Nb/La ratios. But I don’t know whether these are sufficient and robust to support my discussions. So I need your help to recommend some other indicators (or related papers). Thank you advance!
Nb/Th ratios can be used to characterize crustal contamination of basaltic and komatiitic lavas. Generally Nb/Th ratio
Thank you for your helpful suggestions, Professor Jayananda. I will read the article carefully.
Strictly speaking ALL the basalts bear some indication of interaction with crustal lithologies, MORB and OIB comprised.
Your question is quite simple but is quite impossible to be answered correctly. It is now clear that also the basalts from the type locality of OIB (Hawaii) cannot be explained without taking into account the involvement of recycled lithologies (carbonate or Mn-rich sediments, subducted oceanic crust) or simply olivine-poor (or olivine-free) lithologies resulting from complex interaction between frozen basaltic melts (stable in the eclogitic facies at P >1 GPa) with ambient peridotite. The literature is full of papers quoting this. One of the most commonly quoted in literature is Sobolev et al., 2005 (Nature, 434, 590-597), but many many other are available from the web.
The same holds for MORB, whose "garnet" signature has been evidenced for about twenty years (Hirschmann and Stolper, 1996, CMP124, 185-208)
Another matter depends on the "pristine" composition of the melt and the composition of the potential crustal contaminat. Continental crust is not a homogeneous object. If you mix travertine or limestone with a basalt you will obtain a mix product that is completely different if you mix the basalt with an arkose or a granite. Similarly, if you mix a basalt with a granite you will obtain a result that is completely different if you mix a lamproite with a granite. Things are not as simple as they like.
Cheers
michele
Dear Professor Lustrino,
Yes, I toally agree with you and thank you for your helpful answer. But what I really want to convey is that how to identify crust contamination/assimilation during ascent of the basaltic magma, not that in the source region. If there is no such contamination during ascent, I can use the geochemical data to trace the source and tectonic setting.
Best wishes.
Houxiang
Dear Houxiang,
in theory what you ask for is easy, but in practice it is not always so.
Everything depends on the differences between the original mantle melt (i.e., basaltic melt in general) and the potential contaminant. As already written, if the contaminant is a limestone, the results of contamination could be completely opposite to what usually assumed. This is not an anomalous case. For example, see the leucitites from a small volcano around Rome (Alban Hills). Here the interaction of basaltic melt with local crust (essentially Jurassic to Cretaceous limestones) produces hybrid melts extremely poor in SiO2 (differently from what assumed if a basaltic melt interact with an "average" upper crustal rock).
Similarly, the geochemical signatures typically ascribed to interaction of basaltic melts with crustal lithologies at shallow depths (e.g., high Th/Nb, low Ce/Pb, Nb/U and many others) can be related to source characteristics (i.e. mantle metasomatized by subducted rocks). In other words the same features could be interpreted as primary features of the source or features acquired after the production of the basaltic magma.
In general, a negative correlation between SiO2 and HFSE (Ti, Hf, Nb, Ta, Zr) and a positive correlation between SiO2 and 87Sr/86Sr (or the opposite if you use MgO as abscissa) is assumed as a proof for upper crustal contamination.
Note also that you shoud distinguish upper from lower crust. In general lower crust has a basaltic/gabbroid composition, hardly distinguishable from typical basaltic melts (of course).
Cheers,
michele
Dear Professor Lustrino,
Can you simply explain why "a negative correlation between SiO2 and HFSE (Ti, Hf, Nb, Ta, Zr) is assumed as a proof for upper crustal contamination. " I don't really understand this. Any related articles?
Basalts have lower SiO2 and higher HFSE than average upper continental crust. If you contaminate a basaltic melt with solid upper crustal rocks the overal effect will be to have an increase of SiO2 associated with decrease of HFSE content. This is defined a negative correlation.
Of course this is only a general scheme and everything depends on the starting melt composition and on the composition of the assimilating country rock. Moreover is also true that digestion of crustal rocks enhances cooling in the basaltic melt. This speeds up the fractional crystallization, resulting in what is called AFC (Assimilation + Fractional Crystallization).
During fractional crystallization, the concentration of incompatible elements (such as many, not all, the HFSE) tends to increase in the residual melt.
In other words, the simple assimilation of crust would lead to a decrease of HFSE, while the simple fractional crystallization would lead to a decrease of HFSE. In general, however, the decrease of HFSE is more robust than their increase in the hybrid melt.
Cheers,
michele
Dear Professor Lustrino, I have understood your meaning now. Thank you for your answers listed above. Your words also set me thinking deeply and let me know that geoscience is a rigorous subject.
Thank you again.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Inorganic Materials
- Metals
- Alloys
- Steel