Research On Spatio-temporal Changes Of The Groundwater And Monitoring Network Multi-objective Optimization

At present,a quarter of global land area is classified as arid or semi-arid land. Nearly half of the country in the world are directly affected by the drought in some way.Groundwater has become an important source of water in the arid or semi-arid region,sometimes only water source. Scientific monitoring and evaluation of groundwater resources and understanding of its quantity,quality and distributionof time and space has significance for the reasonable protection of groundwater resource and maintain of arid or semi-arid region's ecological environment in order to make groundwater resources serve for human and society sustainably.Groundwater level and quality is an important content of monitoring and evaluation of groundwater resources and the key factor to cause ecological problems such as soil desertification and salinization in the arid or semi-arid areas. Forecasting trends in groundwater level changes and research on spatial variability and distribution of groundwater level and quality is a hot one of groundwater reourses'evaluation.For cost reasons,many areas of groundwater monitoring network is multi-purpose,each of which monitors groundwater level and quality at the same time. It is significant to study multi-objective optimization modeling method of multi-purpose groundwater monitoring network in order to make monitoring network afford groundwater corresponding information required by groundwater resources evaluation.This paper used multi-objective evolutionary algorithm to solve multi-objective optimization model of groundwater monitoring network,which afforded new idea to solve the optimization problem of multi-purpose groundwater monitoring network.Shule River project area is located in northwest China's Gansu province,Hexi Corridor,which is the largest agricultural crop production base of Gansu province and typically arid and semi-arid areas. There are characteristics such as scarce rainfall,the temporal and spatial distribution inequality,strong evaporation,fragile ecological environment,the limited quantity of surface water. Agriculture,the people's living and ecological environment are significantly dependent on the groundwater. Based on the collection and processing of regional hydrological and geological data and the groundwater monitoring data,using the numerical simulation,geostatistics theory and multi-objective evolutionary algorithm,this paper studied the monitoring and evaluation of groundwater resources from the three aspects: groundwater numerical simulation and prediction,groundwater spatial variability and distribution and optimization modeling and sloving of groundwater monitoring network. The main work and research lies:(1) Using Yumen-Tashi basin,Anxi-Dunhuang basin and Huahai basin as study area,this paper used modflow model to simulate groundwater flow of three areas. Under the conditions of the planning,"Hexi Corridor (Shule) medium-term plan to adjust the project report",model calculated and generated 2007-2030 groundwater level and corresponding contour maps of three regions. Forecasted groundwater level change trend of three areas from 2007 to 2030. Based on study results,analysed environment affect caused by groundwater dynamical changes.(2) Based on ArcGIS geostatistical analyst module,this paper did exploratory spatial data analysis and spatial variation features analysis of groundwater depth average value between 2001 and 2006 and groundwater salinity average value in 2001 as well as 2004. Based on ordinary kriging interpolation method,contour map of depth and salinity was produced. Based on the contour map overlayed by the zoning map,the spatial distribution of groundwater level and the space and time change tendency of groundwater salinity changes tendency are analyzed and forecasted.Combined with predicted results,this paper evaluated ecological environment of the project area.(3) Based on geostatistical theory,this paper used the kriging variance minimum of groundwater level and salinity as the evaluation criteria of groundwater monitoring network multi-objective optimization and constructed multi-objective optimization model of groundwater monitoring network of Shule river project area.(4) Multi-objective optimization model of groundwater monitoring network was solved by multi-objective particle swarm algorithm.This paper studied particle updating formula and selection method of global and local extreme value and proposed the improved multi-objective particle swarm optimization algorithm. The application of groundwater network multi-objective optimization of Shule river project area proved the effectiveness of the method.