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

Magma transport in the shallow crust – the dykes of the Chachahuén volcanic complex (Argentina)

Tobias Schmiedel1,2, Steffi Burchardt1,2, Frank Guldstrand3, Tobias Mattsson1, Olivier Galland3, Octavio Palma4, Emma Rhodes1,2, Taylor Witcher1,2, and Bjarne Almqvist5
Tobias Schmiedel et al.
  • 1Uppsala University, Mineralogy, Petrology, Tectonics (MPT), Department of Earth Sciences, Uppsala, Sweden (;
  • 2Centre for Natural Hazards and Disaster Science, Uppsala, Sweden,
  • 3PGP-NJORD Centre, Department of Geosciences, University of Oslo, Oslo, Norway
  • 4Y-TEC – CONICET Av. Del Petroleo, Argentina
  • 5Geophysics, Department of Earth Sciences, Uppsala University, Uppsala, Sweden

Recent eruptions such as the Kilauea 2018 (fissure) eruption on Hawaii are the result of magma intruding into Earth’s crust and ascending towards the surface. Magma is dominantly transported, through the shallow crust in form of vertical sheet intrusions (dykes). Even though dyke propagation and emplacement has been monitored with geodetic and geophysical methods, direct observations of subsurface intrusion processes remain inaccessible due to the hazardous nature of active volcanic and igneous systems. Therefore, we studied the extinct and eroded volcanic system of the Chachahuén volcanic complex (CVC) in Argentina to investigate the scale and physical mechanisms of magma transport in volcanic and igneous plumbing systems.

The Chachahuén volcanic complex is located in the northern part of the Neuquén Basin, east of the southern volcanic zone (SVZ) of the Andes. A decline in volcanic activity during the Quaternary and erosion have exposed the shallow part of the Miocene CVC’s plumbing system, including two major vertical sheet intrusions: (1) the Great Dyke and (2) the Sosa Dyke.

The objective of this ongoing study is to characterize the mechanisms of magma transport within the two exposed dykes to better understand the physical processes during their emplacement. We apply a multiscale approach combining field work and state-of-the-art analytical techniques, i.e., drone/ground-based photogrammetry, Fourier Transform Infrared Spectroscopy (FTIR), Electron Backscatter Diffraction (EBSD) and Anisotropy of Magnetic Susceptibility (AMS), with traditional geological methods, i.e., microstructural analysis and igneous petrology. Thus, we can investigate the effect of magma rheology (small-scale) on the outer shape and morphology of the dykes (large-scale).

Our results using high-resolution 3D outcrop models show a segmentation of the investigated dykes. Each of these dyke segments shows blunt ends. This suggests either the emplacement of a highly viscous magma or a weak brittle host rock. Flow features identified with AMS analysis indicate a dominantly lateral magma transport within the dykes. To estimate the magma viscosity during emplacement FTIR (H2O content of the initial melt), and microstructural analysis (for crystallinity) are performed at the moment. These analyses in combination with a map of the host rock and, the dyke morphologies, will help to characterize the dominantly controlling mechanism(s) of magma emplacements in the CVC. Finally, the new findings from this project will contribute to the general understanding on how the physical properties of the magma affect the shape of magma bodies and magma flow in the Earth’s shallow crust.

How to cite: Schmiedel, T., Burchardt, S., Guldstrand, F., Mattsson, T., Galland, O., Palma, O., Rhodes, E., Witcher, T., and Almqvist, B.: Magma transport in the shallow crust – the dykes of the Chachahuén volcanic complex (Argentina), EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-18651,, 2020

Comments on the presentation

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

Presentation version 1 – uploaded on 30 Apr 2020
  • CC1: Comment on EGU2020-18651, Craig Magee, 01 May 2020

    Hi Tobias,

    Looks like some great and promising work here. Can you clarify three things:


    1) What is the dyke dip at the two AMS transects?

    2) There's some variation in it systematic across and/or along the dykes?

    3) Have you had a chance to test what the magnetic carriers are yet?


    Shame we couldn't discuss this in person!




    • AC1: Reply to CC1, Tobias Schmiedel, 04 May 2020

      Hi Craig,

      Yes, it would be awesome to get your input o this! Below the answers to your questions:

      1) The dyke dip at the transects (or anywhere else along the dyke) is basically vertical (+/- 5 degrees)

      2) The variation (purple and red dots in the TAS) was measured by previous studies. They basically highlight the evolution of the Chachahuen volcanic complex over several episodes of activity. The sampled intrusions (yellow, blue dots in the TAS) are basically a snapshot and hence plot together. However, that said the yellow dot with the Andesite signal is from a margin sample of the Sosa dyke. And the two blue dot represent each margin and centre of the Great dyke. Thus, to answer your question, yes there seems to be a slight change in composition accross the dykes towards a more felsic centre. I can sadly not say anything about the along the dyke variation, since we only did geochemistry on one traverse. But that might actuallt be an interesting point, thanks for the hint!

