The New Austrian Tunnelling Method

7 The New Austrian Tunnelling Method

¨ The New Austrian Tunnelling Method (NATM, in German: NOT) emerged in 1 the years 1957 to 1965 and was entitled in this way to be distinguished from the Old Austrian Tunnelling Method. The NATM was developed by Austrian ¨ller-Salzburg). Its tunnelling specialists (von Rabcewicz, Pacher, Mu main idea is to head the tunnel conventionally, to apply support (mainly shotcrete) sparingly and to follow the principles of the observational method. The NATM requires the distortion of the ground to be kept to a minimum (in order to avoid softening and thus loss of strength). But at the same time sufficient ground deformations should be allowed in order to mobilise the strength of the ground. Consequently, thick and stiff linings which do not ¨llercompletely abut on the rock, are no longer in use. According to Mu Salzburg2 the main principles of the NATM were guesstimates (mainly by Rˇ ziha, Heim, Andreae), which could not be applied until the techniques for shotcrete and rock monitoring had been developed. As many of the NATM’s recommendations were already in use, it is not easy to differentiate NATM against other tunnelling methods. This has led to a lengthy controversy, which is still underway. The debate does not refer to the content but rather to the name of the NATM because the lack of an exact definition makes it unclear in which cases this name should be used. One attempt to define NATM was made by the Research Society for Road Engineering of the Austrian Union of Engineers and Architects,3 who published a complex and not very illuminating definition composed of not less than 5 basic principles and explanations, 3 general principles and 8 specific 1 L. M¨ uller-Salzburg und E. Fecker: Grundgedanken und Grunds¨ atze der ‘Neuen ¨ Osterreichischen Tunnelbauweise’, in Felsmechanik Kolloquium Karlsruhe 1978, Trans Tech Publications, Clausthal 1978, 247-262 2 L. M¨ uller-Salzburg, Der Felsbau, dritter Band: Tunnelbau, p. 562, Enke-Verlag 1978 3 ¨ Schriftenreihe der Forschungsgesellschaft f¨ ur das Straßenwesen im Osterreichischen Ingenieur- und Architektenverein, Heft 74, 1980

172

7 The New Austrian Tunnelling Method

principles.4 The definition states that the NATM activates a bearing ring in the surrounding ground. However, this statement has been widely criticised, because any cavity in the rock is — at least partially — supported by the rock itself, no matter if this cavity has been headed according to NATM or not.5 Another debate surrounding the NATM is the Austrian designation,6 which has been questioned with regard to the first application of shotcrete. But von Rabcewicz himself confirms that shotcrete was first applied during the construction of the Swiss Lodagno-Losogno-tunnels between 1951 and 1955. The Austrians were quick to follow. Shotcrete and rockbolts were systematically applied in the pressure tunnel Wenns of the Prutz-Imst water power plant from 1953 onwards.7 There is no doubt that Austrian engineers contributed with courageous and pioneering applications to the propagation and foundation of NATM. The most widely held view is that NATM has come to represent conventional heading with shotcrete support.8 Confusion over the designation ’NATM’ was also apparent in the HSE Review9 which emerged after two inrushes of NATM headings in London clay in 1994. Two notations were used: ’N.A.T.M.’ for the method according to the Austrian Union of Engineers and Architects, and ’NATM’ for tunnels headed with open face excavation and supported as soon as possible with shotcrete, anchors, nails and bolts. NATM’s recommendations were launched originally as empirical guidelines which can be interpreted today in terms of theoretical analysis. The theoretical foundations were always missing.10 But even without such scrutiny, NATM has achieved remarkable successes (Tauern tunnel, Arlberg tunnel, Inntal tunnel, metro Frankfurt, Schweikheim tunnel, Tarbela caverns). When applying the NATM in urban areas with a soft ground, one of its rules has to be relaxed: deformations to mobilise rock strength should be limited as otherwise the surface settlements can become excessive. Nevertheless, NATM was successfully applied in the metro construction in Frankfurt 4

The perception of this definition as ’impenetrable shroud of complexity’ (A. Muir-Wood, Tunnelling: Management by design. Spon, London 2000) appears thus understandable. 5 ¨ K. Kov´ ari: Gibt es eine NOT?, XLII. Geomechanik Kolloquium 1993, Salzburg 6 Note that in Norway the NATM is known as ’Norwegian Method of Tunnelling’. 7 In US mining, shotcrete has been used since about 1925. 8 see Tunnels & Tunnelling, September 1995, p. 5, for the history of NATM see also J. Spang: Die Geschichte des Spritzbetons und seiner Anwendung beim untert¨ agigen Hohlraumbau. Taschenbuch f¨ ur den Tunnelbau 1996, p. 321 ff, Verlag Gl¨ uckauf. 9 Health and Safety Executive: Safety of New Austrian Tunnelling Method (NATM) tunnels. A review of sprayed concrete lined tunnels with particular reference to London clay. HMSO, 1996 10 ’We need a scientific foundation, otherwise we have to disappear’, F. Laabmayr in a3BAU 12/1994, p. 88

