吉林西部浅层地下水系统对旱改水工程的响应研究
|
摘要
水是一切生物赖以生存的基本条件和人类生活生产的重要资源。地下水作为陆地上最为重要的淡水形式之一，是水循环的主要参与者和水资源的重要组成部分。在人口日益增加、工农业生产规模日渐扩大的今天，高强度的人类活动已经对水循环和水资源产生了重要的影响。
吉林西部位于松嫩平原的西南部，在气候上属于典型的干旱半干旱地区，多年平均降雨量仅为400-500mm，且分配不均，生态环境十分脆弱。区内工农业生产多以地下水作为供水水源，大量的地下水开采导致部分地区地水位呈现持续下降趋势。此外，由于工业和生活污水的排放以及农药和化肥的过量使用，地下水水质出现恶化。吉林省为增产百亿斤商品粮，规划在吉林西部实施旱改水工程，通过引地表水（嫩江、第二松花江）将部分旱田、荒地改造为水田（包括旱田改水浇地）。大量地表水的引入、土地利用类型的改变和化肥农药使用量的增加，势必导致地下水补、径、排条件的变化，区域地下水污染程度的变化以及生态环境的改变。因此，开展吉林西部浅层地下水系统对旱改水工程的响应研究十分必要。
本文以吉林省科技发展计划项目《吉林省西部旱改水工程对次生盐碱化的影响及其控治技术》（20080456）为依托，在对吉林西部旱改水工程调查分析的基础上，应用地下水模拟仿真技术，通过模型的识别和验证，建立了吉林西部平原区地下水流和水质的数值模拟模型。应用小波神经网络方法对研究区未来的降水量进行了预报。在此基础上，运用所建立的地下水水流和水质数值模拟模型，预测了未来10年两种情景方案下（旱改水工程实施和旱改水工程不实施）地下水水位和水质的响应变化。结合地理信息系统Arcgis空间分析功能的运用，分析了旱改水工程实施对研究区浅层地下水系统水流和水质的影响。为了解掌握旱改水工程实施后的环境和生态效应提供科技支撑。然后，采用LH-OAT全局敏感度分析方法，通过数值模拟模型分析了地下水位和溶质浓度对某些参数（包括渗透系数、给水度、降水入渗系数和纵向弥散度）变化的敏感程度。为今后的观测、试验和调查工作提供指导。
通过以上内容的系统研究，得到以下主要成果：
（1）在资料收集和现场调查的基础上，分析了吉林西部旱改水工程的工程概况、引排水量。并根据盐量均衡原理，对旱改水工程中引嫩入白工程、哈达山引水工程和大安灌区新建工程灌区储盐量和主要离子含量的变化量进行了计算。
分析结果表明：吉林西部旱改水工程引水总量为31.696亿m~3，其中农业灌溉引水总量为22.785亿m~3。灌溉退水总量为8.105亿m~3。旱改水工程规划农田灌溉面积416.32万亩，其中水田296.82万亩，水浇地119.5万亩。三个引水工程灌区年累计储盐量的变化量均为负值，说明从全年来看，灌区处于脱盐状态。其中，哈达山引水工程年累计储盐量的变化量最大，为-730.01kg/hm~2；大安灌区次之为-674.78kg/hm~2；引嫩入白工程最小为-482.84kg/hm~2。灌区内主要离子含量的变化量表现不完全相同，Na+和HCO3在三个灌区处于脱盐状态，Cl在三个灌区均处于积盐状态。
（2）应用地下水数值模拟方法，在详细分析研究区地质和水文地质条件的基础上，对研究区潜水含水层的分布特征、边界条件、初始条件及源汇项进行概化，建立了地下水流系统的概念模型和数学模拟模型。并利用63口潜水观测井的实测资料对模型进行了识别与验证。
识别时段计算水位与观测水位的平均残差值为0.405m，绝对平均残差为1.064m。验证时段计算水位与观测水位的平均残差值为-0.018m，绝对平均残差为0.649m。两个时段拟合误差小于0.5m的观测井占总观测井的比例均达到85%以上，识别验证结果满足精度要求，所建立的地下水流模型切合实际。
（3）应用小波神经网络方法，通过建立降水量预测模型，对吉林省西部未来10年的降水量进行了预报。
（4）基于识别验证后的地下水水流模拟模型，预测了在旱改水工程实施和旱改水工程不实施两种情景方案下，研究区未来10年的地下水位的动态变化特征。
模拟结果表明：在方案一不实施旱改水工程的情景方案下，研究区未来5年地下水流场与初始流场大致吻合，局部地区水位出现小幅下降，下降幅度一般小于0.8m。其中，前郭县的东南部和长岭县的东部高台地区域水位下降幅度相对较大，达到1.2m。未来10年地下水流场与初始流场相比变化也不大，仍有部分区域地下水位小幅下降，下降幅度一般小于1.5m。其中，前郭县的东南部和长岭县的东部高台地区域水位下降幅度达到2.2m。在方案二实施旱改水工程的情景方案下，未来5年地下水流场与初始流场相比，整体变化不明显，但局部旱改水工程影响区域水位出现一定幅度的上升。未来10年旱改水工程影响区域的水位上升幅度增大，最大上升幅度为1.63m。
（5）在上述两种情景方案预报结果的基础上，利用地理信息系统Arcgis的空间分析功能对实施旱改水工程和不实施旱改水工程两种情景方案下，未来十年10年地下水位的变化情况进行对比，分析了旱改水工程对浅层地下水位的影响状况。
对比结果表明：在旱改水工程影响下，灌区内部由于灌溉水的渗漏（包括渠道渗漏和田间渗漏），地下水位上升明显，升高幅度大部分超过1.0m，尤其是大安灌区及五家子灌区，最大升高幅度超过1.5m。而在灌区周边地下水位也有不同程度的升高，距灌区边界越近，水位上升越明显。根据两者对比后的结果表明，在旱改水工程的影响下，地下水位上升幅度超过1m的面积为2158.11km~2，其中，灌区内1125.16km~2，灌区外1032.95km~2。针对旱改水工程建成后引起的地下水位升高状况，需要重点加强对次生盐碱化的防治。特别是灌区外1032.95km~2区域发生次生盐碱化的风险最高，建议在该区域增加地下水的开采，从而控制地下水位的上升，防止次生盐碱化的发生。
（6）在地下水流模拟模型的基础上，建立了研究区水质数值模拟模型，以地下水中的总氮（包括NH+4N、NO+3N和NO+2N）作为模拟因子，运用实测资料对模型进行识别。识别结果表明，模型中源汇项的确定、边界条件的概化以及相关参数的选取切合实际，识别结果满足精度要求，可以用该模型对地下水中的水质变化状况进行预测。
（7）基于识别后的地下水质数值模拟模型，预测了在旱改水工程实施和旱改水工程不实施两种方案下，研究区未来10年浅层地下水系统中总氮的变化状况。
模拟预报结果表明，在方案一不考虑旱改水工程影响的情况下，未来10年研究区内浅层地下水中总氮浓度相比于初始浓度有不同程度的升高。乾安县和大安县交界处污染最为严重，总氮的最高浓度可达80mg/L。其次是通榆县和大安市县城所在地附近，污染也较为严重。在方案二考虑旱改水工程影响的情况下，地下水中总氮浓度相比于初始浓度的升高幅度更为明显。乾安县和大安县交界处，仍为整个研究区内污染最为严重区域，其超标面积也明显扩大。
（8）在上述两种方案模拟预报结果的基础上，仍利用Arcgis的空间分析功能，对比分析了两种情景方案下（旱改水工程实施和旱改水工程不实施）地下中总氮浓度的变化状况，分析了旱改水工程对研究区浅层地下水系统水质所造成的影响。
对比结果表明：旱改水工程实施运行后，灌区附近地下水中总氮的浓度上升明显。新建的大安灌区和五家子灌区地下水中总氮的浓度变化最为显著，最大升高值达到15mg/L；其次是扶余和松原灌区，升高幅度为9mg/L。洮儿河灌区升高幅度最小，为3mg/L。其他地区的浓度无明显变化。建议在灌区运行之后，尽量减少化肥的淋失量，提高化肥的利用率，从而减轻对地下水的污染。
（9）为考察地下水数值模拟模型中参数的不确定性，利用LH-OAT全局敏感度分析法，对模型中渗透系数、给水度、降水入渗系数、纵向弥散度四个参数的敏感性进行了分析，计算了模型输出结果（地下水位和溶质浓度）对上述四个参数的敏感程度。
计算结果表明：在研究区内13个参数分区中，2、6、7、8、9分区的地下水位对四个参数的敏感程度排序为：降水入渗系数（α）>渗透系数（K）>给水度（μ）>纵向弥散度（DL）；1、3、4、5、10、11、12、13分区的地下水位对四个参数的敏感程度排序为：降水入渗系数（α）>给水度（μ）>渗透系数（K）>纵向弥散度（DL）。在对地下水溶质浓度的影响程度方面，各个参数表现不同。纵向弥散度的敏感度在各个参数分区内普遍较大，仅在2、9、10参数分区中，降水入渗系数比纵向弥散度的敏感度大。给水度和渗透系数在各个分区内的敏感度比较相近，但在不同的参数分区内，两者的敏感度大小排序不同。敏感度分析结果表明，在对研究区浅层地下水系统进行数学模拟时，降水入渗系数对模拟计算结果影响较大，因此应注重提高降水入渗系数值的准确性，从而使数值模拟模型的预测结果更接近于实际。
Water is essential to all of living things relying for existence and vital resource to humanproduction and life. As one of the most important form of fresh water on land, groundwater isthe main participant in the hydrologic cycle and an important part of the water resources.With the increase of population and an unprecedented economic prosperity today, strong andsustained human activities has also brought serious influence on the surrounding waterresource and water cycle.
