.. _example-phreeqc-manual-17:

17 - Inverse Modelling with Evaporation
=======================================

Black Sea water evaporated until halite precipitates, and whether a short
list of processes can account for every major ion in the result.

The hypothesis under test is specific: that **evaporation**, precipitation of
**calcite**, **gypsum** and **halite**, and **loss of CO₂** are enough to
explain the difference between the initial water and the evaporated one --
for all the major ions and for bromide.

Bromide is there as the control. It forms no mineral under these conditions,
so it can only be concentrated by evaporation, and it therefore says how much
water was lost independently of everything else. Any model that gets bromide
wrong is wrong, whatever it does for the rest.


Evaporation as a mole transfer
------------------------------

.. raw:: html
   :file: study_17 - Inverse Modelling with Evaporation.html

Inverse modelling balances mole transfers of phases, so evaporation is
included as one: water is treated as a phase that leaves the system, and the
amount removed is solved for alongside the minerals.

The second input repeats the calculation with different constraints.

.. raw:: html
   :file: study_17 - Inverse Modelling with Evaporationb.html


The result
----------

.. figure:: plot_1.svg
   :alt: Major ion molalities and mineral saturation against concentration factor for evaporating Black Sea water
   :align: center

   The major ions and the minerals against concentration factor, on a
   logarithmic scale.

   The ions that form no mineral -- sodium at first, magnesium, chloride --
   rise along parallel straight lines, each in step with the concentration
   factor. That parallel rise *is* evaporation, and a species that follows it
   is being concentrated and nothing else.

   The departures are the chemistry. **Calcite** saturates first and removes
   carbonate. **Gypsum** follows and takes calcium and sulfate out, so their
   curves bend away from the evaporation line. **Halite** is last, at high
   concentration, and when it arrives sodium and chloride stop rising
   together with the rest.

   **Glauberite** and **polyhalite** also appear, which is the part worth
   dwelling on: the hypothesis named three minerals, and the saturation
   calculation finds others reaching equilibrium along the way. Whether the
   hypothesis survives depends on whether they actually precipitated, and
   that is a question for the evaporite mineralogy, not for the model.


What it shows
-------------

That inverse modelling tests a stated hypothesis against data, and that the
test is most informative where it fails. A conservative tracer like bromide
is what anchors it.


Source
------

* Parkhurst, D. L. and Appelo, C. A. J. (2013). *Description of input and
  examples for PHREEQC version 3.* U.S. Geological Survey Techniques and
  Methods, book 6, chapter A43. This is Example 17 of that manual.
* Carpenter, A. B. (1978). *Origin and chemical evolution of brines in
  sedimentary basins.* Oklahoma Geological Survey Circular 79, 60-77. The
  Black Sea evaporation data is from this paper.