Tracing suspended sediment sources in catchments and river systems
Introduction
Recent years have seen a growing awareness of the wider environmental significance of the suspended sediment loads transported by rivers and streams. This includes the importance of fine sediment in the transport of nutrients and contaminants, such as phosphorus (P), pesticides, PCBs, heavy metals and pathogens through fluvial systems (e.g., Shear and Watson, 1977, UNESCO, 1983, Allan, 1986, Warren et al., 2003). Table 1 serves to emphasise the important role of fine sediment in catchment P exports, by showing that sediment-associated transport can account for a major proportion of the P load in UK rivers. Recognition of the wide-ranging environmental significance of fine sediment has generated a need for improved information on the amounts of sediment involved (i.e., loads and concentrations) and on the changes in those amounts through time, consequent upon longer-term changes in land use and other facets of environmental change. However, it also important to obtain information on the main sources of the transported sediment, since sediment source can exert a key control on both the physical and geochemical properties of fine sediment, including its P content, which in turn exert a fundamental control over the magnitude of sediment-associated nutrient and contaminant fluxes.
The suspended sediment load transported by a river or stream will commonly represent a mixture of sediment derived from different locations and from different source types within the contributing catchment. Thus, for example, a relatively small area of the catchment, underlain by a particular rock type or supporting a particular land use, could contribute most of the suspended sediment load at the catchment outlet. Equally, in some catchments, sheet and rill erosion could dominate the sediment supply, whereas in others, channel erosion or gully erosion could represent the primary source. Information on sediment source is of fundamental importance in understanding the suspended sediment dynamics and the sediment budget of a catchment (e.g., Dietrich and Dunne, 1978, Trimble, 1983, Walling, 1988, Walling et al., 2001a, Walling et al., 2002).
Information on sediment source also represents a key requirement from the management perspective, since identification of sediment sources is a key precursor to the design of effective sediment management and control strategies. Whereas soil conservation programmes are primarily concerned with controlling on-site soil loss from agricultural land, sediment control programmes are more concerned with downstream problems and must consider a wider range of potential sources. Resources could be effectively wasted if, for example, control measures focussed on reducing surface erosion, when most of the sediment transported through a river system was contributed by channel and gully erosion. As indicated above, sediment source can exert a fundamental control on the nutrient and contaminant content of fine sediment, since the source of the sediment is likely to influence its physical and chemical properties and its contaminant loading and any management strategy aimed at controlling sediment-associated nutrient and contaminant fluxes would again need to take account of sediment source. In the case of P, for example, it may be important to consider the P content of the various sources and its bioavailability, so that the primary sources of P linked to specific impacts can be targeted. Thus, although the dominant sediment source within a catchment might be channel bank erosion, the P content of this sediment is likely to be substantially lower and less bioavailable than that mobilised from the surface of agricultural land. From a P management perspective, it could therefore prove more effective to control sediment inputs from agricultural land. The precise type of information on sediment source required will depend on the purpose in hand and the nature of the sediment-related problem. However, information on both the source type (e.g., sheet and rill erosion of areas under different land use, gully erosion, channel erosion or mass movements) and the spatial location of the sources (e.g., which tributary or part/parts of the basin) will frequently be required to address the above issues.
Although the need for information on suspended sediment source is clear, it has proved less easy to assemble such data (see Collins and Walling, 2004). Traditional methods for assessing the relative importance of individual source types employ an indirect approach and involve visual observations or measurements of erosional activity, which are in turn used to infer the relative importance of different potential sources. Thus, aerial photography could provide evidence of the incidence of channel and gully erosion and perhaps sheet and rill erosion (e.g., Eriksson et al., 2003), erosion pins could be employed to record the rate of surface lowering or retreat of features such as eroding river banks or gully walls (e.g., Haigh, 1977, Lawler et al., 1999, Stott, 1999) and erosion plots could be used to document rates of soil loss from surface sources (e.g., Soons, 1968, Loughran, 1990). However, this indirect approach faces many problems. Firstly, it will commonly require some a priori assumptions as to the likely sources, and in some environments, these may not be clear. Secondly, the use of erosion pins and erosion plots to provide information on the relative magnitude of erosion rates associated with different potential sources is difficult in anything but small drainage basins, due to the spatial variability of erosion and thus spatial sampling problems (c.f. Peart and Walling, 1988). Thirdly, and perhaps most importantly, the approach only provides information on sediment mobilisation and is unable to take account of the efficiency of sediment delivery to the stream system, for which the information on the source of transported sediment is ultimately required. As an alternative approach, some workers have attempted to infer sediment source contributions using models and prediction procedures. Thus, for example, the Universal Soil Loss Equation (USLE) could be used to estimate sheet and rill erosion from a small catchment and the difference between this estimate and the sediment yield could be attributed, at least in theory, to channel and gully erosion.