Western Jilin is located in the southwest of Songnen plain, with climate belongs to thetypical arid and semi-arid area. Average rainfall is only400-500mm, and is unevenlydistributed. Ecological environment is fragile in the study area. The productions of industrialand agricultural make more use of groundwater as a water source in the region. Water levelsfell significantly because of plentiful groundwater mining in some areas. In addition, becausethe industrial and domestic sewage discharge and excessive use pesticides and fertilizers,groundwater quality changed deterioration. In order to increase ten billion axes of commoditygrain production, Jilin province plans to implement Dryland Changed into Paddy Field projectand diverse surface water (Nen River、Second Songhua River) for development of irrigationpaddy field (including upland changed irrigated) in the western. A lot of introduction ofsurface water, the change of land type, the increase of chemical fertilizers and pesticides, willinevitably lead to the change of recharge, runoff, discharge conditions of groundwater. It islikely to aggravate the regional groundwater pollution and continuously deteriorate to ecologyenvironment. Therefore, it is necessary to carry out the research of response of shallowgroundwater system to Dryland Changed into Paddy Field project in western of Jilin province.
This thesis is sponsored by the “Jilin Province drought water improvement projects thatfocus on the salinization controlling technology”. Based on the detail, investigation andanalysis of drought water in western Jilin province, the groundwater simulation technology was occupied to build hydrogeological conceptual model of groundwater system and thecorresponding mathematical model of flow and solute transport. The characteristics ofdynamic change of groundwater level and water quality after10years in the study area undertwo solutions: with irrigation district construction and without irrigation district constructionwere predicted by using identified numerical simulation model on the basis of theprecipitation forecasts. Combined with the use of geographic information system ArcGISspatial analysis functions, the impact of the drought water improvement projects on shallowgroundwater flow and solute of the study area was assessed. Furthermore, the globalparameters LH-OAT sensitivity analysis method was employed to calculate permeabilitycoefficient, specific yield, precipitation infiltration coefficient and vertical dispersion of thedegree of influence on the level of ground water and solute distribution by numericalsimulation model.
According to the systematic study of both site investigation and modeling work, wemade the following main results:
（1）On the basis of in situ investigation and data collection, the projects summary anddisplacement were analyzed. According to the principle of water and salt balance, changingamount of reservoir salt in irrigation area (Dryland Changed into Paddy Field project: leadingNen river into Bai-cheng city, Hada mountain water diversion project and Da-an newconstruction of irrigation) were calculated.
The analysis results show that the total quantity water diversion is3.1696billion m~3onwater diversion project in western Jilin, including agricultural irrigation water diversion2.2785billion m~3and irrigation return water810.5million m~3. The upland project planningirrigation area of416.32hectares, of which296.82hectares of paddy field,119.5hectares ofirrigated land. Three irrigation area water diversion project in the cumulative of salt storagevariation are negative, and that from the point of view throughout the year, irrigation area in astate of desalination. Among them, Hada mountain diversion project in salt row led thebiggest difference, is-730.01kg/hm~2, Dan-an irrigation is-674.78kg/hm~2secondly, leadingNen river into Bai-cheng city is the smallest of-482.84kg/hm~2. But the performance of themain ions in the soil is not exactly the same, Na+and HCO3-in three irrigation area is in a state of desalination, Cl–is in a state of salt deposition.
（2）Groundwater numerical simulation method coupled with the detail hydrogeologicalconditions were used to generalize aquifer characteristics in terms of, boundary conditions,initial conditions, source sink term and establish conceptual model, mathematical model ofgroundwater flow system. All those results are identified and validated with the measureddata of63observation wells.
The average and absolute residual error between the calculating water level and theobserving water level during identification time is0.405m and1.064m respectively; Whilethe error altered during the validation, which are-0.018m, and0.649m respectively; Thefitting error under these two conditions are both less than0.5m and relative observation wellaccounted for85%and more, which demonstrated that the recognition of test results isacceptable and that the groundwater flow model is reasonable.
（3）The precipitations were predicted through establishing precipitation forecast modelby using wavelet neural network method over the next10years in western Jilin Province.
（4）The groundwater simulation model based on identification test, predicted thecharacteristics of dynamic change of groundwater level after10years in the study area undertwo solutions: with irrigation district construction and without irrigation district construction.
The results showed that groundwater flow field in the study area will be roughlyconsistent with initial flow field after five years, the water level in a small local area willdecrease slightly, and the descend range will be less than0.8meters without considering theeffects of Dryland Changed into Paddy Field project. Groundwater level decline is oppositebigger in the region of the southeast of Qian-guo and the high platform region in the east ofChang-ling, reaching to1.2meters. Groundwater flow field will be changed a little after tenyears, there are still parts of regions underground water level slightly decreased, the descendrange will be less than1.5meters. The descend range in the region of the southeast ofQian-guo and the high platform region in the east of Chang-ling will reach to2.2meters.Considering the effects of Dryland Changed into Paddy Field project, compared with theinitial flow field with5years later, the entirety change of flow field of groundwater level isnot obvious, and the water levels rise in a certain scope in local regional, which affected by Dryland Changed into Paddy Field project. The range of the water level will increase in thearea which influenced byDryland Changed into Paddy Field project after ten years, and themaximum increasingrange is1.63meters.
（5）In order to analysis the influence of groundwater level by Dryland Changed intoPaddy Field project, groundwater level changes under two programs (considering DrylandChanged into Paddy Field project or not) are compared using the spatial analysis function ofArcgis based on the simulation results of the two kinds of solutions in over the next10years.
The results showed that: under the influence of Dryland Changed into Paddy Field project,the groundwater level will rise obviously in the internal of irrigation district due to directseepage of irrigation water, most of the increasing ranges are more than1.0meters, especiallythe Da-an and Zhen-lai irrigation district, the maximum increasing range are more than1.5meters. The groundwater level in surrounding irrigation areas also has the varying degreerises, the nearer to the border irrigation is the more obvious. According to the comparisonresults, under the influence of Dryland Changed into Paddy Field project, the area increasingrange of groundwater level more than1meters is2158.11km~2, among them, in the outside ofirrigation area is1125.16km~2, the inside of irrigation area is1032.95km~2. In connection withstatus of groundwater level rise caused by Dryland Changed into Paddy Field project, themeasure of preventing secondary salinization should be strengthened. Especially the highestrisk of secondary salinization is in outside of irrigated district which area is1032.95km~2. Forthe purpose of controlling groundwater reason ably exploitation and preventing theoccurrence of secondary salinization，it is suggested that the exploitation of groundwatershould increase in the area that water level rise obviously.
（6）Based on the groundwater flow simulation model, the groundwater solute transportsimulation model was built. Then, the total nitrogen (NH4+N、N03--N and N02--N) ingroundwater was used as simulated factor in transport simulation model. The model wasidentified through the measured data. The identified results showed that the determination ofsource and sink terms, the generalization of the boundary conditions and the selection ofinterrelated parameter in model were reasonable, and the effect of identified is good. Themodel can be used to predict the groundwater solute transport.
（7）We predict total nitrogen distribution of shallow groundwater in the next10yearsbased on the identification of groundwater transport model under the projects of both buildingirrigation and without building irrigation.
Simulation results showed that the nitrogen concentration in the next10years ofprograms1, which take no consideration of Dryland changed into paddy field project is higherthan that in2007.The highest concentration locates in the junction of Qian-an and Da-an city.Its concentration is80mg/L. In addition, the concentration near Tong-yu county and Da-ancity is also high. The nitrogen concentration in the next10years of program2, which takesconsideration of Dryland changed into paddy field project, is more significant than that in2007. The heaviest place also locates in the junction of Qian-an and Da-an city, whichexceeded the scope obviously.
（8）We do comparative analysis of groundwater total nitrogen concentration distributionon two programs (considering Dryland Changed into Paddy Field project or not) on the basisof the simulation results using ArcGIS and the spatial analysis. Furthermore, we illustrate theimpact of the drought water improvement projects shallow groundwater in the study areaenvironment.
The comparison results showed that: after running of dry land changed into paddy fieldproject in the new irrigation area, groundwater total nitrogen concentration increased veryobvious, especially in the area of new Da-an Irrigation District and Wujiazi irrigation. What’smore, the largest increase in value reached15mg/L. Moreover, it is9mg/L in Fu-yu andSong-yuan irrigation. The lowest concentration is3mg/L in Tao’er River Irrigation. Nosignificant concentration changes in other regions. After the completion of the irrigation area,it is encouraged to carry out reasonable scientific fertilization, which can improve foodproduction with less fertilizer leaching. Thus, it can improve the efficiency of fertilizer useand ease the pollution of groundwater.
