Laboratoire d'Etude des Transferts en Hydrologie et Environnement, LTHE, UMR 5564, CNRS, INPG, IRD, UJF, B.P. 53, 38041, Grenoble Cedex 9, France.
Waste Manag. 2010 Mar;30(3):452-64. doi: 10.1016/j.wasman.2009.10.002. Epub 2009 Nov 12.
Leachate recirculation is a key process in the scope of operating municipal waste landfills as bioreactors, which aims to increase the moisture content to optimize the biodegradation in landfills. Given that liquid flows exhibit a complex behaviour in very heterogeneous porous media, in situ monitoring methods are required. Surface time-lapse electrical resistivity tomography (ERT) is usually proposed. Using numerical modelling with typical 2D and 3D injection plume patterns and 2D and 3D inversion codes, we show that wrong changes of resistivity can be calculated at depth if standard parameters are used for time-lapse ERT inversion. Major artefacts typically exhibit significant increases of resistivity (more than +30%) which can be misinterpreted as gas migration within the waste. In order to eliminate these artefacts, we tested an advanced time-lapse ERT procedure that includes (i) two advanced inversion tools and (ii) two alternative array geometries. The first advanced tool uses invariant regions in the model. The second advanced tool uses an inversion with a "minimum length" constraint. The alternative arrays focus on (i) a pole-dipole array (2D case), and (ii) a star array (3D case). The results show that these two advanced inversion tools and the two alternative arrays remove almost completely the artefacts within +/-5% both for 2D and 3D situations. As a field application, time-lapse ERT is applied using the star array during a 3D leachate injection in a non-hazardous municipal waste landfill. To evaluate the robustness of the two advanced tools, a synthetic model including both true decrease and increase of resistivity is built. The advanced time-lapse ERT procedure eliminates unwanted artefacts, while keeping a satisfactory image of true resistivity variations. This study demonstrates that significant and robust improvements can be obtained for time-lapse ERT monitoring of leachate recirculation in waste landfills.
渗滤液回灌是作为生物反应器运行城市垃圾填埋场的关键过程，旨在增加水分含量以优化垃圾填埋场中的生物降解。由于液体在非常不均匀的多孔介质中表现出复杂的行为，因此需要原位监测方法。表面时变电阻率层析成像（ERT）通常被提出。使用具有典型 2D 和 3D 注入羽流模式和 2D 和 3D 反演代码的数值模拟，我们表明，如果使用标准参数进行时变 ERT 反演，则可能会在深度计算出错误的电阻率变化。主要伪影通常表现出明显的电阻率增加（超过+30%），这可能被误解为废物内的气体迁移。为了消除这些伪影，我们测试了一种先进的时变 ERT 程序，该程序包括（i）两个先进的反演工具和（ii）两种替代的阵列几何形状。第一个先进的工具使用模型中的不变区域。第二个先进的工具使用具有“最小长度”约束的反演。替代的阵列集中于（i）偶极-偶极阵列（2D 情况）和（ii）星型阵列（3D 情况）。结果表明，这两种先进的反演工具和两种替代的阵列几乎完全消除了 2D 和 3D 情况下的 +/-5%以内的伪影。作为现场应用，在非危险城市垃圾填埋场的 3D 渗滤液注入期间，使用星型阵列进行时变 ERT。为了评估两个先进工具的稳健性，构建了一个包括电阻率真实降低和增加的合成模型。先进的时变 ERT 程序消除了不需要的伪影，同时保持了真实电阻率变化的令人满意的图像。这项研究表明，在垃圾填埋场渗滤液回灌的时变 ERT 监测中可以获得显著且稳健的改进。
