Introduction to the special issue ‘Tracer Applications in Sediment Research’

Abstract

Hydrological ProcessesVolume 27, Issue 6 p. 775-780 PrefaceFree Access Introduction to the special issue ‘Tracer Applications in Sediment Research’ First published: 04 January 2013 https://doi.org/10.1002/hyp.9701Citations: 9AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Recent years have witnessed the increasing use of tracer techniques in sediment-related investigations in fluvial systems (Foster, 2002), and such techniques are now being successfully exploited in both small catchments and larger river basins in many areas of the world. The use of tracers in sediment research has a relatively long history, but important developments occurred in the middle years of the 20th century. For example, the Vigil Network programme, which can in many respects be seen as the forerunner of contemporary catchment sediment monitoring programmes, aimed at documenting the erosional impact of fluvial processes, promoted the use of painted stones to provide information on bed load transport thresholds and transport distances almost 50 years ago (e.g. Leopold et al., 1966; Schick, 1971). The use of painted stones has now been largely superseded by more sophisticated technology for tracing the movement of individual clasts, including the implanting of radioactive material, magnets and radio transmitters and the use of luminescent particles (Ergenzinger and de Jong, 2003; Habersack, 2003). Such technology is able to provide additional information on step lengths and rest periods. Radioactive tracers and more particularly the seeding of sediment with radioactive particles were used in a similar way to study sand transport in rivers (Crickmore and Lean, 1962). The addition of radioactive (e.g. Fe-59 and Co-60) and fluorescent materials to the soil surface was also successfully used to study patterns of erosion on slopes (e.g. Toth and Alderfer, 1960; Wooldridge, 1965; Young and Holt, 1968), although in most cases such work was undertaken on experimental plots, often in the laboratory, rather than in the wider landscape. More recently, rare earth elements (REEs), which are readily measured using ICPMS, have been used in soil erosion investigations to overcome the restrictions on the use of radioactive tracers (e.g. Zhang et al., 2001). A key feature of the early tracing work briefly reviewed above was the emphasis on the principle of what Black et al. (2007) have referred to as ‘particle tracking’. In simple terms, particles with similar properties are added to the material in transport, and, by tracking the movement of these particles, it is possible to derive information on transport processes and fluxes more generally. Although this approach to sediment tracing has proved successful in a range of contexts and is able to provide valuable information on sediment mobilisation and transport processes, it also has a number of potential disadvantages. These primarily relate to the need to add particles to the natural system, so that the fate of these particles can be traced. This can prove costly where the tracer material is expensive and may severely restrict the spatial scale at which the approach can be applied. Most studies involving REEs have, for example, been restricted to plots or individual fields and, when considering the mobilisation and transport of sediment on slopes, it is difficult, if not impossible, to apply the particle tracking approach at the landscape scale. The need to ensure that the particles added to the system have similar properties to the ambient material and behave in a near- identical manner, as well as providing a distinctive and readily identifiable non-hazardous tracer, can also introduce problems. In addition, it will frequently be necessary to continue to add tracer particles to the system over the study period, and this can limit the timescale of the study and thus its representativeness. More recent work on the application of tracers in sediment research within fluvial systems has been characterized by three important features, which distinguish it from earlier approaches. The first is the increasing emphasis on fine, rather than coarse, sediment, the second is the directing of attention to the entire sediment cascade or catchment sediment budget and the overall landscape, rather than just sediment transport in the river channel or sediment mobilisation from the slopes, and the third is the use of ‘natural’ or pre-existing tracers, which avoid the need to add tracers to the system and thereby overcome many of the potential disadvantages of the ‘particle tracking’ approach referred to above. The increasing emphasis on fine sediment in more recent sediment tracing investigations can be seen as the response to a growing awareness of the importance of fine sediment in degrading aquatic habitats (e.g. Wood and Armitage, 1997; Prosser et al., 2001; McKergow et al., 2005; Kemp et al., 2011; Collins et al., 2011; Chalov, 2011) and in the transport and fate of nutrients and contaminants within both terrestrial and aquatic ecosystems and receiving coastal waters (e.g. Stone, 2000; Warren et al., 2003; Gordeev et al., 2004). Relatively small changes in fine sediment fluxes, caused, for example, by land use change and other human impacts, can have serious implications downstream. In order to reduce or prevent associated environmental problems and to preserve the ecological status of aquatic ecosystems, it is necessary to develop an improved understanding of the mobilisation, transfer and fate of fine sediment within catchments, to support the development and implementation of effective fine sediment control and management strategies. In many, if not most, environments, coarse sediment has more limited ecological significance than fine sediment, and its behaviour is arguably easier to quantify and model numerically. Sediment