Question about Using “RESTART” in PFLOTRAN for Mineral Dissolution

9 views
Skip to first unread message

平田光一

unread,
Jan 23, 2026, 10:51:58 AMJan 23
to pflotra...@googlegroups.com, 茂木寛之, 大塚悠一
Dear members of the PFLOTRAN mailing list,

I have a question regarding the use of the “RESTART” card in PFLOTRAN for simulating mineral dissolution. To clarify my inquiry, I will describe a very simple scenario where only calcite dissolves in water.

In the initial run, I perform a simulation with the following initial conditions for concentrations (almost pure water) and calcite volume fraction, modeling calcite dissolution. (Primary: H+, HCO3-, Ca++. Secondary: OH-, CO3–, CO2(aq), CaCO3(aq), CaHCO3+, CaOH+)
++++++++++++++++++++
CONSTRAINT initial_constraint
  CONCENTRATIONS
    H+     1.d-7      F
    HCO3-  1.d-12     F
    Ca++   1.d-12     F
  /
  MINERALS
    Calcite 1.d-5 1.d0 m^2/m^3
  /
END
++++++++++++++++++++
During the initial run, I use the CHECKPOINT card to generate a file named “pflotran-restart.h5”(HDF5 file) at the end of the calculation, which I intend to use for the second run.

After the initial run is completed, I want to perform a second run. In this second run, I wish to set only the concentrations of dissolved species to the initial state from the first run  (almost pure water), but I want the mineral (calcite) volume fraction to be carried over from the end of the initial run.

Question 1:
Do I need to edit the output HDF5 file (e.g., “pflotran-restart.h5”) generated at the end of the initial run for this purpose?

Question 2:
If editing is necessary, which parts should I modify, and how should I do it? Specifically, should I modify the sections that store concentration data within the HDF5 file to match the initial conditions? Are there any other parts of the file that require modification?


Thank you very much for your assistance.


Koichi Hirata, Ph.D.
Tokyo Electric Power Services CO.,Ltd.
Email: koichi-hirata@tepsco.co.jp
--
------------------------------------------------------------
平田 光一 / Koichi Hirata
東電設計株式会社
新領域研究開発推進室 高度解析ソリューショングループ
Tel : 090-3439-0143
Email: koichi-hirata@tepsco.co.jp
-------------------------------------------------------------

Hammond, Glenn E

unread,
Jan 26, 2026, 1:20:20 PM (11 days ago) Jan 26
to pflotra...@googlegroups.com, 茂木寛之, 大塚悠一
Koichi,

You can overwrite the primary species concentrations in the solution vector in the restart file, but that is complicated. A better option is to transfer the mineral volume fractions from the first simulation to a cell indexed HDF5 formatted dataset and use those as input in a second simulation. The following regression test illustrates how specify initial mineral volume fractions through datasets:


And the capability is documented (including an example cell_indexed_dataset.py script) at:


Regards,

Glenn

From: pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of 平田光一 <koichi...@tepsco.co.jp>
Date: Friday, January 23, 2026 at 7:52 AM
To: pflotra...@googlegroups.com <pflotra...@googlegroups.com>
Cc: 茂木寛之 <mot...@tepsco.co.jp>, 大塚悠一 <yuichi...@tepsco.co.jp>
Subject: [pflotran-users: 8651] Question about Using “RESTART” in PFLOTRAN for Mineral Dissolution

Check twice before you click! This email originated from outside PNNL.
--
You received this message because you are subscribed to the Google Groups "pflotran-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/CAFy52D548aMErt-64yTezgQt1iH-niR-hWiLAntkirpOkpfERQ%40mail.gmail.com.
Reply all
Reply to author
Forward
0 new messages