（9）In order to determine the uncertainty of the parameters in the groundwater numericalmodel that had been established, the sensitivity of four parameters: permeability coefficient,specific yield, precipitation infiltration coefficient and longitudinal dispersion coefficient ofthe model were analyzed using the LH-OAT global analysis method, the impact of different parameters zones on the groundwater flow and water quality were also calculated.
Results show that: in the13parameters partition zones of the study area, in2,6,7,8,9zones, the rank of the four parameters sorted by influence degree on the groundwater level is:precipitation infiltration coefficient (α)> permeability coefficient (K)> specific yield (u)>longitudinal dispersion degree (DL); in1,3,4,5,10,11,12,13partition zones, the rank of thefour parameters is precipitation infiltration coefficient (α)> specific yield (u)> permeabilitycoefficient (K)> longitudinal dispersion degree (DL).The influence degree of differentparameters partitions on groundwater solute transport were different. The sensitivity of thelongitudinal dispersion are large in common, but the sensitivity of precipitation infiltrationcoefficient is larger than that of longitudinal dispersion in2,9,10parameters partition zones;the sensitivity of specific yield and permeability coefficient in every partition zones aresimilar, but in different parameters partition zones, the sensitivity have different orders sortedby their value. According to simulation results, in the research of the shallow groundwatersystem in the studied area, the exploration and test job of the precipitation infiltrationcoefficient should be added to improve the accuracy of the parameters, which could make theresults of numerical simulation of the model more realistic.
吉林西部位于松嫩平原的西南部，在气候上属于典型的干旱半干旱地区，多年平均降雨量仅为400-500mm，且分配不均，生态环境十分脆弱。区内工农业生产多以地下水作为供水水源，大量的地下水开采导致部分地区地水位呈现持续下降趋势。此外，由于工业和生活污水的排放以及农药和化肥的过量使用，地下水水质出现恶化。吉林省为增产百亿斤商品粮，规划在吉林西部实施旱改水工程，通过引地表水（嫩江、第二松花江）将部分旱田、荒地改造为水田（包括旱田改水浇地）。大量地表水的引入、土地利用类型的改变和化肥农药使用量的增加，势必导致地下水补、径、排条件的变化，区域地下水污染程度的变化以及生态环境的改变。因此，开展吉林西部浅层地下水系统对旱改水工程的响应研究十分必要。
本文以吉林省科技发展计划项目《吉林省西部旱改水工程对次生盐碱化的影响及其控治技术》（20080456）为依托，在对吉林西部旱改水工程调查分析的基础上，应用地下水模拟仿真技术，通过模型的识别和验证，建立了吉林西部平原区地下水流和水质的数值模拟模型。应用小波神经网络方法对研究区未来的降水量进行了预报。在此基础上，运用所建立的地下水水流和水质数值模拟模型，预测了未来10年两种情景方案下（旱改水工程实施和旱改水工程不实施）地下水水位和水质的响应变化。结合地理信息系统Arcgis空间分析功能的运用，分析了旱改水工程实施对研究区浅层地下水系统水流和水质的影响。为了解掌握旱改水工程实施后的环境和生态效应提供科技支撑。然后，采用LH-OAT全局敏感度分析方法，通过数值模拟模型分析了地下水位和溶质浓度对某些参数（包括渗透系数、给水度、降水入渗系数和纵向弥散度）变化的敏感程度。为今后的观测、试验和调查工作提供指导。
通过以上内容的系统研究，得到以下主要成果：
（1）在资料收集和现场调查的基础上，分析了吉林西部旱改水工程的工程概况、引排水量。并根据盐量均衡原理，对旱改水工程中引嫩入白工程、哈达山引水工程和大安灌区新建工程灌区储盐量和主要离子含量的变化量进行了计算。
分析结果表明：吉林西部旱改水工程引水总量为31.696亿m~3，其中农业灌溉引水总量为22.785亿m~3。灌溉退水总量为8.105亿m~3。旱改水工程规划农田灌溉面积416.32万亩，其中水田296.82万亩，水浇地119.5万亩。三个引水工程灌区年累计储盐量的变化量均为负值，说明从全年来看，灌区处于脱盐状态。其中，哈达山引水工程年累计储盐量的变化量最大，为-730.01kg/hm~2；大安灌区次之为-674.78kg/hm~2；引嫩入白工程最小为-482.84kg/hm~2。灌区内主要离子含量的变化量表现不完全相同，Na+和HCO3在三个灌区处于脱盐状态，Cl在三个灌区均处于积盐状态。
（2）应用地下水数值模拟方法，在详细分析研究区地质和水文地质条件的基础上，对研究区潜水含水层的分布特征、边界条件、初始条件及源汇项进行概化，建立了地下水流系统的概念模型和数学模拟模型。并利用63口潜水观测井的实测资料对模型进行了识别与验证。
识别时段计算水位与观测水位的平均残差值为0.405m，绝对平均残差为1.064m。验证时段计算水位与观测水位的平均残差值为-0.018m，绝对平均残差为0.649m。两个时段拟合误差小于0.5m的观测井占总观测井的比例均达到85%以上，识别验证结果满足精度要求，所建立的地下水流模型切合实际。
（3）应用小波神经网络方法，通过建立降水量预测模型，对吉林省西部未来10年的降水量进行了预报。
（4）基于识别验证后的地下水水流模拟模型，预测了在旱改水工程实施和旱改水工程不实施两种情景方案下，研究区未来10年的地下水位的动态变化特征。
模拟结果表明：在方案一不实施旱改水工程的情景方案下，研究区未来5年地下水流场与初始流场大致吻合，局部地区水位出现小幅下降，下降幅度一般小于0.8m。其中，前郭县的东南部和长岭县的东部高台地区域水位下降幅度相对较大，达到1.2m。未来10年地下水流场与初始流场相比变化也不大，仍有部分区域地下水位小幅下降，下降幅度一般小于1.5m。其中，前郭县的东南部和长岭县的东部高台地区域水位下降幅度达到2.2m。在方案二实施旱改水工程的情景方案下，未来5年地下水流场与初始流场相比，整体变化不明显，但局部旱改水工程影响区域水位出现一定幅度的上升。未来10年旱改水工程影响区域的水位上升幅度增大，最大上升幅度为1.63m。
（5）在上述两种情景方案预报结果的基础上，利用地理信息系统Arcgis的空间分析功能对实施旱改水工程和不实施旱改水工程两种情景方案下，未来十年10年地下水位的变化情况进行对比，分析了旱改水工程对浅层地下水位的影响状况。
对比结果表明：在旱改水工程影响下，灌区内部由于灌溉水的渗漏（包括渠道渗漏和田间渗漏），地下水位上升明显，升高幅度大部分超过1.0m，尤其是大安灌区及五家子灌区，最大升高幅度超过1.5m。而在灌区周边地下水位也有不同程度的升高，距灌区边界越近，水位上升越明显。根据两者对比后的结果表明，在旱改水工程的影响下，地下水位上升幅度超过1m的面积为2158.11km~2，其中，灌区内1125.16km~2，灌区外1032.95km~2。针对旱改水工程建成后引起的地下水位升高状况，需要重点加强对次生盐碱化的防治。特别是灌区外1032.95km~2区域发生次生盐碱化的风险最高，建议在该区域增加地下水的开采，从而控制地下水位的上升，防止次生盐碱化的发生。
（6）在地下水流模拟模型的基础上，建立了研究区水质数值模拟模型，以地下水中的总氮（包括NH+4N、NO+3N和NO+2N）作为模拟因子，运用实测资料对模型进行识别。识别结果表明，模型中源汇项的确定、边界条件的概化以及相关参数的选取切合实际，识别结果满足精度要求，可以用该模型对地下水中的水质变化状况进行预测。
（7）基于识别后的地下水质数值模拟模型，预测了在旱改水工程实施和旱改水工程不实施两种方案下，研究区未来10年浅层地下水系统中总氮的变化状况。
模拟预报结果表明，在方案一不考虑旱改水工程影响的情况下，未来10年研究区内浅层地下水中总氮浓度相比于初始浓度有不同程度的升高。乾安县和大安县交界处污染最为严重，总氮的最高浓度可达80mg/L。其次是通榆县和大安市县城所在地附近，污染也较为严重。在方案二考虑旱改水工程影响的情况下，地下水中总氮浓度相比于初始浓度的升高幅度更为明显。乾安县和大安县交界处，仍为整个研究区内污染最为严重区域，其超标面积也明显扩大。
（8）在上述两种方案模拟预报结果的基础上，仍利用Arcgis的空间分析功能，对比分析了两种情景方案下（旱改水工程实施和旱改水工程不实施）地下中总氮浓度的变化状况，分析了旱改水工程对研究区浅层地下水系统水质所造成的影响。
对比结果表明：旱改水工程实施运行后，灌区附近地下水中总氮的浓度上升明显。新建的大安灌区和五家子灌区地下水中总氮的浓度变化最为显著，最大升高值达到15mg/L；其次是扶余和松原灌区，升高幅度为9mg/L。洮儿河灌区升高幅度最小，为3mg/L。其他地区的浓度无明显变化。建议在灌区运行之后，尽量减少化肥的淋失量，提高化肥的利用率，从而减轻对地下水的污染。
（9）为考察地下水数值模拟模型中参数的不确定性，利用LH-OAT全局敏感度分析法，对模型中渗透系数、给水度、降水入渗系数、纵向弥散度四个参数的敏感性进行了分析，计算了模型输出结果（地下水位和溶质浓度）对上述四个参数的敏感程度。
计算结果表明：在研究区内13个参数分区中，2、6、7、8、9分区的地下水位对四个参数的敏感程度排序为：降水入渗系数（α）>渗透系数（K）>给水度（μ）>纵向弥散度（DL）；1、3、4、5、10、11、12、13分区的地下水位对四个参数的敏感程度排序为：降水入渗系数（α）>给水度（μ）>渗透系数（K）>纵向弥散度（DL）。在对地下水溶质浓度的影响程度方面，各个参数表现不同。纵向弥散度的敏感度在各个参数分区内普遍较大，仅在2、9、10参数分区中，降水入渗系数比纵向弥散度的敏感度大。给水度和渗透系数在各个分区内的敏感度比较相近，但在不同的参数分区内，两者的敏感度大小排序不同。敏感度分析结果表明，在对研究区浅层地下水系统进行数学模拟时，降水入渗系数对模拟计算结果影响较大，因此应注重提高降水入渗系数值的准确性，从而使数值模拟模型的预测结果更接近于实际。
Water is essential to all of living things relying for existence and vital resource to humanproduction and life. As one of the most important form of fresh water on land, groundwater isthe main participant in the hydrologic cycle and an important part of the water resources.With the increase of population and an unprecedented economic prosperity today, strong andsustained human activities has also brought serious influence on the surrounding waterresource and water cycle.