tracing techniques can provide valuable information on the dynamics of fine sediment mobilisation and transfer in fluvial systems. In this context, it is important to recognise that sediment-related environmental problems may extend well beyond areas with high erosion rates and high sediment fluxes. It is frequently catchments with low sediment loads and low sediment concentrations that are the most sensitive to small changes in fine sediment flux. In catchments with low erosion rates and sediment fluxes, sediment sources, transfer pathways and sinks can be difficult to assess visually, meaning that sediment tracing techniques can prove particularly valuable for understanding the system. As indicated above, much recent work involving sediment tracing has been driven by a need to assemble the information required characterize the sediment budget of a catchment, which is commonly the starting point for developing sediment control and management strategies (e.g. Wilkinson et al., 2005; Walling and Collins, 2008; Gellis and Walling, 2011). Sediment tracing techniques can provide an essentially unique means of assembling the required information on sediment sources and sediment sinks and spatially distributed sediment redistribution rates (Walling, 2006). As such, attention is frequently directed to the entire catchment and the movement of sediment from its various sources to the catchment outlet and the storage of sediment in intervening sinks. This catchment-wide perspective distinguishes recent work from the more traditional approaches which frequently focussed on individual channel reaches or slopes and seldom directed attention to sinks and stores. Recent work aimed at using tracers to quantify rates of overbank sedimentation on river floodplains has, for example, demonstrated the potential importance of river floodplains as sediment sinks and therefore in controlling the magnitude of the sediment output from a river basin (e.g. Walling et al., 1999; Hughes et al., 2009). The need to document catchment sediment budgets and the associated move to a catchment-wide perspective can be seen as being closely linked to a shift to the use of ‘natural’ or pre-existing tracers, to support investigations at the larger landscape or catchment scale. As the area of interest increases in size, it becomes effectively impossible to seed large areas with tracer material and to track the movement of the tracer through the system. Attention therefore turns to natural properties that can be used to trace the movement of sediment, termed here natural tracers. Perhaps, the most obvious application of natural tracers is the use of the geochemical properties of fine sediment to trace its source within a catchment using source fingerprinting techniques (Olley and Caitcheon, 2000; Walling, 2005). In this approach, the properties of the sediment are compared with those of potential sources and the relative contributions of those sources are established using a mixing or unmixing model. The potential sources could be defined in terms of either spatial location within the catchment or source ‘type’ (e.g. sheet and rill erosion, gully erosion, channel erosion, mass movements etc). A wide range of geochemical properties, including major and trace element chemistry, mineralogy, radionuclides, mineral magnetic properties, the isotopic composition of the organic fraction and sediment colour, have been successfully used as source fingerprints (Mukundan et al., 2012). Although in some cases they should not be seen as truly ‘natural’, fallout radionuclides, and particularly caesium-137 (137Cs), excess lead-210 (210Pbex) and beryllium-7 (7Be), have also been widely exploited as tracers in recent years. In this case, the radionuclide reaches the land surface as fallout and is firmly and rapidly fixed by the surface soil. If it can be assumed that the initial fallout input was spatially uniform, information on the post fallout redistribution of the radionuclide, as reflected by spatial variability of the radionuclide inventory, can be used to derive estimates of soil erosion and deposition rates. The different half-lives of different fallout radionuclides afford a basis for estimating erosion and deposition rates for different time windows (Mabit et al., 2008). In the case of 137Cs, this time window extends back to the main period of bomb fallout in the late 1950s and 1960s, and in the case of Chernobyl fallout to 1986 (Golosov et al., 2010). For 210Pbex the window is longer (ca. 60–100 years), whereas for 7Be, it covers only a few weeks. The activity of fallout radionuclides in surface soil or sediment has also frequently been shown to provide a good source fingerprint, since, whereas the catchment surface will be exposed to fallout, channel banks and subsurface material are likely to receive little or no fallout and will be characterized by low or zero activity. Furthermore, the surface of cultivated and non-cultivated land can generally be discriminated by virtue of the mixing of the radionuclide into the plough layer of cultivated soils by tillage and associated reduction of the surface activity of cultivated soil (Walling and Woodward, 1992; Wallbrink and Murray, 1993; Wallbrink et al., 1998, 1999). Although 210Pbex and 7Be are of natural geogenic and cosmogenic origin, respectively, 137Cs is man-made and is a product of past bomb testing in the 1950s and 1960s or accidental large-scale releases from nuclear reactors, such as that at Chernobyl in 1986 and the more recent releases from the Fukushima nuclear power station in Japan in 2011. However, the basic principle underlying the use of these three fallout radionuclides is essentially the same. They can be seen as tracers applied to the land surface ‘free of charge’ as a result of ‘natural’ fallout processes. These fallout radionuclides are often also used as chronometers to provide