Erroneous Concepts behind NATM

K. Kovári
Swiss Federal Institute of Technology Zurich

Lecture given at the Rabcewicz-Geomechanical Colloquium in Salzburg, Octobre 14, 1993

 Erroneous Concepts behind NATM

K. Kovári
Professor of tunnelling
Swiss Federal Institute of Technology Zurich 

Summary

The New Austrian Tunnelling Method (NATM) represents in the words of one of its main
authors, L. Müller, a "structure of thought". It will be shown in the following paper that
this structure rests not on an established theoretical foundation, but rather on two fun-
damental misconceptions. The first concerns the definition of the NATM itself, according
to whose concept the rock mass (ground) becomes part of the support structure. Upon
closer examination this concept is revealed to be unfounded because the ground
inevitably becomes part of the support structure in any tunnel independently of the
selected method of construction. The second fundamental error lies in the claim that
the NATM theory can optimize the design of the tunnel lining following the so-called
Fenner-Pacher ground reaction curve. Since both theoretical and empirical justification
is lacking for the existence of the latter, the central claim concerning the optimization of
the lining design is groundless. In the final part of the paper, the question is discussed
of why such misconceptions were able to gain such credence in the engineering
community and why the NATM-theory was able to survive for so long without being
refuted. The reason is that the NATM operates with ambiguous or undefined terms
making it difficult to prove its incorrectness with logical argumentations.

1. The Official Definition of the New Austrian Tunnelling Method

In 1980 the Austrian National Committee on “Underground Construction” of the Interna-

tional Tunnelling Association (ITA) published in 10 languages an official definition [1] of
the New Austrian Tunnelling Method, which is as follows:

“The New Austrian Tunnelling Method (NATM) is based on a concept whereby the
ground (rock or soil) surrounding an underground opening becomes a load bearing
structural component through activation of a ring -like body of supporting ground”.

The latter is referred to in the following simply as the "ground ring".


 Lecture given at the Rabcewicz-Geomechanical Colloquium in Salzburg, Octobre 14, 1993

1
This definition contains three principal statements:

- the ground becomes a load bearing structural component,

- to achieve this a particular concept is required,
- the concept consists of the activation of a ground ring.

In this and the following chapters emphasis will be given to the published material of
the authors of the NATM, i.e. L. Rabcewicz, L. Müller and F. Pacher (often referred to
as the "fathers" of the NATM). Reference will also be made to the work of the second
generation of NATM exponents.

1.1 The ground as load bearing component

The essence of the above definition is that, under the NATM, the ground (i.e. rock or
soil mass) itself becomes part of the load-bearing system. The definition emphasises
that property of the NATM which distinguishes it from all related concepts, i.e. from
other tunnelling methods. Presumably, under this method alone the ground becomes a
structural component. From this basic statement other well-known formulations have
been derived and frequently reported in NATM literature. For example, under the NATM:

- the support capacity of the ground is brought into play,

- the ground supports itself,
- the main load bearing component is the surrounding ground,
- the ground is transformed from a loading to a supporting medium,
- the self-supporting capacity of the ground is exploited,
- one works with and not against the ground, etc...

The claim that the NATM alone allows the ground to act as a structurally supporting
component is basically false. In reality, tunnelling without the structural action of the
ground is inconceivable. Whether the engineer employs a technical measure to support
the underground opening or not makes no difference to the inherent support action of
the ground. By trusting instinctively the mechanical laws governing a rock mass, man
has occupied caves since early times and has made underground openings of various
types for his own purposes. One can show in fact that the idea of the ground as a
structural element is inherent to the concept of a tunnel. This knowledge is logically in-
dependent of experience, since it follows directly from fundamental mechanical princi-
ples.

Thus the NATM claims for itself what is a universal necessity (the structural support ac-
tion of the ground) which is also effective for all other methods of tunnelling and
characterizes tunnelling per se. In order to elucidate the logical error of the definition of
the NATM let us consider the following simple example of a similar false definition:

“The New Swimming Technique is based on the concept that by activation of uplift the
water becomes a supporting medium”.

