Lead in synthetic and municipal drinking water varies by field versus laboratory analysis

Water lead measurements by two field analyzers, relying on anodic stripping voltammetry (ASV) and fluorescence spectroscopy, were compared to reference laboratory measurements by inductively coupled plasma mass spectrometry (ICP-MS) in progressively complex datasets (phases A, B, C), to assess field analyzer performance. Under controlled laboratory quantitative tests of dissolved lead within the field analysis range and optimal temperature range, lead recoveries by ASV ranged within 85-106 % of reference laboratory values (corresponding linear model: y = 0.96x, r² = 0.99), compared to lower lead recoveries of 60-80 % by fluorescence (y = 0.69x, r² = 0.99) in phase A. Field analyzer performance deteriorated in three opportunistic laboratory datasets compiled for phase B that contained dissolved lead (ASV: y = 0.80x, r² = 0.98; no fluorescence data). Further lead underestimations were observed in five field datasets compiled for phase C, some of which contained known particulate lead (ASV: y = 0.54x, r² = 0.76; fluorescence: y = 0.06x, r² = 0.38). Deteriorating performance between phases was presumably due to the increasingly complex water matrices and lead particulates present in some phase C subsets (phase A < phase B < phase C). Phase C field samples had lead concentrations that were out-of-range, including a 5 % and 31 % false negative rate by ASV and by fluorescence, respectively. The range of results relevant to the diverse nature of compiled datasets, suggests that unless ideal conditions are known to be present (i.e., the lead content of water is dissolved within the field analysis range and optimal water temperature range), these field lead analyses might only be used as a water screening tool. Given the unknown conditions in many field settings, combined with the lead concentration underestimations including the false negative rates reported herein for field datasets, caution is encouraged when employing ASV and particularly fluorescence field analysis.