Statistical Analysis for Extracting Features on the Groundwater Level Dynamics
This work presents statistical and geostatistical approaches to describe different features of the groundwater level dynamics. Principle component analysis was used to extract the most important to describe groundwater level components. These components appeared to present different features in correlation with the precipitation data, associated with different physical mechanisms.Estimation and modeling cross-correlation structures between principle components and precipitation allows to improve a geostatistical spatio-temporal groundwater level model.Independent components (sources) provide the other view on the principle components and the various description of the groundwater level dynamics. The independent components identify locations of possible sources causing all groundwater level dynamics. In the current work independent components were estimated basing on the Central limit theorem with a kurtosis as a measure of normal (Gaussian) distribution. Analysis of matrix of transformation to independent sources allows to describe connections between wells and nearby surface reservoirs.The current work is illustrated by analysis of the real data connected with radioactive storing facilities.
Anton Nuzhny Elena Savelieva Alexey Jastrebkov
Nuclear Safety Institute RAS, 52, B. Tulskaya, 113191 Moscow, Russia
国际会议
The 12th Conference of the International Association for Mathematical Geology(第12届国际数学地质大会)
北京
英文
723-726
2007-08-26（万方平台首次上网日期，不代表论文的发表时间）