The collection of precipitation for chemical analysis

This paper is one of a series of publications on the collection and chemical analysis of precipitation. While the first paper presented the results from an intercalibration of 10 different designs of precipitation collectors (Galloway and Likens, 1976), this paper presents results of experiments on the collection efficiency of collectors, preservation of samples, monthly variation of wet and dry deposition, sample storage, length of sampling period, choice of collector location and sources of variability in precipitation data. The primary conclusions are that dry fallout should usually be excluded from rain and snow samples, sampling should be on an event basis, all biocides tested have some detrimental effect on sample composition, and that the major sources of variability in precipitation data are due to sampling errors and choice of collector location.