Articles | Volume 9, issue 5
https://doi.org/10.5194/esurf-9-1091-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/esurf-9-1091-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
03 Sep 2021
Research article |
| 03 Sep 2021
| 03 Sep 2021
Inverse modeling of turbidity currents using an artificial neural network approach: verification for field application
Graduate School of Science, Kyoto University, Kitashirakawa Oiwakecho, Sakyo-ku, Kyoto, 606-8502 Japan
Kento Nakao
Baseload Power Japan, SHINTORA-DORI CORE 3F, 4-1-1, Shimbashi, Minato-ku, Tokyo, 105-0004 Japan
Related authors
Rimali Mitra, Hajime Naruse, and Tomoya Abe
Nat. Hazards Earth Syst. Sci., 24, 429–444, https://doi.org/10.5194/nhess-24-429-2024, https://doi.org/10.5194/nhess-24-429-2024, 2024
Short summary
Short summary
This study estimates the behavior of the 2011 Tohoku-oki tsunami from its deposit distributed in the Joban coastal area. In this study, the flow characteristics of the tsunami were reconstructed using the DNN (deep neural network) inverse model, suggesting that the tsunami inundation occurred in the very high-velocity condition.
Koji Ohata, Hajime Naruse, and Norihiro Izumi
Earth Surf. Dynam., 11, 961–977, https://doi.org/10.5194/esurf-11-961-2023, https://doi.org/10.5194/esurf-11-961-2023, 2023
Short summary
Short summary
We investigated the influence of sediment transport modes on the formation of bedforms using theoretical analysis. The results of the theoretical analysis were verified with published data of plane beds obtained by fieldwork and laboratory experiments. We found that suspended sand particles can promote the formation of plane beds on a fine-grained bed, which suggests that the presence of suspended particles suppresses the development of dunes under submarine sediment-laden gravity currents.
Koji Ohata, Hajime Naruse, and Norihiro Izumi
Earth Surf. Dynam. Discuss., https://doi.org/10.5194/esurf-2021-60, https://doi.org/10.5194/esurf-2021-60, 2021
Publication in ESurf not foreseen
Short summary
Short summary
We investigated the influence of sediment transport modes on the formation of bedforms using theoretical analysis. The results of the theoretical analysis were verified with published data of plane beds obtained by fieldwork and laboratory experiments. We found that suspended sand particles can promote the formation of plane beds on a fine-grained bed, which suggests that the presence of suspended particles suppresses the development of dunes under submarine sediment-laden gravity currents.
Rimali Mitra, Hajime Naruse, and Shigehiro Fujino
Nat. Hazards Earth Syst. Sci., 21, 1667–1683, https://doi.org/10.5194/nhess-21-1667-2021, https://doi.org/10.5194/nhess-21-1667-2021, 2021
Short summary
Short summary
A case study on the 2004 Indian Ocean tsunami was conducted at the Phra Thong island, Thailand, using a deep neural network (DNN) inverse model. The model estimated tsunami characteristics from the deposits at Phra Thong island. The uncertainty quantification of the result was evaluated. The predicted flow conditions and the depositional characteristics were compared with the reported observed values. This DNN model can serve as an essential tool for tsunami hazard mitigation at coastal cities.
Rimali Mitra, Hajime Naruse, and Tomoya Abe
Nat. Hazards Earth Syst. Sci., 24, 429–444, https://doi.org/10.5194/nhess-24-429-2024, https://doi.org/10.5194/nhess-24-429-2024, 2024
Short summary
Short summary
This study estimates the behavior of the 2011 Tohoku-oki tsunami from its deposit distributed in the Joban coastal area. In this study, the flow characteristics of the tsunami were reconstructed using the DNN (deep neural network) inverse model, suggesting that the tsunami inundation occurred in the very high-velocity condition.
Koji Ohata, Hajime Naruse, and Norihiro Izumi
Earth Surf. Dynam., 11, 961–977, https://doi.org/10.5194/esurf-11-961-2023, https://doi.org/10.5194/esurf-11-961-2023, 2023
Short summary
Short summary
We investigated the influence of sediment transport modes on the formation of bedforms using theoretical analysis. The results of the theoretical analysis were verified with published data of plane beds obtained by fieldwork and laboratory experiments. We found that suspended sand particles can promote the formation of plane beds on a fine-grained bed, which suggests that the presence of suspended particles suppresses the development of dunes under submarine sediment-laden gravity currents.
Koji Ohata, Hajime Naruse, and Norihiro Izumi
Earth Surf. Dynam. Discuss., https://doi.org/10.5194/esurf-2021-60, https://doi.org/10.5194/esurf-2021-60, 2021
Publication in ESurf not foreseen
Short summary
Short summary
We investigated the influence of sediment transport modes on the formation of bedforms using theoretical analysis. The results of the theoretical analysis were verified with published data of plane beds obtained by fieldwork and laboratory experiments. We found that suspended sand particles can promote the formation of plane beds on a fine-grained bed, which suggests that the presence of suspended particles suppresses the development of dunes under submarine sediment-laden gravity currents.
