Insights into river bank erosion processes derived from analysis of negative erosion‐pin recordings: observations from three recent UK studies

Abstract Recent studies of river bank erosion in three catchments in the UK have been characterized by the persistent occurrence of negative erosion‐pin results. The cause of these negative recordings is considered with reference to field data from the Afon Trannon, Nant Tanllwyth and River Arrow, and to a laboratory study of freeze–thaw and desiccation processes. It seems that there is potential for, and in some cases evidence of, a number of different circumstances that generate negative results, but none of these alone is sufficient to explain all incidents. Factors considered include: deposition of sediment during high flows; soil fall from the upper parts of the bank on to lower erosion pins; loosening of the soil surface and expansion/contraction of the soil mass with fluctuations in temperature and moisture content; movement of the erosion‐pin within the bank and human interference. Each has its own implications for the use of erosion pins. Further issues arise when including negative data in subsequent data analysis, and it is demonstrated that attempts to correlate erosion rates with hydro‐meteorological data in order to ascertain causes of erosion will be influenced by the way in which negative data are handled. It is thus suggested that any study of river bank erosion using erosion pins should state whether or not negative data were obtained, and if so, how they were included in data analysis. Failure to include this information could lead to comparison of mean erosion rates that reflect bank processes very differently. The studies presented here offer a clear example of the value of ‘anomalous’ field data: results which do not appear to fit expected patterns can reveal as much about the processes in operation as those that do. Copyright © 2002 John Wiley & Sons, Ltd.