Inverse modelling of groundwater flow around a large-scale underground cavern system considering the excavation-induced hydraulic conductivity variation

This study proposed an inverse modelling procedure for evaluating the anisotropic hydraulic conductivity and its variation induced by excavation in fractured rocks by integrating a strain-dependent hydraulic conductivity model. The time-series measurements of both hydraulic head and discharge were used to construct the objective function for improving the reliability, which was solved with a combined method of orthogonal design, transient groundwater flow modelling, artificial neural network and genetic algorithm-based optimization for reducing the computational cost. The proposed methodology proves its effectiveness by successful inverse modelling of the groundwater flow around the underground caverns at the Jinping-I Hydropower Station.     

基于不同地质统计方法的渗透系数场对污染物运移的影响

渗透系数场的空间变异性是影响污染物运移结果的决定因素,而地质统计方法是解决渗透系数空间变异性的主要技术手段。本文利用野外场地实测数据,采用普通克里格法和指示克里格法、顺序高斯模拟法和顺序指示模拟法四种地质统计方法,插值估测和模拟再现随机渗透系数场,进而对比研究四种渗透系数场对大尺度污染物运移的影响。研究结果表明,污染羽的质心位置（一阶矩）主要由渗透系数的平均值来决定;污染羽在空间上的展布范围（二阶矩）主要受渗透系数空间变异方差的影响;条件模拟克服了估计法的平滑效果,较好地再现真实曲线的波动性,渗透系数（ lnK）估计方差与污染羽空间二阶矩随着条件模拟次数的增加而减小,并且顺序指示模拟程度更加明显。     

Groundwater contaminant source identification based on QS-ILUES

When groundwater pollution occurs,to come up with an efficient remediation plan,it is particularly important to collect information of contaminant source(location and source strength)and hydraulic conductivity field of the site accurately and quickly.However,the information can not be obtained by direct observation,and can only be derived from limited measurement data.Data assimilation of observations such as head and concentration is often used to estimate parameters of contaminant source.As for hydraulic conductivity field,especially for complex non-Gaussian field,it can be directly estimated by geostatistics method based on limited hard data,while the accuracy is often not high.Better estimation of hydraulic conductivity can be achieved by solving inverse groundwater problem.Therefore,in this study,the multi-point geostatistics method Quick Sampling(QS)is proposed and introduced for the first time and combined with the iterative local updating ensemble smoother(ILUES)to develop a new data assimilation framework QS-ILUES.It helps to solve the contaminant source parameters and non-Gaussian hydraulic conductivity field simultaneously by assimilating hydraulic head and pollutant concentration data.While the pilot points are utilized to reduce the dimension of hydraulic conductivity field,the influence of pilot points’layout and the ensemble size of ILUES algorithm on the inverse simulation results are further explored.     

运用地下水对潮汐的响应识别压力传导系数

在地下水扩散方程中,压力传导系数是描述地下水运动的重要参数。传统的方法是通过抽水或注水给地下水系统一个扰动,监测地下水水位的响应,由此计算含水层的压力传导系数。文章提出用潮汐衰减率方法识别含水层压力传导系数,其适用于滨海区承压含水层的参数识别。在推导出解析解的基础上,通过数值拟合、最小二乘法、牛顿迭代法求得含水层的压力传导系数,提出潮汐衰减率的概念,建立余切函数与潮汐衰减率的线性关系,用线性关系中的斜率和截距识别压力传导系数。用潮汐衰减率方法识别出的压力传导系数与其实际值相等,说明该方法是正确有效的。潮汐信号衰减率与海水振荡的余切函数线性相关。数值仿真表明,该方法可以准确地估算出含水层的压力传导系数。潮汐衰减率方法具有少打井,经济高效等优点。潮汐衰减率方法为实际工程应用提供了可靠的理论基础,它可以用于部分实际工程中。该方法的局限性在于需要提供含水层的部分参数,如含水层的长度、海水波动振幅、频率等。     

Estimation of regional-scale groundwater flow properties in the Bengal Basin of India and Bangladesh

