EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Ignimbrite flare-ups in the Central Andes: Crustal sources and processes of magma generation

Gerhard Wörner1, Elena Belousova2, Simon Turner2, Jelte Kemann2, Axel K Schmitt3, Axel Gerdes4, and Shan de Silva5
Gerhard Wörner et al.
  • 1Universität Göttingen, GZG, Abt. Geochemie, Göttingen, Germany
  • 2Earth and Planetary Sciences, Macquarie University, NSW 2109, Australia
  • 3Universität Heidelberg, Institut für Geowissenschaften, Heidelberg, Germany
  • 4Inst. f. Geowissenschaften, Goethe Universität Frankfurt, 60438 Frankfurt, Germany
  • 5College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, Oregon 97331, USA

Silicic magmatism in the Central Andes forms rhyolitic to dacitic volcanic deposits that range from large-volume ignimbrites (>1000 km3) to small local dome eruptions. The mass proportion between mantle-derived magmatic contributions to crustal melting was previously estimated to range from 20 to 70 % based on Sr-O isotope data obtained on separated feldspar and quartz contained as crystal cargo. New O-Hf isotope data from in-situ ion-probe and laser ablation measurements of U-Pb-dated zircons further constrain type, proportion, and processes of crustal input into silicic magmas. Variations in time and space of these geochemical parameters are documented here using representative samples that cover the entire Central Andes over 20 Ma and 800 km distance. Systematic covariations in isotope tracers relate to increasing crustal thickening through time during Andean orogenesis. Collectively, Sr-Nd-Pb-Hf-O isotopic signatures vary in space and time and temporally reflect increasing crustal input during ignimbrite flare-ups as the crust becomes thermally matures. Spatial variations derive from different crustal domains in the Central Andes and reflect the different age and composition of crustal components.

Remarkably, inherited zircon representing basement involved in crustal assimilation is exceedingly rare over the entire province. This most probably reflects high temperatures that exceed zircon saturation temperatures of crustal melts in ignimbrite-forming magmas. This observation distinguishes silicic ignimbrite-forming magmatism from typical granitoid-forming magmatism in orogenic settings where abundant older zircons inherited from the crust are commonly found.

How to cite: Wörner, G., Belousova, E., Turner, S., Kemann, J., Schmitt, A. K., Gerdes, A., and de Silva, S.: Ignimbrite flare-ups in the Central Andes: Crustal sources and processes of magma generation, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-3519,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 03 May 2020
  • CC1: Comment on EGU2020-3519, Agustin Cardona, 04 May 2020

    Fantastic review. You mention on the abstract that the limited inherited zircons in the ignimbrites may be related to high temperatures, and point a difference with granitoids. You mean that there are no granitoids connected to ignimbrites or that the residence time of magmas in the plutonic system may allow for more extensive incorporation of older crustal zircons. 

    • AC1: Reply to CC1, Gerhard Wörner, 04 May 2020


      I am sure that there are large intrusive bodies related to these ignimbrites. These bodies, after thermal decay and crstallization will eventualy form batholiths, just as we see older batholiths in the Cordilleras. These then will very likely have older inherited zircons (as we see in the older exposed ones).

      If the mush rejuvenation model applies, then then the extractedand erupted melts  come from those parts of the intrusive bodies that were heated sufficiently to be reactivated and form eruptible melts. These then do not have zircons because in hotter melts they are dissolved. Those parts of the intrusives that were not heated (probably the lartger parts of the entire batholith) maintained their zircons.

      This is the "model", but it is in conflict with the idea of Bachman and co that ALL intrusive batholiths are the residue of melt extraction (with these melts forming ignimbrites).


      So, the other model would be the "physical" filter: Any melt extraction would leave behind most of the crystals from the remelted mush. However, there is also a conflict because  the very argument for rejuventation process comes from the observation that eruoted magmas are full of "antecrysts" = older crystals entrained from the mush. If so, why are there no old zircons entrained ??

      Same applies for the isotopic heterogeneity of the zircons, which must have crystallized shortly before eruption from isotopically wildly heterogeneous melts. Why should magmas that are remobilized from mushes be isotopically so different ??

      All I can say is that our data are profoundly challenging the paradigm of melt-from-mush extraction.