Department Surface Waters - Research and Management


Modeling Early DIAgenesis in sediments using MATLAB

Model Description

MEDIALAB is a multi-component early diagenesis model capable of simulating full set of kinetic-equilibrium biogeochemical reactions coupled with one-dimensional transport processes for aquatic sediments. It calculates spatial-temporal concentrations and fluxes of chemical species as well as rates of all the possible biogeochemical pathways. Sedimentation, molecular diffusion, bioturbation and bioirrigation are the processes included in transport component of associated partial differential equations (PDE). Biogeochemical transformations consist of organic matter degradation through spectrum of electron acceptors, secondary reactions, sorption reactions, acid-base reactions as well as dissolution and precipitation reactions.

Model Features

The major feature of MEDIALAB is enhancement of MATLAB’s symbolic programming capability which leads to automatic generation of early diagenesis equations. MEDIALAB is executed through the MATLAB home screen. The execution of MEDIALAB requires an active installation of MATLAB (Version 7.6 release R2008a or later).  Stoichiometric matrix, reaction, transport and boundary condition functions are automatically computed and passed to PDE solver. By avoiding the hurdles of traditional programming languages such as rewriting and compilation of source codes, the user can focus on the definition, performance and application of the model.
Another advantage of MEDIALAB is that it gives the user the possibility of including transient boundary conditions in a straightforward way. The transient fluxes or concentrations of species can be imposed through time-dependant mathematical functions or real-world temporal field data. It is particularly useful when simulating the fate of compounds whose inputs are changing due to, for example, anthropogenic activity, or when dealing with systems where the bottom water chemistry varies over time.

Model Calibration

To calibrate the model we have used the measurements of anions and cations from sediment cores collected from Lake Baldegg in Switzerland which were collected in 2012 from the deepest location (66 m). Depth profiles of nineteen species were simulated by including thirty three mixed kinetic-equilibrium biogeochemical processes as well as imposing the fluxes of organic and inorganic matters along with solute concentrations at the SWI as dynamic boundary conditions. Lake Baldegg represents an interesting case where low oxygen concentrations and meromictic condition were established between 1960 and 1980, as a result of high productivity. The lake has been artificially aerated since 1982. Depletion of dissolved oxygen (O2) in the hypolimnia of the lake and phosphorous cycling were investigated before and after initiation of aeration and oxygenation in 1982.

History of the model

Developing sediment early diagenesis model, initially called MATsedLAB, started in 2008 as part of Babak Shafei’s Ph.D. thesis project at Georgia Institute of Technology under supervision of Prof. Philippe Van Cappellen. The model was successfully applied in simulating arsenic diagenesis in sediments of lake Tantare in Canada (Couture et al. 2010, Shafei et al. 2010), modeling an acid mine drainage contamination of Sancho reservoir in Spain (Torres et al. 2014) and non-steady state modeling of a major flood event in the Rhone prodelta (Pastor et al. 2011). Despite the profound success of the model in tackling various early diagenesis problems, MATsedLAB was not benefited from MATLAB’s symbolic programming capabilities and automatic PDE generator. In fact, PDE components i.e. transport, reaction and boundary condition vector functions had to be manually computed by the user and passed to PDE solver. Thus, adding or removing species and reactions could be a tedious task and required a rigorous modification of the equations. MEDIALAB, however, provides a full flexibility of defining and modifying set of the equations.

Projects based on MATsedLAB

  • Shafei B, Couture R-M, and Van Cappellen P. Arsenic sorption in aquatic sediments: Equilibrium, kinetic and mixed modeling approaches. Geochim. Cosmochim. Acta 2010 (74) A938-A938.

  • Couture R-M, Shafei B, Van Cappellen P, Tessier A, Gobeil C. Non-steady state modeling of arsenic diagenesis in lake sediments.  Environ. Sci. Technol. 2010 (44) 197-203.

  • Pastor L, Deflandre B, Shafei B, Couture R-M, Cathalot C, Metzger E, Van Cappellen P, Viollier E. Non-steady state modeling of early diagenetic processes following a major flood event in the Rhône prodelta (NW Mediterranean Sea, France). Gordon Research 1-Conference. Chemical Oceanography 2011.

  • Torres E, Couture RM, Shafei B, Canovas CR, Van Cappellen P, Ayora C. Sediment Diagenesis Modelling in an AMD Contaminated Reservoir. Macla 2011 (15)