Western Jilin is located in the southwest of Songnen plain, with climate belongs to thetypical arid and semi-arid area. Average rainfall is only400-500mm, and is unevenlydistributed. Ecological environment is fragile in the study area. The productions of industrialand agricultural make more use of groundwater as a water source in the region. Water levelsfell significantly because of plentiful groundwater mining in some areas. In addition, becausethe industrial and domestic sewage discharge and excessive use pesticides and fertilizers,groundwater quality changed deterioration. In order to increase ten billion axes of commoditygrain production, Jilin province plans to implement Dryland Changed into Paddy Field projectand diverse surface water (Nen River、Second Songhua River) for development of irrigationpaddy field (including upland changed irrigated) in the western. A lot of introduction ofsurface water, the change of land type, the increase of chemical fertilizers and pesticides, willinevitably lead to the change of recharge, runoff, discharge conditions of groundwater. It islikely to aggravate the regional groundwater pollution and continuously deteriorate to ecologyenvironment. Therefore, it is necessary to carry out the research of response of shallowgroundwater system to Dryland Changed into Paddy Field project in western of Jilin province.
This thesis is sponsored by the “Jilin Province drought water improvement projects thatfocus on the salinization controlling technology”. Based on the detail, investigation andanalysis of drought water in western Jilin province, the groundwater simulation technology was occupied to build hydrogeological conceptual model of groundwater system and thecorresponding mathematical model of flow and solute transport. The characteristics ofdynamic change of groundwater level and water quality after10years in the study area undertwo solutions: with irrigation district construction and without irrigation district constructionwere predicted by using identified numerical simulation model on the basis of theprecipitation forecasts. Combined with the use of geographic information system ArcGISspatial analysis functions, the impact of the drought water improvement projects on shallowgroundwater flow and solute of the study area was assessed. Furthermore, the globalparameters LH-OAT sensitivity analysis method was employed to calculate permeabilitycoefficient, specific yield, precipitation infiltration coefficient and vertical dispersion of thedegree of influence on the level of ground water and solute distribution by numericalsimulation model.
According to the systematic study of both site investigation and modeling work, wemade the following main results:
（1）On the basis of in situ investigation and data collection, the projects summary anddisplacement were analyzed. According to the principle of water and salt balance, changingamount of reservoir salt in irrigation area (Dryland Changed into Paddy Field project: leadingNen river into Bai-cheng city, Hada mountain water diversion project and Da-an newconstruction of irrigation) were calculated.
The analysis results show that the total quantity water diversion is3.1696billion m~3onwater diversion project in western Jilin, including agricultural irrigation water diversion2.2785billion m~3and irrigation return water810.5million m~3. The upland project planningirrigation area of416.32hectares, of which296.82hectares of paddy field,119.5hectares ofirrigated land. Three irrigation area water diversion project in the cumulative of salt storagevariation are negative, and that from the point of view throughout the year, irrigation area in astate of desalination. Among them, Hada mountain diversion project in salt row led thebiggest difference, is-730.01kg/hm~2, Dan-an irrigation is-674.78kg/hm~2secondly, leadingNen river into Bai-cheng city is the smallest of-482.84kg/hm~2. But the performance of themain ions in the soil is not exactly the same, Na+and HCO3-in three irrigation area is in a state of desalination, Cl–is in a state of salt deposition.
（2）Groundwater numerical simulation method coupled with the detail hydrogeologicalconditions were used to generalize aquifer characteristics in terms of, boundary conditions,initial conditions, source sink term and establish conceptual model, mathematical model ofgroundwater flow system. All those results are identified and validated with the measureddata of63observation wells.
The average and absolute residual error between the calculating water level and theobserving water level during identification time is0.405m and1.064m respectively; Whilethe error altered during the validation, which are-0.018m, and0.649m respectively; Thefitting error under these two conditions are both less than0.5m and relative observation wellaccounted for85%and more, which demonstrated that the recognition of test results isacceptable and that the groundwater flow model is reasonable.
（3）The precipitations were predicted through establishing precipitation forecast modelby using wavelet neural network method over the next10years in western Jilin Province.
（4）The groundwater simulation model based on identification test, predicted thecharacteristics of dynamic change of groundwater level after10years in the study area undertwo solutions: with irrigation district construction and without irrigation district construction.
The results showed that groundwater flow field in the study area will be roughlyconsistent with initial flow field after five years, the water level in a small local area willdecrease slightly, and the descend range will be less than0.8meters without considering theeffects of Dryland Changed into Paddy Field project. Groundwater level decline is oppositebigger in the region of the southeast of Qian-guo and the high platform region in the east ofChang-ling, reaching to1.2meters. Groundwater flow field will be changed a little after tenyears, there are still parts of regions underground water level slightly decreased, the descendrange will be less than1.5meters. The descend range in the region of the southeast ofQian-guo and the high platform region in the east of Chang-ling will reach to2.2meters.Considering the effects of Dryland Changed into Paddy Field project, compared with theinitial flow field with5years later, the entirety change of flow field of groundwater level isnot obvious, and the water levels rise in a certain scope in local regional, which affected by Dryland Changed into Paddy Field project. The range of the water level will increase in thearea which influenced byDryland Changed into Paddy Field project after ten years, and themaximum increasingrange is1.63meters.
（5）In order to analysis the influence of groundwater level by Dryland Changed intoPaddy Field project, groundwater level changes under two programs (considering DrylandChanged into Paddy Field project or not) are compared using the spatial analysis function ofArcgis based on the simulation results of the two kinds of solutions in over the next10years.
The results showed that: under the influence of Dryland Changed into Paddy Field project,the groundwater level will rise obviously in the internal of irrigation district due to directseepage of irrigation water, most of the increasing ranges are more than1.0meters, especiallythe Da-an and Zhen-lai irrigation district, the maximum increasing range are more than1.5meters. The groundwater level in surrounding irrigation areas also has the varying degreerises, the nearer to the border irrigation is the more obvious. According to the comparisonresults, under the influence of Dryland Changed into Paddy Field project, the area increasingrange of groundwater level more than1meters is2158.11km~2, among them, in the outside ofirrigation area is1125.16km~2, the inside of irrigation area is1032.95km~2. In connection withstatus of groundwater level rise caused by Dryland Changed into Paddy Field project, themeasure of preventing secondary salinization should be strengthened. Especially the highestrisk of secondary salinization is in outside of irrigated district which area is1032.95km~2. Forthe purpose of controlling groundwater reason ably exploitation and preventing theoccurrence of secondary salinization，it is suggested that the exploitation of groundwatershould increase in the area that water level rise obviously.
（6）Based on the groundwater flow simulation model, the groundwater solute transportsimulation model was built. Then, the total nitrogen (NH4+N、N03--N and N02--N) ingroundwater was used as simulated factor in transport simulation model. The model wasidentified through the measured data. The identified results showed that the determination ofsource and sink terms, the generalization of the boundary conditions and the selection ofinterrelated parameter in model were reasonable, and the effect of identified is good. Themodel can be used to predict the groundwater solute transport.
（7）We predict total nitrogen distribution of shallow groundwater in the next10yearsbased on the identification of groundwater transport model under the projects of both buildingirrigation and without building irrigation.
