Ensemble models from machine learning: an example of wave runup and coastal dune erosion

Beuzen, Tomas; Goldstein, Evan B.; Splinter, Kristen D.

doi:10.5194/nhess-19-2295-2019

Cited by 39 publications

(17 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Dune vegetation helps to trap dune directed aeolian sediment supply [15,16] and also stabilizes the dune from erosion by wind [17] and waves [18]. Waves are typically considered from an erosive perspective in relation to dune systems e.g., [19][20][21][22][23][24][25]; however, there is a recognition that wave-driven onshore transport can provide sediment reservoirs that subsequently can nourish dunes through aeolian transport [26][27][28]. Spatial variation in foredune evolution has been noted over a range of scales [29][30][31][32].…”

Section: Introductionmentioning

confidence: 99%

Spatial Variation in Coastal Dune Evolution in a High Tidal Range Environment

et al. 2020

View full text Add to dashboard Cite

Coastal dunes have global importance as ecological habitats, recreational areas, and vital natural coastal protection. Dunes evolve due to variations in the supply and removal of sediment via both wind and waves, and on stabilization through vegetation colonization and growth. One aspect of dune evolution that is poorly understood is the longshore variation in dune response to morphodynamic forcing, which can occur over small spatial scales. In this paper, a fixed wing unmanned aerial vehicle (UAV), is used to measure the longshore variation in evolution of a dune system in a megatidal environment. Dune sections to the east and west of the study site are prograding whereas the central portion is static or eroding. The measured variation in dune response is compared to mesoscale intertidal bar migration and short-term measurements of longshore variation in wave characteristics during two storms. Intertidal sand bar migration is measured using satellite imagery: crescentic intertidal bars are present in front of the accreting portion of the beach to the west and migrate onshore at a rate of 0.1–0.2 m/day; episodically the eastern end of the bar detaches from the main bar and migrates eastward to attach near the eastern end of the study area; bypassing the central eroding section. Statistically significant longshore variation in intertidal wave heights were measured using beachface mounted pressure transducers: the largest significant wave heights are found in front of the dune section suffering erosion. Spectral differences were noted with more narrow-banded spectra in this area but differences are not statistically significant. These observations demonstrate the importance of three-dimensionality in intertidal beach morphology on longshore variation in dune evolution; both through longshore variation in onshore sediment supply and through causing longshore variation in near-dune significant wave heights.

show abstract

Section: Introductionmentioning

confidence: 99%

Spatial Variation in Coastal Dune Evolution in a High Tidal Range Environment

et al. 2020

View full text Add to dashboard Cite

show abstract

“…NOAA Wavewatch III, Deltares Global Flood Forecasting and Information System, Aviso Global Tide FES model) can provide deepwater wave, tide and surge information along coastlines where these are not measured locally, however the estimation of wave setup and runup is still likely to be dependent on calculations of sitespecific wave transformation in the nearshore (da Silva et al, 2020). Probabilistic methods such as ensembles (Beuzen et al, 2019a), Monte Carlo (Davidson et al, 2017) and Bayesian networks (Beuzen et al, 2018) are practical approaches that enable uncertainty in local morphology and storm hydrodynamics to be appropriately considered. As the availability of routine coastal observations spanning regional scales continues to expand and modelling tools improve, implementation of the Storm Hazard Matrix within the context of operational Early Warning Systems has the potential to deliver forecasts of coastal storm hazards spanning both wave-dominated and surge-dominated coasts.…”

Section: Synthesis and Future Directionsmentioning

confidence: 99%

A Storm Hazard Matrix combining coastal flooding and beach erosion

Leaman¹,

Harley²,

Splinter³

et al. 2022

Preprint

View full text Add to dashboard Cite

Coastal storms cause widespread damage to property, infrastructure, economic activity and the environment. Along open sandy coastlines, two of the primary coastal storm hazards are coastal flooding by elevated ocean water levels and beach erosion as the result of storm wave action. At continental margins characterized by a shallow, wide continental shelf, coastal storms are more commonly associated with amplified storm surge and the damaging impacts caused by flooding of low-lying land. In contrast, along margins where the continental shelf is narrow and deep, coastal storm impacts are more often characterized by extensive beach erosion, due to the typically lower magnitude of storm surge but a higher proportion of deepwater wave energy reaching the shoreline. A new Storm Hazard Matrix is presented that integrates these two distinct but inherently linked open coast hazards. The approach is based on the combination of two hazard scales. The first is a ‘coastal flooding hazard scale’ that follows an established framework in which hazards are predominately driven by the vertical increase in ocean water levels during storms. The second is a storm wave ‘beach erosion hazard scale’ where hazards are predominately driven by the horizontal recession of the sandy beach and dune. The resulting framework comprises a total of nine unique combinations of flooding/erosion storm hazard regimes, from which six unified, qualitative indicators of the total storm hazard level ranging from ‘Low’ to ‘Extreme’ are defined. Real-world application of the Storm Hazard Matrix is explored at contrasting coastlines for two major storm events, encompassing an extratropical cyclone that impacted the coastline of southeast Australia in June 2016, and Hurricane Ivan that impacted the Gulf Coast of the United States in 2004. The new approach is shown to identify and distinguish between the severity of localized coastal flooding and/or coastal erosion, as well as provide enhanced insight to the nature, magnitude and alongshore variation of coastal storm hazards along the impacted coastline. Within the context of disaster risk reduction, preparedness and operational early warning, implementation of the Storm Hazard Matrix has the potential to deliver robust evaluations of storm hazards spanning a wider variety of both wave-dominated and surge-dominated coasts.

show abstract

“…Given the wealth of available ML algorithms in open-access software, geomorphologists have an unprecedented set of available tools for data exploration and hypothesis testing. Machine learning allows us to teach a computer to learn by example, usefully approximating quantities from readily obtainable data that are otherwise hard to sense , parameterize [Ni et al, 2021, Beuzen et al, 2019, Tinoco et al, 2015, flag for quality control [Sugiura and Hosoda, 2020], or to visualize or make automated inference on high-dimensional datasets that a human could not Stockdon, 2012, Chmiel et al, 2021], especially for phenomena without well-developed theory [Fox et al, 2015, Goldstein and. However, the generation of the right type of examples for the machine to learn, or enough of sufficient quality, is a challenge that requires the development of specialist data labeling tools.…”

Section: The Need For Data Labeling Tools For Earth Surface Processes Researchmentioning

confidence: 99%

Human-in-the-Loop Segmentation of Earth Surface Imagery

Buscombe¹,

Goldstein²,

Sherwood³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Segmentation, or the classification of pixels (grid cells) in imagery, is ubiquitously applied in the natural sciences. Manual methods are often prohibitively time-consuming, especially those images consisting of small objects and/or significant spatial heterogeneity of colors or textures. Labeling complicated regions of transition that in Earth surface imagery are represented by collections of mixed-pixels, -textures, and -spectral signatures, can be especially error-prone because it is difficult to reliably unmix, identify and delineate consistently. However, the success of supervised machine learning (ML) approaches is entirely dependent on good label data. We describe a fast, semi-automated, method for interactive segmentation of N-dimensional (x,y,N) images into two-dimensional (x,y) label images. It uses human-in-the-loop ML to achieve consensus between the labeler and a model in an iterative workflow. The technique is reproducible; the sequence of decisions made by human labeler and ML algorithms can be encoded to file, so the entire process can be played back and new outputs generated with alternative decisions and/or algorithms. We illustrate the scientific potential of segmentation of imagery of diverse settings and image types using six case studies from river, estuarine, and open coast environments. These photographic and non-photographic imagery consist of 1- and 3-bands on regular and irregular grids ranging from centimeters to tens of meters. We demonstrate high levels of agreement in label images generated by several labelers on the same imagery, and make suggestions to achieve consensus and measure uncertainty, ideal for widespread application in training supervised ML for image segmentation.

show abstract

Ensemble models from machine learning: an example of wave runup and coastal dune erosion

Cited by 39 publications

References 55 publications

Spatial Variation in Coastal Dune Evolution in a High Tidal Range Environment

Spatial Variation in Coastal Dune Evolution in a High Tidal Range Environment

A Storm Hazard Matrix combining coastal flooding and beach erosion

Human-in-the-Loop Segmentation of Earth Surface Imagery

Contact Info

Product

Resources

About