John N. Hutchinson

The Fourth Glossop Lecture

In this lecture, the process of reading the ground is explored, some of the necessary background is defined and ways of improving our present and future competence in this area are outlined. It acknowledges the world-wide scope of this activity, but employs mainly the narrower canvas of mass movements in NW European conditions to illustrate its nature. Some basic tools which can assist in this process are first noted, good databases, the techniques of initial site appraisal, comprising desk study, site reconnaissance and mapping (with the aid of stereoscopic air photo interpretation and other remote sensing techniques), terrain modelling and the use of associated well-logged and sampled trial trenches. The importance of the classification of mass movements, not least to develop an agreed terminology, is touched upon and the great value, even at this early stage, of the influence line approach in assessing rapidly the effects of proposed cuts and fills is noted. An attempt is then made to identify the relevant vocabulary for reading the ground, that is those physical phenomena which bear on site appraisal. These are divided into bedrock elements, of which lithology and tectonics are highlighted, and Quaternary elements, particularly those that occurred in areas of cold climate, i.e. past freezing and thawing, sea-level changes, hydrogeological features and glacial, periglacial, fluvial and marine erosional and depositional features. Abandoned cliffs and inland scarps in chalk and some clays are then identified as characteristic landforms, and their nature and development are also explored. Case records of successful and unsuccessful earthworks on clayey scarps are also reviewed, in relation to the quality of initial site appraisal undertaken. In conclusion, significant weaknesses in our current education and training in this area are identified, specifically a near-absence of geomorphology and insufficient Quaternary geology, and proposals made to remedy these.

Potentially dangerous surges in an Antrim mudslide

Summary Measurements of surface movements and pore-water pressures in a mudslide at Minnis North, north-eastern Ireland, are presented. This feature comprises steep feeder mudslides leading down to a more gently inclined accumulation mudslide, from which a steep front slope descends to the coast road. Frequent surges of the mudslide, which discharges considerable quantities of debris on to this road, are described and discussed. In one of these, pore-water pressures and surface movements were monitored up to failure and the surge itself witnessed by one of the authors. It is concluded that the surges generally originate through‘undrained loading’ of the rear of the accumulation slide by the feeder slides. Attention is drawn particularly to the danger inherent in situations where large volumes of underconsolidated and therefore potentially highly mobile material can collect above a steep front slope and to the difficulty, in the case studied, of obtaining warning of failure. Evidence is also presented to show that this danger is increased if surface streams are present on the mudslide.

A landslide in periglacially disturbed Etruria Marl at Bury Hill, Staffordshire

Summary The investigation of a landslide on a slope of Etruria Marl at a site just south of the Weichselian limit in Staffordshire is described and discussed. Carefully logged trial trenches and shafts show that the stability of the slope is controlled by a 6 to 10 m thick mantle of periglacially disturbed and soliflucted Etruria Marl, and descriptions of the lithology, mineralogy and structure of this mantle are given. The landslide is shown to have involved the renewal of movement on a pre-existing slip surface. Ground water conditions at failure are reconstructed and two and three-dimensional back analyses made. At the appropriate level of normal effective stress, the estimate of residual shear strength from back analysis is shown to be lower than indicated by conventional direct shear and triaxial tests and probably higher than the value given by ring shear tests.

The reactivation of a landslide during the construction of the Ok Ma tailings dam, Papua New Guinea

The Ok Ma dam (Ok is a local word meaning river) was to form part of the permanent tailings disposal system designed for the Ok Tedi gold and copper opencast mine located at Mount Fubilan in the Star Mountains of Papua New Guinea. During the construction works a landslide involving approximately 35 million cubic metres of soil and rock moved downslope to partially fill the foundation excavations of the dam. The failure of this landslide was the start of an insurance litigation that finally reached the Supreme Court of Papua New Guinea in November 1989. The geotechnical, geomorphological and engineering geological data that were available at the time of the failure are presented together with an evaluation of the key deductions from the case study.