Quantitative evaluation of management strategies for long-term supply of safe groundwater for drinking from the Bengal Basin aquifer (India and Bangladesh) requires estimation of the large-scale hydrogeologic properties that control flow. The Basin consists of a stratified, heterogeneous sequence of sediments with aquitards that may separate aquifers locally, but evidence does not support existence of regional confining units. Considered at a large scale, the Basin may be aptly described as a single aquifer with higher horizontal than vertical hydraulic conductivity. Though data are sparse, estimation of regional-scale aquifer properties is possible from three existing data types: hydraulic heads, ¹⁴C concentrations, and driller logs. Estimation is carried out with inverse groundwater modeling using measured heads, by model calibration using estimated water ages based on ¹⁴C, and by statistical analysis of driller logs. Similar estimates of hydraulic conductivities result from all three data types; a resulting typical value of vertical anisotropy (ratio of horizontal to vertical conductivity) is 10⁴. The vertical anisotropy estimate is supported by simulation of flow through geostatistical fields consistent with driller log data. The high estimated value of vertical anisotropy in hydraulic conductivity indicates that even disconnected aquitards, if numerous, can strongly control the equivalent hydraulic parameters of an aquifer system.     

Improving groundwater-flow modeling using optimal zoning methods

Hydraulic conductivity sometimes exhibits complicated spatial variation over a site. A thorough understanding of the spatial distributions of hydraulic conductivity helps to make deterministic models of groundwater more accurate. This study presents a novel procedure that combines simulated annealing algorithms (SA) and the shortest distance method (SD) with the modular three-dimensional groundwater flow model (MODFLOW). The procedure is applied to a hypothetical site with groundwater-monitoring wells to minimize the difference between simulated and observed hydraulic head for optimal zoning of the spatial distribution of hydraulic conductivity. The results of this optimal zoning method indicate that this new procedure not only improves the efficiency of optimization, but also increases the probability of finding the global optimum, minimizing the errors of the hydraulic head simulated by MODFLOW in two scenarios, one with known and the other with unknown hydraulic conductivity. The results also illustrated that the procedure can effectively determine and delineate hydrogeological zones.     

The hydraulic conductivity field and groundwater flow in the unconfined aquifer system of the Keta Strip,Ghana

This study investigates the hydraulic conductivity field and the groundwater flow pattern as predicted by a calibrated steady state groundwater flow model for the Keta Strip, southeastern Ghana. The hydraulic conductivity field is an important parameter in evaluating aquifer properties in space, and in general basin-wide groundwater resources evaluation and management. This study finds that the general hydraulic conductivity of the unconsolidated unconfined aquifer system of the Keta Strip ranges between 2 m/d and 20 m/d, with an average of 15 m/d. The spatial variation in horizontal hydraulic conductivity appears to take the trend in the variations in the nature of the material in space. Calibrated groundwater recharge suggests that 6.9–34% of annual precipitation recharges the shallow aquifer system. This amount of recharge is significant and suggests high fortunes in terms of groundwater resources development for agriculture and industrial activities in the area. A spatial distribution of groundwater recharge from precipitation is presented in this study. The spatial pattern appears to take the form of the distribution in horizontal hydraulic conductivity, and suggests that the vertical hydraulic conductivity takes the same pattern of spatial variation as the horizontal hydraulic conductivity. This is consistent with observations in other areas. The resulting groundwater flow is dominated by local flow systems as the unconfined system is quite shallow. A general northeast – southwest flow pattern has been observed in the study area.     

Investigating the role of hydromechanical coupling on flow and transport in shallow fractured-rock aquifers

Fractured-rock aquifers display spatially and temporally variable hydraulic conductivity generally attributed to variable fracture intensity and connectivity. Empirical evidence suggests fracture aperture and hydraulic conductivity are sensitive to in situ stress. This study investigates the sensitivity of fractured-rock hydraulic conductivity, groundwater flow paths, and advection-dominated transport to variable shear and normal fracture stiffness magnitudes for a range of deviatoric stress states. Fracture aperture and hydraulic conductivity are solved for analytically using empirical hydromechanical coupling equations; groundwater flow paths and ages are then solved for numerically using groundwater flow and advection-dispersion equations in a traditional Toth basin. Results suggest hydraulic conductivity alteration is dominated by fracture normal closure, resulting in decreasing hydraulic conductivity and increasing groundwater age with depth, and decreased depth of long flow paths with decreasing normal stiffness. Shear dilation has minimal effect on hydraulic conductivity alteration for stress states investigated here. Results are interpreted to suggest that fracture normal stiffness influences hydraulic conductivity of hydraulically active fractures and, thus, affects flow and transport in shallow (<1 km) fractured-rock aquifers. It is suggested that observed depth-dependent hydraulic conductivity trends in fractured-rock aquifers throughout the world may be partly a manifestation of hydromechanical phenomena.     

