Chemistry and properties of natural organic matter: An agent-based model

by Dr. Stephen E Cabaniss
University of New Mexico

ABSTRACT
The production, transformation and mineralization of natural organic matter (NOM) has been simulated using an agent-based model. Beginning with natural product precursor compounds and allowing them to react under 'environmental' conditions via a stochastic kinetic model, the simulated NOM has bulk composition (elemental and functional group) and other properties (molecular weight, aromaticity) similar to that observed in freshwaters and soils. To be useful in a geochemical or ecological context, the model must also be able to predict environmentally important properties like acidity, metal ion binding, and microbial utilization rates. Acidity and Cu(II) binding can now be estimated using a series of empirical structure-activity relationships developed using simple molecules without any ‘calibration’ to environmental data. Results to date show that proton and Cu(II) binding by simulated NOM is similar to that observed for isolated NOM samples and predicted rates of microbial NOM utilization are quantitatively similar to observed field rates. Potential application of this model to reactive transport systems will be discussed.