Setting treatment of water

[jschools]

Hi Paula,

I am working on a scenario involving the melting of Martian mantle and the crystallization of the resulting melt under lithospheric conditions. Up until now I have only performed anhydrous calculations, but now I want to see the effects of water on the conditions of melting and melt composition. I just wanted to double check on how I should set up the calculation, given the variety of ways the MELTS family can handle water. I believe that since I want to see changes in melting behavior I should go ahead and put water in the major elements (Initial Composition: H2O) with MELTS or pMELTS. Is this correct?

Thanks,
Joe

[asimow]

Dear Joe,

It depends a little bit on what you need to do.

Certainly, the simplest thing to do is to add H2O as a major oxide with the Initial Composition: H2O line in the input file (mass of H2O given in grams). This is appropriate if you are imagining more than a few hundred ppm in the source for the melting calculation and for all concentrations for the crystallization calculation. However, you should realize that in this case MELTS or pMELTS will only know about some of the potential hosts for H2O in the system: dissolved in melt, stabilized in nominally hydrous phases (e.g. mica, amphibole, a few others), or in H2O vapor phase if saturated. It will not know about trace water bound in the nominally anhydrous phases. That means you can't really study the damp solidus of a mantle assemblage. There isn't usually enough water or other essential components (e.g., K) to stabilize a hydrous phase, so if ANY water is present it will stabilize melt all the way down to the water-saturated solidus. If this is not an issue for what you are trying to do, then go ahead and do it this way.

If you are trying to locate the damp solidus, or study the effects of very low degree melting below the dry solidus, then you will need to use the pHMELTS option to exchange H2O between the trace element reservoir and the major element reservoir, by setting ALPHAMELTS_TRACE_ELEMENTS and ALPHAMELTS_DO_TRACE_H2O to true and adding Initial Trace: H2O (mass of water given in ppm) to the input file.

-- P

[Paula]

Hi Joe,

Paul wrote essentially what I was going to write... thanks, Paul. I would add two things.

- If ALPHAMELTS_DO_TRACE_H2O is set then you can use "Initial Composition: H2O" and/or "Initial Trace: H2O". If you use both then the total H2O will be be the sum of the two values. (If ALPHAMELTS_DO_TRACE_H2O is not set then these two water budgets will be separate.)

- Despite the pHMELTS name, the ALPHAMELTS_DO_TRACE_H2O option can be used with either MELTS or pMELTS. It sounds like you will be using pMELTS as the underlying liquid model but others may find that MELTS is more appropriate.

Cheers,
Paula