Attempts to obtain information on the spatial source of the sediment transported by a river commonly involves less uncertainty, since, theoretically, it would be possible to monitor the sediment load at a large number of points within a river network and therefore evaluate the relative importance of different tributaries or different parts of the catchment as sediment sources. However, this approach is commonly precluded by both practical and cost constraints. Furthermore, storage of sediment within the channel system could introduce problems in terms of relating the downstream flux to the fluxes from individual tributaries.
Faced with the many problems and constraints associated with the use of traditional approaches to obtaining information on sediment source and the growing need for such information, in the 1970s, a number of workers attempted to exploit the potential of an alternative direct approach to quantifying suspended sediment sources, based on source tracing or ‘fingerprinting’ (e.g., Klages and Hsieh, 1975, Wall and Wilding, 1976, Walling et al., 1979). 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 equivalent values for potential sources, in order to identify the likely source of that sediment. Early work successfully used geochemical (e.g., Wall and Wilding, 1976), mineralogical (e.g., Klages and Hsieh, 1975) and mineral magnetic (e.g., Walling et al., 1979, Oldfield et al., 1985) properties for source fingerprinting. However, the scope of these studies was generally limited in terms of providing, firstly, only a qualitative indication of the likely importance of particular sources and, secondly, only broad discrimination between a small number of potential sources, typically surface and subsurface/channel bank materials.
Section snippets
The development of sediment source tracing or fingerprinting procedures
Subsequent development of the source tracing or fingerprinting approach directed attention to a number of important methodological aspects, with a view on refining the approach and improving the reliability of the results obtained. The first area of development involved the search for fingerprint properties that were capable of clearly discriminating several potential sources. Geochemical, mineralogic and mineral magnetic properties of soils and sediments continued to be used, but sediment
Sediment sources in the Upper Torridge catchment
The Upper Torridge catchment in Devon, UK, drains an area of 258 km2 above the Environment Agency flow gauging station at Rockhay Bridge. With its moderate relief (maximum altitude 220 m), relatively high annual precipitation and runoff (ca. 1250 mm and 900 mm, respectively) and heavy soils, the land use of the catchment is dominated by pasture (ca. 80%) with arable land and woodland accounting for ca. 16% and 4%, respectively. Livestock grazing (cattle and sheep) constitutes the main farming
The prospect
The case studies and results presented above clearly demonstrate the potential for using source tracing or fingerprinting techniques to obtain information on suspended sediment sources within river basins, both in the UK and more generally. However, there remains a need to refine and further develop several aspects of the approach and scope clearly exists to extend the application of the source fingerprinting approach to embrace other aspects of the catchment sediment system. Both aspects are
Acknowledgements
The findings presented in this paper draw heavily on work undertaken in collaboration with several co-workers, including Adrian Collins, Qingping He, Dan Nicholls, Phil Owens, Mark Russell and Jamie Woodward; their contributions and collaboration and the support for the work provided by the Natural Environment Research Council, MAFF/DEFRA and the Environment Agency are gratefully acknowledged.
References (88)
- et al.
Fingerprinting suspended sediment sources in a large urban river system
Sci Total Environ
(2003) - et al.
Particulate phosphorus transport by sub-surface drainage from agricultural land in the UK. Environmental significance at the catchment and national scale
Sci Total Environ
(2001) - et al.
Selecting fingerprint properties for discriminating potential suspended sediment sources in river basins
J Hydrol
(2002) - et al.
A strontium isotopic investigation of the origin of suspended particulate matter (SPM) in the Murray–Darling river system, Australia
Chem Geol
(1995) - et al.
Source identification for suspended sediments
J Hydrol
(1980) - et al.
Interpreting particle size effects in the adsorption of 137Cs and unsupported 210Pb by mineral soils and sediments
J Environ Radioact
(1996) - et al.
The use of floodplain sediment cores to investigate recent historical changes in overbank sedimentation rates and sediment sources in the catchment of the River Ouse, Yorkshire, UK
Catena
(1999) - et al.
Suspended sediment sources in two small lowland agricultural catchments in the UK
J Hydrol
(2001) - et al.
Sediment source modelling: unmixing of artificial magnetisation and natural radioactivity measurements
Earth Planet Sci Lett
(1994) - et al.
Use of mineral magnetic measurements to fingerprint suspended sediment sources: approaches and techniques for data analysis
J Hydrol
(1997)