Probabilistic coastal cliff collapse hazard from repeated terrestrial laser surveys: case study from Mesnil Val (Normandy, northern France)

Dewez, Thomas; Rohmer, Jérémy; Regard, Vincent; Cnudde, Céline

doi:10.2112/si65-119.1

Cited by 72 publications

(101 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Mesnil Val is a coastal chalk cliff site which has hosted a series of studies on rock fall and cliff collapse over the years (Dewez and Rohmer, 2013;Dewez et al, 2007;Regard et al, 2012;Senfaute et al, 2009Senfaute et al, , 2005. It is located at the northern end of the Normandy chalk cliffs, along the French coast of the English Channel.…”

Section: Study Sitementioning

confidence: 99%

“…But such average retreat rate does not inform on the size of single collapse events. This is why probabilistic hazard analysis was undertaken, initially by means of Terrestrial Laser Scanners (Dewez et al, 2009(Dewez et al, , 2007Dewez et al, 2013) and here using UAV photogrammetric surveys…”

Section: Study Sitementioning

confidence: 99%

“…Small rockfalls occur far more frequently than larger rockfall, the relationship being controlled by a power-law (e.g. Dewez et al 2013). To establish an empirical probabilistic rockfall hazard relationship on a set of representative rockfall magnitudes, in a matter of a few years to be practical for rockfall risk managers, capturing small rockfalls is paramount because they occur often enough to be seen rapidly.…”

Section: Comparison Of Uav Surveysmentioning

confidence: 99%

“…In the last decade, advances in rock fall hazards has widely benefitted from the topographic measurement capacity of Terrestrial Laser Scanners (Abellán et al, 2010;Dewez and Rohmer, 2013;Rosser et al, 2014). When surveys are repeated at regular time intervals of a few weeks or months on a cliff face, topographic changes reveal the scars of rock falls.…”

Section: Introductionmentioning

confidence: 99%

“…Dewez and Rohmer, 2013;Rohmer and Dewez, 2015). A method to compute the probability of cliff collapse from TLS data was proposed by Dewez and Rohmer, (2013) based on a data set collected by TLS between 2005 and 2008 on the coastal chalk cliff of Mesnil Val, in Normandy. The use of this method is obvious to land managers and public safety authorities.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Cliff Collapse Hazard From Repeated Multicopter Uav Acquisitions: Return on Experience

Dewez¹,

Leroux²,

Morelli³

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Cliff collapse poses a serious hazard to infrastructure and passers-by. Obtaining information such as magnitude-frequency relationship for a specific site is of great help to adapt appropriate mitigation measures. While it is possible to monitor hundreds-ofmeter-long cliff sites with ground based techniques (e.g. lidar or photogrammetry), it is both time consuming and scientifically limiting to focus on short cliff sections. In the project SUAVE, we sought to investigate whether an octocopter UAV photogrammetric survey would perform sufficiently well in order to repeatedly survey cliff face geometry and derive rock fall inventories amenable to probabilistic rock fall hazard computation. An experiment was therefore run on a well-studied site of the chalk coast of Normandy, in Mesnil Val, along the English Channel (Northern France). Two campaigns were organized in January and June 2015 which surveyed about 60 ha of coastline, including the 80-m-high cliff face, the chalk platform at its foot, and the hinterland in a matter of 4 hours from start to finish. To conform with UAV regulations, the flight was flown in 3 legs for a total of about 30 minutes in the air. A total of 868 and 1106 photos were respectively shot with a Sony NEX 7 with fixed focal 16mm. Three lines of sight were combined: horizontal shots for cliff face imaging, 45°-oblique views to tie plateau/platform photos with cliff face images, and regular vertical shots. Photogrammetrically derived dense point clouds were produced with Agisoft Photoscan at ultrahigh density (median density is 1 point every 1.7cm). Point cloud density proved a critical parameter to reproduce faithfully the chalk face's geometry. Tuning down the density parameter to "high" or "medium", though efficient from a computational point of view, generated artefacts along chalk bed edges (i.e. smoothing the sharp gradient) and ultimately creating ghost volumes when computing cloud to cloud differences. Yet, from a hazard point of view, this is where small rock fall will most likely occur. Absolute orientation of both point clouds proved unsufficient despite the 30 black and white quadrants ground control point DGPS surveyed. Additional ICP was necessary to reach centimeter-level accuracy and segment rock fall scars corresponding to the expected average daily rock fall volume (ca. 0.013 m3).

show abstract

Section: Study Sitementioning

confidence: 99%

Section: Study Sitementioning

confidence: 99%

Section: Comparison Of Uav Surveysmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Cliff Collapse Hazard From Repeated Multicopter Uav Acquisitions: Return on Experience

Dewez¹,

Leroux²,

Morelli³

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

Spatial and temporal variations in rockfall determined from TLS measurements in a deglaciated valley, Switzerland

et al. 2015

View full text Add to dashboard Cite

Steep U-shaped valleys produced by glaciation are commonly eroded by rockfall. In this study we investigate modern and long-term (power law predicted) rates of rock wall retreat in a deglaciated valley. Our emphasis is on the 5.2 km 2 calcareous cliffs covering in the Lauterbrunnen Valley, Switzerland.Terrestrial laser scans were collected during nine field campaigns over 18 months to provide a continuous coverage of most of the valley walls. Results indicate a total of 122 rockfalls with volumes ranging between 0.06 ± 0.01 and 119.34 ± 1.07 m 3. Different size groups of rockfall events were correlated with environmental factors (e.g., freeze-thaw cycles/temperature, precipitation, and seismicity) using a linear regression with variable lag times of 0-6 months. The highest correlation factor (r = 0.6, P = 0.08, 90% significance level) is observed for freeze-thaw cycles and rockfall events smaller than 1 m 3 with a 2 month delay between temperature extremes and rockfall. Frequency-magnitude relationships for rockfall events were calculated to predict less frequent larger rockfall events that did not occur during the observation period. A Monte Carlo analysis was applied to the frequency-magnitude relationship to evaluate the sensitivity of results to their stochastic nature. An average power law exponent of 1.71 ± 0.09 was calculated using linear regression. The calculated exponents are similar to other studies in calcareous rock settings conducted over different observation durations. Finally, we determine an average total eroded volume of 2:0

show abstract

Mapping Coastal Erosion of a Mediterranean Cliff with a Boat‐Borne Laser Scanner: Performance, Processing, and Cliff Erosion Rate

Giuliano

Dewez

Lebourg

et al. 2022

3D Digital Geological Models

View full text Add to dashboard Cite

Terrestrial laser scanners (TLS) are well known for providing an efficient means to monitor coastal cliff erosion. Cliffs along micro-tidal coasts, however, have often escaped quantification because the narrow or absent coastal platforms do not offer stable and embracing vantage points. To circumvent this issue, mobile laser scanning surveys from a boat can be used. We present a case study from Provence-Alpes-Côte-d'Azur (Mediterranean coastsouthern France) to quantify cliff erosion in such micro-tidal environments. Three surveys were sub-contracted over a period of 17 months to monitor a 3.5-km-long cliff of Carry-le-Rouet (15km west of Marseille). Data quality was check independently using man-made planar walls positioned above the cliff face, to retrieve survey precision and change detection thresholds.Boat-borne mobile lidar system was capable of describing planar features with a precision of 3-4 cm (epoch 1 and 2) and improved to 2.6 cm (epoch 3) with point densities around 100 pts/m². Absolute positioning accuracy varied between 0.1 cm and 0.3 cm. Because the coastline is very sinuous, we describe a method to unfold the point clouds using a continuous analytical surface made of vertical planes joined by arcs of cylinders and perform the analysis in 2.5D.Change was detected using a unique, conservative, threshold of 14 cm (99% quantile estimated on plane change) on grids of 10 x 10 cm pixels. Integrated over the entire cliff face, the average annual cliff recession rate at Carry-le-Rouet is 1.1cm/year. In 17 months, erosion was three times more effective in sandstone and marls layers than in calcarenites and conglomerates.Erosion varies vertically with erosion three times more effective in the lower 25 m of the cliffs than above. Despite imperfections, boat-borne laser scanning system are capable of delivering meaningful erosion data even in this low erosion context.

show abstract

Probabilistic coastal cliff collapse hazard from repeated terrestrial laser surveys: case study from Mesnil Val (Normandy, northern France)

Cited by 72 publications

References 15 publications

Cliff Collapse Hazard From Repeated Multicopter Uav Acquisitions: Return on Experience

Cliff Collapse Hazard From Repeated Multicopter Uav Acquisitions: Return on Experience

Spatial and temporal variations in rockfall determined from TLS measurements in a deglaciated valley, Switzerland

Mapping Coastal Erosion of a Mediterranean Cliff with a Boat‐Borne Laser Scanner: Performance, Processing, and Cliff Erosion Rate

Contact Info

Product

Resources

About