information on the chronology of recent sediment deposits, by virtue of the known temporal pattern of 137Cs fallout or the natural decay of 210Pbex and 7Be through time (e.g. Stokes and Walling, 2003; Belyaev et al., 2011). The ratio of 210Pb to 137Cs activity has also been used to provide an estimate of the residence time of sediment within a river system (Wallbrink et al., 2002). As such, fallout radionuclides provide particularly valuable and essentially ‘general purpose’ tracers for investigating several different facets of catchment sediment dynamics, including sediment source, erosion and deposition rates and sediment accumulation on river floodplains and in other sinks. As indicated above, the current status of tracer studies in sediment investigations can be seen as reflecting a number of important drivers. The key role of fine sediment in degrading aquatic habitats and thereby influencing ecological status has focussed attention on fine sediment. The need to develop an improved understanding of the mobilisation, transport and fate of fine sediment has directed attention to the catchment sediment budget as a key management tool, and sediment tracing techniques have provided a means of obtaining the information necessary to characterize such budgets. The scientific community has responded to these drivers by developing a wide range of tracer techniques which are capable of being applied at the catchment scale. For the most part, these techniques are founded on the use of ‘natural’ tracers, and the development and refinement of the use of such tracers can be seen both as representing a response to the need identified above and as being facilitated by recent technological advances, particularly those related to geochemical analysis. Such analytical advances have greatly expanded the scope for analysing large numbers of samples for a wide range of geochemical properties, including radionuclides, and particularly properties which offer important potential as sediment source tracers and in tracing soil and sediment redistribution. The first decade of the current century has seen very important advances and developments in sediment tracing techniques, which have included a move from their being primarily a research tool towards their more routine application as a management tool. In early July 2011, the International Commission on Continental Erosion of the International Association of Hydrological Sciences (IAHS) organised a 2-day workshop on ‘Tracer Applications in Sediment Research’ within the IAHS General Assembly held in Melbourne, Australia. This workshop was convened by the three authors of this preface. Its aim was to bring together researchers working on the application of tracer techniques in sediment investigations in different areas of the world, to provide an overview of current sediment tracing research, to discuss problems and opportunities for further advances and to promote collaboration and sharing of ideas. The workshop was very successful and, since the presentations were seen to provide a valuable overview of current international work in the field, including the different questions being addressed and the approaches used, it seemed appropriate to publish the contributions in a journal special issue. As editors, we were pleased that Hydrological Processes agreed to do this. This special issue contains 14 papers based on the presentations made at the Melbourne workshop, which together provide a useful demonstration of the scope of current work on sediment tracing. Most report the results of specific studies, but some provide more generic coverage, linked to the use of particular tracers or techniques or to specific applications. The 14 papers can conveniently be divided into two main groups. The first group include contributions which focus primarily on the use of fallout radionuclides to quantify soil and sediment redistribution rates in catchments. The second group are primarily concerned with sediment source tracing and establishing the relative contribution of a number of potential sediment sources to the sediment output from a catchment. The five papers dealing with the use of fallout radionuclides to quantify soil and sediment redistribution rates cover a range of topics. That by Golosov et al. (2013) considers the use of Chernobyl-derived 137Cs fallout to document soil and sediment redistribution in catchments, using examples from European Russia. More than 25 years have now elapsed since the Chernobyl accident and the associated 137Cs fallout, which in most locations in European Russia was greatly in excess of the earlier bomb fallout, has provided a valuable opportunity to quantify erosion and deposition rates over this period. There is a need to recognise potential problems associated with the spatial variability of the initial fallout, but the authors provide several examples of the successful use of the approach to document erosion rates along slope transects, rates and patterns of soil redistribution within small catchments and deposition on river floodplains. The contribution from Porto et al. (2013) focuses more explicitly on the use of fallout radionuclide measurements to construct a sediment budget for a small catchment. The study involved a small forested catchment in southern Italy and made use of both 137Cs and 210Pbex measurements. A key feature of the study was the use of a substantial number of sampling sites distributed essentially randomly across the study catchment to provide representative information on the incidence of erosion and deposition within the catchment and the magnitude of the rates involved. These data were combined with information on the sediment output from the catchment to establish a sediment budget for the study catchment. Belyaev et al. (2013) direct attention to the use of 137Cs to quantify rates of sedimentation on river floodplains and