Simulation results showed that the nitrogen concentration in the next10years ofprograms1, which take no consideration of Dryland changed into paddy field project is higherthan that in2007.The highest concentration locates in the junction of Qian-an and Da-an city.Its concentration is80mg/L. In addition, the concentration near Tong-yu county and Da-ancity is also high. The nitrogen concentration in the next10years of program2, which takesconsideration of Dryland changed into paddy field project, is more significant than that in2007. The heaviest place also locates in the junction of Qian-an and Da-an city, whichexceeded the scope obviously.
（8）We do comparative analysis of groundwater total nitrogen concentration distributionon two programs (considering Dryland Changed into Paddy Field project or not) on the basisof the simulation results using ArcGIS and the spatial analysis. Furthermore, we illustrate theimpact of the drought water improvement projects shallow groundwater in the study areaenvironment.
The comparison results showed that: after running of dry land changed into paddy fieldproject in the new irrigation area, groundwater total nitrogen concentration increased veryobvious, especially in the area of new Da-an Irrigation District and Wujiazi irrigation. What’smore, the largest increase in value reached15mg/L. Moreover, it is9mg/L in Fu-yu andSong-yuan irrigation. The lowest concentration is3mg/L in Tao’er River Irrigation. Nosignificant concentration changes in other regions. After the completion of the irrigation area,it is encouraged to carry out reasonable scientific fertilization, which can improve foodproduction with less fertilizer leaching. Thus, it can improve the efficiency of fertilizer useand ease the pollution of groundwater.
（9）In order to determine the uncertainty of the parameters in the groundwater numericalmodel that had been established, the sensitivity of four parameters: permeability coefficient,specific yield, precipitation infiltration coefficient and longitudinal dispersion coefficient ofthe model were analyzed using the LH-OAT global analysis method, the impact of different parameters zones on the groundwater flow and water quality were also calculated.
Results show that: in the13parameters partition zones of the study area, in2,6,7,8,9zones, the rank of the four parameters sorted by influence degree on the groundwater level is:precipitation infiltration coefficient (α)> permeability coefficient (K)> specific yield (u)>longitudinal dispersion degree (DL); in1,3,4,5,10,11,12,13partition zones, the rank of thefour parameters is precipitation infiltration coefficient (α)> specific yield (u)> permeabilitycoefficient (K)> longitudinal dispersion degree (DL).The influence degree of differentparameters partitions on groundwater solute transport were different. The sensitivity of thelongitudinal dispersion are large in common, but the sensitivity of precipitation infiltrationcoefficient is larger than that of longitudinal dispersion in2,9,10parameters partition zones;the sensitivity of specific yield and permeability coefficient in every partition zones aresimilar, but in different parameters partition zones, the sensitivity have different orders sortedby their value. According to simulation results, in the research of the shallow groundwatersystem in the studied area, the exploration and test job of the precipitation infiltrationcoefficient should be added to improve the accuracy of the parameters, which could make theresults of numerical simulation of the model more realistic.
引文
Anderson M P, Woessner W W. Applied groundwater modeling [M]. London: Academic Press,Inc.,1992.
Ayse M, Ahmet Y. A fuzzylogic approach to assess groundwater pollution levels belowagricultural fields [J]. Environmental Monitoring and Assessment,2006,118:337–354.
Brebbia C A, Wrobel L C. The boundary element method: Computer methods in fluids [J].London: Pentech Press, Ltd.,1980:26-48.
Bredehoeft J D. Finite difference approximations to the equations of ground-water flow[J].Water Resources Research,1969,5(2):531-534.
Diersch H J G, Kolditz O. Variable-density flow and transport in porous media: approachesand challenges [J]. Advances in Water Resources,2002,25:899–944.
Freeze R A, Witherspoon P A. Theoretical analysis of regional ground water flow3:Quantitative interpretations [J]. Water Resour Res,1968,4:581–590.
Freeze R A, Witherspoon P A. Theoretical analysis of regional groundwater flow1: Analyticaland numerical solutions to the mathematical model [J]. Water Resour Res,1966,2:641–656.
Freeze R A, Witherspoon P A. Theoretical analysis of regional groundwater flow2: Effect ofwater table configuration and subsurface permeability variations [J]. Water Resour Res,1967,3(2):623–634.
Guiggiani M, Gigante A. A general algorithm for multidimensional cauchy principal valueintegrals in the boundary element method [J]. Journal of Applied Mechanics,1990,57(4):906-915.
Holvoet K, van Griensven A, Seuntjens P, et al. Sensitivity analysis for hydrology andpesticide supply towards the river in SWAT [J]. Physics and Chemistry of the Earth,2005,30(8):518-526.
Huyakorn P S, Pinder G F. Computational methods in subsurface flow [M]. Academic Press,Orlando, Florida, Inc.,1986.
Jury W A. Simulation of solute transport using a transfer function model [J]. Water ResourcesResearch,1982,18(2):363-368.
Kandeler E, Eder G, Sobotik M. Microbial biomass, N-mineralization, and the activities ofvarious enzymes in relation to nitrate leaching and root distribution in a slurry-amendedgrassland [J]. Biology and fertility of soils,1994,18(l):7-12.
Kloosterman F H, Stuurman R J, Van der Meijden R. Groundwater flow systems analysis on aregional and nation-wide scale in the netherlands: the use of flow systems analysis inwetland management [J]. Water Science and Technology,1995,31:375–378.
Konikow L F, Bredehoeft J D. Computer model of two-dimensional solute transport anddispersion in ground water [M]. US Government Printing Office,1978.
Morris M D. Factorial sampling plans for preliminary computational experiments [J].American Statistical Association and the American Society for Quality Control,1991,33:161-174.
Rogers C C M, Beven K J, Morris E M, et al. Sensitivity analysis, calibration and predictiveuncertainty of the institute of hydrology distributed model [J]. Journal of hydrology,1985,81(1):179-191.
Simms M. Dolomitization by groundwater—flow systems in carbonate platforms [J]. GulfCoast Assoc,1984:411–420.
Sobol I M. Sensitivity estimates for nonlinear mathematical model math [J]. ModelingComput Exp,1993(1):407-414.
Stallman R W. Numerical analysis of regional water levels todefine aquifer hydrology [J]. AmGeophys Union Trans,1956,37(4):451–460.
Stocke J T, Morris S L, Gioia I M, et al. The Einstein observatory extendedmedium-sensitivity survey [J]. The Astrophysical Journal Supplement Series,1991,76:813-874.
Sykes J F, Wilson J L, Andrews R W. Sensitivity analysis for steady state groundwater flowusing adjoin to perators [J]. Water Resources Research,1985,21(3):359-371.
Tóth J. A theoretical analysis of groundwater flow in small drainage basins [J]. Geophys Res J,1963,68:4785–4812.
Tóth J. A theory of groundwater motion in small drainage basins in central Alberta, Canada[J]. Geophys Res J,1962,67(11):4375–4387.
Tóth J. Groundwater as a geologic agent: an overview of the causes, processes, andmanifestations [J]. Hydrogeol J,1999,7:1–14.
Trescott P C. and Larson S P. Comparison of iterative methods of solving two-dimensionalgroundwater flow equations [J]. Water Resour Res,1977,13(1).
Tyson H N, Weber E M. Groundwater management for the nation’s future [J]. ComputerSimulations of Groundwater Basins,1964,90:59–77.
Van Griensven A, Meixner T, Grunwald S, et al. A global sensitivity analysis tool for theparameters of multi-variable catchment models [J]. Journal of Hydrology,2006,324(1):10-23.
Willis R, Finney B A. Planning model for optimal control of saltwater intrusion [J]. Journal ofWater Resources Planning and Management,1988,114(2):163-178.
Yeh G T. Computational subsurface hydrology: Fluid flows [D]. Kumwer Academic,Pennsylvania State University,1999.
Zheng C, Bennett G D. Applied contaminant transport modeling [M]. New York: John Wileyand Sons,1995.
鲍卫锋,黄介生,陈玲,等.地表水地下水联合运用模拟模型[J].中国农村水利水电,2003,6:45-47.
卞玉梅.吉林省西部地下水动态研究[D].吉林大学,2006.
陈崇希,李国敏.地下水溶质运移理论及模型[M].武汉:中国地质大学出版社,1996.
陈家军.边界元法在地下水补给量计算中的应用[J].长安大学学报(地球科学版),1989,(01):50-63.
陈钟祥.渗流力学的近况、动向和展望(续)[J].力学学报,1974,(02):94-103(55).
范伟.吉林省平原区地下水功能评价[D].吉林大学,2007.
冯波.吉林省西部地下水化学特征演化研究[D].吉林大学,2007.
冯启申,朱琰,李彦伟.地表水水质模型概述[J].安全与环境工程,2010,(02):1-4.
傅荟璇，赵红.MATLAB神经网络应用设计[M].北京：机械工业出版社，2010，19-38，193-207.
高慧琴,杨明明,黑亮,叶伟聪. MODFLOW和FEFLOW在国内地下水数值模拟中的应用[J].地下水,2012,(04):13-15.
宫辉力,李培,崔桂莲,魏学君,许斌.地下水系统准三维流场数值模拟与流场宏观调控[J].系统工程理论与实践,1999,(02):136-140.
韩磊,赵盼,闫金兵.农村河道水环境恶化成因及对策[J].天津农业科学,2010,(03):124-126.
韩曼.地下水封石油洞库渗流场及溶质运移模拟研究[D].中国海洋大学,2007.
贺国平,邵景力,崔亚莉,张德强. FEFLOW在地下水流模拟方面的应用[J].成都理工大学学报(自然科学版),2003,(04):356-361.
胡轶,谢水波,蒋明,陈泽昂,刘奇,杨晶. Visual Modflow及其在地下水模拟中的应用[J].南华大学学报(自然科学版),2006,(02):1-5.
黄清华,张万昌. SWAT模型参数敏感性分析及应用[J].干旱区地理,2010,(01):8-15.
江曙光.中国水污染现状及防治对策[J].现代农业科技,2010,(07):313-315.
江思珉,朱国荣,王浩然,孙斌堂. FAC方法在地下水数值模拟中的应用[J].水利学报,2006,(11):1389-1392.
吉林环境科学院.吉林省百亿斤商品粮能力建设总体规划环境影响报告书[R].长春：吉林环境科学院,2008.
蒋辉.河南沁阳市地下水水质的数值模拟[J].水文地质工程地质,1996,(02):14-15(20).
孔凡哲,宋晓猛,占车生,叶爱中.水文模型参数敏感性快速定量评估的RSMSobol方法[J].地理学报,2011,(09):1270-1280.
兰盈盈,肖长来,梁秀娟.吉林省西部主要生态环境地质问题及对策[J].吉林水利,2004,(04):1-4.
李存法,马自芬,刘秀婷.地下水非稳定渗流分析的显式有限单元法[J].地下水,2003,(03):150-151+171.
李军,梅琨,张明华.水质模型的全局参数敏感性分析[J].浙江农业学报,2012,(02):314-320.
李鹏强.火电厂地下水污染溶质运移数值模拟研究[D].太原理工大学,2012.
李琴,叶水根,高盼.基于GMS的北京市房山平原区地下水数值模拟研究[J].中国农村水利水电,2012,(05):1-5(10).
李守波.黑河下游地下水波动带地下水时空动态GIS辅助模拟研究[D].兰州:兰州大学,2006.
李文文,王开章,李晓.浅层地下水污染敏感性评价——以泰安市为例[J].安全与环境工程,2009,(04):18-22.
李艳平,李兰,朱灿,等.地表水地下水的交互作用与耦合模拟[J].长江科学院院报,2006,23(5):17-19.
梁杏,张人权,牛宏,靳孟贵,孙蓉琳.地下水流系统理论与研究方法的发展[J].地质科技情报,2012,(05):143-151.
刘春玲.吉林西部地下水开发风险评价[D].吉林大学,2007.
刘明新,陈钟祥.单重和双重孔隙介质二相二维渗流的一个新的数值模拟方法[J].力学学报,1984,(03):225-233.
刘派.基于HydroGeoSphere技术的吉林省西部地下水系统随机模拟研究[D].吉林大学,2011.
刘晓峰,陈通.投资项目因素敏感性分析方法的新探索[J].统计与决策,2008,(06):38-40.
龙玉桥. HydroGeoSphere在白石水库最小下泄流量研究中的应用[D].吉林大学,2008.
卢文喜,刘派,徐威,辛欣.基于HydroGeoSphere的地下水数值模拟及参数灵敏度分析[J].水电能源科学,2011,(06):64-67.
卢文喜.地下水系统的模拟预测和优化管理[M].北京:科学出版社,1999:20-29.
鲁程鹏,束龙仓,刘丽红,刘佩贵,董贵明.基于灵敏度分析的地下水数值模拟精度适应性评价[J].河海大学学报(自然科学版),2010,(01):26-30.
陆桂华,何海.全球水循环研究进展[J].水科学进展,2006,(03):419-424.
吕贤弼.地下水污染的数值模拟及应用实例[J].环境科学,1987,(04):28-33(48).
罗兰.我国地下水污染现状与防治对策研究[J].中国地质大学学报(社会科学版),2008,(02):72-75.
马喆.吉林西部低平原盐渍化水盐运移影响因素研究[D].吉林大学,2007.
梅一,吴吉春.基于FEMWATER的东北某地抽水蓄能电站渗流模型研究[J].水文地质工程地质,2008,(06):1-5.
齐培培.基于贝叶斯网络的水质污染评价及预测[D].武汉理工大学,2009.
钱会,王毅颖,宋秀玲.地下水流数值模拟中不应忽视的几个工作程序[J].勘察科学技术,2004,(01):40-43.
秦雅飞.吉林西部地下水模拟预报及生态效应探讨[D].吉林大学,2008.
任理.混合拉普拉斯变换有限单元法计算二维地下水溶质运移问题[J].水利学报,1997,(04):18-26.
任启伟,陈洋波,周浩澜,徐会军.基于Sobol法的TOPMODEL模型全局敏感性分析[J].人民长江,2010,(19):91-94(107).
施小清,姜蓓蕾. FEMWATER在地下水流数值模拟中运用[J].工程勘察,2008,(04):27-32.
沈娟,高强.吉林省水稻施肥现状的调查分析[J].吉林农业科学,2011,36(2):40-43(59).
束龙仓,王茂枚,刘瑞国,陈国浩.地下水数值模拟中的参数灵敏度分析[J].河海大学学报(自然科学版),2007,(05):491-495.
宋晓猛,孔凡哲,占车生,韩继伟.基于统计理论方法的水文模型参数敏感性分析[J].水科学进展,2012,(05):642-649.
孙志浩,王开章.济宁市湖东平原浅层地下水溶质运移数值模拟[J].人民黄河,2010,(04):83-84(86).
谭文清,孙春,胡婧敏,马力,赵彦宁. GMS在地下水污染质运移数值模拟预测中的应用[J].东北水利水电,2008,(05):54-55+59+72.
田玲玲.直接边界元法及其在地下水渗流问题中的应用[J].西华大学学报(自然科学版),2006,(01):44-45+49.
田雨,雷晓辉,蒋云钟,李薇.水文模型参数敏感性分析方法研究评述[J].水文,2010,(04):9-12(62).
汪馨竹.吉林省西部地下水埋深时空变化研究[D].吉林大学,2009.
王秉忱,杨天行,王宝金,等.地下水污染地下水水质模拟方法[M].北京:北京师范学院出版社,1985.
王大纯,张人权,史毅红,等.水文地质学基础[M].北京:地质出版社,1995.
王浩,陆垂裕,秦大庸,桑学锋,李扬,肖伟华.地下水数值计算与应用研究进展综述[J].地学前缘,2010,(06):1-12.
王旭升.自流井有限差分模拟的校正模型[J].地球科学(中国地质大学学报),2008,(01):112-116.
王占兴,宿青山,林绍志,张庆云,曹玉清.白城地区地下水及第四纪地质[M].北京:地质出版社,1985.
王喆,卢丽,夏日元.基于GMS的北京西郊垃圾场地下水溶质运移模拟[J].人民黄河,2012,(11):85-87(90).
王真真.基于敏感性分析的项目风险评估方法研究[D].湖南大学,2006.
肖长来.吉林省西部地下水资源评价与水资源可持续开发利用研究[D].吉林大学,2001.
谢树森,王军海.哈达山水利枢纽工程建设市场需求分析[J].吉林水利,2011,(01):60-63.
辛欣.吉林西部地下水的模拟预报及管理模型探讨[D].吉林大学,2008.
徐恒力,肖国强,李红.人为活动条件下河北平原第四系地下水系统的演变[J].地质科技情报,2002,(01):7-13.
薛禹群,谢春红.地下水数值模拟[M].北京:科学出版社,2007.
薛禹群.中国地下水数值模拟的现状与展望[J].高校地质学报,2010,(01):1-6.
杨春杰.吉林省粮食生产可持续发展研究[D].吉林大学,2010.
杨威.基于FEFLOW的吉林西部地下水数值模拟研究[D].吉林大学,2007.
姚华夏.全球尺度水文学:陆地表面过程模拟的进展[J].地球科学进展,1993,(02):62-65.
叶淑君,吴吉春,薛禹群.多尺度有限单元法求解非均质多孔介质中的三维地下水流问题[J].地球科学进展,2004,(03):437-442.
尹雅芳,刘德深,李晶,苗迎,吴旺发.中国地下水污染防治的研究进展[J].环境科学与管理,2011,(06):27-30.
袁峰明.土壤中的氮素淋失，氮素、产量、环境[M].北京:中国农业科技出版社,1996:196-216．
翟远征,王金生,苏小四,左锐.地下水数值模拟中的参数敏感性分析[J].人民黄河,2010,(12):99-101.
詹红丽.封闭圩区地下水溶质运移模拟研究[D].河海大学,2002.
张大龙,闫卫,于虎广,金喆.基于Visual Modflow的邯郸县地下水位动态研究[J].地下水,2012,(06):103-106.