Rimali Mitra, Hajime Naruse, and Shigehiro Fujino
Nat. Hazards Earth Syst. Sci., 21, 1667–1683, https://doi.org/10.5194/nhess-21-1667-2021, https://doi.org/10.5194/nhess-21-1667-2021, 2021
Short summary
Short summary
A case study on the 2004 Indian Ocean tsunami was conducted at the Phra Thong island, Thailand, using a deep neural network (DNN) inverse model. The model estimated tsunami characteristics from the deposits at Phra Thong island. The uncertainty quantification of the result was evaluated. The predicted flow conditions and the depositional characteristics were compared with the reported observed values. This DNN model can serve as an essential tool for tsunami hazard mitigation at coastal cities.
Related subject area
Cross-cutting themes: Quantitative and statistical methods in Earth surface dynamics
Introducing standardized field methods for fracture-focused surface process research
Full four-dimensional change analysis of topographic point cloud time series using Kalman filtering
Comparison of rainfall generators with regionalisation for the estimation of rainfall erosivity at ungauged sites
Automated quantification of floating wood pieces in rivers from video monitoring: a new software tool and validation
Particle size dynamics in abrading pebble populations
Computing water flow through complex landscapes – Part 3: Fill–Spill–Merge: flow routing in depression hierarchies
A photogrammetry-based approach for soil bulk density measurements with an emphasis on applications to cosmogenic nuclide analysis
Dominant process zones in a mixed fluvial–tidal delta are morphologically distinct
Identifying sediment transport mechanisms from grain size–shape distributions, applied to aeolian sediments
Determining the optimal grid resolution for topographic analysis on an airborne lidar dataset
Systematic identification of external influences in multi-year microseismic recordings using convolutional neural networks
Earth's surface mass transport derived from GRACE, evaluated by GPS, ICESat, hydrological modeling and altimetry satellite orbits
The R package “eseis” – a software toolbox for environmental seismology
Bayesian inversion of a CRN depth profile to infer Quaternary erosion of the northwestern Campine Plateau (NE Belgium)
A new CT scan methodology to characterize a small aggregation gravel clast contained in a soft sediment matrix
Creative computing with Landlab: an open-source toolkit for building, coupling, and exploring two-dimensional numerical models of Earth-surface dynamics
An introduction to learning algorithms and potential applications in geomorphometry and Earth surface dynamics
Sensitivity analysis and implications for surface processes from a hydrological modelling approach in the Gunt catchment, high Pamir Mountains
Constraining the stream power law: a novel approach combining a landscape evolution model and an inversion method
Martha Cary Eppes, Alex Rinehart, Jennifer Aldred, Samantha Berberich, Maxwell P. Dahlquist, Sarah G. Evans, Russell Keanini, Stephen E. Laubach, Faye Moser, Mehdi Morovati, Steven Porson, Monica Rasmussen, and Uri Shaanan
Earth Surf. Dynam., 12, 35–66, https://doi.org/10.5194/esurf-12-35-2024, https://doi.org/10.5194/esurf-12-35-2024, 2024
Short summary
Short summary
All rocks have fractures (cracks) that can influence virtually every process acting on Earth's surface where humans live. Yet, scientists have not standardized their methods for collecting fracture data. Here we draw on past work across geo-disciplines and propose a list of baseline data for fracture-focused surface process research. We detail the rationale and methods for collecting them. We hope their wide adoption will improve future methods and knowledge of rock fracture overall.
Lukas Winiwarter, Katharina Anders, Daniel Czerwonka-Schröder, and Bernhard Höfle
Earth Surf. Dynam., 11, 593–613, https://doi.org/10.5194/esurf-11-593-2023, https://doi.org/10.5194/esurf-11-593-2023, 2023
Short summary
Short summary
We present a method to extract surface change information from 4D time series of topographic point clouds recorded with a terrestrial laser scanner. The method uses sensor information to spatially and temporally smooth the data, reducing uncertainties. The Kalman filter used for the temporal smoothing also allows us to interpolate over data gaps or extrapolate into the future. Clustering areas where change histories are similar allows us to identify processes that may have the same causes.
Ross Pidoto, Nejc Bezak, Hannes Müller-Thomy, Bora Shehu, Ana Claudia Callau-Beyer, Katarina Zabret, and Uwe Haberlandt
Earth Surf. Dynam., 10, 851–863, https://doi.org/10.5194/esurf-10-851-2022, https://doi.org/10.5194/esurf-10-851-2022, 2022
Short summary
Short summary
Erosion is a threat for soils with rainfall as the driving force. The annual rainfall erosivity factor quantifies rainfall impact by analysing high-resolution rainfall time series (~ 5 min). Due to a lack of measuring stations, alternatives for its estimation are analysed in this study. The best results are obtained for regionalisation of the erosivity factor itself. However, the identified minimum of 60-year time series length suggests using rainfall generators as in this study as well.
