Evaluation of whole-water churn splitters for suspended-sediment sample collection and analysis

The U.S. Geological Survey (USGS) collects a wide range of whole-water samples to test for the many physical and chemical constituents that represent stream conditions at the time of sampling to assess the quality of the Nation’s waters. During sampling efforts, in which a suspended-sediment concentration is one result among a broader suite of constituents, a sample is sometimes composited into a churn splitter and then subdivided for analysis. Five churn splitters—comprising three sizes and two different materials—used by the USGS were tested for single-withdrawal accuracy from one-half of full capacity and for multiple-withdrawal accuracy at varied volumes of fullness to see if churn splitters introduce bias during the collection of sediment samples. Both tests were similar to previously conducted tests for consistency, but the tests conducted in this report also attempted to answer questions that arose during previous evaluations of churn splitters. The purpose of this report is to inform sediment analysts about the capabilities and limitations of all available churn-splitter sizes and materials used by the USGS for the analysis of suspended sediment.

The results indicate that suspended-sediment samples and constituents absorbed into suspended sediment may have substantial bias errors when withdrawn from churn splitters. Results were affected by the settling velocity of sediment particles relative to the resuspension velocities induced by the churning, the effects of prior withdrawals of the water-sediment mixture, and the remaining volume in a churn splitter after samples were withdrawn.