张宏仁,李俊亭.有限差分法与有限单元法在渗流问题中的对比[J].水文地质工程地质,1979,(02):50-55.
张俊华,于福荣,李志萍. FEFLOW在黄龙工业园地下水数值模拟中的应用[J].人民黄河,2011,(06):69-70.
张力春.吉林省西部水资源可持续利用的优化配置研究[D].吉林大学,2006.
张荔.分布式水文模型构建及在渭河流域水环境解析中的应用研究[D].西安建筑科技大学,2010.
张民.哈达山水利枢纽工程大坝安全监测技术研究[D].吉林大学,2011.
张瑜芳,张蔚榛.垂向一维均质土壤水分运动的数值模拟在不同地下水位控制和蒸发条件下的运用[J].工程勘察,1984,(06):1-4.
张玉珍.九龙江上游五川流域农业非点源污染研究[D].厦门大学,2003,6:98-100．
张玉珍.农田不同土地利用氮素渗漏量的研究[J].福州大学学报(自然科学版),2006,(04):620-624.
赵海卿.吉林西部平原区地下水生态水位及水量调控研究[D].中国地质大学(北京),2012.
郑佳.白城市地下水数值模拟与评价[D].吉林大学,2006.
周彦章,周志芳,傅志敏.地源热泵系统地下水热量运移模拟参数敏感性分析[J].水动力学研究与进展A辑,2010,(02):262-269.
朱查松,罗震东,胡继元.基于生态敏感性分析的城市非建设用地划分研究[J].城市发展研究,2008,(04):30-35.
Ayse M, Ahmet Y. A fuzzylogic approach to assess groundwater pollution levels belowagricultural fields [J]. Environmental Monitoring and Assessment,2006,118:337–354.
Brebbia C A, Wrobel L C. The boundary element method: Computer methods in fluids [J].London: Pentech Press, Ltd.,1980:26-48.
Bredehoeft J D. Finite difference approximations to the equations of ground-water flow[J].Water Resources Research,1969,5(2):531-534.
Diersch H J G, Kolditz O. Variable-density flow and transport in porous media: approachesand challenges [J]. Advances in Water Resources,2002,25:899–944.
Freeze R A, Witherspoon P A. Theoretical analysis of regional ground water flow3:Quantitative interpretations [J]. Water Resour Res,1968,4:581–590.
Freeze R A, Witherspoon P A. Theoretical analysis of regional groundwater flow1: Analyticaland numerical solutions to the mathematical model [J]. Water Resour Res,1966,2:641–656.
Freeze R A, Witherspoon P A. Theoretical analysis of regional groundwater flow2: Effect ofwater table configuration and subsurface permeability variations [J]. Water Resour Res,1967,3(2):623–634.
Guiggiani M, Gigante A. A general algorithm for multidimensional cauchy principal valueintegrals in the boundary element method [J]. Journal of Applied Mechanics,1990,57(4):906-915.
Holvoet K, van Griensven A, Seuntjens P, et al. Sensitivity analysis for hydrology andpesticide supply towards the river in SWAT [J]. Physics and Chemistry of the Earth,2005,30(8):518-526.
Huyakorn P S, Pinder G F. Computational methods in subsurface flow [M]. Academic Press,Orlando, Florida, Inc.,1986.
Jury W A. Simulation of solute transport using a transfer function model [J]. Water ResourcesResearch,1982,18(2):363-368.
Kandeler E, Eder G, Sobotik M. Microbial biomass, N-mineralization, and the activities ofvarious enzymes in relation to nitrate leaching and root distribution in a slurry-amendedgrassland [J]. Biology and fertility of soils,1994,18(l):7-12.
Kloosterman F H, Stuurman R J, Van der Meijden R. Groundwater flow systems analysis on aregional and nation-wide scale in the netherlands: the use of flow systems analysis inwetland management [J]. Water Science and Technology,1995,31:375–378.
Konikow L F, Bredehoeft J D. Computer model of two-dimensional solute transport anddispersion in ground water [M]. US Government Printing Office,1978.
Morris M D. Factorial sampling plans for preliminary computational experiments [J].American Statistical Association and the American Society for Quality Control,1991,33:161-174.
Rogers C C M, Beven K J, Morris E M, et al. Sensitivity analysis, calibration and predictiveuncertainty of the institute of hydrology distributed model [J]. Journal of hydrology,1985,81(1):179-191.
Simms M. Dolomitization by groundwater—flow systems in carbonate platforms [J]. GulfCoast Assoc,1984:411–420.
Sobol I M. Sensitivity estimates for nonlinear mathematical model math [J]. ModelingComput Exp,1993(1):407-414.
Stallman R W. Numerical analysis of regional water levels todefine aquifer hydrology [J]. AmGeophys Union Trans,1956,37(4):451–460.
Stocke J T, Morris S L, Gioia I M, et al. The Einstein observatory extendedmedium-sensitivity survey [J]. The Astrophysical Journal Supplement Series,1991,76:813-874.
Sykes J F, Wilson J L, Andrews R W. Sensitivity analysis for steady state groundwater flowusing adjoin to perators [J]. Water Resources Research,1985,21(3):359-371.
Tóth J. A theoretical analysis of groundwater flow in small drainage basins [J]. Geophys Res J,1963,68:4785–4812.
Tóth J. A theory of groundwater motion in small drainage basins in central Alberta, Canada[J]. Geophys Res J,1962,67(11):4375–4387.
Tóth J. Groundwater as a geologic agent: an overview of the causes, processes, andmanifestations [J]. Hydrogeol J,1999,7:1–14.
Trescott P C. and Larson S P. Comparison of iterative methods of solving two-dimensionalgroundwater flow equations [J]. Water Resour Res,1977,13(1).
Tyson H N, Weber E M. Groundwater management for the nation’s future [J]. ComputerSimulations of Groundwater Basins,1964,90:59–77.
Van Griensven A, Meixner T, Grunwald S, et al. A global sensitivity analysis tool for theparameters of multi-variable catchment models [J]. Journal of Hydrology,2006,324(1):10-23.
Willis R, Finney B A. Planning model for optimal control of saltwater intrusion [J]. Journal ofWater Resources Planning and Management,1988,114(2):163-178.
Yeh G T. Computational subsurface hydrology: Fluid flows [D]. Kumwer Academic,Pennsylvania State University,1999.
Zheng C, Bennett G D. Applied contaminant transport modeling [M]. New York: John Wileyand Sons,1995.
鲍卫锋,黄介生,陈玲,等.地表水地下水联合运用模拟模型[J].中国农村水利水电,2003,6:45-47.
卞玉梅.吉林省西部地下水动态研究[D].吉林大学,2006.
陈崇希,李国敏.地下水溶质运移理论及模型[M].武汉:中国地质大学出版社,1996.
陈家军.边界元法在地下水补给量计算中的应用[J].长安大学学报(地球科学版),1989,(01):50-63.
陈钟祥.渗流力学的近况、动向和展望(续)[J].力学学报,1974,(02):94-103(55).
范伟.吉林省平原区地下水功能评价[D].吉林大学,2007.
冯波.吉林省西部地下水化学特征演化研究[D].吉林大学,2007.
冯启申,朱琰,李彦伟.地表水水质模型概述[J].安全与环境工程,2010,(02):1-4.
傅荟璇，赵红.MATLAB神经网络应用设计[M].北京：机械工业出版社，2010，19-38，193-207.
高慧琴,杨明明,黑亮,叶伟聪. MODFLOW和FEFLOW在国内地下水数值模拟中的应用[J].地下水,2012,(04):13-15.
宫辉力,李培,崔桂莲,魏学君,许斌.地下水系统准三维流场数值模拟与流场宏观调控[J].系统工程理论与实践,1999,(02):136-140.
韩磊,赵盼,闫金兵.农村河道水环境恶化成因及对策[J].天津农业科学,2010,(03):124-126.
韩曼.地下水封石油洞库渗流场及溶质运移模拟研究[D].中国海洋大学,2007.
贺国平,邵景力,崔亚莉,张德强. FEFLOW在地下水流模拟方面的应用[J].成都理工大学学报(自然科学版),2003,(04):356-361.
胡轶,谢水波,蒋明,陈泽昂,刘奇,杨晶. Visual Modflow及其在地下水模拟中的应用[J].南华大学学报(自然科学版),2006,(02):1-5.
黄清华,张万昌. SWAT模型参数敏感性分析及应用[J].干旱区地理,2010,(01):8-15.
江曙光.中国水污染现状及防治对策[J].现代农业科技,2010,(07):313-315.
江思珉,朱国荣,王浩然,孙斌堂. FAC方法在地下水数值模拟中的应用[J].水利学报,2006,(11):1389-1392.
吉林环境科学院.吉林省百亿斤商品粮能力建设总体规划环境影响报告书[R].长春：吉林环境科学院,2008.
蒋辉.河南沁阳市地下水水质的数值模拟[J].水文地质工程地质,1996,(02):14-15(20).
孔凡哲,宋晓猛,占车生,叶爱中.水文模型参数敏感性快速定量评估的RSMSobol方法[J].地理学报,2011,(09):1270-1280.
