Probability of Detecting Perchlorate under Natural Conditions in Deep Groundwater in California and the Southwestern United States

详细信息    查看全文

• 作者：Miranda S. Fram ; Kenneth Belitz
• 刊名：Environmental Science & Technology
• 出版年：2011
• 出版时间：February 15, 2011
• 年：2011
• 卷：45
• 期：4
• 页码：1271-1277
• 全文大小：892K
• 年卷期：v.45,no.4(February 15, 2011)
• ISSN：1520-5851

文摘

We use data from 1626 groundwater samples collected in California, primarily from public drinking water supply wells, to investigate the distribution of perchlorate in deep groundwater under natural conditions. The wells were sampled for the California Groundwater Ambient Monitoring and Assessment Priority Basin Project. We develop a logistic regression model for predicting probabilities of detecting perchlorate at concentrations greater than multiple threshold concentrations as a function of climate (represented by an aridity index) and potential anthropogenic contributions of perchlorate (quantified as an anthropogenic score, AS). AS is a composite categorical variable including terms for nitrate, pesticides, and volatile organic compounds. Incorporating water-quality parameters in AS permits identification of perturbation of natural occurrence patterns by flushing of natural perchlorate salts from unsaturated zones by irrigation recharge as well as addition of perchlorate from industrial and agricultural sources. The data and model results indicate low concentrations (0.1鈭?.5 渭g/L) of perchlorate occur under natural conditions in groundwater across a wide range of climates, beyond the arid to semiarid climates in which they mostly have been previously reported. The probability of detecting perchlorate at concentrations greater than 0.1 渭g/L under natural conditions ranges from 50鈭?0% in semiarid to arid regions of California and the Southwestern United States to 5鈭?5% in the wettest regions sampled (the Northern California coast). The probability of concentrations above 1 渭g/L under natural conditions is low (generally <3%).