Erosion and Sediment Transport Monitoring Programmes in River Basins (Proceedings of the Oslo Symposium, August 1992). IAHS Publ. no. 210, 1992. 153 Use of radiometric fingerprints to derive information on suspended sediment sources D.E. WALLING & J.C. WOODWARD Department of Geography, University of Exeter, Exeter, EX4 4RJ, UK ABSTRACT There is an increasing need for information on suspended sediment sources which is difficult to meet using traditional monitoring techniques. The fingerprinting technique offers an alternative approach, but it is important to select appropriate fingerprinting properties. Fallout radionuclide concentrations appear to offer considerable potential as fingerprinting properties, since they arc essentially independent of lithology and soil type and can clearly distinguish surficial and channel sources. Examples of the use of 137 Cs, unsupported 210 Pb and 7 Bc concentrations to decipher suspended sediment sources in two small drainage basins in Devon, UK are presented. INTRODUCTION Recent concern for the role of suspended sediment in non-point pollution from land use activities and in the transport of nutrients and contaminants has highlighted the need for sediment monitoring programmes to provide information on sediment sources as well as concentrations and loads. This need has also been further promoted by the desire to establish sediment budgets for drainage basins, by attempts to develop distributed sediment yield models and by attempts to develop more meaningful gcomorphological interpretations of sediment yield data in terms of landscape evolution. Although in some circumstances it is important to ascertain the precise spatial location of sediment sources within a drainage basin, more often the requirement is for information on the type of source involved and, for example, whether the sediment has originated primarily from erosion of cultivated land, pasture or forest, or from channel or gully erosion, irrespective of the precise spatial location of the source. Detailed information concerning the types and relative importance of sediment sources within a drainage basin is, however, difficult to assemble for all but very small basins using traditional monitoring techniques (cf. Peart & Walling, 1986). The "fingerprinting" technique offers an alternative approach to determining sediment source which would appear to offer very considerable potential (cf. Wall & Wilding, 1976; Oldficld et ai, 1979; Peart & Walling, 1986,1988). In essence, this method involves firstly, the selection of a physical or chemical property which clearly differentiates potential source materials and, secondly, comparison of measurements of the same property obtained from suspended sediment with the equivalent values for the potential sources. In the case of a simple distinction between surficial and channel sources, the property could be one that differentiated topsoil from the underlying parent material and bed rock. The essential simplicity of the fingerprinting technique is, however, complicated by a number of potential problems, including enrichment of the sediment relative to the source material and transformation of sediment properties within the fluvial system (cf. Peart, 1990). Furthermore, it is important to identify physical or chemical properties which vary over a substantial range and are therefore able to distinguish different sources in an unequivocal manner. Peart & Walling (1988) advocated the use of several alternative diagnostic properties, rather than a single indicator, in order to establish the consistency of the results obtained. The quest for diagnostic properties capable of distinguishing a range of source types has embraced a large number of sediment properties, including sediment colour (Grimshaw & Lewin, 1980), clay mineralogy (Wall & Wilding, 1976), sediment chemistry (Peart & Walling,