Abstract

A fully quantitative cellular automaton (CA) biofilm model was developed. The model describes substrate and biomass as discrete particles? existing and interacting in a specified physical domain. Substrate particles move by random walks, simulating molecular diffusion. Microbial particles grow attached to a surface or to other microbial particles, consume substrate particles, and duplicate if a sufficient amount of substrate is consumed. The dynamics of the system are simulated using stochastic processes that represent the occurrence of specific events, such as substrate diffusion, substrate utilization, biofilm growth, and biofilm decay and detachment. The ability of the CA model to predict substrate gradients and fluxes was evaluated by comparing model simulations to predictions from a traditional differential equations model. One and 2D CA models were evaluated. In general, CA model predictions of steady-state flux, biofilm thickness, and substrate gradients inside the biofilm fitted well the differential equations model results; the 2D model had a better agreement at high substrate concentrations. Fully quantitative CA biofilm models offer an alternative approach to simulate biofilm activity and development. Specific advantages of CA modeling include the ability to simulate growth of heterogeneous biofilms with irregular boundary conditions, and the possibility of developing computationally efficient parallel processing algorithms for the quantitative simulation of biofilms in two and three dimensions.