Longcang Shu

Sensitivity analysis of groundwater level in Jinci Spring Basin (China) based on artificial neural network modeling

Jinci Spring in Shanxi, north China, is a major local water source. It dried up in April 1994 due to groundwater overexploitation. The groundwater system is complex, involving many nonlinear and uncertain factors. Artificial neural network (ANN) models are statistical techniques to study parameter nonlinear relationships of groundwater systems. However, ANN models offer little explanatory insight into the mechanisms of prediction models. Sensitivity analysis can overcome this shortcoming. In this study, a back-propagation neural network model was built based on the relationship between groundwater level and its sensitivity factors in Jinci Spring Basin; these sensitivity factors included precipitation, river seepage, mining drainage, groundwater withdrawals and lateral discharge to the associated Quaternary aquifer. All the sensitivity factors were analyzed with Garson’s algorithm based on the connection weights of the neural network model. The concept of “sensitivity range” was proposed to describe the value range of the input variables to which the output variables are most sensitive. The sensitivity ranges were analyzed by a local sensitivity approach. The results showed that coal mining drainage is the most sensitive anthropogenic factor, having a large effect on groundwater level of the Jinci Spring Basin.RésuméLa Source Jinci dans le Shanxi, nord de la Chine, est une importante ressource en eau locale. Elle s’est asséchée en avril 1994 du fait de la surexploitation des eaux souterraines. Le système aquifère est complexe, impliquant de nombreux facteurs non-linéaires et incertains. Les modèles par réseaux neuronaux artificiels (RNA) sont des techniques statistiques qui permettent d’étudier les paramètres des relations non-linéaires des systèmes aquifères. Toutefois, les modèles RNA n’offrent qu’une faible partie de la vision explicative des mécanismes des modèles prédictifs. Les analyses de sensibilité peuvent surmonter ce défaut. Dans cette étude, un modèle par réseaux neuronaux à rétro propagation a été construit à partir de la relation entre le niveau des eaux souterraines et de ses facteurs de sensibilité dans le Bassin de la Source Jinci. Ces facteurs comprennent : les précipitations, l’infiltration des eaux de la rivière, le drainage par les mines, les prélèvements dans la nappe et la vidange latérale vers l’aquifère quaternaire associé. Tous les facteurs de sensibilité ont été analysés avec l’algorithme de Garson basé sur les pondérations de connexion du réseau neuronal. Le concept de «plage de sensibilité» a été proposé pour décrire la plage de valeurs des variables d’entrées auxquelles les variables de sorties sont les plus sensibles. Les plages de sensibilités ont été étudiées par une analyse de sensibilité locale. Les résultats montrent que le drainage des mines de charbon est le facteur anthropique le plus sensible, ayant un impact fort sur les niveaux souterrains du Bassin de la Source Jinci.ResumenEl manantial Jinci en Shanxi, norte de China, es una fuente importante local de agua. Se secó en abril de 1994 debido a la sobreexplotación del agua subterránea. El sistema de agua subterránea es complejo, involucra muchos