HR: 0800h
AN: H51F-0437 [Abstracts]
TI: Subscale Mixing Parameters for Solute Transport
AU: * Attinger, S
EM: sabine.attinger@ufz.de
AF: UFZ Centre for Environmental Research, Permoserstrasse 15, Leipzig, 04318
Germany
AB:
Mixing in natural formations is dominated by the heterogeneous structure of the material. To capture observed phenomena
accurately, one has to discretize numerical transport models by grid cells of smaller size than the scale of heterogeneity.
The computational effort to solve field scale problems is very demanding and it is common to diminish the computational
resolution by choosing coarser grids. Variability is lost in this case and one faces e.g. the problem how to model the impact
of unresolved velocity fluctuations upon transport not only for asymptotic scales but for finite or interacting scales as
well. Standard upscaling procedures such as homogenization and stochastic modeling compensate unresolved effects by
introduction of macrodispersive fluxes. Both methods average out all heterogeneities resulting in a total loss of spatial
variability in the flow and transport parameters. Making use of the spatial filtering method Coarse Graining large scale
fluctuations of the transport velocities are still resolved, whereas only subscale fluctuations are smoothed by the spatial
filtering procedure and efficiently modelled by a scale dependent dispersive flux. To this end, Coarse Graining is capable to
transfer solute transport equations to arbitrary grid resolution and thus to intermediate or finite scales. Employing
perturbation theory approximations, we explicitly calculate scale dependent real and ensemble dispersion coefficients and to
quantify subscale mixing effects. These finite scale results differ from those obtained in the standard or asymptotic
approach. It demonstrates that standard or asymptotic upscaling does not consistently account for finite scale effects: it
tends to overestimate not only the ensemble mixing but also the real mixing characterizing the single realization. The
results are relevant in all cases parameterizing subscale effects in numerical transport models with finite grid sizes.
DE: 1719 Hydrology
DE: 1805 Computational hydrology
DE: 1829 Groundwater hydrology
DE: 1831 Groundwater quality
DE: 1832 Groundwater transport
SC: Hydrology [H]
MN: Fall Meeting 2005