Estimation of fault-zone conductance by calibration of a regional groundwater flow model: Desert Hot Springs,California

The hydraulic conductance of a large fault zone has been estimated by calibrating a regional groundwater flow model. Drops in groundwater elevations of over 80 m have been observed along a 15-km length of the Mission Creek fault, California, USA. The large drops in elevation are attributed to the reduced hydraulic conductivity of the fault materials. A conceptual and numerical model of the two hydrologic subbasins in Desert Hot Springs, separated by the Mission Creek fault, was developed. The model was used to estimate the hydraulic conductance along the fault. The parameter estimation involved calibrating the model with observed groundwater elevations from over 40 locations over a 60-year period. The fault hydraulic conductances were estimated assuming a linear trend in the fault length, yielding variations in the fault hydraulic conductance of about an order of magnitude along the fault length (2?×?10^?11–4?×?10^?10 1/s). When an average fault thickness of 35 m is assumed, the fault hydraulic conductivity values are estimated to be from three to five orders of magnitude lower than the surrounding materials. A sensitivity analysis indicated that assumptions made in the conceptual model do not significantly affect estimated fault hydraulic conductances.     

考虑参数空间变异性的非饱和土坡可靠度分析

在考虑多个土体参数空间变异性的基础上,提出了基于拉丁超立方抽样的非饱和土坡稳定可靠度分析的非侵入式随机有限元法。利用Hermite随机多项式展开拟合边坡安全系数与输入参数间的隐式函数关系,采用拉丁超立方抽样技术产生输入参数样本点,通过Karhunen-Loève展开方法离散土体渗透系数、有效黏聚力和内摩擦角随机场,并编写了计算程序NISFEM-KL-LHS。研究了该方法在稳定渗流条件下非饱和土坡可靠度分析中的应用。结果表明：非侵入式随机有限元法为考虑多个土体参数空间变异性的非饱和土坡可靠度问题提供了一种有效的分析工具。土体渗透系数空间变异性和坡面降雨强度对边坡地下水位和最危险滑动面位置均有明显的影响。当降雨强度与饱和渗透系数的比值大于0.01时,边坡失效概率急剧增加。当土体参数变异性或者参数间负相关性较大时,忽略土体参数空间变异性会明显高估边坡失效概率。     

渗透系数空间变异性研究

水文地质参数的空间变异性是随机理论研究的基础,而渗透系数是最为重要的水文地质参数。国外有关渗透参数空间变异性的研究工作已开展很多,但渗透系数究竟服从什么分布目前尚无确切答案。利用Borden含水层试验数据,对渗透系数的空间变异性进行探讨,结果表明若处理方法得当,渗透系数应服从对数正态分布。同时,还对今后野外开展含水层渗透系数空间变异性试验研究提出几点建议。     

Probability of plume interception using conditional simulation of hydraulic head and inverse modeling

A procedure to estimate the probability of intercepting a contaminant groundwater plume for monitoring network design has been developed and demonstrated. The objective of the procedure is to use all available information in a method that accounts for the heterogeneity of the aquifer and the paucity of data. The major components of the procedure are geostatistical conditional simulation and parameter estimation that are used sequentially to generate flow paths from a suspected contaminant source location to a designated monitoring transect. From the flow paths, a histogram is constructed that represents the spatial probability distribution of plume centerlines. With an independent estimate of the plume width, a relationship between the total cost and the probability of detecting a plume can be made. The method uses geostatistical information from hydraulic head measurements and is conditioned by the data and the physics of groundwater flow. This procedure was developed specifically for the design of monitoring systems at sites where very few, if any, hydraulic conductivity data are available.     

含水岩组概化的累积导水系数法

地下水流系统理论和数值模拟技术分别是水文地质学的基本理论和技术方法,含水岩组的概化是地下水流系统分析和地下水数值模拟的重要基础,直接影响着数值模拟和水流系统分析的精度和可信度.为提高含水岩组概化的精度和可信度,提出一种含水岩组概化的新方法,即累积导水系数法.依据岩层厚度与渗透系数乘积累积值随深度的变化,以及水文地质剖面岩性分布的整体特征,概化含水介质结构.以玛纳斯河流域为例,应用该方法概化流域内的岩性剖面,结合GMS软件中TINS模块构建水文地质结构模型.结果表明,应用该方法概化后的含水层结构具有较好的合理性和仿真性,建立的三维模型很好地显示了研究区含水介质的空间展布特征,为建立地下水流模型奠定了良好的基础.      