7.1 HSE Review

173

clay between 1969 and 1971 and proved itself economical when compared with shield heading. Successful applications followed in N¨ urnberg, Bochum, Bonn, Stuttgart, Vienna. Since 1992 NATM has also been applied in London clay, which has similarities with Frankfurt clay. In October 1994 two inrushes occurred which gave rise to detailed investigations to NATM (see HSE Review and ICE design and practice guide11 ). The successful completion of the Heathrow baggage transfer tunnel and the R¨ omerberg tunnel12 are further vindications of NATM. There have also been some drawbacks (M¨ unchenOrleansplatz, M¨ unchen-Trudering, Heathrow and the new rail track from Hannover to W¨ urzburg), which should be attributed not to the method itself but rather to its improper application or other reasons. A list of 39 inrushes or daylight collapses of NATM headings is given in the HSE Review. If we look at NATM in the wider sense — i.e. as the Austrian tunnelling school, we can appreciate all the more the globally respected expertise of Austrian tunnelling specialists. Their philosophy has been sharpened by the highly variable geology of the Alps and relies not so much on previous site investigations as on the flexibility to find the correct support measures on the spot. In conclusion, probably the best definition of NATM belongs to H. Lauffer:13 NATM is a tunnelling method in which excavation and support procedures, as well as measures to improve the ground — which should be distorted as low as possible, — depend on observations of deformation and are continuously adjusted to the encountered conditions. Consequently, NATM contrasts the design and construct principle, where the construction has to proceed as originally designed.

7.1 HSE Review The HSE Review mainly addresses the NATM applications in urban areas with soft ground. Its conclusion is that NATM is indeed a safe construction method, as long as some principles are taken into account. The increasing number of accidents on NATM construction sites can be attributed to several reasons: application in difficult ground, improper application, shortcomings of the NATM, insufficient control, over-confidence in the method and more open reporting of failures. However, it cannot be deduced that NATM is less safe compared with other methods. Tunnel construction sites can hardly be 11

ICE design and practice guide. Sprayed concrete linings (NATM) for tunnels in soft ground. Thomas Telford, London, 1996 12 see Tunnels & Tunnelling, July 1995, 17-18, and H. Lutz: Driving the R¨ omerberg tunnel given slight Overburden, Tunnel 4, 1995, 18-21 13 personal communication

174

7 The New Austrian Tunnelling Method

compared with each other as each has its own typical conditions. Limited existing investigations reveal a comparable frequency of accidents in NATM and non-NATM sites. Among others, the following conclusions are drawn: Reduction of risk: Daylight collapses in urban areas can have grave consequences. To minimise the risk one should, if necessary, change the tunnel alignment to avoid sensitive ground. Other measures are to relocate bus stops and traffic lights and hindering access to endangered sites.14 Qualified personnel: NATM requires the in situ fabrication of shotcrete, which is a highly complex procedure. To assure safety, only trained personnel should be employed, no training phase during construction must be allowed for. Management: Poor site management is a main reason for accidents. Good management contributes to: • coping with unforeseen events • proper application of the observational method • elimination of human errors. It is important to inform the involved persons of hazards. Compatibility and cooperation of the teams should be assured. Collapses: Most of the inrushes in NATM headings occur in the unsupported face. Therefore, a sufficient stand-up time is necessary. However, observations are of little use if the collapse is unannounced.15 A feature of the NATM is that it is unable to provide a support for a suddenly destabilised face. The probability of a daylight collapse in shallow tunnels is particularly high if watersaturated permeable layers of low strength are encountered. In absence of groundwater, the inrushing earth masses form a heap that prevents further soil inrush. In most cases, there is sufficient warning for the personnel to escape. The largest hazard for persons is due to blocks falling from the unsupported face. Reports on such accidents are rare, but there is some indication that on average one in 15 inrushes causes injury. Geology: Bad geologic conditions are often blamed for inrushes. In particular, unexpected erosion structures, such as lenses or old wells filled with watersaturated cohesionless material, can be troublesome. Therefore, a forwards exploration and a thorough geological record of the face are recommended. Observational method: This method requires: 14

see also W. Schiele: Findings from the Underground Shield Drive for the Munich Underground Lot 1 West 5, Tunnel 6/1996, 23-30 15 In this context the law of the Japanese seismologist K. Mogi should be mentioned, according to which the fracture process strongly depends on the degree of heterogeneity of materials: the more heterogeneous a material is, the more warnings one gets before collapse (cited in D. Sornette: Critical Phenomena in Natural Sciences, Springer, 2000)

7.1 HSE Review

175

1. Determination of acceptable limits for the behaviour of a construction 2. Verification that these limits will (with sufficient probability) not be exceeded 3. Establishing a monitoring programme that gives sufficient warning of whether these limits are kept 4. Providing measures for the case that these limits are exceeded. To date, items No. 1 and 2 have not been considered in a convincing way by the NATM literature, i.e. the decision of which deformations are acceptable is left to the experience and intuition of the engineer in charge.16 Contemporary measuring programmes usually produce an overwhelming amount of data which are very difficult to grasp. An appropriate processing and graphic representation of the data is therefore highly advisable. Protection measures: The following measures can be applied to increase the stability of the excavated cavity: • heap • sidewall drift • elephant foots • anchors in the crown • forepoling • drainage and pressure relief • temporary ring closure • thicker shotcrete lining • larger crown curvature • additional ribs • compressed air support • reduction of advance step • reduction of partial face cross sections • earlier construction of the cast concrete lining.

16

K. Kov´ ari and P. Lunardi point to this shortage of NATM (On the observational method in tunnelling. Proceedings of GeoEng 2000, Melbourne, Australia, 2000, Vol. 1, 692-707). However, their explanatory statement “NATM can be disregarded as an observational method” because “Pacher’s concept violates the fundamental principles of the conservation of energy” is not tractable.