factores no lineares e inciertos. Los modelos de red neuronales artificiales (ANN) son técnicas estadísticas para estudiar relaciones no lineares paramétricas de los sistemas de agua subterránea. Sin embargo, los modelos ANN ofrecen poco visión explicativa de los mecanismos de los modelos predictivos. El análisis de sensibilidad puede superar esta deficiencia. En este estudio, se construyó un modelo de redes neuronales de retro-propagación basado en la región entre niveles de agua subterránea y sus factores de sensibilidad en la cuenca del manantial Jinci; estos factores de sensibilidad incluyeron a la precipitación, filtración del río, drenaje de minas, extracción de agua subterránea y descarga lateral para el acuífero Cuaternario asociado. Todos los factores de sensibilidad fueron analizados con el algoritmo de Garson basado en los pesos de conexión del modelo de la red neuronal. Se propuso el concepto de “intervalo de sensibilidad” para describir el intervalo de las variables de entrada a las cuales las variables de salida son más sensibles. Los intervalos de sensibilidad fueron analizados por una aproximación de sensibilidad local. Los resultados mostraron que el drenaje de una mina de carbón es el factor antropogénico más sensible, y que tiene un gran efecto sobre los niveles de agua subterránea de la cuenca del manantial Jinci.摘要中国华北山西省的晋祠泉是当地的主要水源。由于地下水的过量开采, 在1994年4月晋祠泉枯竭了。地下水系统是很复杂的, 受到许多非线性的和不确定的因素的影响。人工神经网络模型利用统计学的方法来研究地下水系统参数之间的非线性关系。但是, 人工神经网络模型对预测模型的机制揭示的并不多。敏感性分析可以克服这些不足。本次研究中, 基于晋祠泉流域地下水位和敏感性因素之间的关系建立了反向传播的神经网络模型, 这些敏感性因素包括降雨,河流渗漏, 矿区排水, 地下水开采和向有水力联系的第四系含水层的侧向排泄。所有这些敏感性因素都是利用基于神经网络模型连接权重的加森算法进行分析的。“敏感性范围”这个概念是用来描述输入变量值的变化范围的,输出变量对这些输入变量是最敏感的。敏感性范围是利用局部敏感性分析方法进行分析的。分析结果表明, 煤矿排水是最敏感的人为因素, 对晋祠泉流域地下水位的影响非常大。ResumoA Nascente de Jinci, em Shanxi, no norte da China, é uma fonte principal de água doce local. Devido à sobreexploração de águas subterrâneas, essa fonte esgotou-se em Abril de 1994. O sistema hidrogeológico é complexo, envolvendo muitos fatores incertos e não lineares. As redes neuronais artificiais (RNA) são técnicas estatísticas que permitem estudar as relações não-lineares de sistemas hidrogeológicos. No entanto, os modelos RNA fornecem uma visão incompleta dos mecanismos dos modelos de previsão. A análise de sensibilidade pode ultrapassar este problema. Neste estudo, foi construída uma RNA de propagação retroativa (back-propagation), baseada na relação entre o nível piezométrico e os fatores de sensibilidade na Bacia da Nascente de Jinci; estes fatores de sensibilidade incluem a precipitação, o escoamento subsuperficial, a drenagem por explorações mineiras, a extração de água subterrânea e a descarga lateral para um aquífero Quaternário associado. Todos os fatores de sensibilidade foram analisados com o algoritmo de Garson, baseado em ponderadores de conexão do modelo de rede neuronal. É proposto o conceito de “faixa de sensibilidade” para descrever a gama de valores das variáveis de entrada para as quais as variáveis de saída são as mais sensíveis. As gamas de sensibilidade foram analisadas com base numa abordagem de sensibilidade local. Os resultados mostraram que a drenagem proveniente da exploração das minas de carvão é o fator antropogénico mais sensível, influenciando fortemente o nível de águas subterrâneas na Bacia da Nascente de Jinci.