磁共振探测估算含水层渗透系数的原位试验研究

通过一个完整的磁共振探测（MRS）、钻探、抽水试验过程,开展了MRS估算含水层渗透系数的原位试验研究。在分析MRS估算渗透系数准确性的基础上,系统剖析了引起准确性差异的主要因素,并对进一步提高估算效果提出了建议。结果显示：（1）采用现阶段常用的参数设置,与抽水试验计算值相比,MRS估算渗透系数的差值为抽水试验计算值的17.59%。（2）MRS推断的含水层顶板、底板埋深与钻孔揭示的信息相比,差值分别为4.11m、1.03m,表明其对估算渗透系数准确性的影响较小。（3）CP值是影响估算准确性的重要因素,其值为10-9数量级符合大多数地层的特点。另外,通过在已知渗透系数的钻孔附近进行MRS,从而获取CP参考值,应用该值估算的渗透系数准确性高于常用的参数设置。（4）指数a、b设置方面,应用Seevers公式（a=1、b=2）的估算效果优于Kenyon公式（a=4、b=2）。本成果有助于提高MRS估算渗透系数方法在野外条件下的适用性。     

Large-scale hydraulic conductivity distribution in an unconfined carbonate aquifer using the ocean tidal propagation

The hydraulic conductivity of an unconfined carbonate aquifer at the uplifted atoll of Minami-Daito, Japan, was evaluated by a combination of cross-spectral analysis, analytical solution, and density-dependent groundwater modeling based on observed groundwater levels in 15 wells and at sea level. The island area was divided into 10 subregions based on island morphology and on inland propagation of ocean tides. The hydraulic conductivity was obtained for each subregion using analytical solutions based on phase lags of M₂ constituents of ocean tides at each well by assuming two aquifer thicknesses (300 and 1,800 m) and two effective porosities (0.1 and 0.3). The density-dependent groundwater model evaluated the hydraulic conductivity of the subregions by reproducing observed groundwater levels. The hydraulic conductivity in the subregions was estimated as 3.46?×?10^?3 to 6.35?×?10^?2 m/s for aquifer thickness of 300 m and effective porosity of 0.1, and as 1.73?×?10^?3 to 3.17?×?10^?2 m/s for aquifer thickness of 1,800 m and the effective porosity of 0.3. It was higher in southern and northern areas, and higher in interior lowland than in the western and eastern areas. Fissures and dolomite distributions on the island control differences of the omnidirectional ocean tidal propagation and cause these differences in hydraulic conductivity. The method used for this study may also be applicable to other small islands that have few or no data for hydraulic conductivity.

     

Groundwater dewatering optimization in the Shengli no. 1 open-pit coalmine, Inner Mongolia, China

Impacted by groundwater, faults, lithology, and other factors, the slope of the Shengli no. 1 open-pit coalmine has been in a state of instability. Among these factors, groundwater in the quaternary aquifer is considered primary. Thus, an optimum design for dewatering well pumping rates is presented in this paper. A two-dimensional groundwater simulation model is built to characterize the groundwater flow of the study area. A steady-state model was applied to the observed data (head and discharge) to verify and calibrate the groundwater model. The pilot point method, with a regularization option provided by parameter estimation, was used to identify the hydraulic conductivity field. Afterward, a groundwater optimization model is integrated with the calibrated simulation model to realize groundwater dewatering optimization in the studied open-pit coalmine, and an optimization method called modified Pareto dominance-based real-coded genetic algorithm is adopted. Taking into account the safety of the mine, slope and dewatering wells, seepage discharge is added to objective function and the maximum aquifer saturated thickness is set as the constraint condition in the optimization model. The results indicate that the dewatering optimization procedure developed in this paper can serve as a useful template and framework for solving mining related water problems.     

