 |
| Titel |
On the impact of the conductivity microstructure on longitudinal and transverse macrodispersivities |
| VerfasserIn |
Mariaines Di Dato, Felipe de Barros, Aldo Fiori, Alberto Bellin |
| Konferenz |
EGU General Assembly 2016
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 18 (2016) |
| Datensatznummer |
250125323
|
| Publikation (Nr.) |
 EGU/EGU2016-4892.pdf |
|
|
|
|
|
| Zusammenfassung |
| In groundwater flows, the spread of pollutants is dominated by large variations of
hydraulic conductivity. In order to model the heterogeneity of the conductivity
field, we represent the porous medium as an ensemble of inclusions of different
conductivity placed randomly in a homogeneous matrix. By this method, we can
obtain an analytical expression for the velocity field for any degree of heterogeneity,
depending on the particular geometry of the inclusions, which characterizes the
“medium microstructure”. In the present work we considered spheroidal inclusions.
Considering pure advection and isotropic media, we aim to analyze how different
microstructures affect plume dispersion, by comparing porous formations with the same
spatial statistics. This way, we calculate longitudinal and transversal dispersivity for
binary media, as a function of the conductivity contrast, and for unimodal media, as
a function of the variance of the logconductivity field. In the binary case, i.e for
inclusions of uniform conductivity contrasting with the conductivity of the underlying
matrix, longitudinal and transversal dispersivities display an antisymmetrical behavior
depending on the value of conductivity contrast. In particular, the larger longitudinal
dispersivity is generated by low conductive inclusions, which stuck the contaminant and
stretch the plume. On the other hand, high conductive spheroids produce the larger
transversal dispersivity, which is therefore mostly influenced by the flow compression.
In the unimodal case, i.e, for inclusions with constant conductivity drawn from a
lognormal distribution, we observe that the impact of microstructure depends on the
heterogeneity degree of the formation. In fact for low heterogeneity, the shape of the
inclusions doesn’t influence the dispersivity value and the results are consistent with the
first order approximation. As the variance of the logconductivity increases, the
effect of microstructure emerges clearly. In particular, we observe that, in highly
heterogeneous media, the larger dispersivity is produced by thinner inclusions. |
|
|
|
|
|
|