Organic carbon burial by river meandering partially offsets bank-erosion carbon fluxes in a discontinuous permafrost floodplain

Arctic river systems erode permafrost in their banks and mobilize particulate organic carbon (OC). Meandering rivers can entrain particulate OC from permafrost many meters below the depth of annual thaw, potentially enabling OC oxidation and the production of greenhouse gases. However, the amount and fate of permafrost OC that is mobilized by river erosion is uncertain. To constrain OC fluxes due to riverbank erosion and deposition, we collected riverbank and floodplain sediment samples along the Koyukuk River, which meanders through discontinuous permafrost in central Alaska. We measured sediment total OC (TOC), radiocarbon content, water content, bulk density, grain size, and floodplain stratigraphy. Radiocarbon abundance and TOC were higher in samples dominated by silt as compared to sand, which we used to map OC content onto floodplain stratigraphy and estimate carbon fluxes due to river meandering. Results showed that sediment being eroded from cutbanks and deposited as point bars had similar OC stocks (mean ± 1SD of 125.3 ± 13.1 kgOC m−2 in cutbanks versus 114.0 ± 15.7 kgOC m−2 in point bars) whether or not the banks contained permafrost. We also observed radiocarbon-depleted biospheric OC in both cutbanks and permafrost-free point bars. These results indicate that a significant fraction of aged biospheric OC that is liberated from floodplains by bank erosion is subsequently re-deposited in point bars, rather than being oxidized. The process of aging, erosion, and re-deposition of floodplain organic material may be intrinsic to river-floodplain dynamics, regardless of permafrost content.

Geomorphic Apophenia: Inferring meaning where there could be none?

<p>Apophenia describes the experience of seeing meaningful patterns or connections in random or meaningless data. Francis Bacon was one of the first to identify its role as a "human understanding is of its own nature prone to suppose the existence of more order and regularity in the world than it finds". Since then, experiments using streams of randomly generated binary sequences show a propensity for people to believe random data fluctuates more than it actually does. A more mainstream example of this is gamblers fallacy, where lucky or unlucky streaks are identified in the random selection of a roulette wheel. Furthermore, humans can also be influenced by a pre-existing ideas or a narrative that they then transpose into their findings leading to tending to support a hypothesis instead of disproving (confirmation bias).  </p><p>As much of geomorphological science involves the interpretation of data, we argue that the persuasiveness of a narrative and human difficulties in recognizing genuinely random data could lead to apophenia. This presentation examines where apophenia might affect geomorphology, using examples from sediment stratigraphy, signal shredding, river meandering and the numerical modelling of landscape systems. In particular, we focus on how seductive it can be to link changes in landscape to drivers when there are potentially hazardous gaps in the data we are using.</p><p>In Geomorphology correlation has for long been substituted by causation. However, with emerging data rich methods including structure from motion, seismology, remote sensing and numerical modelling, former ‘classic’ techniques of qualitative interpretation can give way to quantitative hypothesis testing.</p>

Mapping Fluvial Landforms with Deep Similarity Learning

<p>Semantic image classification as practised in Earth Observation is poorly suited to mapping fluvial landforms which are often composed of multiple landcover types such as water, riparian vegetation and exposed sediment. Deep learning methods developed in the field of computer vision for the purpose of image classification (ie the attribution of a single label to an image such as cat/dog/etc) are in fact more suited to such landform mapping tasks. Notably, Convolutional Neural Networks (CNN) have excelled at the task of labelling images. However, CNN are notorious for requiring very large training sets that are laborious and costly to assemble. Similarity learning is a sub-field of deep learning and is better known for one-shot and few-shot learning methods. These approaches aim to reduce the need for large training sets by using CNN architectures to compare a single, or few, known examples of an instance to a new image and determining if the new image is similar to the provided examples. Similarity learning rests on the concept of image embeddings which are condensed higher-dimension vector representations of an image generated by a CNN. Ideally, and if a CNN is suitably trained, image embeddings will form clusters according to image classes, even if some of these classes were never used in the initial CNN training.</p><p> </p><p>In this paper, we use similarity learning for the purpose of fluvial landform mapping from Sentinel-2 imagery. We use the True Color Image product with a spatial resolution of 10 meters and begin by manually extracting tiles of 128x128 pixels for 4 classes: non-river, meandering reaches, anastomosing reaches and braiding reaches. We use the DenseNet121 CNN topped with a densely connected layer of 8 nodes which will produce embeddings as 8-dimension vectors. We then train this network with only 3 classes (non-river, meandering and anastomosing) using a categorical cross-entropy loss function. Our first result is that when applied to our image tiles, the embeddings produced by the trained CNN deliver 4 clusters. Despite not being used in the network training, the braiding river reach tiles have produced embeddings that form a distinct cluster. We then use this CNN to perform few-shot learning with a Siamese triplet architecture that will classify a new tile based on only 3 examples of each class. Here we find that tiles from the non-river, meandering and anastomising class were classified with F1 scores of 72%, 87% and 84%, respectively. The braiding river tiles were classified to an F1 score of 80%. Whilst these performances are lesser than the 90%+ performances expected from conventional CNN, the prediction of a new class of objects (braiding reaches) with only 3 samples to 80% F1 is unprecedented in river remote sensing. We will conclude the paper by extending the method to mapping fluvial landforms on entire Sentinel-2 tiles and we will show how we can use advanced cluster analyses of image embeddings to identify landform classes in an image without making a priori decisions on the classes that are present in the image.</p>