Hossein Ghaffarian, Pierre Lemaire, Zhang Zhi, Laure Tougne, Bruce MacVicar, and Hervé Piégay
Earth Surf. Dynam., 9, 519–537, https://doi.org/10.5194/esurf-9-519-2021, https://doi.org/10.5194/esurf-9-519-2021, 2021
Short summary
Short summary
Quantifying wood fluxes in rivers would improve our understanding of the key processes in river ecology and morphology. In this work, we introduce new software for the automatic detection of wood pieces in rivers. The results show 93.5 % and 86.5 % accuracy for piece number and volume, respectively.
András A. Sipos, Gábor Domokos, and János Török
Earth Surf. Dynam., 9, 235–251, https://doi.org/10.5194/esurf-9-235-2021, https://doi.org/10.5194/esurf-9-235-2021, 2021
Short summary
Short summary
Abrasion of sedimentary particles is widely associated with mutual collisions. Utilizing results of individual, geometric abrasion theory and techniques adopted in statistical physics, a new model for predicting the collective mass evolution of large numbers of particles is introduced. Our model uncovers a startling fundamental feature of collective particle dynamics: collisional abrasion may either focus size distributions or it may act in the opposite direction by dispersing the distribution.
Richard Barnes, Kerry L. Callaghan, and Andrew D. Wickert
Earth Surf. Dynam., 9, 105–121, https://doi.org/10.5194/esurf-9-105-2021, https://doi.org/10.5194/esurf-9-105-2021, 2021
Short summary
Short summary
Existing ways of modeling the flow of water amongst landscape depressions such as swamps and lakes take a long time to run. However, as our previous work explains, depressions can be quickly organized into a data structure – the depression hierarchy. This paper explains how the depression hierarchy can be used to quickly simulate the realistic filling of depressions including how they spill over into each other and, if they become full enough, how they merge into one another.
Joel Mohren, Steven A. Binnie, Gregor M. Rink, Katharina Knödgen, Carlos Miranda, Nora Tilly, and Tibor J. Dunai
Earth Surf. Dynam., 8, 995–1020, https://doi.org/10.5194/esurf-8-995-2020, https://doi.org/10.5194/esurf-8-995-2020, 2020
Short summary
Short summary
In this study, we comprehensively test a method to derive soil densities under fieldwork conditions. The method is mainly based on images taken from consumer-grade cameras. The obtained soil/sediment densities reflect
truevalues by generally > 95 %, even if a smartphone is used for imaging. All computing steps can be conducted using freeware programs. Soil density is an important variable in the analysis of terrestrial cosmogenic nuclides, for example to infer long-term soil production rates.
Mariela Perignon, Jordan Adams, Irina Overeem, and Paola Passalacqua
Earth Surf. Dynam., 8, 809–824, https://doi.org/10.5194/esurf-8-809-2020, https://doi.org/10.5194/esurf-8-809-2020, 2020
Short summary
Short summary
We propose a machine learning approach for the classification and analysis of large delta systems. The approach uses remotely sensed data, channel network extraction, and the analysis of 10 metrics to identify clusters of islands with similar characteristics. The 12 clusters are grouped in six main classes related to morphological processes acting on the system. The approach allows us to identify spatial patterns in large river deltas to inform modeling and the collection of field observations.
Johannes Albert van Hateren, Unze van Buuren, Sebastiaan Martinus Arens, Ronald Theodorus van Balen, and Maarten Arnoud Prins
Earth Surf. Dynam., 8, 527–553, https://doi.org/10.5194/esurf-8-527-2020, https://doi.org/10.5194/esurf-8-527-2020, 2020
Short summary
Short summary
In this paper, we introduce a new technique that can be used to identify how sediments were transported to their place of deposition (transport mode). The traditional method is based on the size of sediment grains, ours on the size and the shape. A test of the method on windblown sediments indicates that it can be used to identify the transport mode with less ambiguity, and therefore it improves our ability to extract information, such as climate from the past, from sediment deposits.
Taylor Smith, Aljoscha Rheinwalt, and Bodo Bookhagen
Earth Surf. Dynam., 7, 475–489, https://doi.org/10.5194/esurf-7-475-2019, https://doi.org/10.5194/esurf-7-475-2019, 2019
Short summary
Short summary
Representing the surface of the Earth on an equally spaced grid leads to errors and uncertainties in derived slope and aspect. Using synthetic data, we develop a quality metric that can be used to compare the uncertainties in different datasets. We then apply this method to a real-world lidar dataset, and find that 1 m data have larger error bounds than lower-resolution data. The highest data resolution is not always the best choice – it is important to consider the quality of the data.
Matthias Meyer, Samuel Weber, Jan Beutel, and Lothar Thiele
Earth Surf. Dynam., 7, 171–190, https://doi.org/10.5194/esurf-7-171-2019, https://doi.org/10.5194/esurf-7-171-2019, 2019
Short summary
Short summary
Monitoring rock slopes for a long time helps to understand the impact of climate change on the alpine environment. Measurements of seismic signals are often affected by external influences, e.g., unwanted anthropogenic noise. In the presented work, these influences are automatically identified and removed to enable proper geoscientific analysis. The methods presented are based on machine learning and intentionally kept generic so that they can be equally applied in other (more generic) settings.
