Investigating the role of connectivity and scale in assessing the sources of sediment in an agricultural watershed in the Canadian prairies using sediment source fingerprinting

Purpose

Sediments adversely impact the quality of surface waters and are a significant source of contaminants, such as nutrients and pesticides, in agricultural watersheds. The development of effective beneficial management practices (BMPs) to minimize these impacts requires a sound understanding of the sources of sediments. The objectives of this study were: (1) to determine the sources of sediment in an agricultural watershed in the Canadian prairies using sediment source fingerprinting and; (2) to assess the results of the sediment fingerprinting study within the context of the scale of observation and the hydro-geomorphic connectivity of the watershed.

Materials and methods

Geochemical and radionuclide fingerprints were used to discriminate between the three potential sediment sources identified: topsoil, streambanks, and shale bedrock. Suspended and bed sediment samples were collected over the course of 3 years at six locations along the main stem of the creek, ranging from 3rd- (48 ha) to 7th-order (7441 ha) drainage basins. Four sediment fingerprint properties were selected that met statistical- and process-based selection criteria and the Stable Isotope Analysis in R model was used to estimate the proportion of sediment derived from each source at each sampling location in the watershed.

Results and discussion

The suspended sediments in the upper reaches were dominated by topsoil sources (64%–85%), whereas the suspended sediments moving through the lower reaches and being exported from the watershed had a higher proportion of sediment coming from streambank (32%–51%) and shale bedrock (29%–40%) sources. The switch in the sources of sediment between the headwaters and the watershed outlet are due to: (1) changes in sediment storage and connectivity; (2) a transition in the dominant erosion processes from topsoil to streambank erosion; and (3) the incision of the stream through the shale bedrock as it crosses the Manitoba Escarpment.

Conclusions

The results of this sediment fingerprinting study demonstrated that there was a switch in sediment sources between the headwaters and the outlet of the watershed. This research highlights the importance of the sampling location, in relation to the scale and geomorphic connectivity of the watershed, on the interpretation of results derived from the sediment fingerprinting technique, particularly in terms of developing suitable watershed BMPs to protect surface waters.     

Sediment source analysis in the Linganore Creek watershed, Maryland, USA, using the sediment fingerprinting approach: 2008 to 2010

Purpose

Fine-grained sediment is an important pollutant in streams and estuaries, including the Chesapeake Bay in the USA. The objective of this study was to determine the sources of fine-grained sediment using the sediment fingerprinting approach in the Linganore Creek watershed, a tributary to the Chesapeake Bay.

Materials and methods

The sediment fingerprinting approach was used in the agricultural and forested, 147-km² Linganore Creek watershed, Maryland from 1 August 2008 to 31 December 2010 to determine the relative percentage contribution from different potential sources of fine-grained sediment. Fine-grained suspended sediment samples (<63 μm) were collected during storm events in Linganore Creek using an automatic sampler and manual isokinetic samplers. Source samples were collected from 40 stream bank sites, 24 agricultural (cropland and pasture) sites, and 19 forested sites. Suspended sediment and source samples were analyzed for elements and stable isotopes.

Results and discussion

Results of sediment fingerprinting for 194 samples collected in 36 separate storm events indicate that stream banks contributed 53% of the annual fine-grained suspended sediment load, agriculture contributed 44%, and forests contributed 3%. Peak flows and sediment loads of the storms correlate to stream bank erosion. The highest peak flows occurred in the winter and, along with freeze–thaw activity, contributed to winter months showing the highest rate of stream bank erosion. Peak flow was negatively correlated to sediment sources from agricultural lands which had the greatest contribution in non-winter months. Caution should be observed when trying to interpret the relation between sediment sources and individual storms using the sediment fingerprinting approach. Because the sediment fingerprinting results from individual storms may not include the temporal aspects of the sourced sediment, sediment that is in storage from previous events, remobilized and sampled during the current event, will reflect previous storm characteristics. Stream bank sediment is delivered directly to the channel during an event, whereas the delivery of upland sediment to the stream is lower due to storage on hillslopes and/or in channels, sediment from stream banks are more likely to be related to the characteristics of the sampled storm event.

Conclusions

Stream banks and agricultural lands are both important sources of fine-grained sediment in the Linganore Creek watershed. Peak flows and sediment loads for the 36 storms show a significant relation to sediment sources from stream bank erosion. Attempting to link upland sediment sources to flow and seasonal characteristics is difficult since much of the upland sediment eroded in an event goes into storage. By averaging sediment sources over several storms, it may be possible to determine not only the sediment sources that are directly contributed during the current event but also sediment from previous events that was in storage and remobilized.     

Fingerprinting the sources of suspended sediment delivery to a large municipal drinking water reservoir: Falls Lake, Neuse River, North Carolina, USA

Purpose

We employ a geochemical-fingerprinting approach to estimate the source of suspended sediments collected from tributaries entering Falls Lake, a 50-km² drinking water reservoir on the Neuse River, North Carolina, USA. Many of the major tributaries to the lake are on North Carolina’s 303(d) list for impaired streams, and in 2008, the lake was added to that list because of high values of turbidity, likely sourced from tributary streams.