Cambering and valley bulging, periglacial solifluction and LateglacialColeoptera at Dowdeswell, near Cheltenham

Engineering works at Dowdeswell Dam, Gloucestershire, where the River Chelt emerges from the Cotswold scarp, enabled investigation of some of the local Quaternary features. The log of the dam cut-off trench (1884) shows the presence of a hitherto unreported valley bulge. It is of the type known as a crumple, exhibiting two sharp anticlinal folds superimposed upon a much gentler valley anticline. The former bulge had, as usual, been eroded away by the river. A trial pit at the foot of the northern valley side showed 2.86 m of silty clay head over a thin Oolitic breccia resting on the eroded surface of the Charmouth Mudstone Formation. Resting on the breccia was a thin discontinuous layer of organic clay. This was found to contain 19 taxa of Coleoptera, of which 11 could be identified to species or species group. The assemblage indicates that cold conditions obtained, with an open, barren landscape and a mean annual air temperature of −3°C. The deposit is assigned to the Younger Dryas period. The head mantling the valley slopes, at least in the area of Trial Pit 1, was probably emplaced predominantly during this cold period. The valley bulging was much earlier, evidence from other sites indicating an age of about oxygen isotope Stage 12.

Rapid failures of colliery spoil heaps in the South Wales Coalfield

Abstract The exploitation of the South Wales Coalfield, particularly from the last quarter of the nineteenth century onward, was accompanied by numerous instances of instability within spoil heaps and their foundation materials. Sudden failures which were sufficiently rapid to overwhelm property and services and, in some instances, to threaten life, occurred on at least twenty three occasions, details of which are provided. It is believed that thesel included sixteen flow slides, five debris slides and two failures caused by outbursts of groundwater. At five sites, debris flow was a secondary failure mechanism. The locations of these failures are mostly clustered in those parts of the coalfield with the highest relief and with the highest rainfall, although antecedent rainfall conditions for the failures were variable. Most are shown to be associated with active tipping faces, but one flow slide is believed to have occurred on a tip four years after its abandonment and an outburst failure on a tip fifteen years old. The occurrence of rapid failures is shown to mirror the development of the coalfield but has ceased largely as a result of legislation to improve tipping practice, which was enacted following the Aberfan flow slide.

Some landslides in Airedale, Yorkshire, and their incidence in relation to palaeoclimate compared with that indicated generally for southern Britain and NW Europe

Abstract Four landslides in the Aire valley near Bingley are described, with geotechnical data, and dated. The bedrock consists of an Upper Carboniferous (Namurian) sequence of alternating grits and mudstones of low dip. During the Quaternary, the area was affected by ice sheets and valley glaciers and by periglacial solifluction. During retreat of the Late Devensian valley glacier, terminal moraines were emplaced, with proglacial lakes upstream and, on breaching, strong erosion downstream, particularly in the Bingley area. All of the Airedale slides are inferred to have been triggered in part by climatic changes and, in the case of the two nearest the river, also by associated erosion. Radiocarbon dating indicates that the slides occurred as follows: (a), Scarths Fields successive rotational slides: Dimlington Stadial–Windermere Interstadial transition; (b), Scarths Fields periglacial solifluction sheet: Loch Lomond Stadial; (c), Holme House Wood compound rockslide: mid-Holocene Interglacial (early Atlantic); (d), Altar Lane periglacial mudslide: Late Holocene Interglacial (early Sub-Atlantic). A comparison of the approximate dates of the above mass movements with the periods of enhanced landslide activity established generally for southern Britain indicates a fairly close concordance.

Discussion on ‘Large landslides in London Clay at Herne Bay, Kent’ by E.N. Bromhead: J. N. Hutchinson writes

Bromheads paper provides useful information, derived from the back-analysis of existing slips in the Herne Bay area, on the values of residual strength of the London Clay mobilized in the field. As the slips treated are generally rather deep-seated, the data extend our knowledge of the field residual failure envelope for the London Clay into a higher range of normal effective stress. In his summary plot (Fig. 10) of this envelope, however, the author does not make use of some existing data in the low and medium normal effective stress range. Field residual strengths derived from the back-analysis of shallower slips in the London Clay are available from sites at Guildford, Hadleigh, Sud-bury Hill and Dawes Road. These data, plotted with Bromheads values in Fig. 1a, permit the construction of a well substantiated field residual envelope for the London Clay up to a value of normal effective stress of about 160kN/m2. The shape of the residual failure envelope in the vicinity of the origin of the Mohr diagram and the magnitude of any small intercept on the shear strength axis remain, of course, uncertain. The field residual envelope shown in Fig. la can be approximated, up to a normal effective stress of about 70 kN/m2, by the parameters c′r = o, φ′r = 13.3° (Hutchinson & Gostelow 1976) or by c′r = 0.5 kN/m2, φ′ r = 12.9°. The envelope thus defined lies slightly lower than that shown by the author for this stress range. At higher normal