Determination of the anisotropy of an upper streambed layer in east-central Nebraska, USA

Information on the anisotropy of streambed hydraulic conductivity (K) is a necessity for analyses of water exchange and solute transport in the hyporheic zone. An approach is proposed for the determination of K, developed from existing in-situ permeameter test methods. The approach is based on determination of vertical and horizontal hydraulic conductivity of streambed sediments on-site and eliminates the effects of vertical flow in the hyporheic zone and stream-stage fluctuation, which normally influence in situ permeameter tests. The approach was applied to seven study sites on four tributaries of the Platte River in east-central Nebraska, USA. On-site permeameter tests conducted on about 172 streambed cores for the determination of vertical hydraulic conductivity (Kv) and horizontal hydraulic conductivity (Kh) at the study sites indicate that the study sites have relatively small anisotropic ratios, ranging from 0.74 to 2.40. The ratios of Kh to Kv from individual locations within a study site show greater variation than the anisotropic ratios from the mean or median K at each of the study sites.RésuméL’information sur l’anisotropie de la conductivité hydraulique (K) est nécessaire pour les analyses d’échange d’eau et de transport de soluté dans la zone hyporhéique. Une démarche est proposée pour la détermination de K, dérivée de méthodes existantes de test au perméamètre in situ. La démarche est basée sur la détermination sur site des conductivités hydrauliques verticale et horizontale des sédiments du lit et élimine les effets du flux vertical dans la zone hyporhéique et de la fluctuation du niveau d’eau, qui influencent normalement les tests perméamètre in situ. La démarche a été menée sur sept sites d’étude sur quatre tributaires de la Platte River, Centre-Est Nebraska, USA. Des tests de perméabilité sur site réalisés sur environ 172 carottes pour la détermination des conductivités hydrauliques verticale (Kv) et horizontale (Kh) indiquent des ratios d’anisotropie relativement petits, s’échelonnant de 0.74 à 2.40. Les ratios de Kh sur Kv mesurés sur différents emplacements d’un même site montrent une plus grande variabilité que les ratios d’anisotropie de la moyenne ou de la médiane de K sur chacun de ces sites d’étude.ResumenLa información sobre la anisotropía de la conductividad hidráulica de un cauce (K) es una necesidad para los análisis del intercambio de agua y transporte de solutos en una zona hiporreica. Se propone un método para la determinación de K, desarrollado a partir de métodos de prueba in situ en permeámetros existentes. El método está basado en la determinación de la conductividad hidráulica vertical y horizontal de los sedimentos del cauce en el sitio y elimina los efectos del flujo vertical en la zona hiporreica y las fluctuaciones del estado de la corriente, la cual normalmente influye en las pruebas in situ con los permeámetros. El método fue aplicado en siete sitios de estudio en cuatro tributarios del Platte River en el centro este de Nebraska, EEUU. En los lugares de las pruebas en los permeámetros se llevaron a cabo 172 testigos en el cauce para la determinación de la conductividad hidráulica vertical (Kv) y la conductividad hidráulica horizontal (Kh) en los sitios de estudios indican que los sitios de estudio tienen un cociente de anisotropía relativamente pequeña, oscilando de 0.74 a 2.40. Los cocientes de Kh a Kv a partir de lugares individuales dentro del sitio de estudio muestran mayores variaciones que los cocientes de anisotropía a partir de la K media o mediana en cada uno de los sitios estudiados.摘要河床渗透系数的各向异性比是进行河流潜流带水量交换和溶质运移分析时必不可少的重要信息。本文提出一种测定河床渗透系数的改进渗透方法,该方法是基于原位渗透试验的改进。改进试验中采用非原位操作方法,避免了河水位波动和河流潜流带垂向流对原位试验的影响。本研究采用改进的非原位渗透试验方法分别应用在美国内布拉斯加州中东部普拉特河的四条支流上的七个试验场地。非原位渗透试验共计测定172组河床砂样的垂向渗透系数(Kv)和水平渗透系数(Kh)。结果表明这些试验场地具有较弱的各向异性特征,各向异性比介于0.74和2.40之间。在任意试验场地内,每个试验位置上的Kh/Kv值与使用该场地的Kh与Kv的平均值(或中位数)得到的各向异性比相比,具有更大的变异性。ResumoA informação acerca da anisotropia da condutividade hidráulica (K) do sedimento do leito de rios é uma necessidade para as análises de trocas de água e transporte de solutos na zona hiporreica. Propõe-se uma metodologia para determinar K, desenvolvida a partir de métodos existentes de ensaios de permeâmetro in situ. A metodologia é baseada na determinação, no local, da condutividade hidráulica vertical e horizontal de sedimentos do leito de rios e elimina os efeitos do fluxo vertical na zona hiporreica e da flutuação do nível do rio, que normalmente influenciam os ensaios de permeâmetro in situ. A metodologia foi aplicada a sete casos de estudo em quatro afluentes do rio Platte no Centro-este de Nebraska, nos EUA. Os ensaios de permeâmetro in situ, feitos em cerca de 172 sondagens em sedimentos dos leitos de rios, para a determinação da condutividade hidráulica vertical (Kv) e condutividade hidráulica horizontal (Kh) indicam que as zonas de estudo têm rácios de anisotropia relativamente baixos, variando entre 0.74 e 2.40. Os rácios entre Kh e Kv de locais individuais dentro de uma zona de estudo mostram uma variação maior do que os rácios de anisotropia da K média ou mediana em cada uma das zonas de estudo.