present the results of a study of the floodplain of the River Plava in European Russia. This was in an area that received a high input of Chernobyl fallout, and they make use of the associated 137Cs fallout to document sedimentation rates over the ca 25 years that have elapsed since the accident. They use the Chernobyl fallout both as a chronometer to locate the 1986 floodplain surface in sediment profiles and thereby estimate the sedimentation rate and also to estimate the mass of sediment deposited on the floodplain by documenting the increase in inventory. By extrapolating the results obtained for individual sampling points across the floodplain reach, they are able to estimate the total amount of sediment deposited and the importance of floodplain sedimentation as a sink within the sediment budget of the River Plava basin. Bai et al. (2013) use a similar approach involving measurements of both 137Cs and 210Pbex to establish the chronology of sediment deposits within a small karst depression in southwest China, which represents the outlet of a small catchment and traps the sediment mobilised from its catchment by erosion. The chronology provided by the radionuclides is used to reconstruct the recent erosional history of the catchment and to assess the impact of the severe deforestation that occurred in 1979. The radionuclides are also used as tracers to link the sediment sources to the deposits in the depression. The final contribution in this group of papers by Walling (2013) is more generic and provides an overview of the use of 7Be as a sediment tracer. This radionuclide has been used much less extensively than 137Cs and 210 Pbex and this situation is linked to both a number of limitations associated with its short half-life use and the more recent recognition of its potential. Its potential is assessed in terms of its use both to quantify short-term soil and sediment redistribution rates (e.g. individual events) and as a sediment source tracer. The nine papers contained in the group dealing with sediment source tracing provide a valuable overview of the potential of this approach and the various refinements that have been introduced into such investigations in recent year. The first paper by Theuring et al. (2013) focuses on the meso-scale Kharaa River basin in Northern Mongolia and provides a good example of how sediment source fingerprinting techniques have been used to establish the main sources of the fine sediment which is seen as exerting stress on the local aquatic ecosystems. The radionuclides 137Cs, 7Be and 210Pb were used as source fingerprints to establish the relative importance of channel, gully and upland erosion as sediment sources during both the spring snowmelt and summer rainfall events. The resulting information on sediment sources was combined with measurements of the sediment load of the river and application of the SedNet model, to establish the key features of the sediment budget of the catchment. The reliability of the results provided by source fingerprinting techniques is heavily dependent on the rigour of the statistical and numerical procedures used for source discrimination and for source apportionment. The following three papers provide further useful case studies involving sediment source tracing, but also report a range of refinements to the procedures, aimed at increasing the reliability of the results. The contribution provided by Collins et al. (2013a) again focuses on establishing the relative importance of a range of potential sediment sources, in this case for the Wensum catchment, in Norfolk UK. However, attention is directed to the source of fine channel bed sediment, rather than the suspended sediment load, and the paper is important for presenting a clearly defined protocol for implementing the source discrimination and source apportionment procedures commonly used in such studies. These procedures have been refined to take explicit account of the uncertainty involved. A combination of statistical tests is used for discriminating source end members, and the numerical mass balance modelling used for source apportionment incorporates weightings, to account for within-source tracer variations and tracer-specific discriminatory power, as well as a combination of local and genetic algorithm optimisation routines, coupled with Monte Carlo uncertainty analysis. Olley et al. (2013) direct particular attention to the need to take account of the spatial variability of source material properties when defining the fingerprints of individual sources. Their study, undertaken in a number of small catchments in South-east Queensland, Australia, explores the variability of 137Cs and 210Pbex activity in potential source materials and demonstrates that the end member fingerprints are best represented by probability distributions of radionuclide activity and that these distributions might not be normal. These probability distributions are used to characterise the potential sources in the mixing model used for source apportionment. In this study area, the channel system is shown to be the primary sediment source, and any attempt to reduce sediment fluxes will need to target the river channels. The third study, undertaken by D'Haen et al. (2013), differs from those presented in the other papers, in that it focuses on past catchment response and aims to establish the provenance of the Holocene floodplain deposits. The study was undertaken in the Büğdüz catchment in southwest Turkey, and emphasis was placed on determining the contribution of different lithological units within the catchment. The source units were discriminated using geochemical fingerprints. In this case, uncertainty was addressed by using a Bayesian mixing model incorporating Markov chain Monte Carlo random walks for source apportionment, with the uncertainty in both the input data and the results being taken into