心墙堆石坝渗透稳定可靠性分析的随机响应面法

将大规模渗流有限元计算与随机响应面法相结合,对双江口心墙堆石坝进行渗透稳定可靠性分析。在基于随机响应面法的可靠度分析框架内,堆石坝稳定渗流有限元计算过程和可靠度分析过程分开独立进行,通过对心墙渗透坡降较大区域的节点建立统一的渗透稳定功能函数,采用渗流有限元分析方法和随机响应面法,计算出该区域每个节点处的渗透破坏失效概率,并将最大失效概率作为心墙的失效概率。最后,分析了心墙渗透系数、覆盖层渗透系数、上游水位与心墙具有最大失效概率节点处渗透坡降的相关关系,以及心墙渗透系数和上游水位的变异性对心墙渗透破坏失效概率的影响。计算结果表明,随机响应面法3阶Hermite展开就能够保证良好的计算精度,且计算耗时较小;双江口堆石坝心墙具有最大失效概率节点处的渗透坡降与上游水位密切相关,而与心墙本身的渗透系数呈弱负相关关系,与覆盖层渗透系数的相关性不显著;随着上游水位变异性的增大,心墙失效概率急剧增大,而这种效应对于心墙渗透系数并不明显。研究成果为随机响应面法在实际工程中的应用奠定了一定的基础。     

基于NSMC方法的某核工程场地裂隙地下水数值模拟不确定分析

裂隙介质渗透结构表现为高度的非均质性与各项异性。为了科学有效地预测某核工程场地裂隙地下水的流动规律,揭示裂隙岩体地下水的渗流特性,笔者等采用Pilot Point调参方法与null space Monte Carlo方法（NSMC）,开展了裂隙岩体渗透结构的不确定性分析研究,构建了符合实际水文地质条件的多个渗流数值模型集合。结果表明：该方法获得的各个实现地下水位模拟结果能够与实际观测数据较好吻合,可反映工程场地裂隙地下水动力特征与流动趋势;各个实现的参数化渗透结构在空间上存在一定的差异性,但整体变化趋势是保持一致的,渗透参数的不确定性表现为在实测数据分布区域相对较低,钻孔空白区域相对较高;该方法可以弥补单一、确定性模拟结果在表征裂隙介质渗透结构方面的局限性,有效地降低模型参数的不确定性与随机性。此方法对进一步提升裂隙岩体渗流模拟精度与预测能力,深化裂隙地下水迁移规律的认识具有重要的意义。     

基于随机模拟的浑河冲洪积扇地区地下水压采风险预报

在分析随机因素对水位模拟结果影响程度的基础上建立地下水流随机模拟模型,可为地下水资源风险管理和决策提供重要依据。以浑河冲洪积扇地区为研究区,基于蒙特卡罗原理建立了区域地下水流随机预报模型,对压采条件下的地下水位上升进行风险预测和评价。参数灵敏度分析结果表明,地下水水位对含水层渗透系数的变化最敏感,其次是给水度,而对河床沉积物渗透系数和降雨入渗补给系数的灵敏性较差,且渗透系数和给水度在其率定值附近增加或减少时,灵敏度系数随之增加或减小。研究表明,压缩开采地下水资源能够有效缓解地下水水位下降带来的环境问题,地下水开采量以每年5%的速度压采时,区内地下水水位平均上涨3.3 m,但水位恢复的同时也可能诱发局部地下工程渗水,且地下建筑物的设计安全水位越低,渗水风险越大。     

A study of the contaminated banks of the Mahoning River, Northeastern Ohio, USA: characterization of the contaminated bank sediments and river water–groundwater interactions

The Mahoning River is one of the five most contaminated rivers in the U.S. This study characterized the contaminated sediments in the river banks and investigated the hydraulic interconnection between shallow aquifer in the banks with the river water. The study was conducted along the most polluted section of the river, which is 50-km long, using over 50 monitoring wells. The characterization part of the study investigated the sedimentology, hydraulic conductivity, and spatial distribution of the contaminated sediments. Results of the characterization revealed that the contaminated sediments consist of fine-grained sand, silt, mud, and clay. The spatial distribution of the contaminated sediment is heterogeneous and positively correlates with the hydraulic conductivity values, i.e., the greatest contamination occurs in high conductivity areas. Hydraulic conductivity was determined by the Hazen formula using 82 sediment samples. Bioremediation, which is one of the remedial options considered for the banks, is found to be hydraulically feasible because of sufficient hydraulic conductivity values (≥10^?4 cm/s) that ensure reasonable rates of nutrient delivery. Monitoring of water levels in the river and groundwater for a 10-month period shows that flow occurs from the river to groundwater and vice versa. The exchange of flow is influenced by rainfall. Flow of groundwater to the river will continually transport the dissolved contaminants in groundwater to the river. Therefore, findings of this study show that one of the remedial options that proposes dredging of channel sediments and permits no action for bank sediments cannot be chosen due to river water–groundwater interactions.