关键词: 地下水')">

地下水, 背景值, 预筛选法, 数理统计法, 基于图谱的探索性数据分析,
')">组合方法

Abstract: Selecting an appropriate method for assessing natural background levels (NBL) in groundwater is crucial for objectively understanding the NBL in groundwater. On the basis of reviewing the development process of NBL in groundwater, this article focused on summarizing the current methods for the assessment of NBL in groundwater with their advantages and disadvantages, and pointed out the development trend of the assessment for NBL in groundwater in the future. The methods for assessing NBL in groundwater can be roughly divided into five categories: Methods based on uncontaminated groundwater samples, pre-selection methods, statistics methods, exploratory data analysis methods based on graphs, and the combination of multi-methods. The method based on uncontaminated groundwater samples has been rarely adopted due to its inherent limitations. Pre-selection and statistics method are two commonly used independent methods in the study of groundwater NBL. Generally, the former one is more subjective than the latter one. Nowadays, the combination of multiple methods becomes a development trend in the research of groundwater NBL assessment, because some limitations of one method for assessing groundwater NBL may be avoided by the combination with another method. Among them, the combination of pre-selection and statistics methods is the most common way of combination. In addition, the combination of emerging exploratory data analysis methods based on graphs with pre-selection methods and/or statistics methods is commonly better than the combination of pre-selection and statistics methods. However, the use of the former combination often requires a good understanding of the hydrogeochemical characteristics in study areas. Therefore, the former ones for assessing groundwater NBL are less convenience and universality in comparison with the latter ones.

Key words: groundwater')">

groundwater, natural background levels, pre-selection methods, statistics methods, exploratory data analysis based on graphs, combination of methods