An integrated risk assessment model of township-scaled land subsidence based on an evidential reasoning algorithm and fuzzy set theory.

Land subsidence risk assessment (LSRA) is a multi-attribute decision analysis (MADA) problem and is often characterized by both quantitative and qualitative attributes with various types of uncertainty. Therefore, the problem needs to be modeled and analyzed using methods that can handle uncertainty. In this article, we propose an integrated assessment model based on the evidential reasoning (ER) algorithm and fuzzy set theory. The assessment model is structured as a hierarchical framework that regards land subsidence risk as a composite of two key factors: hazard and vulnerability. These factors can be described by a set of basic indicators defined by assessment grades with attributes for transforming both numerical data and subjective judgments into a belief structure. The factor-level attributes of hazard and vulnerability are combined using the ER algorithm, which is based on the information from a belief structure calculated by the Dempster-Shafer (D-S) theory, and a distributed fuzzy belief structure calculated by fuzzy set theory. The results from the combined algorithms yield distributed assessment grade matrices. The application of the model to the Xixi-Chengnan area, China, illustrates its usefulness and validity for LSRA. The model utilizes a combination of all types of evidence, including all assessment information--quantitative or qualitative, complete or incomplete, and precise or imprecise--to provide assessment grades that define risk assessment on the basis of hazard and vulnerability. The results will enable risk managers to apply different risk prevention measures and mitigation planning based on the calculated risk states.

Composite Subsidence Vulnerability Assessment Based on an Index Model and Index Decomposition Method

ABSTRACT The threat of damage to buildings and other infrastructures resulting from land subsidence associated with groundwater pumping in urbanized areas is an ongoing problem requiring assessment. An important goal of subsidence vulnerability assessment is to construct a composite subsidence vulnerability index (SVI) that is represented by a set of indicators that focuses on four different thematic factors: physical, social, economic, and environmental vulnerability. These indicators are evaluated on the basis of indicator selection principles and then weighted by their contribution rate to the overall index. The weights reflect different measures assigned to the township-specific conditions. A complete and composite subsidence vulnerability assessment is developed in which future vulnerability management decision-making processes can be readily made. The vulnerability assessment includes not only the construction of the SVI, which involves selecting, assigning value to, weighting, and aggregating the vulnerability indicators, but also the presentation of the SVI decomposition. Research results demonstrate that a composite subsidence vulnerability assessment method can be made by first constructing and then decomposition-presenting the overall SVI. This allows for the relative comparison of subsidence vulnerability and the identification of the main vulnerable indicators; thus providing subsidence risk, which represents an important step toward vulnerability management of water resources.

Modeling the groundwater recharge in karst aquifers by using a reservoir model.

The estimation of the groundwater recharge in a karstic system becomes an important challenge due to the great hydrodynamic variability in both time and space. This paper proposes a two reservoir conceptual model to simulate inflow into both the conduit system and the fissure network system based on the analysis of the spring hydrograph. The structure of the model and the governing equations are proposed on the basis of the physical considerations, with the assumption that flow at the outlet of the reservoirs obeys a linear threshold function. The model is applied on the Houzhai karstic underground river basin where it successfully reflects the temporal recharge distribution. The simulated accumulation recharge is 34.29 mm, which is reasonable in relation to the actual rainfall of 92.8 mm. The variations of water volume in two reservoirs represent the storage and transform characteristics of the karst aquifer system. However, this model is particularly well suited to simulate the recharge event after intensive rainfall.

