Introducing GibbsStudio
========================

About
-----------------

GibbsStudio is a geochemical modelling framework based on the geochemical simulator PHREEQC :cite:`parkhurst2013description` that increases the user's **productivity** and **extends the functionality** of the original program:

* User's productivity is boosted by drastically reducing the postprocessing work, visualizing the simulation results directly in reusable plots and tables.
* The PHREEQC input is extended with the option to add parameters. These parameters then can be used to run PHREEQC in Monte Carlo simulations, parameter fitting or other parametric simulations.


Technical Details
------------------

GibbsStudio is written in C++ and interacts directly with an embedded PHREEQC. Instead of using the PhreeqcLib Module :cite:`charlton2011modules`, the library to deal with PHREEQC from external programs, provided by the USGS, GibbsStudio uses its own custom PHREEQC wrapper module. This tight coupling between these codes enables extended monitoring during the simulation and access to internal PHREEQC variables.

PHREEQC is bundled internally with GibbsStudio, so no separate installation is needed. New versions of PHREEQC are incorporated into GibbsStudio with regular software updates. The PHREEQC version in your copy is shown in *Help → About…*.

GibbsStudio optimizes simulations on multi-core processors by running multiple PHREEQC simulations concurrently on different execution threads.


What is in it
-------------

GibbsStudio is built from modules, and *Help → Plugins…* lists the ones your
copy has. Some need a professional licence; a module without one stays in the
list, inactive.

**Phreeqc**
   Geochemical simulation: the databases, the chemical components and the
   studies that solve them. The core of the application.

**Akva**
   Water chemistry: import water analyses, check their ion balance, classify
   each water by type and Piper facies, compute irrigation indices (SAR,
   Na %, RSC, USSL class), and draw Piper, Schoeller and USSL (Wilcox)
   diagrams.

**Statistics**
   Monte Carlo studies, univariate statistics and principal component
   analysis over any results table.

**Optimization**
   Fits model parameters to data — the inverse of running a simulation
   forward.

**Predominance**
   Predominance and stability diagrams, including studies that track the
   boundaries between fields.

**Database**
   Comparing databases, and seeing what differs between them.

Each is described in its own chapter.