A Deep Learning Approach for Estimation of the Nearshore Bathymetry

Benshila, Rachid; Thoumyre, Grégoire; Najar, Mahmoud Al; Ondoa, G. Abessolo; Almar, Rafaël; Bergsma, Erwin W. J.; Hugonnard, Guillaume; Labracherie, L.; Lavie, Benjamin; Ragonneau, Tom; Simon, Ehouarn; Vieuble, Bastien; Wilson, Dennis G.

doi:10.2112/si95-197.1

Cited by 19 publications

(12 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial intelligence (AI) brings new perspectives in terms of processing of dense EO datasets in a drastically reduced time, but also in solving complex coastal environmental problems (Shulz et al 2018) such as monitoring of worldwide coastal sea surface temperature and salinity at high-resolution (Medina-Lopez and Ureña-Fuentes 2019), classification of land use/cover (Aghighi et al 2014), seagrass distribution in coastal waters (Perez et al 2018), bathymetry inversion (Danilo and Melgani 2019;Benshila et al 2020) and sediment transport and morphological evolution (Goldstein et al 2019). All these methods are now applicable to spatial time series and can be transposed to most of EO satellite sensors when applied to coastal variables.…”

Section: Discussionmentioning

confidence: 99%

Earth Observations for Monitoring Marine Coastal Hazards and Their Drivers

et al. 2020

View full text Add to dashboard Cite

Coastal zones have large social, economic and environmental values. They are more densely populated than the hinterland and concentrate large economic assets, critical infrastructures and human activities such as tourism, fisheries, navigation. Furthermore, coastal oceans are home to a wealth of living marine resources and very productive ecosystems. Yet, coastal zones are exposed to various natural and anthropogenic hazards. To reduce the risks associated with marine hazards, sustained coastal zone monitoring programs, forecasting and early warning systems are increasingly needed. Earth observations (EO), and in particular satellite remote sensing, provide invaluable information: satellite-borne sensors allow an effective monitoring of the quasi-global ocean, with synoptic views of large areas, good spatial and temporal resolution, and sustained time-series covering several years to decades. However, satellite observations do not always meet the precision required by users, in particular in dynamic coastal zones, characterized by shorter-scale variability. A variety of sensors are used to directly monitor the coastal zone and their observations can also be integrated into numerical models to provide a full 4D monitoring of the ocean and forecasts. Here, we review how EO, and more particularly satellite observations, can monitor coastal hazards and their drivers. These include coastal flooding, shoreline changes, maritime security, marine pollution, water quality, and marine ecology shifts on the one hand, and several physical characteristics (bathymetry, topography, vertical land motion) of coastal zones, meteorological and oceanic (metocean) variables that can act as forcing factors for coastal hazards on the other hand.

show abstract

Section: Discussionmentioning

confidence: 99%

Earth Observations for Monitoring Marine Coastal Hazards and Their Drivers

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Cloud computing services such as Google Earth Engine (Li et al, 2021), Microsoft Azure, Amazon AWS, and Copernicus DIAS, then make it possible for multiple new users to compute and access this new type of information. A recently increasing number of works make use of machine learning for SDB (Sagawa et al, 2019), bringing great expectations to solve satellite-based bathymetry issues in areas of complex physics and environmental parameters, by merging different methods and speeding up computation time (Danilo and Melgani, 2016;Benshila et al, 2020). Coastal morphology changes over a wide range of timescales (from storm events, seasonal and interannual variability to longer-term adaptation to changing environmental conditions), in particular in response to changing incoming wave regimes (Karunarathna et al, 2016;Bergsma et al, 2019) and human interventions.…”

Section: Perspectives and Recommendationsmentioning

confidence: 99%

Pan-European Satellite-Derived Coastal Bathymetry—Review, User Needs and Future Services

et al. 2021

Self Cite

View full text Add to dashboard Cite

Low-lying coastal zones are home to around 10% of the world’s population and to many megacities. Coastal zones are largely vulnerable to the dynamics of natural and human-induced changes. Accurate large-scale measurements of key parameters, such as bathymetry, are needed to understand and predict coastal changes. However, nearly 50% of the world’s coastal waters remain unsurveyed and for a large number of coastal areas of interest, bathymetric information is unavailable or is often decades old. This lack of information is due to the high costs in time, money and safety involved in collecting these data using conventional echo sounder on ships or LiDAR on aircrafts. Europe is no exception, as European seas are not adequately surveyed according to the International Hydrographic Organisation. Bathymetry influences ocean waves and currents, thereby shaping sediment transport which may alter coastal morphology over time. This paper discusses state-of-the-art coastal bathymetry retrieval methods and data, user requirements and key drivers for many maritime sectors in Europe, including advances in Satellite-Derived Bathymetry (SDB). By leveraging satellite constellations, cloud services and by combining complementary methods, SDB appears as an effective emerging tool with the best compromise in time, coverage and investment to map coastal bathymetry and its temporal evolution.

show abstract

“…The most convincing application of deep learning to coastal SDB currently appears to be from [39], which uses reflectance values from Sentinel-2 Level 2A images to estimate coastal water depth with high precision in clear waters (1.48 m RMSE). While machine learning and deep learning applications for satellite-derived bathymetry have-until now-mainly been applied to color-based approaches, great expectations come from the combination of different methods, in particular, based on wave information [40,41].…”

Section: Introductionmentioning

confidence: 99%

Coastal Bathymetry Estimation from Sentinel-2 Satellite Imagery: Comparing Deep Learning and Physics-Based Approaches

et al. 2022

Self Cite

View full text Add to dashboard Cite

The ability to monitor the evolution of the coastal zone over time is an important factor in coastal knowledge, development, planning, risk mitigation, and overall coastal zone management. While traditional bathymetry surveys using echo-sounding techniques are expensive and time consuming, remote sensing tools have recently emerged as reliable and inexpensive data sources that can be used to estimate bathymetry using depth inversion models. Deep learning is a growing field of artificial intelligence that allows for the automatic construction of models from data and has been successfully used for various Earth observation and model inversion applications. In this work, we make use of publicly available Sentinel-2 satellite imagery and multiple bathymetry surveys to train a deep learning-based bathymetry estimation model. We explore for the first time two complementary approaches, based on color information but also wave kinematics, as inputs to the deep learning model. This offers the possibility to derive bathymetry not only in clear waters as previously done with deep learning models but also at common turbid coastal zones. We show competitive results with a state-of-the-art physical inversion method for satellite-derived bathymetry, Satellite to Shores (S2Shores), demonstrating a promising direction for worldwide applicability of deep learning models to inverse bathymetry from satellite imagery and a novel use of deep learning models in Earth observation.

show abstract

A Deep Learning Approach for Estimation of the Nearshore Bathymetry

Cited by 19 publications

References 0 publications

Earth Observations for Monitoring Marine Coastal Hazards and Their Drivers

Earth Observations for Monitoring Marine Coastal Hazards and Their Drivers

Pan-European Satellite-Derived Coastal Bathymetry—Review, User Needs and Future Services

Coastal Bathymetry Estimation from Sentinel-2 Satellite Imagery: Comparing Deep Learning and Physics-Based Approaches

Contact Info

Product

Resources

About