Large-scale low-δ18O magmas have been interpreted to have formed by remelting or assimilation of high-temperature hydrothermally altered rocks, such as those found in Iceland basalts and Yellowstone rhyolites. Formation of low-δ18O magma, in most cases, seems to related to rift or extensional tectonic environment. I just wonder whether low-δ18O may be formed at subduction zone and how it happens. Discussion or any relevant reference is appreciated.
Hi Ya-Nan, I believe that much of the high-temperature hydrothermal alteration of the ocean crust happens near Mid-ocean ridges where there is still some heat and pores have not been completely occluded by secondary minerals.
The low d18O values (as low as zero) are simply due to the high temperature of equilibration, and so it may be the case that these conditions are met at subduction zones, however, I believe the water:rock ratio would likely be lower at subduction zones, and therefore low d18O values would probably be inherited from prior alteration at a spreading center (If there is no evidence for possible alteration by meteoric water in your samples), but I may be wrong about this.
A good reference is Muehlenbachs and Clayton, 1976 "Oxygen isotope composition of the oceanic crust and its bearing on seawater", and Muehlenbachs 1998. Hope this helps!
Michael, thanks for your timly reply. High temperature of equilibration between sea water and oceanic crust would decrease the δ18O of the latter. But is it possible that the low δ18O signature be transferred to to igneous rocks formed at subduction zone through remelting of the oceanic part with low-δ18O?
The work by Pieter Vroon on the Banda Arc has pointed towards the fact that the d18O of arc magmas shows little deviation from mantle values, unless high-level crustal contamination has taken place. The likely reason for this is that oxygen is a major element in the mantle. So, if you adhere to the idea that arc magmas are formed by melting of the fluxed mantle, then only a fraction of the oxygen will be coming from fluids or melts from the subducted materials, and the bulk from the mantle. Therefore, the oxygen isotopic composition of early-formed magmatic minerals is generally indistinguishable from that of the mantle. Of course, you could also produce magmas be remelting of continental crust in continental arc settings, and they could show a deviating oxygen isotopic signature - but that doesn't seem to be the thing you are after.
I have the paper that I was thinking of as a PDF, so I attach it here, just in case...
Hi Ya-Nan,
It is also known that the world-wide serpentinites, eclogites and slab-derived melts may be strongly depleted in 18O and have lower than typical mantle δ18O values (down to ~3‰) (e.g., Bindeman, 2008, attached). That fact may be useful for you as well. I believe that in situ isotopic techniques (like SIMS) regularly reveal stronger inter-crystal isotopic heterogeneity compared to that obtained based on bulk-rock- and grain-scale analyses. So, if you use such in situ isotopic data, they can tell you more complete story about your magmas.
Subduction settings do not inherently produce low-d18O magmas similar to those found at Yellowstone and Iceland, although there are examples of volcanoes that are formed through subduction and produce low-d18O magmas. Bindeman and others have found some of these in Kamchatka, although the magma d18O values are not as low as are seen at Yellowstone. The low-d18O magmas of Kamchatka are typically associated with larger caldera-forming eruptions that produce ignimbrites, which are associated with similar processes (e.g. rifting or caldera formation) that are found at Yellowstone and Iceland. Therefore, it is not the process of subduction that produces the low-d18O magmas, but instead the later incorporation of low-d18O country rock or previously erupted and hydrothermally altered rock that produce the low-d18O magmas. I attached an article by Bindeman et al. (2010) that includes a lot of d18O data on Kamchatkan volcanoes if you are interested.
Anastassia and Angela, thanks for your detailed explanation. I will go through the references you provide. I have done in situ oxygen isotope analyses on zircon grains seperated from some pyroclastics. Abundant (30%) low d18O values less than 4 per mill are revealed. Since almost all low d18O magmas are found on non-subduction related environment, I supposed I could use the low d18O values of magmatic zircons to indicate their formation of tectonic settings. Thanks for all your advice.
I would not go so far as to try and deduce the tectonic setting from the fact that you find a low d18O signature: as explained by Anastassia and Angela, it's the process that is important. The process could potentially also operate in a subduction setting. It is of course more likely to operate in larger volcanic centres, which are more commonly found in rift-type settings. But you may have other (field, structural) evidence to argue for a non-subduction setting. Have you tried doing trace element analyses on your zircons? I found that the Th/Nb ratios of zircons matched those of the whole rocks, and this may give you an indication whether you are dealing with a subduction setting too (see publication).
Article Early Neoproterozoic metagabbro-tonalite-trondhjemite of Sør...
Marlina, you are right. I am considering collecting other evidence such as sedementary facies analysis and structural records to argure for the subduction or non-subduction setting. Thank you for the advice. I will certainly go through your paper.
Dear Ya-Nan Yang,
i'm attaching an example of SSZ magmas with low O signature. As you supposed, these magmas are no so common. But also, and as stated in the above answers, is not correct to suposse a tectonic setting on the basis of such information. Maybe a whole-rock geochemistry of the piroclastic rocks will be much more usefull on unraveling the tectonic setting of magma generation.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs