【作者】 徐海珍； 李国敏； 张寿全； 董艳辉； 贺国平；

【Author】 XU Hai-zhen~1,LI Guo-min~1,ZHANG Shou-quan~(1,2),DONG Yan-hui~1,HE Guo-ping~3 1.Institute of Geology and Geophysics,Chinese Academy of Sciences,Beijing 100029,China 2.Beijing Municipal Commission of Development and Reform,Beijing 100031,China 3.Beijing Institute of Water planning and design,Beijing 100044,China

【机构】 中国科学院地质与地球物理研究所； 北京市发展和改革委员会； 北京市水利规划设计研究院；

【摘要】 为解决确定性模型可能为地下水源地划分带来的保护不足或过保护问题,探索采用基于正交拉丁超立方抽样法的地下水模型研究,并结合粒子反向追踪技术,进行地下水流随机模拟和对应保护区的概率统计分析,最终得到具有分布概率的保护区范围。为验证该方法的实用性与可靠性,选取北京市平谷区王都庄应急水源地进行实例应用研究。在考虑渗透系数和降雨补给量随机抽样组合基础上进行地下水流动随机模拟,并完成对应保护区划分和加权统计。与确定性模拟划分结果相对比:一级保护区中,确定性模拟划分结果仅相当于随机模拟结果中统计概率70%以上的范围;对于二级保护区,确定法对应保护面积相当于随机模拟结果中50%概率以上的范围。实例研究表明,基于正交拉丁超立方抽样的随机模拟法可完成多个不确定性输入变量条件下的随机模拟,且在小样本数的基础上获得理想结果,适用于复杂水文地质条件下的地下水源地保护区划分研究。更多还原

【Abstract】 Protection zone delineation plays an important role in the protection of groundwater source field’ and at this stage deterministic numerical simulation is the main delineation method in China.However’ the uncertainty that stems from the heterogeneity and variability of the hydrogeologic properties and supplying sources calls for a stochastic approach in which one or more properties are modeled as a random space function.Stochastic simulation based on the orthogonal Latin hypercube sampling OLHS method was applied to the delineation of well head protection area.OLHS is a stratified random sampling method with good uniformity and representativeness. In the work presented here’ the steps of protection area delineation with stochastic simulation were introduced first’ then the Wangduzhuang well field located in Pinggu District of Beijing was taken as an example’ and an OLHS approach was adopted in which conductivity(K) and recharge(R) were modeled as uncertain input variables.Many equally likely realizations of the random field were generated and used in deterministic numerical capture zone delineation.The ensemble of possible capture zones thus obtained is statistically treated to infer a capture zone probability distribution CAPD.By comparing the results with the deterministic model’ CAPD turned to be more reliable.The deterministic protection area is equivalent to the range above 70%of statistic probability.Stochastic simulation method based on OLHS is more suitable and accurate for protection area delineation of large and complex groundwater well field.更多还原