18 March 2021
Retrogressive breach failures
Posted by Dave Petley
Retrogressive breach failures
Earlier this week I posted a piece about the remarkable failure at Knappensee in eastern Germany, which was caught on video. The failure, on the banks of a flooded open cast coal mine, was retrogressive, generating a substantial displacement wave. Loyal reader Konrad Beinssen from Australia posted a description of this type of failure, which is termed a “retrogressive breach failure”. As I wasn’t aware of this particular failure type, I thought I would post his comment here.
This type of soil failure now known as a ‘retrogressive breach failure’ is not uncommon in coastal, river and lake settings, worldwide. Notable locations are the lower Mississippi, Inskip and Amity Points in Eastern Australia and in the sandy estuaries of the Netherlands.
A small triggering event in submerged (saturated) sand starts the process of ‘breaching’, well known to operators in the dredging industry. A near vertical wall of sand (the breach) migrates upslope. This wall is temporally stabilised by dilation of the sand skeleton immediately behind the breach causing suction (strain hardening). Granular material (such as sand) is shed from the breach as it migrates upslope and this generates a density current which transports grains offshore before they loose momentum and settle to the bottom.
As the breach arrives at the shore and migrates further inland, it undermines material above the water table, which topples in and adds further to the density (turbidity) current.
Konrad points us to a paper (Mastbergen et al. 2019), which is open access, which describes this failure mechanism in more detail. It includes this explanatory diagram:

A schematic diagram of the retogressive breach failure concept. From Mastbergen et al. (2019).
.
The paper includes a number of other examples of these types of failure. It includes the Inskip landslide in Australia, which I also featured on this blog.
.
Reference
Mastbergen, D.R.; Beinssen, K.; Nédélec, Y. 2019. “Watching the Beach Steadily Disappearing: The Evolution of Understanding of Retrogressive Breach Failures” Journal of Marine Science and Engineering, 7 (10):368. https://doi.org/10.3390/jmse7100368.
I am wondering if this process describes what has been occurring along the Murray River in Australia over many years?
Yes, river bank collapses along the Murray River can be explained by the RBF process.
Firstly, they can be observed at the Murray Mouth where the dredges operate. Scallops in the beach adjacent to dredge operations are common. Here this form of erosion is of no consequence.
Further up the river, observed scallops in the sandy river banks can also be explained by the RBF process. I would assume that river bank sand is well packed and therefore ‘dilatant’ when subjected to shear. (The ‘shear dilation’ effect is demonstrated by the drying around a footprint on a saturated beach).
The retrogression rate of the Knappensee failure was unusually fast. This may be explained by the retrogressing breach (steep underwater sand wall) encountering loosely packed pockets of sand as it migrated upslope. Loosely packed sand under shear can collapse rather than dilate. In this case, pore pressure rises sometimes to the point of liquefaction. The whole weight of the sand body is then carried by the pore water and so it behaves as a fluid. Very rapid collapse then happens. We will probably never fully know the complexities of this event.
I am reminded of the ’60-’70’s explanation for tubidite deposits in submarine, near-coastal marine geology. Longshore drift moves surf-zone sediment to a zone of accumulation, the unconsolidated shelf deposit fails, triggered by storm waves, tsunami or local seismic shaking, flows through a submarine canyon as turbid flow, deposited as a turbidite fan on the deep sea floor,
A major retrogressive beach landslide occurred on 17 September 1997 and lasted for one hour in a newly dredged slope for port development in the city of Tianjin, northern China. Details can be found in the following two papers.
1. S Li, ZQ Yue, LG Tham, CF Lee, SW Yan, 2005. Slope failure in under-consolidated soft soils during the development of a port in Tianjin, China. Part 1, field investigation, Canadian Geotechnical Journal. 42(1) : 147-165.
2. S Li, ZQ Yue, LG Tham, CF Lee, SW Yan, 2005. Slope failure in under-consolidated soft soils during the development of a port in Tianjin, China. Part 2, analytical study, Canadian Geotechnical Journal. 42(1) : 166-183
Go back a few years:
https://blogs.agu.org/landslideblog/2010/10/29/an-amazing-landslide-video-from-manaus-in-brazil/
https://youtu.be/9cf6nR4X894 This video is speeded up
This slide is, I believe, a result of loose hydraulic fill (not quick clay).