MASSACR

Modeling Altered Seafloor: Simulation and Climatic Response

BACKGROUND

MASSACR improves upon its predecessor, SWAGCUNT, with more adaptable PDE solvers, cleaner output, and improved functionality for all of your porous-media-flow-coupled-to-aqueous-geochemistry needs. MASSACR incorporates new insights (anderson et al. 2012, 2013) into off-axis hydrothermal systems that have not been previously considered in a geochemical context. not that i know of anyway.

FLUID DYNAMICS

massacre.f90: main method, PDE solving subprograms, message passing central hub (distribution to slave processors and collection by root processor), output writing. reactive transport happens here (done by root processor) instead of in its own module.

globals.f90: includes stuff that the model needs to be available globally, including meshgrid points and timesteps in various fine+coarse combinations, physical properties of fluid/rock, and solvers for sets of linear equations (e.g. banded matrix solver, tridiagonal solver)

initialize.f90: contains just one subroutine, init(), called at the beginning of the main method that sets up the material properties set in globals.f90 and boundary/ initial conditions. fluid/rock properties that require extensive calculation are born here instead of the globals module.

GEOCHEMISTRY

alteration.f90: this is a module that contains all of the necessary functionality for basalt alteration that can be coupled to the fluid dynamic model by running alter() for one or all or some grid cell(s). it is designed to accept relevant information such as cell temperature, timestep size, amounts of primary consituents, and amounts of secondary alteration products which are then used as inputs to PHREEQC. the output is PHREEQC SELECTED-OUTPUT and is not parsed in alter().

POST-PROCESSING

plotPressure.py: plot streamfunctions, isotherms, and several benchmark values that correspond to output in hydrothermal circulation simulations by snelgrove and forster, 1996.

plotAnimation.py: for making animations for talks and navahnavahnavah.com

ONE-BOX MODELS

i made a bunch of models for screwing around purposes. each does something slightly different and the directory contains individual post-processing and visualization scripts.

phreeqout/: original basalt alteration experiments (based on carbon sequestration simulations by pham et al., 2012) using aqueous geochemistry package PHREEQC interfaced to FORTRAN using IPHREEQC.

batch/: single box model of basalt alterating and carbonate precipitation including mixing of seawater-derived hydrothermal fluid with fresh deep ocean seawater. batch.f90 and batchControl.f90 handle the mixing differently. batchControl.f90 is probably better.

cell/: pretty much the same as batchControl but with better output/post-processing and ultimately for the development of MASSACR's geochemical model.