Kocaeli University, Faculty of Eng., Dept. of Geophysical Engineering, 41001 Kocaeli, Turkey.
Kocaeli University, Faculty of Eng., Dept. of Geophysical Engineering, 41001 Kocaeli, Turkey.
J Contam Hydrol. 2024 Nov;267:104443. doi: 10.1016/j.jconhyd.2024.104443. Epub 2024 Oct 10.
This study presents the monitoring of the contamination and remediation processes of an aquifer system using the ground-penetrating radar (GPR) method. To achieve this, periodic GPR measurements were performed on a flow cell, which was established to simulate aquifer contamination and remediation scenarios. The exterior of the flow cell was made of plexiglas, and a small hole was created at the top center to inject Dense Non-aqueous Phase Liquid (DNAPL) contamination. To study contaminant distribution and remediation processes, DNAPL Trichloroethylene (TCE) was used as the groundwater contaminant. Methyl-beta-cyclodextrin (MCD) was used as remediation agent. The interior of the flow cell was packed with two different sand grain size distributions: low permeability sand along the bottom and a higher permeability sand along the top. Common offset GPR measurements were performed along the top plane of the flow cell. The GPR measurements were conducted periodically over a total duration of 66 days, encompassing the phases prior to injecting DNAPL TCE, after injection of DNAPL TCE, and during the remediation process using MCD. Time-lapse GPR data were analyzed using 1D and 2D plots. In addition, to evaluate and monitor the contamination and remediation processes within the flow cell, central traces of the time-lapse GPR dataset were inverted by full-waveform inversion to estimate the dielectric properties of the layers and the depth by Particle Swarm Optimization (PSO). To utilize PSO, a forward model calculation using 1D finite difference in time domain adapted to PSO algorithm to generate theoretical GPR traces to compare the theoretical model GPR traces to the measured GPR traces as a "goodness of fit" metric. The electrical conductivity, relative dielectric permittivity, relative magnetic permeability, and depth of the layer were estimated by PSO of the GPR data. The results of the inversion process indicated a significant change in the estimated electrical conductivity values for the post TCE DNAPL contamination stage and the remediation stage. In addition, to support the inversion results, contaminant mass removal was quantified using moment analysis from dissolved-phase contaminant concentrations collected from the flow cell over time. According to the mass recovery calculation by moment analysis, which was collected via an outlet from the flow cell, approximately 38 % of the injected DNAPL TCE mass was removed by enhanced dissolution from the representative aquifer domain of the flow cell during the remediation stage of the experiment.
本研究采用探地雷达(GPR)方法监测含水层系统的污染和修复过程。为此,在流动池上进行了定期的 GPR 测量,该流动池是为模拟含水层污染和修复场景而建立的。流动池的外部由有机玻璃制成,在顶部中心开有一个小孔,用于注入高密度非水相液体(DNAPL)污染。为了研究污染物分布和修复过程,将地下水污染物三氯乙烯(TCE)用作 DNAPL,将甲基-β-环糊精(MCD)用作修复剂。流动池的内部用两种不同的砂粒尺寸分布填充:底部是低渗透率砂,顶部是高渗透率砂。对流动池的顶平面进行了共偏移 GPR 测量。GPR 测量在总共 66 天的时间内定期进行,涵盖了注入 DNAPL TCE 之前、注入 DNAPL TCE 之后以及使用 MCD 进行修复过程的各个阶段。使用 1D 和 2D 图分析时移 GPR 数据。此外,为了评估和监测流动池内的污染和修复过程,通过全波形反演对时移 GPR 数据集的中心迹线进行反演,以估计各层的介电特性和深度,优化算法采用粒子群优化(PSO)。为了利用 PSO,使用 1D 有限差分时间域来适应 PSO 算法生成理论 GPR 迹线,以比较理论模型 GPR 迹线和实测 GPR 迹线作为“拟合优度”指标。通过 GPR 数据的 PSO 来估计电导率、相对介电常数、相对磁导率和层深。反演过程的结果表明,在 TCE 后 DNAPL 污染阶段和修复阶段,估计的电导率值发生了显著变化。此外,为了支持反演结果,根据随时间从流动池收集的溶解相污染物浓度的矩分析,量化了污染物质量去除。根据通过流动池出口收集的矩分析的质量回收计算,在实验的修复阶段,大约 38%的注入 DNAPL TCE 质量通过增强从流动池的代表性含水层域中的溶解而去除。