Influence of particle distribution on filter coefficient in the initial stage of filtration

Under the condition that the size of suspended particles is nonhomogeneous, we studied how filter grain and suspended particles affect filter coefficient in the early stage. A stochastic method was used to study the variation of the initial filter coefficient. Through physical experiment, the collected data include the initial inflow and outflow concentration, the size distributions of particles in suspension and so on. By introducing the standard capture probability function P(d) and the characteristic length of filter bed Lc, this separated the grains’ and particles’ influence on filter coefficient. An example showed that we could use P(d), Lc to ascertain the result of effluent distribution, ratio of Cout to Cin in any depths and equivalent filter coefficient λm. We also studied the filter coefficient λl during the experiment and the first-order derivative of λl.

Laboratory Analog Analysis of Spring Recession Curve in a Karst Aquifer with Fracture and Conduit Domains

AbstractKarst aquifers differ from other types of hydrogeological systems because of their complex behavior, which originates from strong heterogeneity. A karst spring carries an imprint of hydrologic information for the karst aquifer. The shape of the outflow hydrograph recorded at a spring is a unique reflection of the aquifer’s response. A karstic aquifer consisting of a fracture network and a conduit was proposed, and the spring recession curves generated from the designed karst aquifer were analyzed in this study. The purpose of the study was to discuss the influence of the main spring conduit diameter and the saturated thickness of the aquifer on spring recession curves. A combination of 7 groups of outlet pipes with different diameters and 7 initial water levels were used to simulate changes in diameter and saturated thickness, respectively. Thus, a total of 49 experimental tests were carried out in the laboratory karst aquifer. The results indicate that spring recession curves can be separated int...

Interpretation of Pumping Test with Radial Collector Well Using a Reservoir Model

AbstractThis study proposes a reservoir model for evaluation of aquifer parameters from a long duration pumping test conducted with a radial collector pumping well and nine observation wells in an unconfined aquifer in the Tailan River basin of China. The proposed model, based on the concept of double continuum, was used to conceptualize the pumping test site into conduit and porous reservoirs coupled by a linear flow exchange for simulating flow during the pumping test. The set of model equations developed from the concept were solved by an iterative method. The model-simulated hydraulic heads agree reasonably well with the observation heads in both the pumping and observation wells at an average normalized root mean square error of 10.99 and 8.06%, respectively, during pumping but were weaker in the recovery period. This notwithstanding, the specific yield estimates compare well with the range obtained for a numerical modeling of the entire aquifer basin. Significantly, the model was applied successfull...

Application of gray relational method to the time-lag between spring discharge and precipitation

Karst groundwater system is a complicated system with inexact data, short sample and unclear boundary conditions. Gray theory has emerged in recent years as a new tool to deal with problems of small samples and insufficient information. In this paper, the gray relational method was applied to Jinci spring to investigate the time-lag between spring flow and precipitation. The results showed that the average groundwater residence time of Jinci spring is about 7 years. Jinci spring in north China is a system of slow-response spring. The spring discharges in Jinci spring basin lag behind precipitation by several years. A slow-response to precipitation should be considered one of characteristics of karst springs in north-China.

Analysis of Karst spring discharge in semiarid of China

Jinci spring is one of the most noted karst spring in the semiarid part of northern China, and the spring dried up in 1994 due to long time groundwater overexploitation. Statistics analysis method was used in this paper to capture the relationship between the spring discharge and its influence factors, and the analysis is based on the long time series of observation data from 1956 to 1994. It is shown that the groundwater abstraction was the major factor of the depletion of spring discharge, and groundwater overexploitation is mainly because the lack of surface water resource and mining drainage of this area. Reduction of precipitation is another key factor controlling the evolution of the spring discharge, and the time-lag of rainfall recharge is 2 year in Jinci spring area. Groundwater level decline is also a signal of groundwater overexploitation, and the spring discharge is a function of groundwater level in the karst area. Karst spring discharge curve was analyzed to show the relation of spring discharge and water level, and a function was deduced based on the observation data records. This research can be a scientific guideline to the Jinci spring reflowing management.