Christian Gruber, Sergei Rudenko, Andreas Groh, Dimitrios Ampatzidis, and Elisa Fagiolini
Earth Surf. Dynam., 6, 1203–1218, https://doi.org/10.5194/esurf-6-1203-2018, https://doi.org/10.5194/esurf-6-1203-2018, 2018
Short summary
Short summary
By using a set of evaluation methods involving GPS, ICESat, hydrological modelling and altimetry satellite orbits, we show that the novel radial basis function (RBF) processing technique can be used for processing the Gravity Recovery and Climate Experiment (GRACE) data yielding global gravity field models which fit independent reference values at the same level as commonly accepted global geopotential models based on spherical harmonics.
Michael Dietze
Earth Surf. Dynam., 6, 669–686, https://doi.org/10.5194/esurf-6-669-2018, https://doi.org/10.5194/esurf-6-669-2018, 2018
Short summary
Short summary
Environmental seismology is the study of the seismic signals emitted by Earth surface processes. This emerging research field is at the intersection of many Earth science disciplines. The overarching scope requires free integrative software that is accepted across scientific disciplines, such as R. The article introduces the R package "eseis" and illustrates its conceptual structure, available functions, and worked examples.
Eric Laloy, Koen Beerten, Veerle Vanacker, Marcus Christl, Bart Rogiers, and Laurent Wouters
Earth Surf. Dynam., 5, 331–345, https://doi.org/10.5194/esurf-5-331-2017, https://doi.org/10.5194/esurf-5-331-2017, 2017
Short summary
Short summary
Over very long timescales, 100 000 years or more, landscapes may drastically change. Sediments preserved in these landscapes have a cosmogenic radionuclide inventory that tell us when and how fast such changes took place. In this paper, we provide first evidence of an elevated long-term erosion rate of the northwestern Campine Plateau (lowland Europe), which can be explained by the loose nature of the subsoil.
Laurent Fouinat, Pierre Sabatier, Jérôme Poulenard, Jean-Louis Reyss, Xavier Montet, and Fabien Arnaud
Earth Surf. Dynam., 5, 199–209, https://doi.org/10.5194/esurf-5-199-2017, https://doi.org/10.5194/esurf-5-199-2017, 2017
Short summary
Short summary
This study focuses on the creation of a novel CT scan methodology at the crossroads between medical imagery and earth sciences. Using specific density signatures, pebbles and/or organic matter characterizing wet avalanche deposits can be quantified in lake sediments. Starting from AD 1880, we were able to identify eight periods of higher avalanche activity from sediment cores. The use of CT scans, alongside existing approaches, opens up new possibilities in a wide variety of geoscience studies.
Daniel E. J. Hobley, Jordan M. Adams, Sai Siddhartha Nudurupati, Eric W. H. Hutton, Nicole M. Gasparini, Erkan Istanbulluoglu, and Gregory E. Tucker
Earth Surf. Dynam., 5, 21–46, https://doi.org/10.5194/esurf-5-21-2017, https://doi.org/10.5194/esurf-5-21-2017, 2017
Short summary
Short summary
Many geoscientists use computer models to understand changes in the Earth's system. However, typically each scientist will build their own model from scratch. This paper describes Landlab, a new piece of open-source software designed to simplify creation and use of models of the Earth's surface. It provides off-the-shelf tools to work with models more efficiently, with less duplication of effort. The paper explains and justifies how Landlab works, and describes some models built with it.
Andrew Valentine and Lara Kalnins
Earth Surf. Dynam., 4, 445–460, https://doi.org/10.5194/esurf-4-445-2016, https://doi.org/10.5194/esurf-4-445-2016, 2016
Short summary
Short summary
Learning algorithms are powerful tools for understanding and working with large data sets, particularly in situations where any underlying physical models may be complex and poorly understood. Such situations are common in geomorphology. We provide an accessible overview of the various approaches that fall under the umbrella of "learning algorithms", discuss some potential applications within geomorphometry and/or geomorphology, and offer advice on practical considerations.
E. Pohl, M. Knoche, R. Gloaguen, C. Andermann, and P. Krause
Earth Surf. Dynam., 3, 333–362, https://doi.org/10.5194/esurf-3-333-2015, https://doi.org/10.5194/esurf-3-333-2015, 2015
Short summary
Short summary
A semi-distributed hydrological model is used to analyse the hydrological cycle of a glaciated high-mountain catchment in the Pamirs.
We overcome data scarcity by utilising various raster data sets as meteorological input. Temperature in combination with the amount of snow provided in winter play the key role in the annual cycle.
This implies that expected Earth surface processes along precipitation and altitude gradients differ substantially.