      3) Regarding the magnetic carriers, I would love some discussions. I did the AMS measurements and also did the Magnetothermometry (I believe that is what it is called, where you heat and cool the sample and measure its susceptibility). I probably have a mix of multiple maghemite, magnetite a bit of hematite, which could be related to the samples undergoing weathering, I suppose. However, there is also a signal at even higher temperatures than 600 C, which I have not found a solution for yet. I would welcome any input on how to interpret the Thermomagnetometry curves (Curie-temp analysis). What I can say, however, is that there seems to be an overall symmetry in the signal across the dyke (margin-centre-margin).

      I hope that could answer your questions, thanks for the idea, and if you have any input on resources to interpret the thermomagnetometry curves, please let me know.

      Best regards,


      • CC2: Reply to AC1, Craig Magee, 04 May 2020

        Thanks Tobias!

        If you can send me the thermo-mag plots I can have a look...not that I'm an expert in this! Will McCarthy at St Andrews would also be a good person to ask on this. Something above 600 sounds very suspicious... For the maghemite and haematite, they could be due to oxidation during the experiment if it was conducted in an argon atmosphere.

        Great work though! Hopefully I'll catch it tomorrow but I'm convening at the same time.

  • CC3: Comment on EGU2020-18651, Stefano Urbani, 05 May 2020

    Hi Tobias,

    very interesting topic and results! I want to ask some details on the AMS results of the Sosa dyke.

    - Do you see any variation of the AMS fabric across the dyke from the margin to the center (e.g. a switch from normal to inverse)? At a first look on traverse 2 it seems that k1 becomes vertical from one margin to the other.

    - The clustering of the AMS axes is astonishing so it seems that no overprinting of the primary fabric related to magma flow has occurred. This is pretty surprising to me since the dyke is relatively thick thus I would have expected large AMS variations due to cooling effects. Any constrain on the emplacement depth of the dyke? A shallow dyke that cooled fast may explain the almost perfect clustering of the axes (especially at the dyke centre) despite the huge thickness.

    - Is there any specific criteria in selecting the position of the two traverses? Do they correspond to different emplacenment depths?

    Thanks a lot for your answers.



    • AC2: Reply to CC3, Tobias Schmiedel, 06 May 2020

      Hi Stefano,

      Thanks for your interest in our work and your questions. I also have a question to you at the end.

      1) At the moment I can not give you a definite answer regarding the normal or inverse charater of the AMS, since I am still on the analysis of the data. However, yes the verticallity is most-likely due to the traverse 2 being very closely located to a dyke segment tip. Thus, no lateral movement of the magma possible and so a vertical component is recorden in the AMS, since incomming magma will be diverted up/downwards. That , is at least the explanation right now.

      2) Yes, it is! I can support your assumtion. The Sosa dyke seems to have a very shallow emplacement depth. The exposed surface level (today) was probably only a couple of 100 m below the surface during the time of emplacement. The thickness of the dyke is most-likely a result of its relatively high-viscosity. I am still on the way to confirm that, but I would expect the emplacement temperatures well below 850 C. 

      3) The traverses were choosen according to the lateral location in the dyke. Their level should more or less correspont to the same level of emplacement depth. However, traverse 1 was chosen as representative for the middle part of a dyke segment. Whereas, traverse 2 was chosen because of its representation of a dyke segment end/tip.

      I hope I could answer your questions.


      On a different note, I know you worked with dyke emplacement models. Did these models ever represent thick, high viscous dyke emplacement or mainly low-viscous magmas?

      Best regards,


      • CC4: Reply to AC2, Stefano Urbani, 06 May 2020

        Hi Tobias,

        Thanks a lot for your answers I will certainly follow the progress of this work.

        The models I published so far are better scaled for relatively thin (3-4 meters thick), low viscosity dykes like those found in the extinct rifts of Iceland (see. Urbani et al., 2015 and Gudmundsson, 1983) or like the Bardarbunga 2014-2015 dyke. I am currently working on models involving high-viscous fluids analogues for magma but I am still at an early stage. Meanwhile, you may find interesting the work of  Jazmín Chávez and Mariano Cerca (Abstract code: D1353 EGU2020-13768). 

        By the way, with some collegues we are close to submit a paper on AMS data of the Eastern Iceland dykes which represent a sort of "opposite" end member of the dykes you are working on i.e. deep (about 1.5 km), thin, low-viscosity dykes versus shallow, thick, high-viscosity ones. Therefore, if we manage to publish, it will be very inetersting to compare them with the results of your work. I can anticipate that the AMS axes are much more scattered (sometimes completely messy :) ) with frequent change from normal to inverse fabric. Good luck for your next plans!!



        • AC3: Reply to CC4, Tobias Schmiedel, 06 May 2020

          Hi Stefano,

          yes that sounds great, I will make sure to follow the Iceland dyke work to compare it to our endmember case. Success with the high-viscous dykes! I am looking forward to see some analogue thick-dykes in future. :)