Empirical Studies in Alluvial Streams

Meander flow takes place in one single channel which oscillates more orless regularly with amplitudes that tend to increase with time. Meandersare found in beds of fine sediments with gentle slopes. In this study, effortwill be made to investigate meanders’ turbulent boundary layer and toimprove the present knowledge about the river meandering phenomena. Itis assumed that the development of the perturbations which develop intomeanders or braids, is longer than the width of the channel. Interaction between the flow and mobile boundaries produces channel patterns which areclassified as meandering or braided. It is therefore long compared with theripples or dunes which cover the bed of such a channel and whose wavelength is shorter than the width of the channel. The variation of resistance toflow and rate of transport of bed material with velocity are discussed brieflyand taken into account. Meander flow and meander shear stress distribution of the channel are described. The basis is a steady, two-dimensionalmodel of flow in an alluvial channel with variable curvature. The meanderdevelopment is described by forcing a travelling, small-amplitude channelalignment wave on the system, and determining the growth characteristicsof the wave. Laboratory data are used to verify the formulas.

The effect of flow variability on the river meandering dynamics

<p>The dynamics of a meandering river has been widely investigated by the scientific community. However, the effects of discharge variability on the meander evolution is still an open question. In this work, we present numerical simulations of the short-term evolution of a plane river morphology (the Ikeda, Parker and Sawai model is used to describe the stream hydrodynamics) forced by a stochastic flow discharge (simulated by a compound Poisson process). The comparison of the simulation outcomes with those obtained for the same river under a constant discharge (equal to the mean of the stochastic process) shows interesting results. The discharge variability slows down both the formation of the meanders and the occurrence of the cutoff events, and induces lower meander curvilinear wavelengths and excess bank velocities. A theoretical analysis of the relationship between the channel erosion rate and the river discharge for the Kinoshita curve confirms the obtained numerical results.</p>

Geomorphic Apophenia: Inferring meaning where there could be none?

<p>Apophenia describes the experience of seeing meaningful patterns or connections in random or meaningless data. Francis Bacon was one of the first to identify its role as a "human understanding is of its own nature prone to suppose the existence of more order and regularity in the world than it finds". Since then, experiments using streams of randomly generated binary sequences show a propensity for people to believe random data fluctuates more than it actually does. A more mainstream example of this is <em>gamblers fallacy</em>, where lucky or unlucky streaks are identified in the random selection of a roulette wheel. Furthermore, humans can also be influenced by a pre-existing ideas or a narrative that they then transpose into their findings leading to tending to support a hypothesis instead of disproving (<em>confirmation bias</em>). </p><p>As much of geomorphological science involves the interpretation of data, we argue that the persuasiveness of a narrative and human difficulties in recognizing genuinely random data could lead to apophenia. This presentation examines where <em>apophenia</em> might affect geomorphology, using examples from sediment stratigraphy, signal shredding, river meandering and the numerical modelling of landscape systems. In particular, we focus on how seductive it can be to link changes in landscape to drivers when there are potentially hazardous gaps in the data we are using.</p><p>In Geomorphology correlation has for long been substituted by causation. However, with emerging data rich methods including structure from motion, seismology, remote sensing and numerical modelling, former ‘classic’ techniques of qualitative interpretation can give way to quantitative hypothesis testing.</p><p> </p>

Investigation of the lateral channel migration: A case study of the South Morava River (Serbia)

The morphological evolution of the fluvial relief in the lowland areas is determined by the dynamic of the lateral channel migration process. River meandering and lateral channel migration represent continuous, dynamic and complex processes, which intensity modifies alluvial plains. Accordingly, it is a current topic observed from the domain of various scientific disciplines and practices, including the geographical aspect of the study. Directly or indirectly, variations of natural and anthropogenic processes affect changes in the lateral migration intensity, which is later manifested through permanent consequences for the environment. The aim of this paper is to investigate the process of lateral channel migration, through the review and interpretation of theoretical and methodological concepts and results of contemporary scientific literature. In this paper, on specific sections of the South Morava River (Serbia), the values of maximum lateral migration over different time periods are determined. Three representative river sectors were singled out, spatial and temporal dynamic was determined, while the process of lateral channel migration was presented quantitatively and graphically. Based on the obtained data, a comparative analysis showed significant riverbank changes for the observed meanders, within the period 1924-2020. Special emphasis is on the analysis of the lateral channel migration in the last decade of the mentioned period. The obtained results can be further used in order to develop and implement plans of water and land management, environmental protection and socio-economic development strategies.