T. Croissant and J. Braun
Earth Surf. Dynam., 2, 155–166, https://doi.org/10.5194/esurf-2-155-2014, https://doi.org/10.5194/esurf-2-155-2014, 2014
Cited articles
Allen, J. R. L.: Sedimentary structures: their character and physical basis,
vol. 1, Elsevier Scientific Publishing Company, Amsterdam, 1982. a
Allen, J. R. L.: The Bouma division A and the possible duration of turbidity
currents, J. Sediment. Res., 61, 291–295, 1991. a
Amy, L. A. and Talling, P. J.: Anatomy of turbidites and linked debrites based
on long distance (120 × 30 km) bed correlation, Marnoso Arenacea
Formation, Northern Apennines, Italy, Sedimentology, 53, 161–212,
https://doi.org/10.1111/j.1365-3091.2005.00756.x, 2006. a
Amy, L. A., Kneller, B., and McCaffrey, W.: Evaluating the links between
turbidite characteristics and gross system architecture: upscaling insights
from the turbidite sheet-system of Peïra Cava, SE France, in: Deep
Water Reservoirs of the World, Gulf Coast Section SEPM Foundation 20th Annual
Research Conference, 3–6 December 2000, SEPM, Houston, TX (CD-ROM), 1, p. 15, 2000. a
Amy, L. A., McCaffrey, W. D., and Kneller, B. C.: The influence of a lateral
basin-slope on the depositional patterns of natural and experimental
turbidity currents, Geological Society, London, Special Publications, 221,
311–330, https://doi.org/10.1144/GSL.SP.2004.221.01.17, 2004. a
Azpiroz-Zabala, M., Cartigny, M. J. B., Talling, P. J., Parsons, D. R., Sumner,
E. J., Clare, M. A., Simmons, S. M., Cooper, C., and Pope, E. L.: Newly
recognized turbidity current structure can explain prolonged flushing of
submarine canyons, Science Advances, 3, e1700200,
https://doi.org/10.1126/sciadv.1700200, 2017. a
Baas, J. H., van Dam, R. L., and Storms, J. E.: Duration of deposition from
decelerating high-density turbidity currents, Sediment. Geol., 136,
71–88, https://doi.org/10.1016/S0037-0738(00)00088-9, 2000. a
Bartolini, C., Gehin, C., and Stanley, D. J.: Morphology and recent sediments
of the Western Alboran Basin in the Mediterranean Sea, Mar. Geol., 13,
159–223, https://doi.org/10.1016/0025-3227(72)90089-8, 1972. a
Bishop, C. M.: Pattern recognition and machine learning, Springer Verlag, New York, 2006. a
Bornhold, B. D. and Lilkey, O. H.: Bioclastic Turbidite Sedimentation in
Columbus Basin, Bahamas, GSA Bulletin, 82, 1341–1354,
https://doi.org/10.1130/0016-7606(1971)82[1341:BTSICB]2.0.CO;2, 1971. a
Bouma, A. H.: Sedimentology of some flysch deposits: a graphic approach to
facies interpretation, Elsevier Scientific Publishing Company, Amsterdam, 168,
1962. a
Bowen, A., Normark, W., and Piper, D.: Modelling of turbidity currents on Navy
submarine fan, California continental borderland, Sedimentology, 31,
169–185, https://doi.org/10.1111/j.1365-3091.1984.tb01957.x, 1984. a
Buscail, R. and Germain, C.: Present-day organic matter sedimentation on the
NW Mediterranean margin: Importance of off-shelf export, Limnol.
Oceanogr., 42, 217–229, https://doi.org/10.4319/lo.1997.42.2.0217, 1997. a
Clare, M. A., Talling, P. J., Challenor, P., Malgesini, G., and Hunt, J.:
Distal turbidites reveal a common distribution for large (>0.1 km3)
submarine landslide recurrence, Geology, 42, 263–266, 2014. a
Crank, J.: Free and moving boundary problems, Clarendon Press, Oxford, 1984. a
Dietrich, W. E.: Settling velocity of natural particles, Water Resour.
Res., 18, 1615–1626, https://doi.org/10.1029/WR018i006p01615, 1982. a
Dorrell, R., Amy, L., Peakall, J., and McCaffrey, W.: Particle size
distribution controls the threshold between net sediment erosion and
deposition in suspended load dominated flows, Geophys. Res. Lett.,
45, 1443–1452, 2018. a
Duchi, J., Hazan, E., and Singer, Y.: Adaptive subgradient methods for online
learning and stochastic optimization, J. Mach. Learn. Res.,
12, 2121–2159, 2011. a
Falcini, F., Marini, M., Milli, S., and Moscatelli, M.: An inverse problem to
infer paleoflow conditions from turbidites, J. Geophys. Res.-Oceans, 114, C10019, https://doi.org/10.1029/2009JC005294, 2009. a, b
Fildani, A., Normark, W. R., Kostic, S., and Parker, G.: Channel formation by flow stripping: Large‐scale scour features along the Monterey East Channel and their relation to sediment waves, Sedimentology, 53, 1265–1287, 2006. a
Garcia, M. and Parker, G.: Entrainment of bed sediment into suspension, J.
Hydraul. Eng., 117, 414–435,
https://doi.org/10.1061/(ASCE)0733-9429(1991)117:4(414), 1991. a
Garcia, M. H.: Depositing and eroding sediment-driven flows: turbidity
currents, Tech. Rep. 306, St. Anthony Falls Hydraulic Laboratory,
available at: http://hdl.handle.net/11299/108513 (last access: 28 August 2021), 1990. a
Glorot, X. and Bengio, Y.: Understanding the difficulty of training deep
feedforward neural networks, in: Proceedings of the thirteenth international
conference on artificial intelligence and statistics, Proceedings of Machine Learning Research, 9, 249–256, 2010. a
Glorot, X., Bordes, A., and Bengio, Y.: Deep sparse rectifier neural networks,
in: Proceedings of the fourteenth international conference on artificial
intelligence and statistics, 15, 315–323, available at: http://proceedings.mlr.press/v15/glorot11a/glorot11a.pdf (last access: 31 August 2021), 2011. a, b
Goldfinger, C., Nelson, C. H., and Johnson, J. E.: Holocene earthquake records
from the Cascadia subduction zone and northern San Andreas fault based on
precise dating of offshore turbidites, Annu. Rev. Earth Pl.
Sc., 31, 555–577, https://doi.org/10.1146/annurev.earth.31.100901.141246, 2003. a
Harms, J. and Fahnestock, R.: Stratification, bed forms, and flow phenomena
(with an example from the Rio Grande), in: Primary Sedimentary Structures
(SP12), edited by: Middleton, G. V., Special Publications of
SEPM, Tulsa, 84–115, 1960. a
Hastie, T., Tibshirani, R., and Friedman, J.: The elements of statistical
learning, 2nd edition, Springer, New York, 2009. a
Hecht-Nielsen, R.: Kolmogorov’s mapping neural network existence theorem, in:
Proceedings of the international conference on Neural Networks, 3,
IEEE Press New York, 11–14, 1987. a
Hesse, R.: Long-Distance Continuity of Turbidites: Possible Evidence for an
Early Cretaceous Trench–Abyssal Plain in the East Alps, GSA Bulletin, 85,
859–870, https://doi.org/10.1130/0016-7606(1974)85<859:LCOTPE>2.0.CO;2, 1974. a
Heussner, S., Durrieu de Madron, X., Radakovitch, O., Beaufort, L., E.
Biscaye, P., Carbonne, J., Nicole Delsaut, Etcheber, H., and Monaco, A.:
Spatial and temporal patterns of downward particle fluxes on the continental
slope of the Bay of Biscay (northeastern Atlantic), Deep-Sea Res. Pt.
II, 46, 2101–2146,
https://doi.org/10.1016/S0967-0645(99)00057-0, 1999. a
Hirano, M.: River-bed degradation with armoring, Proceedings of the Japan
Society of Civil Engineers, 1971, 55–65, 1971. a
Hirayama, J. and Nakajima, T.: Analytical study of turbidites, Otadai
Formation, Boso Peninsula, Japan, Sedimentology, 24, 747–779,
https://doi.org/10.1111/j.1365-3091.1977.tb01914.x, 1977. a, b
Hiscott, R. N.: Loss of capacity, not competence, as the fundamental process
governing deposition from turbidity currents, J. Sediment. Res., 64, 209–214, 1994. a
Hughes Clarke, J. E.: First wide-angle view of channelized turbidity currents
links migrating cyclic steps to flow characteristics, Nat. Commun.,
7, 1–13, https://doi.org/10.1038/ncomms11896, 2016. a, b
Ishihara, Y., Miyata, Y., and Tokuhashi, S.: Time-series analysis of turbidite
sequence in the upper part of the Awa Group in the Boso Peninsula, central
Japan, Journal of Geological Society of Japan, 103, 579–589, 1997. a
Jaffe, B. E. and Gelfenbaum, G.: A simple model for calculating tsunami flow
speed from tsunami deposits, Sediment. Geol., 200, 347–361, 2007. a
Kendrick, J.: Turbidite reservoir architecture in the Gulf of Mexico-insights
from field development, in: EAGE/AAPG 3rd Research Symposium-Developing and
Managing Turbidite Reservoirs, cp–100, European Association of
Geoscientists & Engineers, 3–9 October 1998, Almeria, Spain, 1998. a
Kostic, S. and Parker, G.: The response of turbidity currents to a canyon–fun
transition: Internal hydraulic jumps and depositional signatures, J.
Hydraul. Res., 44, 631–653, https://doi.org/10.1080/00221686.2006.9521713, 2006. a, b, c, d
Krizhevsky, A., Sutskever, I., and Hinton, G. E.: Imagenet classification with
deep convolutional neural networks, in: Advances in neural information
processing systems, Curran Associates Inc., New York, 1097–1105, 2012. a
Kubo, Y.: How to estimate the paleocurrent velocities from turbidites: some
examples of the Plio-Pleistocene Kakegawa Gorup, J. Geol.
Soc. Jpn., 101, 925–932, https://doi.org/10.5575/geosoc.101.925, 1995. a
Laloy, E., Hérault, R., Jacques, D., and Linde, N.: Training-image based
geostatistical inversion using a spatial generative adversarial neural
network, Water Resour. Res., 54, 381–406, 2018. a
LeCun, Y., Bengio, Y., and Hinton, G.: Deep learning, Nature, 521, 436–444,
https://doi.org/10.1038/nature14539, 2015. a, b
Lesshafft, L., Meiburg, E., Kneller, B., and Marsden, A.: Towards inverse
modeling of turbidity currents: The inverse lock-exchange problem, Comput.
Geosci., 37, 521–529, https://doi.org/10.1016/j.cageo.2010.09.015, 2011. a, b, c
Mitra, R., Naruse, H., and Abe, T.: Estimation of Tsunami Characteristics from
Deposits: Inverse Modeling using a Deep-Learning Neural Network, J.
Geophys. Res.-Earth, 125, e2020JF005583,
https://doi.org/10.1029/2020JF005583, 2020. a, b
Nair, V. and Hinton, G. E.: Rectified Linear Units Improve Restricted
Boltzmann Machines, International Conference on Machine Learning (ICML 2010), Haifa, Israel, 21–24 June 2010,
807–814, available at:
https://icml.cc/Conferences/2010/papers/432.pdf (last access: 28 August 2021), 2010. a
Naruse, H.: nninv1d, Zenodo [code], https://doi.org/10.5281/zenodo.5167581, 2021a. a
Naruse, H.: Training and test data set of thickness distribution of turbidites, Zenodo [data set], https://doi.org/10.5281/zenodo.5167615, 2021b. a
Naruse, H. and Abe, T.: Inverse Tsunami Flow Modeling Including Nonequilibrium
Sediment Transport, With Application to Deposits From the 2011 Tohoku-Oki
Tsunami, J. Geophys. Res.-Earth, 122, 2159–2182,
https://doi.org/10.1002/2017JF004226, 2017. a
Naruse, H., Sequeiros, O., Garcia, M., Parker, G., Endo, N., Kataoka, K.,
Yokokawa, M., Muto, T., Dohmen-Janssen, C., and Hulscher, S.:
Self-accelerating turbidity currents at laboratory scale, River, Coastal and
Estuarine Morphodynamics: London, Taylor & Francis, London, 473–476, 2007. a
Nesterov, Y. E.: A method for solving the convex programming problem with
convergence rate O(
Ohata, K., Naruse, H., Yokokawa, M., and Viparelli, E.: New bedform phase
diagrams and discriminant functions for formative conditions of bedforms in
open-channel flows, J. Geophys. Res.-Earth, 122,
2139–2158, 2017. a
Pantopoulos, G., Vakalas, I., Maravelis, A., and Zelilidis, A.: Statistical
analysis of turbidite bed thickness patterns from the Alpine fold and thrust
belt of western and southeastern Greece, Sediment. Geol., 294, 37–57,
https://doi.org/10.1016/j.sedgeo.2013.05.007, 2013. a
Parker, G., Fukushima, Y., and Pantin, H. M.: Self-accelerating turbidity
currents, J. Fluid Mech., 171, 145–181,
https://doi.org/10.1017/S0022112086001404, 1986. a, b, c, d
Parker, G., Garcia, M., Fukushima, Y., and Yu, W.: Experiments on turbidity
currents over an erodible bed, J. Hydraul. Res., 25, 123–147,
https://doi.org/10.1080/00221688709499292, 1987. a
Parkinson, S. D., Funke, S. W., Hill, J., Piggott, M. D., and Allison, P. A.: Application of the adjoint approach to optimise the initial conditions of a turbidity current with the AdjointTurbidity 1.0 model, Geosci. Model Dev., 10, 1051–1068, https://doi.org/10.5194/gmd-10-1051-2017, 2017. a, b
Paull, C. K., Talling, P. J., Maier, K. L., Parsons, D., Xu, J., Caress, D. W., Gwiazda, R., Lundsten, E. M., Anderson, K., Barry, J. P., Chaffey, M., O’Reilly, T., Rosenberger, K. J., Gales, J. A., Kieft, B., McGann, M., Simmons, S. M., McCann, M., Sumner, E. J., Clare, M. A., and Cartigny, M. J.: Powerful
turbidity currents driven by dense basal layers, Nat. Commun., 9,
1–9, 2018. a, b, c
Pilkey, O. H., Locker, S. D., and Cleary, W. J.: Comparison of Sand-Layer
Geometry on Flat Floors of 10 Modern Depositional Basins1, AAPG Bull.,
64, 841–856, https://doi.org/10.1306/2F9193C3-16CE-11D7-8645000102C1865D, 1980. a
Raschka, S. and Mirjalili, V.: Python machine learning: Machine learning and
deep learning with Python, scikit-learn, and TensorFlow 2, Packt Publishing
Ltd, Birmingham, UK, 2019. a
Rossano, S., Mastrolorenzo, G., De Natale, G., and Pingue, F.: Computer
simulation of pyroclastic flow movement: An inverse approach, Geophys.
