Recent posts

#71

The MELTS family of algorithms, and scientific discussion / Peridotites melting and volati...

Last post by HaranHennig - September 02, 2024, 02:08:55 AM

Hey All, Paula & Paul!
I'm trying to model some melting of peridotites and pyroxenites using pMELTS under the influence of a CO₂ controlled environment (and much less H₂O one, as inferred from melt inclusions volatile analysis).

The system seems to be running quite stable with results that meet the collected data (!), but only when I sum up CO₂ & H₂O to be at least 1000 ppm (input as 0.1 wt.% major oxide to influence melting conditions, for example CO₂= 0.08 ; H₂O= 0.02 or any other combination), which is more than my current estimations for the mantle volatile values (CO₂~400 ppm, H₂O~20 ppm).
Also, although I am more interested in the trace elements results, the liquid composition of the major oxides seems to be going really crazy. This is especially true if CO₂ is modeled to be the only volatile species present, which makes me doubt the validity of the results.

I would really appreciate if you could share some of your knowledge on how to approach these kind of calculations, and to maybe provide some key notes for validation of the outputs.

--At the moment I'm running the UBUNTU alphaMELTS 1.9 version.
Melting at 30-40 kb, continuous melting with 0.02 minF, isobaric for the moment. No exsolution of water or multiple liquids.--

Cheers,
Haran.

#72

Interactive operation / Re: Don't Open the melts-rhyol...

Last post by Paula - August 08, 2024, 05:51:20 PM

All I can suggest is that you either email Mark Ghiorso (ghiorso@ofm-research.org) and see if he knows of something else to try, or have a go with easyMelts. There is more info in the README and the easyMelts version of the 'tutorial':

https://magmasource.caltech.edu/gitlist/easyMelts.git/tree/master/
https://magmasource.caltech.edu/forum/index.php/topic,992.0.html

Paula

#73

Interactive operation / Re: Don't Open the melts-rhyol...

Last post by Pei fuping - August 04, 2024, 03:49:21 PM

Dear Paula,

I have tried by your method, and the following paragraphs still appear.

(---> Initializing data structures using selected calculation mode...
---> Building interface...),

entered the command "export DYLD_LIBRARY_PATH=/opt/X11/lib/flat_namespace:$DYLD_LIBRARY_PATH" in the .bash_profile

and the interface still does not appear.

Best wishes,

Pei

#74

Website issues, and Miscellaneous / Re: Installing Version 2.3.0

Last post by Paula - August 03, 2024, 05:57:29 PM

Version 2.3.1 has the corrected install script, plus updated version numbers and some bug fixes:

https://github.com/magmasource/alphaMELTS/releases/tag/v2.3.1

Paula

#75

Interactive operation / Re: Don't Open the melts-rhyol...

Last post by Paula - August 03, 2024, 03:14:02 PM

The early easyMelts versions were tested on El Capitan. This version should work on your machine:
 
https://magmasource.caltech.edu/gitlist/easyMelts.git/blob/92711c57cca31f2d143f3ffe62093d63c0c61ed4/easyMelts_0.2.4_beta-MacOS.zip

If you want to stick to the old GUI then I would experiment with a different version of Xquartz. Version 2.7.8 was the first to support El Capitan so you can't go older than that but 2.8.5 claims to support macOS 10.9+: https://www.xquartz.org/

I think you need at least 2.7.10 or later for the flat_namespace fix to work: https://www.xquartz.org/releases/XQuartz-2.7.10.html

Paula

#76

Interactive operation / Re: Can't open the melts-rhyol...

Last post by Pei fuping - August 02, 2024, 10:48:37 PM

Dear Paul,

Thanks for your reply. My apple computer is OS X EI Capitan, and Intel Iris Graphics 6100 1536 MB, as well as 2.9GHz Intel Core i5. I have installed Xquartz 2.7.8 and Melts-rhyolite-Mac-Os-10.8+. Since there is no new version of the computer, I still want to try again.

The operation condition is as follows:
rhyolite-MELTS (1.0.1) pMELTS (5.6.1) - (May 21 2015 - 18:06:14)
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
---> Reading environment variables...
<><> There are three releases of rhyolite-MELTS and a version of pMELTS included in this package:
    (*) rhyolite-MELTS v. 1.0.2 (original version, with corrections) - old H2O model, no mixed fluids.
    (*) rhyolite-MELTS v, 1.1.0 (mixed fluid version that perserves the ternary minimum) - old H2O model.
    (*) rhyolite-MELTS v. 1.2.0 (mixed fluid version optimal for mafic and alkalic melts) - new H2O model.
    (*) pMELTS v. 5.6.1 (original version, with corrections) - - old H2O model, no mixed fluids.
<><> Visit the website melts.ofm-research.org for information on which version to use for your application.
---> Default calculation mode is rhyolite-MELTS (v. 1.0.2).  Change this? (y or n): n
---> ************************************************************
---> Calculation mode is rhyolite-MELTS (public release v 1.0.2).
---> ************************************************************
---> Initializing data structures using selected calculation mode...
---> Building interface...