Materials and methods

Suspended sediments were collected from four streams with a time-integrated sampler during high-flow events. In addition, composite sediment samples representing potential sources were collected from stream banks, forests, pastures, construction sites, dirt and paved roads, and road cuts within tributary basins. Radiocarbon dating and magnetic susceptibility measurements were used to determine the origin of stream bank alluvial deposits. Sediment samples were analyzed for the concentrations of 55 elements and two radionuclides in order to identify tracers capable of distinguishing between potential sediment sources. The relative sediment source contributions were determined by applying a Monte Carlo simulation that parameterized the geochemical tracer data in a mixing model.

Results and discussion

Radiocarbon and magnetic susceptibility measurements confirmed the presence of “legacy” sediment in the Ellerbe and New Light Creek valley bottoms. Mixing model results demonstrate that stream bank erosion is the largest contributor to the suspended sediment load in New Light Creek (62%), Ellerbe Creek (58%), and Little Lick Creek (33%), and is the second largest contributor in Lick Creek (27%) behind construction sites (43%).

Conclusions

We find that stream bank erosion is the largest nonpoint source contributor to the suspended sediment load in three of the four catchments and is therefore a significant source of turbidity in Falls Lake. The presence of legacy sediment appears to coincide with increased contributions from stream bank erosion in Ellerbe and New Light creeks. Active construction sites and timber harvesting were also significant sources of suspended sediment. Water quality mitigation efforts need to consider nonpoint-source contributions from stream bank erosion of valley bottom sediments aggraded after European settlement.     

Mineralogy and nutrient desorption of suspended sediments during a storm event

Purpose

This study investigated desorption of potassium (K) and phosphorus (P) from soil and river suspended sediments sampled during a storm event in a Brazilian watershed traditionally used for tobacco plantations.

Material and methods

Suspended sediment samples were collected automatically at the outlet of the watershed and were grouped into three phases: beginning (phase a), middle (phase b) and final stages (phase c) of the storm event. Granulometric and mineralogical characterisation of soils (0 to 0.20 m depth) and suspended sediments was determined, and K and P extractions were performed using a cation and anion exchange resin (CAER) membrane. A kinetic modelling approach was used to estimate the amount of K and P desorbed.

Results and discussion

Clay-sized (<2 μm) content of the soils were all <21 %. Kaolinite, smectite (partially with hydroxy-Al interlayer) and a small amount of illite were found in the clay fraction of the different soils. The clay-sized fractions in sediments of phases a, b and c of the storm event were 49, 52 and 72 %, respectively. Smectite (>90 %) and kaolinite (<10 %) were the dominant clay minerals in the suspended sediments. The values of labile P and potentially available P of suspended sediments were higher than those for soils. In sediments, the highest values of labile P (325 mg kg^?1) and labile K (4,458 mg kg^?1) were found in phase c and in phase a, respectively.

Conclusions

Particle size distribution and clay mineralogy of soils differed from those of suspended sediments collected during the storm event. By comparison with the watershed soils, suspended sediments collected during the storm event were enriched in fine particles composed mainly of smectite, and this may explain their P and K desorption behaviour. This suggests particle size and clay species selectivity processes during the transfer of sediment particles from soils to aquatic systems. The amounts of P and K desorbed from the suspended sediments in the three phases of the storm event were much larger than those desorbed from soils. This indicates that rainfall promoted the transfer of these nutrients to the watercourses.     

Toward generalizable sediment fingerprinting with tracers that are conservative and nonconservative over sediment routing timescales

Purpose

The science of sediment fingerprinting has been evolving rapidly over the past decade and is well poised to improve our understanding, not only of sediment sources, but also the routing of sediment through watersheds. Here, we discuss channel–floodplain processes that may convolute or modify the sediment fingerprinting signature of alluvial bank/floodplain sources and explore the use of nonconservative tracers for differentiating sediment derived from surface soil erosion from that of near-channel fluvial erosion.

Materials and methods

We use a mathematical model to demonstrate the theoretical effects of channel–floodplain exchange on conservative and nonconservative tracers. Then, we present flow, sediment gauging data, and geochemical measurements of long- (meteoric beryllium-10, ¹⁰Be) and short-lived (excess lead-210 and cesium-137, ²¹⁰Pb_ex and ¹³⁷Cs, respectively) radionuclide tracers from two study locations: one above, and the other below, a rapidly incising knick zone within the Maple River watershed, southern Minnesota.

Results and discussion

We demonstrate that measurements of ¹⁰Be, ²¹⁰Pb_ex, and ¹³⁷Cs associated with suspended sediment can be used to distinguish between the three primary sediment sources (agricultural uplands, bluffs, and banks) and estimate channel–floodplain exchange. We observe how the sediment sources systematically vary by location and change over the course of a single storm hydrograph. While sediment dynamics for any given event are not necessarily indicative of longer-term trends, the results are consistent with our geomorphic understanding of the system and longer-term observations of sediment dynamics. We advocate for future sediment fingerprinting studies to develop a geomorphic rationale to explain the distribution of the fingerprinting properties for any given study area, with the intent of developing a more generalizable, process-based fingerprinting approach.

Conclusions

We show that measurements of conservative and nonconservative tracers (e.g., long- and short-lived radionuclides) can provide spatially integrated, yet temporally discrete, insights to constrain sediment sources and channel–floodplain exchange at the river network-scale. Fingerprinting that utilizes nonconservative tracers requires that the nonconservative behavior is predictable and verifiable.     

Core-derived historical records of suspended sediment origin in a mesoscale mountainous catchment: the River Bléone, French Alps

Purpose

Quantifying suspended sediment fluxes and dynamics across mountains, and identifying the origin of sediment in severely eroded areas, are of primary importance for the management of water resources. This contribution aims to generalise previous results from suspended sediment fingerprinting obtained during 2007?C2009 in a mesoscale Alpine catchment (the Bléone River; 905?km²) in France, and to assess variability in sediment sources throughout the second half of the twentieth century.

Materials and methods

Sediment fingerprinting, based on elemental geochemistry and radionuclide measurements, was conducted on a sediment core collected in an alluvial floodplain at the basin outlet. This technique was combined with hydro-sedimentary time-series to reconstruct the origin of suspended sediment deposited at this location over the last 50?years.

Results and discussion

Interpretation of sedimentation based on historical hydrological databases corroborates core dating obtained with ¹³⁷Cs and ²¹⁰Pb_xs activity measurements. Black marls and (marly) limestone sources provided the main fraction of sediment throughout the sequence (40 and 22?%, respectively). However, we also found evidence for the occurrence of major floods carrying large quantities of sediment originating from Quaternary deposits and conglomerates (25 and 16?%, respectively). The variability of sediment sources throughout the sequence may reflect the spatial variability of rainfall within the catchment, which in turn reflects its origin. However, the relatively homogeneous sediment composition throughout the sequence confirms that core-derived information is representative of widespread flood events.

Conclusions

These results are consistent with those obtained in previous studies. They also outline the need to take into account the entire grain size range of fine sediment in order to provide an overall picture of sediment sources and transfers within highly erosive catchments. This study also emphasizes the importance of using archival data to validate the results of sediment fingerprinting studies conducted during short contemporary monitoring programmes, and to extend fingerprinting of sediment sources over longer time-scales which include large and widespread floods.     

A streamlined approach for sediment source fingerprinting in a Southern Piedmont watershed, USA

Purpose

Sediment fingerprinting is a relatively recent research technique, capable of determining the origin of suspended sediment. In this study, we investigated sub-basins within a larger watershed we examined previously. The objectives were to determine if there was spatial variation in the origin of the suspended sediments and to test a streamlined fingerprinting approach which would reduce the cost, thereby paving the way for adoption by government agencies.

Materials and methods

Samples were collected from three tributaries, the outlet of the main stem, and at the middle of the main stem. Two methods to collect suspended sediment samples were compared: a mobile continuous-flow centrifuge and automated samplers. A relatively small initial tracer suite consisting of stable isotopes of nitrogen (N) and carbon (C) (¹⁵N and ¹³C), total N (TN), and total C (TC) was tested. Tracer concentrations were obtained through a single mass spectrometry analysis requiring <1 g of sediment.

Results and discussion

Multivariate discriminant analysis showed that three of the four tracers (δ ¹⁵N, δ ¹³C, and TC) from the initial pool were capable of accurate classification of the source samples. A multivariate mixing model showed that banks contributed the majority of sediment throughout all locations sampled and that in tributaries it was an even more dominant source. Despite variations in land use and stream order, the legacy sediments comprising the banks and floodplains were the main factor in impairment for suspended sediment. We found a small but statistically significant difference in δ ¹⁵N and δ ¹³C concentrations collected using automated samplers vs. the mobile centrifuge, but the effect on analysis of sediment source proportions was minimal.

Conclusions

The results of this study indicate that, at least in the study watershed, the majority of sediment in suspension was of streambank origin. A cost-effective tracer suite was identified as well as an attempt to make a streamlined approach to the technique. The streamlined approach cost much less ($7,550 US) than the conventional approach ($46,600 US) and should be suitable for total maximum daily loads analysis by state government agencies in the Southern Piedmont region of the USA.     

Determining the effects of wildfire on sediment sources using 137Cs and unsupported 210Pb: the role of landscape disturbances and driving forces

Purpose

Wildfires represent one of the major natural disturbances within forested landscapes and have potential implications for the quality and function of downstream aquatic ecosystems. This study aimed to determine if a wildfire in a mountainous, forested watershed in British Columbia, Canada, caused a change in the dominant sediment source in the immediate 1?C2?years following the wildfire, and if the sediment sources changed over the medium term (3?C7?years) as the landscape recovered.

Materials and methods

Source materials (surface soil, subsurface soil and channel bank material) and fluvial (suspended and channel bed) sediment samples were collected over the period 2004 to 2010 from a watershed burnt by a wildfire in 2003, and from an adjacent watershed that was not impacted by the fire. Samples were analysed for the fallout radionuclides (FRNs) caesium-137 (¹³⁷Cs) and unsupported lead-210 (²¹⁰Pb_un). An unmixing model was used to calculate the relative source contributions of the fluvial sediment samples.

Results and discussion

¹³⁷Cs and ²¹⁰Pb_un were concentrated in the upper layers of surface soils in both watersheds and were statistically different to concentrations in subsurface and channel bank material. In the burnt watershed, FRN concentrations were greatest in the ash layer. Sediment sources as determined by the unmixing model were 100?% subsurface/channel bank material in the unburnt watershed, while in the burnt watershed 8.5?±?2.5?% was derived from surface soils. In both watersheds, there were no major changes in the relative contributions from surface soil and from subsurface/channel bank material over the period 2004 to 2010. Thus, while the wildfire did cause a change in sediment sources, it was fairly subtle and did not conform to the effects following wildfire described for other studies in contrasting environments, which typically document a major increase in hillslope contributions relative to channel bank sources.

Conclusions

There was a limited response in terms of fine-grained sediment sources (and also sediment fluxes) in the burnt watershed. The reason for this muted response to a severe wildfire is likely to be the lack of precipitation, especially winter precipitation and the associated snowmelt, in the first year following the wildfire. Thus while the landscape was primed for erosion and sediment transport, the lack of a driving force meant that there was a limited immediate post-fire sediment response.     

Partitioning fine sediment loads in a headwater system with intensive agriculture

Purpose

This study was developed to improve understanding of the temporal variability of sediment delivery in a representative, intensively agricultural, headwater system of the U.S. Midwest by identifying the primary sediment source (i.e., uplands or channel banks) to the fine suspended sediment loads of three consecutive runoff events (with the third event being a flash flood) using naturally occurring radionuclides.

Materials and methods

Suspended sediment concentrations (C _s) from discrete and continuous sampling techniques agreed well despite differences in operating principles. The total sediment flux (Q _s) during each event was quantified over a 24-h period from the initiation of the rainfall using the following: (1) measured C _s and flow discharges (Q _w); (2) individual Q _w?CQ _s relationships for each event (herein called individual event relationships); and (3) a cumulative Q _w?CQ _s rating curve. The radionuclide tracers, beryllium-7 (⁷Be) and excess lead-210 (²¹⁰Pb_xs), were used with a simple two end-member mixing model to differentiate eroded upland surface soils and channel-derived sediments in the suspended loads of each event.

Results and discussion

Total load estimates from the measurement-based values and individual event relationships were similar, within 10?%, because they accounted for an observed non-linearity between C _s and Q _w (i.e., a clockwise hysteresis) during the events. The sediment rating curve assumed a linear relationship between C _s and Q _w and under-estimated the loads of the first two events while over-estimating the load of the flood event. The radionuclide partitioning quantified the proportion of eroded upland soils at 67?% for the first event, which was attributed to a ??first flush?? of readily available material from past events. For the subsequent and flood-event loads, 34?% and 21?% were respectively derived from the uplands, because less material was readily available for mobilization. Proportions are based on integrated samples for each event and are consistent with individual samples where available. During the flood event, stream bank mass failure was observed and bank erosion estimates from multiple methods compared favorably with the load results.

Conclusions

The radionuclide analysis showed decreasing proportions of eroded upland soils in the loads of the three successive events that was supported by observed clockwise hysteresis from source material exhaustion. Decreasing slopes observed in successive hysteresis plots for the events suggested that less material was readily available for mobilization following the first event flushing. The results of this study can assist watershed planners in identifying erosion-prone areas and determining optimal management strategies for sediment control.     

Sampling soil and sediment depth profiles at a fine resolution with a new device for determining physical,chemical and biological properties: the Fine Increment Soil Collector (FISC)

Purpose

Many environmental investigations (empirical and modelling) and theories are based on reliable information on the depth distribution of physical, chemical and biological properties in soils and sediments. However, such depth profiles are not easy to determine using current approaches, and, consequently, new devices are needed that are able to sample soils and sediments at fine resolutions.

Materials and methods

We have designed an economic, portable, hand-operated surface soil/sediment sampler—the Fine Increment Soil Collector (FISC)—which allows for the close control of incremental soil/sediment sampling and for easy recovery of the material collected by a simple screw-thread extraction system. This innovative sampling system was developed originally for the beryllium-7 (⁷Be) approach in soil and sediment redistribution research. To ensure reliable estimates of soil erosion and sediment deposition from ⁷Be measurements, the depth distribution of this short-lived fallout radionuclide in soil/sediment at the resolution of millimetres is a crucial requirement. This major challenge of the ⁷Be approach can be met by using the FISC.

Results and discussion

We demonstrate the usefulness of the FISC by characterising the depth distribution of ⁷Be at increments of 2.5 mm for a soil reference site in Austria. The activity concentration of ⁷Be at the uppermost increment (0–2.5 mm) was ca. 14 Bq kg^?1 and displayed decreasing activity with depth. Using most conventional sampling devices (i.e. the scraper-plate system), the most accurate depth increment would have been 10 mm, and the activity concentration at the surface would have been considerably lower. Consequently, coarser sampling would have influenced estimates of ⁷Be-derived soil erosion and deposition. The potential application for other soil/sediment properties, such as nutrients (e.g. phosphorus), contaminants and carbon are also discussed.

Conclusions

By enabling soil and sediment profiles to be sampled at a depth resolution of millimetres, the FISC has the potential to provide key information when addressing several environmental and geoscientific issues, such as the precise depth distributions of soil/sediment nutrients, contaminants and biological properties.     

Hydrosedimentology of nested subtropical watersheds with native and eucalyptus forests

Purpose

Information on the effects of eucalyptus forests on hydrosedimentological processes is scarce, particularly at the catchment scale. Monitoring and mathematical modeling are efficient scientific tools used to address the lack of information for natural resource management and the representation and prediction of those processes. This study evaluates the effects of eucalyptus cultivation on hydrosedimentological processes in watersheds and to use the Limburg soil erosion model (LISEM) to represent and predict hydrological processes.

Material and methods

The study was conducted in two forested watersheds: the main watershed (94.46 ha) and a nested sub-watershed (38.86 ha), both cultivated with eucalyptus and residual riparian native forest, located in southern Brazil. Hydrosedimentalogical monitoring was conducted from 16th February 2011 to 31st December 2012, and LISEM model calibrations were performed on the bases of six storms events.

Results and discussion

The sediment yield for 2011 was 41.6 Mg km^?2 and 38.5 Mg km^?2 for the watershed and sub-watershed, respectively. An extreme event in 2012 provided greater sediment yield for the sub-watershed (99.8 Mg km^?2) than that for the watershed (51.7 Mg km^?2). Rainfall events with a greater maximum intensity generated rapid discharge and suspended sediment concentration responses in the sub-watershed due to the smaller drainage area and steeper landscape. In the main watershed, the accumulation of flood waves occurred for most events, with less steep hydrographs, and a later occurrence of the discharge peak after that of the sub-watershed. The LISEM adequately reproduced the peak discharge and runoff for the calibrated events; however, the peak time and the shape of the hydrograph were not adequately represented.

Conclusions

The hydrosedimentological patterns of the watershed and sub-watershed, both cultivated with eucalyptus, was characterized by sedimentographs preceding hydrographs during rainfall–runoff events where scale effects occur, with maximum discharge and specific sediment yield greater in the watershed than that in the sub-watershed. Empirical models based on hydrologic variables may be used for estimating the suspended sediment concentration and sediment yield. Therefore, LISEM may be used for the prediction of hydrological variables in these forested watersheds.     

Lead mobilisation in the hyporheic zone and river bank sediments of a contaminated stream: contribution to diffuse pollution

Purpose

Past metal mining has left a legacy of highly contaminated sediments representing a significant diffuse source of contamination to water bodies in the UK and worldwide. This paper presents the results of an integrated approach used to define the role of sediments in contributing to the dissolved lead (Pb) loading to surface water in a mining-impacted catchment.

Materials and methods

The Rookhope Burn catchment, northern England, UK is affected by historical mining and processing of lead ore. Quantitative geochemical loading determinations, measurements of interstitial water chemistry from the stream hyporheic zone and inundation tests of bank sediments were carried out.

Results and discussion

High concentrations of Pb in the sediments from the catchment, identified from the British Geological Survey Geochemical Baseline Survey of the Environment (GBASE) data, demonstrate both the impact of mineralisation and widespread historical mining. The results from stream water show that the stream Pb load increased in the lower part of the catchment, without any apparent or significant contribution of point sources of Pb to the stream. Relative to surface water, the interstitial water of the hyporheic zone contained high concentrations of dissolved Pb in the lower reaches of the Rookhope Burn catchment, downstream of a former mine washing plant. Concentrations of 56???g?l^?1 of dissolved Pb in the interstitial water of the hyporheic zone may be a major cause of the deterioration of fish habitats in the stream and be regarded as a serious risk to the target of good ecological status as defined in the European Water Framework Directive. Inundation tests provide an indication that bank sediments have the potential to contribute dissolved Pb to surface water.

Conclusions

The determination of Pb in the interstitial water and in the inundation water, taken with water Pb mass balance and sediment Pb distribution maps at the catchment scale, implicate the contaminated sediments as a large Pb supply to surface water. Assessment of these diffuse contaminant sources is critical for the successful management of mining-impacted catchments.     

Spectral fingerprinting: sediment source discrimination and contribution modelling of artificial mixtures based on VNIR-SWIR spectral properties

Purpose

Knowledge of the origin of suspended sediment is important for improving our understanding of sediment dynamics and thereupon support of sustainable watershed management. An direct approach to trace the origin of sediments is the fingerprinting technique. It is based on the assumption that potential sediment sources can be discriminated and that the contribution of these sources to the sediment can be determined on the basis of distinctive characteristics (fingerprints). Recent studies indicate that visible–near-infrared (VNIR) and shortwave-infrared (SWIR) reflectance characteristics of soil may be a rapid, inexpensive alternative to traditional fingerprint properties (e.g. geochemistry or mineral magnetism).

Materials and methods

To further explore the applicability of VNIR-SWIR spectral data for sediment tracing purposes, source samples were collected in the Isábena watershed, a 445 km² dryland catchment in the central Spanish Pyrenees. Grab samples of the upper soil layer were collected from the main potential sediment source types along with in situ reflectance spectra. Samples were dried and sieved, and artificial mixtures of known proportions were produced for algorithm validation. Then, spectral readings of potential source and artificial mixture samples were taken in the laboratory. Colour coefficients and physically based parameters were calculated from in situ and laboratory-measured spectra. All parameters passing a number of prerequisite tests were subsequently applied in discriminant function analysis for source discrimination and mixing model analyses for source contribution assessment.

Results and discussion

The three source types (i.e. badlands, forest/grassland and an aggregation of other sources, including agricultural land, shrubland, unpaved roads and open slopes) could be reliably identified based on spectral parameters. Laboratory-measured spectral fingerprints permitted the quantification of source contribution to artificial mixtures, and introduction of source heterogeneity into the mixing model decreased accuracies for some source types. Aggregation of source types that could not be discriminated did not improve mixing model results. Despite providing similar discrimination accuracies as laboratory source parameters, in situ derived source information was found to be insufficient for contribution modelling.

Conclusions

The laboratory mixture experiment provides valuable insights into the capabilities and limitations of spectral fingerprint properties. From this study, we conclude that combinations of spectral properties can be used for mixing model analyses of a restricted number of source groups, whereas more straightforward in situ measured source parameters do not seem suitable. However, modelling results based on laboratory parameters also need to be interpreted with care and should not rely on the estimates of mean values only but should consider uncertainty intervals as well.     

Distinguishing regional and local sources of pollution by trace metals and magnetic particles in fluvial sediments of the Morava River, Czech Republic

Purpose

Regional contamination of southern Moravia (SE part of the Czech Republic) by trace metals and magnetic particles during the twentieth century was quantified in fluvial sediments of the Morava River. The influence of local pollution sources on regional contamination of the river sediments and the effect of sampling site heterogeneity were studied in sediment profiles with different lithologies.

Materials and methods

Hundreds of sediment samples were obtained from regulated channel banks and naturally inundated floodplains and proxy elemental analyses were carried out by energy dispersive X-ray fluorescence spectroscopy (ED XRF) and further calibrated by inductively coupled plasma mass spectrometry (ICP MS). Magnetic susceptibility was determined as a proxy for industrial contamination. The age model for the floodplain sediments was established from ¹³⁷Cs and ²¹⁰Pb dating. Trace metal contamination was assessed by establishing the lithological background values from floodplain profiles and calculating enrichment factors (EF) of trace metals (i.e. Pb, Zn, Cu) and magnetic susceptibility for the entire study area.

Results and discussion

Channel sediments are unsuitable for the reconstruction of historical regional contamination due to their lithological heterogeneity and the “chaotic” influence of local sources of contamination, as well as the possibility of geochemical mobility of pollutants. On the other hand, sediments from regulated river banks qualitatively reflected the actual, local contamination of the river system.

Conclusions

This approach allowed us to distinguish the influence of local sources of contamination by comparison with more spatially averaged contamination signals from distal floodplain profiles. The studied area is weakly contaminated (EF ～1–2), while individual sediment strata from regulated channel banks reflect local sources of contamination and contain up to several times higher concentrations of trace metals.     

Dominant discharges for suspended sediment transport in a highly active Pyrenean river

Purpose

Dominant discharges and associated sediment dynamics of the River Isábena, a 445-km² catchment in the central Pyrenees of Spain that is punctuated by badlands, are analysed.

Materials and methods

Calculations of suspended sediment loads are based on continuous records of discharge and turbidity obtained at the basin outlet for the period 2005–2010.

Results and discussion

Dominant discharges for sediment load (i.e. effective discharge) present a bimodal distribution, with one peak falling in the range of low flows and the other associated to less frequent but higher magnitude floods (i.e. bankfull). The highly suspended sediment availability in the badlands, together with the high connectivity between the badlands and the stream network and the important in-channel fine sediment storage, causes both large and small events to remobilize fines. Baseflows, despite their low competence, generate resuspension and massive sediment loads. Thus, effective discharge (i.e. the discharge which transports most of the sediment) is not solely associated with bankfull (i.e. the discharge that dominates channel form), but to a wider range of discharges. Consequently, this river channel is not specifically adjusted to convey most of the sediment load during high floods, as in many other rivers, but instead large volumes of sediment are transferred downstream at an almost constant rate.

Conclusions

Results suggest that dominant discharge may play a lesser role in terms of (suspended) sediment load in non-supply-limited fluvial systems and/or in rivers that permanently work close to, or at, full transport capacity, as is the case of the Isábena.     

A novel sediment fingerprinting method using filtration: application to the Mara River, East Africa

Purpose

Sediment fingerprinting with elemental tracers is widely used to identify sources of sediment to rivers. However, due to the need to isolate large amounts of suspended sediment, this approach can be difficult to implement in remote locations, such as the Mara River in Kenya, where high (and increasing) sediment loads are of concern.

Materials and methods

We report several innovations that allowed us to carry out sediment fingerprinting in a portion (>6,500 km²) of the Mara River Basin. First, we utilized sediment-laden filters (sediment mass ～0.1 g) for our river samples, rather than the traditional approach of extracting >1 g of sediment from large volumes of water. This allowed us to easily collect flow-weighted samples, and to process and analyze samples without access to centrifugation equipment. We carried out extensive quality control tests to ensure that we could reproducibly measure elemental concentrations of sediment trapped on filters. Second, we modified a readily available Bayesian inference mixing model (Stable Isotope Analysis in R) to create source signatures and to apportion downstream samples to sources. Third, we included hippo feces as a potential source, given the critical role that large wildlife plays in this ecosystem.

Results and discussion

We found that: (1) sediment captured by filtration can be digested and analyzed reproducibly and used in sediment fingerprinting; (2) our four sources (three geographic categories and hippo feces) were reasonably well-separated in their signatures; (3) the three sub-basins all contributed substantially to sediment loading in the Mara; and (4) hippo feces contributed a small, but measurable, proportion of sediment in this system.

Conclusions

Sediment-laden filters can be used successfully in identifying sediment sources through fingerprinting. The modified method of sediment fingerprinting should prove useful in other remote river basins. Our results support the hypothesis that the Upper Mara is important in supplying sediments to the river, while also highlighting the Talek sub-basin as a major contributor.     

Importance of bank erosion for sediment input, storage and export at the catchment scale

Purpose

The importance of bank erosion was quantified during three periods (October 2006–April 2007, May 2007–April 2008 and May 2008–April 2009) in the 486 km² catchment area of River Odense, Denmark. A catchment sediment budget was established including other sediment sources such as tile drains and surface runoff, in-channel and overbank sinks and storage and the resulting bed load and suspended sediment load exported from the catchment.

Material and methods

Bank erosion and sedimentation were measured using ca. 3,000 erosion pins established in 180 pin plots, each consisting of three vertical lines of pins. Thirty-six representative reaches, each with a length of 100 m, were selected by a stratified random procedure in GIS. Bed load and suspended sediment export from the catchment were measured using a bed load sampler and from continuous measurements of turbidity at the outlet gauging station.

Results and discussion

The gross sediment input from bank erosion during the three study periods amounted to 21,100–25,200 t in the River Odense catchment, which is considerably higher than the estimated input of sediment from tile drains and surface runoff, which amounted to 220–500 t and 0–100 t, respectively. The measured bed load (20–490 t) was five to 60 times lower than the suspended sediment export from the catchment (1,240–2,620 t) during the three study periods, with the largest difference occurring in the driest year. Sediment sinks and storage were of high importance for the catchment sediment budget as the measured in-channel storage of sediment on stream banks was as high as 16,200–20,100 t, and the overbank sediment sink was estimated at 360–3,100 t.

Conclusions

Bank erosion was the dominant sediment source (90–94 %) in the River Odense catchment during the three study years. In-channel and overbank sediment sinks and storage dominated the sediment budget as 79–94 % of the sediment input from all sources was not exported from the catchment during the three study years. Such a large attenuation of sediment in river channels and on floodplains is extremely important for fluvial habitats and ecology. Moreover, it has strong implications for attempts to document changes in sediment export following implementation of mitigation measures.     

Vertical distributions of PAHs in the sediments of four lakes in Japan

Purpose

The purpose of this study was to elucidate historical trends, spatial variations, and the sources of polycyclic aromatic hydrocarbons (PAHs) pollution in several Japanese lakes.

Materials and methods

The vertical distributions of PAHs in the core samples of sediments taken at several points in lakes Kasumigaura, Suwa, Kizaki, and Shinji were determined using a gas chromatograph equipped with a mass selective detector and combined with chronological information and the physical/elemental properties of the sediment.

Results and discussion

Seventeen related compounds (congeners) typically had concentration peaks at sediment depths corresponding to the 1960s to 1970s. In Lake Shinji and one bay of Lake Kasumigaura, there was a tendency for PAH concentrations to increase downstream; in contrast, another bay of Lake Kasumigaura showed the reverse trend. During big flood events, the fluxes of PAHs increased due to large inputs of particulate matter, although PAH concentrations were reduced. For the four study lakes and other similar lakes, PAH concentrations of surface sediments were approximately proportional to population densities in the respective watersheds, while the total input of PAHs to the lakes were correlated with their population and watershed area. The source apportionment analysis using isomer ratios for the congener profiles indicated that the principal sources of the PAHs in the lake sediments were gasoline and/or diesel engine exhausts and biomass burning.

Conclusions

The observed concentration peaks showed a deterioration of the chemical quality of atmospheric conditions around 1960?C1970 and a recent tendency for their amelioration. Between-lake differences suggest that the influence of human activity in the watersheds influences sediment PAH concentrations. The PAH sources were identified to be of pyrogenic origin.     

Characterization and optimization of the preparation procedure for solution P-31 NMR analysis of organic phosphorus in sediments

Purpose

The single-step sodium hydroxide?Cethylenediaminetetraacetic acid (NaOH?CEDTA) extraction is currently the most common preparation technique for the measurement of organic phosphorus (P) composition in sediments using solution ³¹P nuclear magnetic resonance (NMR). In this study, detailed investigations were conducted to evaluate the performance of this technique, with an objective of finding an optimal procedure for the measurement of sediment organic P.

Materials and methods

A single-step extraction with NaOH?CEDTA was investigated on two types of sediment, i.e., Fe/Al-rich sediment and calcareous sediment. The influence of different sediment preparation methods, NaOH?CEDTA compositions, solid:solution ratios, extraction time, pre-concentration techniques, and NMR sample frozen storage time on P extraction and solution ³¹P NMR analysis were investigated.

Results and discussion

Both air- and freeze-drying increased organic P extraction for the calcareous sediment. An extraction time of 16?h was sufficient for quantitative recovery of extracted organic P for both Fe/Al-rich and calcareous sediments. The use of a 1:8 solid:solution ratio achieved stronger NMR signals and greater P diversity than the use of a 1:20 ratio. Extraction of the two sediments with 0.25 NaOH?C50?mM EDTA favored ³¹P NMR detection by reducing the relaxation times required and the risk of organic P degradation compared to the use of stronger NaOH?CEDTA solutions. Rotary evaporation was a better technique for pre-concentration of the NaOH?CEDTA extracts than freeze drying. The concentrated extracts could be preserved by freezing (?20?°C) for at least 2?months.

Conclusions

An optimized procedure was developed based on these investigations, including freeze-drying of fresh sediments, extraction with 0.25?M NaOH?C50?mM EDTA for 16?h using a solid:solution ratio of 1:8, pre-concentration of the extract to the level of ??10 times of its original concentration using rotary evaporation, and storage of the NMR sample at ?20?°C until ³¹P NMR analysis.     

Spatial variation and interaction of runoff generation and erosion within a semi-arid, complex terrain catchment: a hierarchical approach

Purpose

Closed erosion plots have been used extensively to investigate soil loss and its spatial variation within a watershed. However, erosion rates measured on closed plots at various locations within a watershed may not reflect the “real world” conditions due to plot boundary problems. The purpose of this study was to identify runoff and sediment sources in a semi-arid, complex terrain catchment by using the data collected from open plots, nested catchments, and tunnel systems.

Materials and methods

The study catchment, in the Loess Plateau of China, was partitioned into various-level geomorphic units. Runoff and sediment discharges were measured from 55 storm events between 1963 and 1968 on open plots and nested catchments. Storm flows were also monitored in 14 rainfall events from the tunnel systems between 1989 and 1990. This study combined the data collected from the two periods to investigate runoff and sediment sources from the different geomorphic units of the catchment.

Results and discussion

On the four open plots (S1, S2, S3, and S4) of the hill slope, total runoff depths of 128.5 mm (S1), 84.3 mm (S2), 101.92 mm (S3), and 141.73 mm (S4) were recorded from all the events over the first period, which correspondingly produced total sediment yields of 3.056 kg m^?2 (S1), 9.058 kg m^?2 (S1), 42.848 kg m^?2 (S3), and 97.256 kg m^?2 (S4). The number of runoff events also varied due to a non-uniformity in runoff generation among the different geomorphic units of the catchment. Tunnel flows generally had higher mean sediment concentrations than catchment outflows. Three nested catchments located from the headwaters (C1) to the mouth of the catchment (C3) generated total runoff depths of 120.02 mm (C1), 143.92 mm (C2), and 149.43 mm (C3), and correspondingly produced sediments yields of 62.01 kg m^?2 (C1), 144.02 kg m^?2 (C2), and 123.92 kg m^?2 (C3) for the first period.

Conclusions

Significant variations in runoff and erosion existed within the catchment. The spatial variation of runoff generation on the hill slopes resulted from the variation of soil infiltration. Sediment produced from the lower hill slope zone was disproportionally higher than that from the upper hill slope zone. Nevertheless, a significant portion of the sediment eroded on the lower slope zone was caused by runoff generated from the upper slope zone. Tunnel erosion also played a significant role in sediment production.