Investigation of multiscale biodegradation processes, a modelling approach

Abstract

This thesis attempts to link processes involved in biodegradation of BTEX (benzene, toluene, ethylbenzene and the xylenes) compounds in groundwater environments at three different observation scales, the micro-, meso- and macroscale, by means of numerical modelling. Different processes, phenomena and characteristics are predominant at each scale. The link between the micro- and mesoscale is performed by using the zero-dimensional model BIOBATCH, whereas the three-dimensional transport model BIO3D is used to link the meso- and macroscale. The assumption is made that small-scale phenomena fully apply at the larger scale where additional processes play a role as well.

A new method was developed to calculate Monod degradation parameters for BTEX compounds using laboratory batch experiments. The problem of non-uniqueness of the calculated parameters was overcome by using several different initial substrate concentrations. With a relative-least-squares technique, unique kinetic degradation parameters were obtained. Calculation of the microbial yield, based on microbial counts at the beginning and the end of the experiments, was crucial for reducing the number of unknowns in the system and therefore for the accurate determination of the kinetic degradation parameters. The new method worked for a constant microbial yield (Chapter 1) and for a case of decreasing microbial yield with an increase in initial substrate concentration (Appendix A).

In order to assess all relevant processes contributing to the transport and degradation of contaminants in the field, a procedure was developed to define an optimal sampling grid to perform a reliable field mass balance by applying numerical modelling and geostatistical methods (Chapter 2). The procedure was used to assess the field behaviour of the slowly degrading compound methyl tertiary butyl ether (MTBE) within the Borden aquifer. By exclusion of the other processes such as sorption, volatilization, abiotic degradation and plant uptake, it is suggested that MTBE biodegradation played a major role in the attenuation of MTBE at Borden.

Furthermore, the phenomenon of changing flow directions on the behaviour of conservative and biodegradable compounds was investigated (Chapter 3). The transient nature of the flow field contributed to the transverse spreading of the plume and therefore enhanced the mixing between the substrate and the electron acceptor which in turn enhances biodegradation. However, the results suggest that in the case of moderate changes of flow direction, a steady-state flow field is justified for many practical applications, thereby avoiding the higher computational costs of a fully transient simulation. Under these conditions, the use of a higher transverse horizontal dispersivity in a steady flow field can adequately forecast plume development.

Finally, linkage of the laboratory and field scale was attempted using a numerical modelling approach with small elements (Chapter 4). Laboratory-derived kinetic degradation and sorption parameters were applied, along with additional physical, chemical and microbiological information, to a dissolved gasoline field experiment at the Borden aquifer (Chapter 4). All additional input parameters were derived from laboratory and field measurements or taken from the literature. The simulated results match the experimental results reasonably well without model calibration.

Based on these results, an extensive sensitivity analysis was performed to estimate the influence of the key controlling factors at the field scale. It is shown that the flow field, in particular, significantly influences the simulated results. Under the field conditions modelled and the assumptions made for the simulation, it could be concluded that laboratory-derived Monod kinetic parameters can adequately describe field-scale degradation processes. By choosing small elements for the simulations, lab-scale processes could be resolved at the field scale, yielding field-scale degradation behaviour without the need for scale relationships to link the laboratory and the field scale. Accurately incorporating the additional processes, phenomena and characteristics such as a) advective and dispersive transport of the contaminants, b) advective and dispersive transport and availability of the electron acceptor, c) mass transfer limitations and d) spatial heterogeneities, at the larger scale and applying well defined lab-scale parameters should accurately describe field-scale processes.