Otherwise, I followed your instructions for a series of operations, and entered the command "export DYLD_LIBRARY_PATH=/opt/X10/lib/flat_namespace:$DYLD_LIBRARY_PATH", but the interface still does not appear.

Furthermore, the following lines occurred.

Last login: Sat Aug  3 13:59:10 on ttys000
appledeMacBook-Pro:~ apple$ source ~/.profile
-bash: /Users/apple/.profile: No such file or directory
appledeMacBook-Pro:~ apple$ source ~/.bash_profile
appledeMacBook-Pro:~ apple$ /rhyolite-MELTS-Mac-OS-10.8+/Melts-rhyolite-public
dyld: Library not loaded: /opt/X11/lib/libX11.6.dylib
  Referenced from: /opt/X11/lib/flat_namespace/libXt.6.dylib
  Reason: Incompatible library version: libXt.6.dylib requires version 11.0.0 or later, but libX11.6.dylib provides version 10.0.0
Trace/BPT trap: 5
appledeMacBook-Pro:~ apple$


Sincerely,

Pei
2024-8-1

#77

Interactive operation / Re: Don't Open the melts-rhyol...

Last post by Paula - July 31, 2024, 12:32:18 PM

My guess is that your Mac has an M2 or M3 chip, and that's why you're having trouble. Only XQuartz 2.8.X supports Apple Silicon natively. It certainly works on M1 chips, but even the latest beta release (https://www.xquartz.org/releases/XQuartz-2.8.6_beta2.html) predates M3 chip Macs and many M2 chips.

The Rhyolite-MELTS GUI is not really supported anymore. I recommend you try easyMelts instead. It has all the features of the GUI, plus working plots. There's more information here: https://magmasource.caltech.edu/alphamelts/links.php

At the moment you can download easyMelts from here: https://magmasource.caltech.edu/gitlist/easyMelts.git/

For now, you'd have to run it through Rosetta 2, but I am planning to make a native Apple Silicon version at some point. When that happens, the downloads will move to here: https://github.com/magmasource/MAGMA

If you want to try to get the Rhyolite-MELTS GUI working then try downloading the 10.08+ version. You will probably need to add the following to your .zshrc file:

Code Select Expand

export DYLD_LIBRARY_PATH=/opt/X11/lib/flat_namespace:$DYLD_LIBRARY_PATH
See here for a more detailed, but slightly out of date, explanation (substitute ".zshrc" for ".profile"):

https://magmasource.caltech.edu/forum/index.php/topic,821.msg1151.html#msg1151

#78

Interactive operation / Don't Open the melts-rhyolite-...

Last post by Pei fuping - July 31, 2024, 07:46:44 AM

When Melts-rhyolite-public is turned on, the system is always in the "Buiding interface..." There is no interface, no stand-alone version of the interface appears, what is going on?

Best wishes!

Pei

#79

General usage / Re: How to parse the output an...

Last post by Freewill - July 29, 2024, 11:14:03 PM

Hi Paula,

Thanks for your replying.

I did run alphaMELTS 1.9 (on Windows 10) by running the Perl script run_alphamelts.command (Perl was installed beforehand) and Phase_main_tbl.txt is one of the output.

I will try different ways of using alphaMELTS as you mentioned.

Best,
Qingwen

#80

General usage / Re: How to parse the output an...

Last post by Paula - July 29, 2024, 04:50:52 PM

Is this alphaMELTS 1.9 output? If so, you can use column_pick.command to select the table columns with the chemical formulas. There is lots of information about this in the documentation. But run_alphamelts.command needs to process the Phase_main_tbl.txt file first. That should happen automatically when you exit the alphaMELTS menu prompt, but what you've posted is not processed.

Assuming you have Perl installed (on Windows) or are on a Mac or Linux, which come with Perl, then the possible reasons for the failure include:

• You are running the alphamelts executable directly instead of through the run_alphamelts.command script
• You need to run install.command (once) to set up the software and links to the executables and scripts
• You are opening the file before exiting the alphamelts menu (this can cause the access errors, especially on Windows, so then run_alphamelts.command cannot process the file)

Depending what you are trying to do, alphaMELTS for MATLAB/Python, easyMelts or standalone alphaMELTS 2 may make processing the output in the way you need simpler:

• easyMelts produces Graphical User Interface (GUI) style output. You may be able to use that output with the Combine_tbl Excel workbook (see the alphaMELTS links page) though it has a tendency to break with updates to Excel and/or when Windows or Mac security settings change
• The GUI-style output also includes phase end member proportions
• The standalone alphaMELTS 2 app has a "Output: both" option that outputs alphaMELTS-style output and GUI-style output
• alphaMELTS for MALTAB/Python makes it fairly easy to get end member proportions from the regular output
• The alphaMELTS 2 version of column-pick.command is more powerful and easier to invoke than the alphaMELTS 1.9 one