荣成市海水入侵监测及宏观态势预测模拟

发布时间：2018-06-13 18:58

本文选题：海水入侵 + 地球物理探测　；参考：《济南大学》2014年硕士论文

【摘要】：海水入侵问题自从被发现并得到科学界的关注以来已经过一个多世纪的广泛调查和深入研究。目前已经在海水入侵的定义、入侵的机理、调查方法、监测技术和数学模拟等方面取得显著的成绩。荣成市近年来由于过量开采地下水造成水位下降，导致海水向地下淡水体入侵，水质恶化，给当地的饮水安全和工农业生产带来了极大的危害。论文在分析荣成市多年海水入侵监测资料的基础上，对区域海水入侵情况开展调查，用水位监测和水文地球化学分析的技术掌握了海水入侵范围；应用基于MATLAB的BP神经网络进行了模型的训练和海水入侵预测。研究成果主要有：（1）根据水文地球化学检测结果，绘制了Piper三线图，分析了荣成市地下水由内陆、过渡区到海侵区，淡水、微咸水和咸水的地下水化学类型为HCO3-Ca、HCO3·C1-Ca·Mg和Cl·SO4-Mg·Na型水，矿化度逐渐增加，地下水水质变差。（2）对研究区的主要监测井进行了Auslog数据采集，通过对Gamma测线的解译，掌握了区域水文地质、含水层岩性等情况；对研究区两个主要海水入侵剖面的地层进行高密度电阻率仪监测，由解译图像分析了咸淡水界面线和含水层分布等情况。通过调查，荣成市海水入侵层位为潜水微承压含水层，含水层埋深主要在10~30m之间，该含水层以粉砂和中砂为主，含水层的透水性和联系性较好，为海水入侵提供了良好的通道。（3）绘制了荣成市海水入侵面积图及矿化度等值线图。据统计，2013年荣成市海水入侵面积达到81.10km2，2003年到2013年10年间增长了44km2。2011年海水入侵面积较上年激增94%，分析原因可能与该年滨海区地下水大量超采，造成海水沿地下水漏斗侵溯有关。2011年~2013年海水入侵面积增长缓慢，2012年甚至出现回退现象，分析与该年荣成市水资源总量较水平年有大幅提升有关。且当地有关部门实施用水总量控制，严格限制了地下水的开采也一定程度上遏制了海水的侵溯。（4）运用基于MATLAB编程的BP神经网络模型，通过对模型的初步训练，预测了荣成市海水入侵发展的宏观趋势，提出维持现有地下水开采量，加大滨海区含水层的回灌力度等是避免海水入侵面积进一步增长有效方式。论文的研究加强了胶东半岛最东端以基岩为主的海岸地带海水入侵情况的调查，，结合之前在莱州湾进行的广泛的海水入侵调查，使得胶东半岛海岸带的海水入侵分布图更加完整，为以后开展更为深入的海水入侵研究做了先期的准备。文章运用BP神经网络模型模拟了未来可能的海水入侵情况并给出潜在的影响因素，为海水入侵的控制及下一步治理措施的制定打下坚实的基础。
[Abstract]:The problem of seawater intrusion has been investigated and studied extensively for more than a century since it was discovered and paid attention to by the scientific community. At present, remarkable achievements have been made in the definition, mechanism, investigation method, monitoring technology and mathematical simulation of seawater intrusion. In recent years, the water level of Rongcheng City has dropped due to the overexploitation of groundwater, which leads to the invasion of seawater into underground fresh water body and the deterioration of water quality, which brings great harm to the local drinking water safety and industrial and agricultural production. On the basis of analyzing the monitoring data of seawater intrusion in Rongcheng city for many years, the paper investigates the situation of seawater intrusion in the region, and grasps the scope of seawater intrusion by using the techniques of water level monitoring and hydrogeochemical analysis. BP neural network based on MATLAB is applied to model training and seawater intrusion prediction. The main research results are as follows: 1) based on the hydrogeochemical test results, the Piper third line map is drawn, and the groundwater in Rongcheng city is analyzed from inland, transitional area to transgressive area, fresh water. The chemical types of groundwater in brackish water and brackish water are HCO3-CaHCO3C1-Ca mg and ClSO4-MgNa type water, and the mineralization degree increases gradually, and the groundwater quality becomes worse. The main monitoring wells in the study area are collected by Auslog data collection, and the Gamma line is interpreted. The regional hydrogeology and aquifer lithology are mastered, and the strata of two main seawater intrusion profiles in the study area are monitored by high density resistivity meter, and the distribution of salt and fresh water interface and aquifer are analyzed by interpretation image. According to the investigation, the intrusive horizon of seawater in Rongcheng city is the groundwater micro-confined aquifer, the depth of the aquifer is mainly between 10m and 30m, the aquifer is mainly silty sand and medium sand, and the aquifer has good permeability and connection. The area map of seawater intrusion and the contour map of salinity in Rongcheng City were drawn. According to statistics, the area of seawater intrusion reached 81.10km2 in Rongcheng City in 2013, and the area of seawater intrusion increased by 94 percent in the 10 years of 44km2.2011 from 2003 to 2013 compared with the previous year. The reason for the analysis may be due to the excessive exploitation of groundwater in the coastal area in that year. The area of seawater intrusion increased slowly from 2011 to 2013, and even retrogression occurred in 2012. The analysis is related to the increase of total water resources in Rongcheng City in that year compared with the level year. Moreover, the local departments concerned implement the total water use control, which strictly limits the exploitation of groundwater and to some extent restrains the invasion of seawater. 4) using the BP neural network model based on MATLAB programming, through the initial training of the model, The macroscopical trend of seawater intrusion in Rongcheng City is forecasted. It is suggested that the effective way to avoid further increase of seawater intrusion area is to maintain the existing groundwater exploitation amount and increase the recharge intensity of the aquifer in the coastal area. The research in this paper strengthens the investigation of seawater intrusion in the coastal zone of the easternmost part of Jiaodong Peninsula, which is dominated by bedrock, combined with the extensive investigation of seawater intrusion carried out in Laizhou Bay. It makes the distribution map of seawater intrusion in the coastal zone of Jiaodong Peninsula more complete, and makes a preliminary preparation for the further study of seawater intrusion. In this paper, BP neural network model is used to simulate the possible seawater intrusion in the future and the potential influencing factors are given, which will lay a solid foundation for the control of seawater intrusion and the formulation of the next control measures.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：P731.2

【参考文献】