Abstract: Based on the extensive literature research of internationally adapted methodologies for the calculation of minor component background values, and after comparing and analyzing the relative strength of each method, a methodology combining hydrochemistry analysis and mathematical statistics was developed for the analysis of groundwater minor component apparent background values. Firstly, factor analysis was carried out on the minor and major components to establish their relationships. Next, the outliers of the major components were identified with the 3σ rule, and results were checked by using Piper graph, which was also used to determine macroscopically the obvious outliers of minor components related to hydrochemistry. Then the outliers of the minor components were identified by the iterative 2σ technique and probability graph method. Finally, the 95th percentiles for the outliereliminated data were taken as the apparent background values of the minor components. This methodology is useful in understanding the overall statistical characteristics, avoiding errors in threshold values due to subjectivity, and removing outliers efficiently. Comparing with the four frequentlyused mathematical statistics methods, this methodology, applied in the Liujiang Basin study, was shown to be more stable and reliable for calculating apparent background values of minor components, and able to explain the abnormal minor components and background constraints. The results show that the apparent background values of minor components in the Liujiang Basin shallow groundwater were 75.1, 27.4, 0.11, 0.30 and 0.32 mg/L for nitrate, metasilicic, acidaluminum, fluorine and bromine, respectively. The high nitrate value indicates that the extensive farming, husbandry and sewage discharge in the Liujiang Basin have inevitably influenced groundwater compositions in the area.

Key words: minor component, apparent background value, shallow groundwater, hydrochemistry, mathematical statistics, Liujiang Basin