Loading…

Velocity postprocessing schemes for multiscale mixed methods applied to contaminant transport in subsurface flows

We propose two postprocessing procedures ( Patch method and Stitch method) to recover local conservation of velocity fields produced by multiscale approximations that are only conservative in coarse scales. These procedures operate on small overlapping regions and are designed to be implemented in p...

Full description

Saved in:
Bibliographic Details
Published in:Computational geosciences 2020-06, Vol.24 (3), p.1141-1161
Main Authors: Guiraldello, Rafael T., Ausas, Roberto F., Sousa, Fabricio S., Pereira, Felipe, Buscaglia, Gustavo C.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose two postprocessing procedures ( Patch method and Stitch method) to recover local conservation of velocity fields produced by multiscale approximations that are only conservative in coarse scales. These procedures operate on small overlapping regions and are designed to be implemented in parallel, which makes them relatively inexpensive. We investigate the applicability of such methods when tested on single-phase flow problems using the Multiscale Robin Coupled Method (MRCM) in highly heterogeneous permeability fields for modeling the contaminant transport in the subsurface. Numerical simulations are presented aiming to illustrate and compare the performance of the new methods with a standard procedure, the Mean method, in terms of accuracy in contaminant concentration. We show that for a collection of permeability fields taken as log-normal fields, the new postprocessing procedures provide similar or better accuracy than the Mean method. Then, we turn our attention to flows in high-contrast channelized porous formations, where the new methods robustly yield more accurate results and should thus be favored.
ISSN:1420-0597
1573-1499
DOI:10.1007/s10596-019-09930-8