Study On Groundwater Vulnerability Of Sanjiang Plain

The research of groundwater vulnerability is the basic work to protect the groundwater. Forutilizing groundwater resource continually, groundwater vulnerability evaluation is necessary.Useful reference to protect, exploit and utilize on groundwater resource are provided rationally.Based on study profoundly the theories and methods that researched on groundwatervulnerability in the nation and abroad, the groundwater vulnerability of Sanjiang Plain was studied.The study combined with the real circumstance of the area.1. The study succeeded in several facets as follows:(1) The evaluation indexes system of groundwater vulnerability in Sanjiang Plain wasestablished.Under guidance of the establishment principle, method and process, the evaluation indexessystem of groundwater vulnerability in Sanjiang Plain was established. The establishing processbased on the traditional evaluation indexes system for DRASTIC model, and combined closelywith the practical condition of the plain. By the relativity analyzing, the indexes system isreasonably. It had certain reference value for that the groundwater vulnerability researches furtherdevelop.(2) The new methods that evaluate groundwater vulnerability were proposed.Mathematics model was embedded in parametric system methods so the deficiencies in thesingular methods were overcome. The objectivity and scientificity of evaluation result were raised.The methods applied in the thesis were as follow: the entropy weight coefficient model, theprojection pursuit classification model based on real coding based accelerating genetic algorithmand the multipurpose fuzziness mode identify model. The first and the second methods had notbeen applied previously in groundwater vulnerability evaluation. The groundwater vulnerability ofSanjiang Plain was evaluated by the three methods and the vulnerability degree of various subzonewere obtained. The three methods were compared comprehensively, at the same time, theiradvantages and disadvantages were analyzed. The evaluation result can be referenced fordepartment concerned.(3) Agricultural water consumption and groundwater level were predicted in Sanjiang Plain. The groundwater level was an important index in the groundwater vulnerability evaluation.The agriculture water consumption of Sanjiang Plain was predicted by two methods. One wasagriculture water consumption method; the other was groundwater exploitable modulus. The resultexpressed that local water resource cannot be still satisfy with the demand on the whole. The lackpart was supplemented basically by exploiting groundwater, which resulted in groundwater leveldescending continually and groundwater funnel. Equal-dimension gray filling BP neural networksmodel based on real coding based accelerating genetic algorithm was established. By the newmodel, we predicted the groundwater level of Chuangye farm in the Sanjiang Plain. The predictionresult expressed that in five years the exploited amount of groundwater resource will have been stillincreased year by year if the planting area of well irrigation paddy rice has been enlarged as present.It will lead to the groundwater level decline continually. The groundwater level of the area willdecline 0.3m every year on the average from 2007 to 2012. Therefore, the department concernedshould take measures availably to control the amount of exploiting groundwater resource. Themeasures are as following: First and foremost, we should arrange the motor-pumped wellreasonably; In the second place, in order to control the exploited amount of groundwater, we shouldtake full advantage of the water resource on the earth's surface and utilize the new water-saveirrigation measures; Last but not the least, we should reduce properly the planting area of wellirrigation paddy rice. Only if we did these, could the groundwater dynamic balanced recover stepby step and the water resource be utilized continually.2. The innovation:1) The evaluation indexes system of groundwater vulnerability in Sanjiang Plain wasestablished. The establishing process based on the traditional evaluation indexes system forDRASTIC model, and combined closely with the practical condition of the plain.2) The new methods that evaluate groundwater vulnerability were proposed. The methodswere entropy weight coefficient model and the projection pursuit classification model based on realcoding based accelerating genetic algorithm.3) When the groundwater level of Sanjiang Plain was predicted by Grey BP Neural NetworksModel Based on RAGA, there were certain innovations in the combine way of the two methods. Ithad been brought into effect based on that we drawed lessons from other's research fruits andconsidered the advantages and disadvantages of the two methods and its applicable domain.