期刊名称：IOP Conference Series: Earth and Environmental Science
印刷版ISSN：1755-1307
电子版ISSN：1755-1315
出版年度：2019
卷号：237
期号：2
页码：022014
DOI：10.1088/1755-1315/237/2/022014
出版社：IOP Publishing
摘要：In the study of groundwater quantitative dating with tritium isotope, the establishment of tritium concentration model in precipitation is the key link, which provides the input function value for the calculation of groundwater tritium age. The Global Network of Isotopes in Precipitation (GNIP) jointly established by the international atomic energy agency (IAEA) and the world meteorological organization (WMO) provides a lot of data for the study of tritium isotopes in precipitation. Factor analysis is able to reduce variables with complex relations into a few comprehensive factors based on the internal dependence of variables. Therefore, factor analysis has obvious advantages in establishing tritium concentration model of precipitation based on a large number of tritium data in precipitation. In this paper, a tritium concentration model of precipitation in northeast Tarim basin had been established using factor analysis method, and calibrated with real data. The calculated results of model were in a good agreement with real data, showing that the model is reasonable. Meanwhile, considering the influence of nuclear test in the 20th century on the tritium concentration in the atmosphere of the study area, some previous research results were adopted, and the tritium concentration data in precipitation from 1960 to 2009 were finally recovered. In order to further verify the reasonability of the model, the recovered tritium concentration data in precipitation in northwest of Tarim basin were put to piston flow model and mixed flow model to date four groundwater samples in Kuketag, and the results showed that the tritium age of groundwater was consistent with the hydrogeological condition of study area. Therefore, it was considered that the tritium concentration model of precipitation in northeastern tarim basin was reasonable and reliable.
