WATERBUGMODEL · TRAINING FOR MODELLING THE GROUNDWATER ECOSYSTEM

TRAINING FOR MODELLING THE GROUNDWATER ECOSYSTEM

WATERBUGMODEL – training for modelling the groundwater ecosystem Groundwater is the basis of over 50 % of the world-wide drinking water production. Groundwater constitutes the largest reservoir of freshwater in the world, accounting for over 97% of all freshwater available on earth (excluding glaciers and ice caps). Depending on the region, up to 90% of the drinking water is produced from groundwater. At the same time, groundwater is increasingly threatened by pollution. Almost all pollutants can actually be degraded by some microorganisms at least under some conditions, but this can be a very slow process, therefore bioremediation seeks to enhance the rate of degradation, and the long term goal of this project has been to develop mathematical modelling in order to predict the effect of various groundwater management scenarios so that an optimal management solution can be identified before any action is taken. However, there are two key problems that need to be solved before this vision can become a reality. First, sampling of groundwater is particularly challenging and the typically slow changes require monitoring for quite long periods; as a result data are and will remain very scarce. Therefore, the fewer input data a model requires, the better. Here we have used experimental evidence combined with theoretical advances in biogeography to argue that suitable degraders are already present or migrate into the system quickly enough so as not to limit the long-term degradation rate. In a nutshell, the idea is that “Everything gets everywhere – but the environment selects”. The reason for everything getting everywhere is the fact that microbes are very small and numerous and have short generation times, thus they disperse at a very high rate and therefore get everywhere relatively fast. Evidence indeed supports sufficient dispersal rates although a recent finding of only one bacterium, Candidatus Desulforudis audaxviator, in the extreme conditions of fracture water 2.8 km deep down a South African gold mine, raised doubts. However, 16S gene sequences similar to “D. audaxviator” have just been found in Finland. Also the gold mines in SA have meanwhile yielded further bacteria. If everything gets everywhere, organisms able to degrade a newly introduced contamination should be getting to this site fast enough not to limit degradation in the long term. Several groundwater studies have indeed found that bioaugmentation, i.e. introducing pre-grown organisms, did not enhance degradation rates, presumably because organisms capable of degrading the contaminant were already there. Also, we use growth rate data to argue that the composition of the microbial community degrading the pollutant will change more quickly than the pollutant becomes degraded, so that the community composition becomes optimal for degradation within months, while degradation takes decades. This assumption of optimal community composition allows one to seed a simulation of a mathematical model with a large range of simulated microorganisms and let the simulated competition in the model generate the optimal community composition. This means for bioremediation applications that knowing which microorganisms are initially abundant is not necessary as this will change towards the optimal composition of the community. Second, the heterogeneity or patchiness of the system on a small and large scale has a crucial impact on the potential for biodegradation – after all the pollutant must meet the degrader. We have studied the importance of the spatial distribution of organisms degrading pollutants on a small scale in a tubular geometry using Comsol to numerically solve the transport of the pollutant by convection and diffusion and the uptake of the pollutant by biomass, for an example see Fig. 1. As can be seen, locally high biomass density leads to high rates of substrate consumption resulting in low substrate concentrations locally, which reduced degradation rates. This is in contrast to assuming that biomass and substrate are homogenously distributed (the standard approach to groundwater biodegradation modelling), which would overestimate reaction rates which are typically proportional to both biomass and substrate concentrations at low substrate concentrations. Dr Schmidt has received extensive training in groundwater transport modelling and numerical simulation software, biogeography, microbiology and kinetics of biodegradation, individual-based modelling and programming. Much of this training was a result of developing modelling concepts and investigating the impact of patchiness on degradation, which are the subjects of three manuscripts in the process of publication. We have made some advances to the field of groundwater modelling, for example, we have applied biogeography and resource competition theory to propose that the degrader community will become optimal and we have developed a conceptual framework for multiscale modelling that addresses the problem of patchiness on small and large scales. Using predictive mathematical models to evaluate potential bioremediation strategies in order to identify optimal strategies should become the basis for future successful groundwater management and remediation strategies. The socio-economic impact of the project is the development of innovative ideas to improve remediation strategies to reduce groundwater contamination, facilitating the use of groundwater. Our target group are therefore consultant and engineering companies in groundwater remediation.

Текст от CORDIS, на английски · Данни: CORDIS, © Европейски съюз