[aspect-devel] subduction and magmatism

[Juliane Dannberg]

Dear Nikola,


it is certainly possible to model geodynamic processes in the 
lithosphere with ASPECT, and a lot of the current development goes in 
this direction.
I don't have much experience with subduction models, but Anne Glerum's 
models would certainly be a good starting point.


For melting in subduction zones, the question would be how exactly you 
want to deal with the magmatism. Do you just want to know how much melt 
is generated, but without any active influence on the model evolution? 
Or do you want to model two-phase flow, where the melt flows with a 
different velocity than the solid? And, if it is the latter, how do you 
want to deal with fluids/dehydration of the slab (do you want to model 
water etc. as a separate phase, or composition of the different phases)?


I think all of those models would require modifying the code to a 
certain degree. While ASPECT includes a number of simple melting 
parameterizations, at the moment, there is no model for hydrous melting. 
So depending on how you want to treat the melt, you would have to extend 
one of the existing melting parameterizations (for example, ASPECT has 
the anhydrous melting parametrization from Katz et al., 2003, so it 
should be straightforward to modify it so that it also includes the 
hydrous melting described in Katz et al., 2003), or add a new one. You 
could then simply track the water content as a compositional field in 
the model, and compute the amount of melt that is generated for each 
given temperature and pressure. So if this is all you want to look at, I 
feel this should be easy to do.


If you also want to track the migration of melt, I think this would be 
more difficult, as this would probably require tracking the composition 
(and water content) of the melt, and you would also need a hydrous 
melting model that gives you some sort of partitioning for the water 
content and the rest of the composition (so, for a certain pressure, 
temperature and bulk composition, what is the equilibrium melt fraction, 
what is the composition of the solid, what is the composition of the 
melt). Tracking the melt composition is something that we plan to 
include in ASPECT in the near future, but it is not in the current 
development version. However, we're working on that, and if you're 
interested in something like this and want to help implementing it, we 
can certainly think about what exactly you need and how to add this to 
ASPECT together. Finally, if you plan to include brittle deformation in 
your model, and also want to model two-phase flow at the same time, this 
would be a very difficult problem (in general, not just in ASPECT). We 
are working on material descriptions in ASPECT that allow this kind of 
model, but it needs a lot of testing and there are many numerical 
difficulties involved.


So I guess what I am saying is, you can certainly use ASPECT to model 
subduction, magmatism and deformation of the lithosphere. You should 
just be aware that depending on how complex you want your models to be, 
it would require modifying the code. And we are of course happy to help 
with any questions that might come up.


I hope that helps!


Best,
Juliane

----------------------------------------------------------------------
Juliane Dannberg
Project Scientist, UC Davis
jdannberg.github.io <https://jdannberg.github.io/>




On 04/16/2018 07:33 AM, Никола Станковић wrote:
>
> Hello everyone,
>
>
> I would like to ask if it is possible to model geodynamic processes in 
> the lithosphere with ASPECT. I am trying to model subduction, and the 
> associated magmatism. I am aware that this code is primarily developed 
> for simulating convection. However, I have found and successfully used 
> some plugins developed for this code, namely subduction plugin by Anne 
> Glerum, as well as viscoplastic plugin.
>
> I am also aware that ASPECT incorporates partial melting, however I 
> haven't seen published papers dealing with modeling magmatism on a 
> detailed scale. In Dannberg and Heister (2016) the melting is 
> associated with a plume. In Glerum et al (2017) subduction is modeled 
> mechanically, there is no associated magmatism.
>
> What I am asking is, whether it is a good idea to use ASPECT to model 
> subduction process, in greater detail as well as detailed magmatism 
> and deformation of the lithosphere. Also if anyone has succeeded in 
> this, any information about it would be very beneficial to me.
>
>
> Best regards,
>
>
> Nikola Stanković
>
>
>
> _______________________________________________
> Aspect-devel mailing list
> Aspect-devel at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/aspect-devel


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.geodynamics.org/pipermail/aspect-devel/attachments/20180416/d04ca048/attachment-0001.html>