account through use of prior and posterior probability distributions. The results demonstrate the ability of the approach to document both temporal and spatial variability in the provenance of the floodplain deposits. Source fingerprinting investigations rely heavily on the ability of the fingerprint properties to discriminate the potential sources, and a wide range of properties have been used in the many successful studies that have now been reported in the literature. This group of papers includes three contributions which as well as providing further examples of the successful use of sediment source tracing techniques, also describe the use of novel fingerprint properties. The first by Evrard et al. (2013) describes the successful use of Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) spectra, which can provide a rapid and relatively inexpensive means of characterizing soils and sediments. These spectra were used in conjunction with a Partial Least Squares mixing model to estimate the contribution of surface and gully sources to sediment yields within the catchment of the Cointzio reservoir, in the Mexican tropical highlands. Information on sediment source was required to target erosion control measures aimed at reducing sedimentation in the reservoir. The results provided by the DRIFTS spectra were compared with those obtained using a more conventional approach based on geochemical and radiometric fingerprints. In general, the two approaches were found to provide consistent results regarding the relative importance of gullies and cropland/rangeland as sediment sources in different subcatchments. The second paper by Hancock and Revill (2013) focuses on the potential for using compound-specific stable isotopes (CSSIs) and more particularly the δ13C signatures of fatty acid compounds bound to sediment particles to discriminate sediment sources associated with areas under different land use. The work was undertaken in the Logan-Albert catchment in Queensland, Australia, and the results obtained were compared with those provided by more traditional source tracing based on geochemical and radiometric fingerprints. The two sets of results were consistent, but the ability of the CSSI fingerprints to provide additional information on the contribution from areas under different land use (e.g. forest, pasture and cropland) and of different erosion processes (e.g. exposed subsoil) provides clear evidence of the potential of this approach. The paper by Erskine (2013) differs from the other papers in that its emphasis is on tracing the travel distance of sediment mobilised from a localised source area by mass movements occurring during a rainstorm, rather than quantifying the relative contribution of different potential sources. The study was undertaken in the Chichester State Forest in New South Wales, Australia, where the mass movements were linked to poor roading practices. Faced with the need to determine the fate of the sediment mobilised by the mass movements and the downstream travel distance, Munsell Soil Colour was shown to be a powerful tool for characterising the mobilised sediment and tracing its downstream travel distance. The final two papers extend the coverage of the papers comprising this group by addressing additional topics. The first paper, by Smith et al. (2013), provides an assessment of the potential for tracing sediment sources in catchments burnt by wildfire. Such catchments commonly experience a major increase in sediment mobilisation after burning and attention is directed to both sediment sources within burnt areas (e.g. surface or channel) and the relative contribution from burnt and unburnt areas in larger catchments. The efficacy and limitations of fallout radionuclide, geochemical and mineral magnetic properties as source fingerprints in burnt areas are discussed, and the potential for using organic compounds is introduced. As in other applications, the potential benefits of combining source tracing with conventional monitoring of sediment yields and other process measurements are emphasised. The second paper, by Collins et al. (2013b), is particularly interesting in that it reports a successful attempt to combine sediment source fingerprinting with sediment tracking. The work was undertaken in the catchment of the River Glaven in eastern England, and a conventional sediment source tracing investigation was used to establish the relative importance of grassland, arable, channel bank / subsurface and damaged road verges as sources of fine channel bed sediment. These findings were then extended by detailed studies of the relative importance of wheelings and inter-wheelings in contributing the sediment mobilised from cultivated areas and of poached and fluvially eroded channel banks in contributing the sediment mobilised from channel banks. These latter studies were based on particle tracking and involved the use of dual signature fluorescent-magnetic grains which were seeded onto target areas on cultivated land and onto channel bank areas and recovered from the river channel using high strength magnets installed in the channel. Taken together, the 14 papers included in this special issue are seen as providing a clear demonstration of the potential of modern tracing techniques to support investigations of catchment sediment dynamics and sediment budgets. The data provided by such techniques are in many respects unique, since they cannot be provided by more conventional monitoring. As such, tracing techniques add an important dimension to sediment investigations, which in turn are then frequently able to address better key contemporary issues related to catchment management. However, tracing techniques are not a substitute for conventional monitoring. They are generally most powerful when used in combination with such monitoring, which is, for example, able to provide important inform