兰盈盈,肖长来,梁秀娟.吉林省西部主要生态环境地质问题及对策[J].吉林水利,2004,(04):1-4.
李存法,马自芬,刘秀婷.地下水非稳定渗流分析的显式有限单元法[J].地下水,2003,(03):150-151+171.
李军,梅琨,张明华.水质模型的全局参数敏感性分析[J].浙江农业学报,2012,(02):314-320.
李鹏强.火电厂地下水污染溶质运移数值模拟研究[D].太原理工大学,2012.
李琴,叶水根,高盼.基于GMS的北京市房山平原区地下水数值模拟研究[J].中国农村水利水电,2012,(05):1-5(10).
李守波.黑河下游地下水波动带地下水时空动态GIS辅助模拟研究[D].兰州:兰州大学,2006.
李文文,王开章,李晓.浅层地下水污染敏感性评价——以泰安市为例[J].安全与环境工程,2009,(04):18-22.
李艳平,李兰,朱灿,等.地表水地下水的交互作用与耦合模拟[J].长江科学院院报,2006,23(5):17-19.
梁杏,张人权,牛宏,靳孟贵,孙蓉琳.地下水流系统理论与研究方法的发展[J].地质科技情报,2012,(05):143-151.
刘春玲.吉林西部地下水开发风险评价[D].吉林大学,2007.
刘明新,陈钟祥.单重和双重孔隙介质二相二维渗流的一个新的数值模拟方法[J].力学学报,1984,(03):225-233.
刘派.基于HydroGeoSphere技术的吉林省西部地下水系统随机模拟研究[D].吉林大学,2011.
刘晓峰,陈通.投资项目因素敏感性分析方法的新探索[J].统计与决策,2008,(06):38-40.
龙玉桥. HydroGeoSphere在白石水库最小下泄流量研究中的应用[D].吉林大学,2008.
卢文喜,刘派,徐威,辛欣.基于HydroGeoSphere的地下水数值模拟及参数灵敏度分析[J].水电能源科学,2011,(06):64-67.
卢文喜.地下水系统的模拟预测和优化管理[M].北京:科学出版社,1999:20-29.
鲁程鹏,束龙仓,刘丽红,刘佩贵,董贵明.基于灵敏度分析的地下水数值模拟精度适应性评价[J].河海大学学报(自然科学版),2010,(01):26-30.
陆桂华,何海.全球水循环研究进展[J].水科学进展,2006,(03):419-424.
吕贤弼.地下水污染的数值模拟及应用实例[J].环境科学,1987,(04):28-33(48).
罗兰.我国地下水污染现状与防治对策研究[J].中国地质大学学报(社会科学版),2008,(02):72-75.
马喆.吉林西部低平原盐渍化水盐运移影响因素研究[D].吉林大学,2007.
梅一,吴吉春.基于FEMWATER的东北某地抽水蓄能电站渗流模型研究[J].水文地质工程地质,2008,(06):1-5.
齐培培.基于贝叶斯网络的水质污染评价及预测[D].武汉理工大学,2009.
钱会,王毅颖,宋秀玲.地下水流数值模拟中不应忽视的几个工作程序[J].勘察科学技术,2004,(01):40-43.
秦雅飞.吉林西部地下水模拟预报及生态效应探讨[D].吉林大学,2008.
任理.混合拉普拉斯变换有限单元法计算二维地下水溶质运移问题[J].水利学报,1997,(04):18-26.
任启伟,陈洋波,周浩澜,徐会军.基于Sobol法的TOPMODEL模型全局敏感性分析[J].人民长江,2010,(19):91-94(107).
施小清,姜蓓蕾. FEMWATER在地下水流数值模拟中运用[J].工程勘察,2008,(04):27-32.
沈娟,高强.吉林省水稻施肥现状的调查分析[J].吉林农业科学,2011,36(2):40-43(59).
束龙仓,王茂枚,刘瑞国,陈国浩.地下水数值模拟中的参数灵敏度分析[J].河海大学学报(自然科学版),2007,(05):491-495.
宋晓猛,孔凡哲,占车生,韩继伟.基于统计理论方法的水文模型参数敏感性分析[J].水科学进展,2012,(05):642-649.
孙志浩,王开章.济宁市湖东平原浅层地下水溶质运移数值模拟[J].人民黄河,2010,(04):83-84(86).
谭文清,孙春,胡婧敏,马力,赵彦宁. GMS在地下水污染质运移数值模拟预测中的应用[J].东北水利水电,2008,(05):54-55+59+72.
田玲玲.直接边界元法及其在地下水渗流问题中的应用[J].西华大学学报(自然科学版),2006,(01):44-45+49.
田雨,雷晓辉,蒋云钟,李薇.水文模型参数敏感性分析方法研究评述[J].水文,2010,(04):9-12(62).
汪馨竹.吉林省西部地下水埋深时空变化研究[D].吉林大学,2009.
王秉忱,杨天行,王宝金,等.地下水污染地下水水质模拟方法[M].北京:北京师范学院出版社,1985.
王大纯,张人权,史毅红,等.水文地质学基础[M].北京:地质出版社,1995.
王浩,陆垂裕,秦大庸,桑学锋,李扬,肖伟华.地下水数值计算与应用研究进展综述[J].地学前缘,2010,(06):1-12.
王旭升.自流井有限差分模拟的校正模型[J].地球科学(中国地质大学学报),2008,(01):112-116.
王占兴,宿青山,林绍志,张庆云,曹玉清.白城地区地下水及第四纪地质[M].北京:地质出版社,1985.
王喆,卢丽,夏日元.基于GMS的北京西郊垃圾场地下水溶质运移模拟[J].人民黄河,2012,(11):85-87(90).
王真真.基于敏感性分析的项目风险评估方法研究[D].湖南大学,2006.
肖长来.吉林省西部地下水资源评价与水资源可持续开发利用研究[D].吉林大学,2001.
谢树森,王军海.哈达山水利枢纽工程建设市场需求分析[J].吉林水利,2011,(01):60-63.
辛欣.吉林西部地下水的模拟预报及管理模型探讨[D].吉林大学,2008.
徐恒力,肖国强,李红.人为活动条件下河北平原第四系地下水系统的演变[J].地质科技情报,2002,(01):7-13.
薛禹群,谢春红.地下水数值模拟[M].北京:科学出版社,2007.
薛禹群.中国地下水数值模拟的现状与展望[J].高校地质学报,2010,(01):1-6.
杨春杰.吉林省粮食生产可持续发展研究[D].吉林大学,2010.
杨威.基于FEFLOW的吉林西部地下水数值模拟研究[D].吉林大学,2007.
姚华夏.全球尺度水文学:陆地表面过程模拟的进展[J].地球科学进展,1993,(02):62-65.
叶淑君,吴吉春,薛禹群.多尺度有限单元法求解非均质多孔介质中的三维地下水流问题[J].地球科学进展,2004,(03):437-442.
尹雅芳,刘德深,李晶,苗迎,吴旺发.中国地下水污染防治的研究进展[J].环境科学与管理,2011,(06):27-30.
袁峰明.土壤中的氮素淋失，氮素、产量、环境[M].北京:中国农业科技出版社,1996:196-216．
翟远征,王金生,苏小四,左锐.地下水数值模拟中的参数敏感性分析[J].人民黄河,2010,(12):99-101.
詹红丽.封闭圩区地下水溶质运移模拟研究[D].河海大学,2002.
张大龙,闫卫,于虎广,金喆.基于Visual Modflow的邯郸县地下水位动态研究[J].地下水,2012,(06):103-106.
张宏仁,李俊亭.有限差分法与有限单元法在渗流问题中的对比[J].水文地质工程地质,1979,(02):50-55.
张俊华,于福荣,李志萍. FEFLOW在黄龙工业园地下水数值模拟中的应用[J].人民黄河,2011,(06):69-70.
张力春.吉林省西部水资源可持续利用的优化配置研究[D].吉林大学,2006.
张荔.分布式水文模型构建及在渭河流域水环境解析中的应用研究[D].西安建筑科技大学,2010.
张民.哈达山水利枢纽工程大坝安全监测技术研究[D].吉林大学,2011.
张瑜芳,张蔚榛.垂向一维均质土壤水分运动的数值模拟在不同地下水位控制和蒸发条件下的运用[J].工程勘察,1984,(06):1-4.
张玉珍.九龙江上游五川流域农业非点源污染研究[D].厦门大学,2003,6:98-100．
张玉珍.农田不同土地利用氮素渗漏量的研究[J].福州大学学报(自然科学版),2006,(04):620-624.
赵海卿.吉林西部平原区地下水生态水位及水量调控研究[D].中国地质大学(北京),2012.
郑佳.白城市地下水数值模拟与评价[D].吉林大学,2006.
周彦章,周志芳,傅志敏.地源热泵系统地下水热量运移模拟参数敏感性分析[J].水动力学研究与进展A辑,2010,(02):262-269.
朱查松,罗震东,胡继元.基于生态敏感性分析的城市非建设用地划分研究[J].城市发展研究,2008,(04):30-35.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |