Documenting and quantifying sediment transport in natural system, especially over millennial timescale, is still challenging. Among potential new approaches, recent development has shown that luminescence signal could be used to estimate transport parameters in rivers such as virtual velocity of sediments, storage time or sediment sources (McGuire & Rhodes, 2015; Gray et al., 2018; Gray et al., 2019; Sawakuchi et al., 2018; Guyez et al., 2022).

In this study, we focus on the factors controlling post-infrared feldspar luminescence signals (pIRIR) of modern fluvial sediments in upstream areas. The objective is to examine whether pIRIR equivalent dose distributions relate to landscape erosion rates and associated sediment fluxes. To test this hypothesis, we studied catchments in the Southern Alps of New Zealand (SANZ), one of the world’s most active mountain ranges, with extremely high rates of exhumation and erosion.

For eight catchments of the SANZ, we compared the single-grain pIRIR equivalent dose distributions from modern fluvial sediments with catchment-wide erosion rates obtained using measurements of ¹⁰Be cosmogenic nuclide concentration in modern fluvial quartz grains. The latter approach is widely used to quantify catchment-wide erosion rates on millennial time scales.

Using the cosmogenic methods, we found catchment-wide erosion rates ranging from 0.2 to 4.0 mm/yr. The rates increased along the mountain range from South-West to North-East, confirming results by Larsen et al. (2014), and may reflect a tectonic uplift gradient related to northward segmentation of the Alpine fault. In addition, erosion rates on the Western side were higher than the Eastern side, which we attribute to the climatic gradient of the SANZ, related to orographic effect.

We measured single-grain pIRIR equivalent dose (D_e) distributions at the outlet of each catchment. We calculated (1) the fraction of grains whose luminescence signal is saturated (Bonnet et al., 2019; Guyez et al., 2022), (2) the fraction of well-bleached grains. We also characterized the D_e distribution using (3) the central age model (CAM; Galbraith et al., 1999) and (4) the bootstrapped minimum age model (BS-MAM; Cunningham & Wallinga, 2012). We found a relationship between those four proxies and erosion rates obtained from ¹⁰Be, as well as with suspended sediment yield (Adams, 1980; Hicks et al., 2011) and channel steepness index.

Our study shows that single grain pIRIR equivalent dose distributions reflect erosion and sediment fluxes of a catchment. This new property could be further developed with the perspective to use this proxy as a new independent tool to quantify erosion and transport processes in a wide range of fluvial settings on time scales shorter than cosmogenic methods.

How to cite: Guyez, A., Bonnet, S., Reimann, T., wilkinson, C., Carretier, S., Norton, K., and Wallinga, J.: Feldspar luminescence signal of modern fluvial sediments as a proxy for erosion rates?, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3135, https://doi.org/10.5194/egusphere-egu23-3135, 2023.

Localized torrential rain, which has recently increased in frequency due to abnormal climate, accelerates erosion in the river basin and increases sediment transport into the river. The movement of inflowed sediment is one of the most important factors in the development and management of water resources.

Among the mechanisms of sediment transport in rivers, bedload has limitations in direct measurement due to the risk it poses and inaccuracy in the existing measurement methods. Measurement equipment based on new concepts is continuously being developed to overcome these limitations. A representative equipment is a pipe hydrophone, which indirectly measures the bedload discharge by collecting and analyzing acoustic data when soil collides with a metal tube with a built-in microphone.

To estimate the bedload discharge, this study acquired data through indoor experiment and applied them to the learning process of the Convolutional Neural Networks(CNN). First, an indoor hydraulic experiment device was built with a pipe hydrophone installed at the bottom of the water outlet of the indoor waterway. Then, a system for analyzing and displaying graphs for the impact sound of bedload, and data acquisition storage programs therein, was established. Finally, learning for bedload discharge estimation was conducted using CNN, and the accuracy of the estimation was reviewed.

As a result, the F1-score for the accuracy of bedload discharge estimation was 61%, and the accuracy was higher when bedload discharge was 3kg and 10kg, compared to other weight ranges. Considering that the accuracy of 61% is an insufficient level to completely trust the estimated result, more efficient measurement would be possible by combining this method with the previously developed measurement methods in a complementary manner. In future studies, additional experimental data under various conditions will be secured and applied, to increase the accuracy of bedload discharge estimation.

"This research was supported by Basic Science Research Program through the National Research Foundation of Korea(NRF) funded by the Ministry of Education(C20017370001)"

How to cite: Jung, M., Jun, K., Kim, S., and Jang, C.: A Study on the Bedload Discharge Estimation using CNN, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-10413, https://doi.org/10.5194/egusphere-egu23-10413, 2023.

The size of coarse sedimentary particles in fluvial systems is key for quantifying sedimentation and transport conditions in both active and ancient fluvial systems. In particular, the grain size of the bed load in gravel-bed rivers allows inferring information on sediment entrainment or deposition mechanisms, and on the hydraulic conditions controlling them. However, collecting data on such coarse-grained sedimentary particles traditionally involved time-intensive and costly fieldwork, leading to the development of image-based techniques for grain size estimation over the last two decades. Nevertheless, despite much progress and the recent deployment of deep learning methods that were trained on large datasets (i.e., > 100 000 manually annotated grains; Lang et al., 2021; Chen et al., 2022), image-based grain size data is limited to single percentile values, often due to a systematic bias and/or a low accuracy (e.g., Chardon et al., 2020; Mair et al., 2022). Specifically, the core challenge for most existing methods is the accurate segmentation, i.e., the identification and delineation of individual grains, across distinctly different types of data.

Here we present a new approach designated to improve the segmentation in images, which is based on the capability of transfer learning of deep learning models. Such a strategy allows us to re-train existing models for new tasks that are similar to their original purpose. In particular, we use the python-based and open-source tool cellpose (Stringer et al., 2021), which is a state-of-the-art machine-learning model based on neural networks and designed to segment cells in biomedical images. We retrained such a cellpose model on several image datasets of fluvial gravel. The rationale for our approach is based on an inferred geometric similarity between cell nuclei and rock pebbles. Our re-trained models outperform existing methods designed for the segmentation of fluvial pebbles in all datasets, despite an order of magnitude smaller number of training data than currently used in machine learning models. Furthermore, our results show that models trained on specialized datasets for a specific sediment setting yield significantly better results than models trained on larger and more diverse datasets. Fortunately, the model’s flexibility, accessibility, and ability for easy and fast training (Pachitariu and Stringer, 2022) enable the training of task- or image-type-specific models. To facilitate the segmentation power of such models, we built an open-source software tool, ImageGrains. This tool allows for easy use of the models we trained, or of other custom models, as well as streamlined grain size and shape measurements. This allows for fast and nearly automated measurements of large numbers of coarse sedimentary particles with high precision and across vastly different image settings.

References

Chardon, V., et al., 2022: River Res. Appl., 38, 358–367, https://doi.org/10.1002/rra.3910.

Chen, X., et al., 2022: Earth Surf. Dyn., 10, 349–366, https://doi.org/10.5194/esurf-10-349-2022.

Lang, N., et al. 2021: Hydrol. Earth Syst. Sci., 25, 2567–2597, https://doi.org/10.5194/hess-25-2567-2021.

Mair, D., et al. 2022: Earth Surf. Dyn., 10, 953–973, https://doi.org/10.5194/esurf-10-953-2022.

Pachitariu, M. and Stringer, C. 2022: Nat. Methods, 19, 1634–1641, https://doi.org/10.1038/s41592-022-01663-4.

Stringer, C., et al. 2021: Nat. Methods, 18, 100–106, https://doi.org/10.1038/s41592-020-01018-x.

How to cite: Mair, D., Do Prado, A. H., Garefalakis, P., Witz, G., and Schlunegger, F.: Machine learning assisted delineation and measurement of grains in sediment images – the potential of transfer learning, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-6550, https://doi.org/10.5194/egusphere-egu23-6550, 2023.

Constraining the initiation of bedload sediment transport in rivers is of fundamental importance to understanding a range of geomorphic processes. Likewise, on hillslopes, identifying the initiation of movement is a vital first step towards developing early warning systems for hazards such as landslides. Several studies have previously experimented with embedding sensors within cobbles and boulders to capture and characterise their initiation and subsequent movement in the laboratory and in the field (both for hillslopes and riverbeds). However, these sensors have been limited by their battery life and/or lack of wireless sensor communication in their ability to monitor movement in natural settings over extended time periods. Accelerometers have been most widely applied, e.g. to detect bedload movement on a river bed, but can only measure vibrations and partial changes in orientation between stationary periods, which can occur simply during shaking of a cobble in its pocket on the bed. Gyroscopes, which can assist in continuous orientation tracking and therefore identification of actual transport (e.g. rolling of a cobble along a riverbed), have higher power consumption.

On SENSUM (smart SENSing of landscapes Undergoing hazardous hydrogeomorphic Movement), we have leveraged advances in micro-electronics and Internet of Things technologies to develop a low-power inertial measurement sensor that communicates in near real-time via Long Range Wide Area Network (LoRaWAN). The sensor includes accelerometers, gyroscopes and magnetometers and laboratory experiments have already shown their potential to differentiate between sliding and rolling behaviour. We have embedded sensors in natural and manmade boulders (SlideCubes), cobbles and wood debris within several landslide and flood prone sites across the UK. The sensors form part of Wireless Sensor Networks that also consist of LoRaWAN gateways and other sensors such as discharge gauges.

We present field data captured from smart cobbles installed in upland rivers on Dartmoor and Cumbria that demonstrate the potential of SENSUM sensors to detect initiation of bedload transport, i.e. the transition from shaking of a cobble in its pocket to downstream transport by rolling and/or saltation. We also present preliminary data of landslide movement captured by sensors installed in SlideCubes at Lyme Regis and Isle of Wight. Moving forwards, we will use machine learning methods to analyse sensor data on the server in near real-time in order to characterise and alert of hazardous movement.

How to cite: Roskilly, K., Bennett, G., Clark, M., Franco, A., Egedusevic, M., Curtis, R., Jones, J., Whitworth, M., Luo, C., and Manzella, I.: Smart cobbles and boulders for monitoring movement in rivers and on hillslopes, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14870, https://doi.org/10.5194/egusphere-egu23-14870, 2023.

Temperate Alpine glaciers produce substantial quantities of sediment that are exported via active subglacial meltwater channels to their proglacial environments. Measurements of suspended sediment and bedload in proglacial rivers have been used to estimate glacial erosion rates and downstream sediment yields, assuming that eroded sediment is rapidly evacuated by flowing meltwater; that subglacial sediment storage remains constant and that the measurements are unaffected by proglacial filtering effects. Studies generally focus on the suspended sediment fraction of export, due to the challenges involved in monitoring coarse sediment transport. It is not surprising that subglacial sediment transport dynamics are poorly understood, and a limited amount of field and model-based research indicates that subglacial sediment transport may be attenuated in the rapidly thinning and retreating snout marginal zones of many Alpine glaciers. This is likely due to the existence of non-pressurised subglacial channels with highly variable transport competence related to diurnal discharge variability, leading to cycles of alluviation and deposition. The potential attenuation of sediment and the unknown relationship between suspended load and bedload has important consequences for estimates of glacial erosion based on proglacial export measurements. 

Here, we present results from a proof-of-concept for a method to track radio-tagged bedload particles through meltwater channels under shallow temperate glacier ice (<50 m). Active radio transmitters were inserted into natural pebbles and then deployed directly via boreholes into a 10 m wide snout-marginal subglacial channel at the Glacier d'Otemma, Switzerland. A roving antenna at the surface was used daily to estimate the planimetric point location and downstream transport distance of each tagged particle using Kernel Density Estimation (KDE) as it moved downstream through the subglacial channel. In addition, stationary antennas on the glacier surface monitored the passage of the particles through successive reaches of the subglacial and proglacial channel, constraining the timing of particle transport events. The roving and stationary antenna data were combined to create a transport distance model for each particle, which, when applied at scale, may be used in conjunction with river gauging data to examine the drivers and timescales of coarse subglacial sediment transport. We present results that confirm this method as a highly original means of quantifying subglacial sediment transport using particle tracking.

How to cite: Lane, S. N., Jenkin, M., Hofmann, M., Hubbard, B., Mancini, D., Miesen, F. M., and Herman, F.: Tracking coarse sediment in an Alpine subglacial channel using radio-tagged particles, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14522, https://doi.org/10.5194/egusphere-egu23-14522, 2023.

Multi grain-size transport models that simulate transport of various grain sizes along with the bed stratigraphy consider that only the sediment present in an active layer at the top of the substratum participates in sediment transport. The thickness of this well-mixed layer may be fixed but also calculated according to the coarsest grain size it contains or to the shear stress applied at the surface of the substratum. However, this approach puts the emphasis on the conservation of the active layer thickness and on the availability of the various sizes within this layer. This means there is little consideration i) for heterogeneity in grain size distribution when mixing together adjacent stratigraphic layers that differ significantly in composition and ii) for grain sizes that could prevent or slow down removal of the others. To cope with these limitations, we developed an algorithm with the ability to capture the transport of heterogeneous sediments and the related stratigraphic record of erosional and depositional events based on the behavior of the various sizes within the bed layers. We built a multi grain-size module based on the precipiton method: the time spent by a precipiton (volume of water that carries sediment) on a pixel determines the grain-size specific magnitude of deposition and erosion. The newness of our work is that the magnitudes of erosion may be corrected according to the sizes that slow down the erosion of the others (zero or slow erosion rate) and stratigraphic layers with similar composition only may be merged. A few tests were conducted to study the morphological evolution of a 1D-river reach under various conditions (water discharge, sediment source, etc.). A lake was added at the end of the reach to record the various sizes existing the reach over time. At low water discharge when only the threshold of fine grains is exceeded, an armoring layer made of coarse grains develop at the surface of the substrate. At a water discharge when all the grains are in motion, the finer the grains are, the further downstream they are transported. This downstream fining pattern may be associated with changes in the concavity of the river profile. This multi grain-size algorithm not restricted to the precipiton approach has the potential to unravel the role of heterogeneous sediments in the formation of sorting patterns and, therefore, it is to be implemented in the landscape evolution model RiverLab (former Eros). 

How to cite: Le Minor, M., Davy, P., Howarth, J., and Lague, D.: A Precipiton-Based Approach for Multi Grain-Size Transport Models, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-15456, https://doi.org/10.5194/egusphere-egu23-15456, 2023.

The hydraulic structures, such as dams and reservoirs, are built for flood mitigation, drinking & irrigation water supply, and hydropower generation. Despite their positive roles, large dams and reservoirs are well known to trap a significant portion of the incoming sediment fluxes. In turn, sedimentation reduces the reservoir's water storage capacity. The Indra Sagar dam, located in the Narmada River Basin, is the largest reservoir in India (total capacity ~ 12.2 Bm³). Therefore, in this study, our objective is to set up a data-driven, i.e., Generalized Additive Model Location Scale and Shape (GAMLSS) to simulate the impact of the Indira Sagar dam on the downstream sediment transport. The daily sediment and water discharge data are used from 1987 to 2019, from June to November, at upstream and downstream gauge stations. Preliminary analysis reveals a significant alteration in downstream sediment discharge after constructing the Indira Sagar dam. However, the pre-dam period doesn't significantly alter sediment transport behavior. In addition, pre-and post-dam water discharge behaviors do not exhibit considerable alteration. The difference between 5-yearly sediment duration curves reveals around 60% to 95% reduction in high and moderate magnitudes sediment load. Further observation suggests an increase in low sediment magnitude flows downstream after the dam construction from the base period 1989-1993. The significance of the study is that it will help water managers in understanding the dam's water storage capacity, which may be affected due to sediment deposition. It is also crucial to understand the geomorphological changes and implications of less sediment supply in the downstream region. The results obtained from this study will further provide additional insights into evolving flood and drought processes and their forecasting around the dam-affected region. This work is in progress, and further results will be presented at the conference.

How to cite: Prajapati, P., Meena, G., Swarnkar, S., and Jha, S.: How is Indira Sagar Dam Altering the Suspended Sediment Transport in Central Indian Region?, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1568, https://doi.org/10.5194/egusphere-egu23-1568, 2023.

Improving river monitoring techniques is critical given the concomitant impact of climate change, population growth, and pollution over the last years. Turbidity is one of the most significant metrics for water quality characteristics. In river basins, high turbidity values can be indicative of both organic and inorganic materials. Turbidity is often used as a proxy for transport of suspended particles and associated fluxes of hydrophobic pollutants in a wide range of hydrological conditions. However, it is demanding to estimate suspended sediment yields in rivers because of the high variability along stream of suspended sediment concentrations. Traditional methods, such as gravimetric analysis, are time-consuming, expensive, often discontinuous in space and time and influenced by human errors or instrumental limitations.

Remote sensing techniques are a suitable alternative to point measurements. Satellite remote sensing allows to study the spatial and temporal variations of water status parameters, but it is limited by the spatial and temporal resolution of the satellites considered. Low range systems can help increase the resolution of the imagery used for this purpose. In particular, the use of optical cameras can significantly reduce the monitoring cost and exponentially increase the information on water bodies health and hydrological dynamics, offering a large amount of data distributed in time and space. Nonetheless, all optical sensing methods are strongly affected by many environmental constraints (light, good optical transmission, visibility, etc.), which make them currently not always suitable for regular long-term monitoring of turbidity in rivers. 

The main goal of the monitoring procedure identified in this work is to avoid all these constraints, by processing the camera image to use it as a real measurement data. In this work, an image processing procedure has been identified by exploiting the water surface reflectance properties to estimate water turbidity spectral indices related to red and green bands of the light visible spectrum (Miglino et al., 2022). This river monitoring system is under development in different cross sections of the Bode River, one of the best-instrumented catchments in Central Germany.managed by UFZ Helmholtz Centre for Environmental Research. They gather a wide range of environmental data including a long-term time series on water quantity and quality. Preliminary results highlighted interesting similarities between the chromatic variation of the water surface captured by the RGB camera and the real data. 

Keywords: turbidity, sediment transport, image processing, spectral indices, remote sensing, camera, water quality assessment.

References:

Miglino, D., Jomaa, S., Rode, M., Isgro, F., & Manfreda, S. (2022). Monitoring Water Turbidity Using Remote Sensing Techniques. Environmental Sciences Proceedings, 21(1), 63.

How to cite: Miglino, D., Jomaa, S., Rode, M., Isgro, F., Saddi, K. C., and Manfreda, S.: The use of optical camera for river turbidity monitoring, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-14960, https://doi.org/10.5194/egusphere-egu23-14960, 2023.