Res. Lett., 23, 3779–3782, https://doi.org/10.1029/96GL03570, 1996. a
Rumelhart, D. E., Hinton, G. E., and Williams, R. J.: Learning representations
by back-propagating errors, Nature, 323, 533–536,
https://doi.org/10.1038/323533a0, 1986. a
Scheidegger, A. E. and Potter, P. E.: TEXTURAL STUDIES OF GRADED BEDDING.
OBSERVATION AND THEORY, Sedimentology, 5, 289–304,
https://doi.org/10.1111/j.1365-3091.1965.tb01563.x, 1965. a
Sequeiros, O. E., Naruse, H., Endo, N., Garcia, M. H., and Parker, G.:
Experimental study on self-accelerating turbidity currents, J.
Geophys. Res.-Oceans, 114, C05025, https://doi.org/10.1029/2008JC005149,
2009. a
Shalev-Shwartz, S. and Ben-David, S.: Understanding machine learning: From
theory to algorithms, Cambridge University Press, New York, USA, 2014. a
Specht, D. F.: A general regression neural network, IEEE T. Neural
Networ., 2, 568–576, 1991. a
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov,
R.: Dropout: A Simple Way to Prevent Neural Networks from Overfitting,
J. Mach. Learn. Res., 15, 1929–1958, 2014. a
Sun, A. Y.: Discovering State-Parameter Mappings in Subsurface Models Using
Generative Adversarial Networks, Geophys. Res. Lett., 45,
11137–11146, https://doi.org/10.1029/2018GL080404, 2018. a
Talling, P. J.: On the triggers, resulting flow types and frequencies of
subaqueous sediment density flows in different settings, Mar. Geol., 352,
155–182, 2014. a
Talling, P. J., Masson, D. G., Sumner, E. J., and Malgesini, G.: Subaqueous
sediment density flows: Depositional processes and deposit types,
Sedimentology, 59, 1937–2003, https://doi.org/10.1111/j.1365-3091.2012.01353.x, 2012. a
Tieleman, T. and Hinton, G.: Lecture 6.5-rmsprop, coursera: Neural networks for
machine learning, Tech. rep., University of Tronto, Tronto, Canada, 2012. a
Tokuhashi, S.: Three dimensional analysis of a large sandy-flysch body,
Mio-Pliocene Kiyosumi Formation, Boso Peninsula, Japan, 1979. a
Tokuhashi, S.: Two stages of submarine fan sedimentation in an ancient forearc
basin, central Japan, Sedimentary facies in the active plate margin, TERRAPUB, Tokyo, Japan,
439–468, 1989. a
van Tassell, J.: Silver Abyssal Plain Carbonate Turbidite: Flow
Characteristics, J. Geol., 89, 317–333, https://doi.org/10.1086/628594,
1981. a
Vangriesheim, A., Khripounoff, A., and Crassous, P.: Turbidity events observed
in situ along the Congo submarine channel, Deep-Sea Res. Pt. II, 56, 2208–2222,
https://doi.org/10.1016/j.dsr2.2009.04.004, 2009. a
Walker, R. G.: The origin and significance of teh internal sedimentary
structures of turbidites, P. Yorks. Geol. Soc.,
35, 1–32, https://doi.org/10.1144/pygs.35.1.1, 1965. a
Xu, J. P.: Normalized velocity profiles of field-measured turbidity currents,
Geology, 38, 563–566, https://doi.org/10.1130/G30582.1, 2010. a
Xu, J. P., Noble, M. A., and Rosenfeld, L. K.: In-situ measurements of
velocity structure within turbidity currents, Geophys. Res. Lett.,
31, L09311, https://doi.org/10.1029/2004GL019718, 2004. a
Yoneda, J., Masui, A., Konno, Y., Jin, Y., Egawa, K., Kida, M., Ito, T., Nagao,
J., and Tenma, N.: Mechanical properties of hydrate-bearing turbidite
reservoir in the first gas production test site of the Eastern Nankai
Trough, Marine Petrol. Geol., 66, 471–486, 2015. a
Zeiler, M. D.: Adadelta: an adaptive learning rate method, arXiv preprint,
arXiv:1212.5701 (last access: 28 August 2021), 2012. a
Short summary
This paper proposes a method to reconstruct the hydraulic conditions of turbidity currents from turbidites. We investigated the validity and problems of this method in application to actual field datasets using artificial data. Once this method is established, it is expected that the method will elucidate the generation process of turbidity currents and will help to predict the geometry of resultant turbidites in deep-sea environments.
This paper proposes a method to reconstruct the hydraulic conditions of turbidity currents from...
