Seminar: Chaotic Flow as a Treatment Strategy to Enhance Dilution and Break Preferentia Paths in Heterogeneous Aquifers
(Associated Unit CSIC-UPC, Barcelona)
Date: Thursday, September 23, 2021
Starting time: 12:15 (Central European Time)
live in: https://meet.google.com/snb-qdkn-eex
This week’s Guest Speaker: Oriol Bertran, Ph D Student
Title of talk: Chaotic Flow as a Treatment Strategy to Enhance Dilution and Break Preferentia Paths in Heterogeneous Aquifers
The injection of a treatment solution is a common remediation method, and several studies have shown that chaotic flow, induced by engineered injection-extraction (EIE) from wells encircling this solution, is beneficial for its spreading within the aquifer. Most of these numerical studies represent heterogeneity as the standard geostatistical multigaussian field, which is characterized by the isolation of the extreme hydraulic conductivity (K) values, with absence of preferential paths. However, connectivity is important in solute transport, reducing travel times significantly. So using multigaussian fields, that omits connectivity, to predict the fate of a treatment solution in a channeled heterogeneous site can make the remediation method to be inefficient, since the treatment solution might be transported along the channels only. It is convenient to generate fields where high-K values are connected (non-multigaussian fields) to better test the chaotic flow as a treatment strategy in both non-connected and connected fields. To be comparable, both field types share the same histogram and variogram, being the connectivity the only difference between them. In this work we evaluate the influence that chaotic flow has on the dilution of a treatment solution (represented as particles) on several heterogeneous fields and its capability in breaking preferential paths in non-multigaussian fields. For this we (I) defined a centered EIE, that encircles a treatment solution, with several combinations of number of wells and rotation velocities; and (II) performed several simulations of flow and transport for each aforementioned scenario, in randomly generated multigaussian fields and their transformed equivalent non-multigaussians. Finally, we get an optimal combination for both non-connected and connected fields that diminishes the uncertainty of the remediation method that performs the greatest dilution of the treatment solution and overcomes channeling, making it more efficient. This optimum is set by using the established metric of the dilution index of Kitanidis (1994), which measures the increasing in volume of the fluid occupied by the particles, and a metric that measures the temporal variability of K zones visited by the particles.