2
In this example, as also in tunnelling, a mechanism is acting (here uplift, there
structural support) due to a natural law, which is not dependent on a concept or a
method. Therefore both formulations are false, because in defining a sub-class they
use the main characteristics of the class to which it belongs (confusing species with
genus).

This error in the definition of the NATM is illustrated in Figures 1 and 2. For a correct
definition (Fig. 1) the NATM would have to have a characteristic which makes it differ,
definition-wise, from other tunnelling methods. Instead of this, the NATM has a basic
property which is common to all methods of tunnelling. The NATM departs from the
category of construction methods, slips into the definition of tunnelling in general (Fig.
2) and feels itself justified in regarding all other methods of tunnelling as being inherent
to it. In NATM circles, in fact, the question of the criteria to which various construction
methods can be classified under the NATM is discussed in earnest [2]. It is believed
now that the NATM is not merely a method but rather a universal collection of knowl-
edge and skill >3@. The concept of tunnelling is thus replaced by the concept of the
NATM. Thereby, the NATM would represent at one and the same time both the most
comprehensive and the most vacuous concept in tunnelling. According to the laws of
logic, the content of an idea decreases in relation to its size. From this it follows, for ex-
ample, that, regarding the NATM, it is not the construction method that is flexible, but
rather the definition of the NATM, which can be stretched in an arbitrary manner.

From the trivial example of swimming we all know that Archimedes recognized the sup-
porting action of water, i.e. the law of uplift. Who was it then that discovered the law of
the structural support action of the ground in tunnelling?

Already in 1879 Ritter [4] observed that from a certain depth of tunnel the influence of
the overlying rock was insignificant or had no influence on the rock pressure. The rock
mass itself supports the weight of overburden. Three years later (in 1882) Engesser [5]
proposed that an “arching action” is induced in a cohesionless ground mass by a
sagging of the tunnel roof (Fig. 3). Thus the connection between rock deformation and
rock pressure exerted on tunnel linings was recognised and clearly formulated.

TUNNELLING METHOD 1 TUNNELLING METHOD 1

TUNNELLING METHOD 2 TUNNELLING METHOD 2

TUNNEL-
LING NATM NATM .................

................. .................

TUNNELLING METHOD X TUNNELLING METHOD X

Fig.1 Aim of the definition of the NATM Fig.2 Practical outworking of the NATM definition:
the NATM arrogates to itself the essence of

3
 tunnelling
ARCH ACTION "PROTECTIVE ZONE"

Vt

GALLERY

Fig. 3 Arch action in cohesionsless ground Fig.4 Protective zone according to Wiesmann
according to Engesser (1882, [5]) (1909/1912, [6])

Wiesmann [6] described in 1912 the support function of the rock mass as follows:

“If the equilibrium state within a rock mass is disturbed by excavating an underground
opening then the material particles surrounding the opening have to resist that pres-
sure as an excess pressure which before was supported by the excavated material, as
is the case when we make a hole in a wall”.

He further remarks that:

“The tunnelling engineer does not have the task of supporting the opening for the ex-
cess rock pressure. That is done by the protective zone. He has to be concerned with
the preservation of this zone”.

By protective zone Wiesmann understands the rock surrounding the opening in which
stress redistribution occurs, that is - in a twodimensional consideration - the plate (plane
strain condition) with a hole in it (Fig. 4). The protective zone is not sharply bounded.

50 years later in 1962, in the year the NATM was born, Rabcewicz [7] wrote:

“I think that today in the construction of underground openings we have come to real-
ize that the supporting material is actually the rock mass itself”.

and adds:

“To preserve as far as possible and to develop the support properties of the rock
mass is thus the most important task of modern tunnelling”.

Rabcewicz completely failed to see that this view was already well-known and taught in
the textbooks of the day. Therefore, he did not apply this correct observation generally
to the whole of modern tunnelling, but restricted it to the term “New Austrian Tunnelling
Method” introduced by him the same year. As a result, the conceptual difficulties of the

4
NATM were built into it and have not gone away up to the present day. So the NATM
was in fact established already in 1962 on wrong premisses.

The members of the Austrian National Committee on “Underground Construction”,

inclu-ding Müller and Pacher, also did not notice when drafting the official definition of
the NATM [1] how deeply the support function of the ground was anchored in the con-
sciousness of engineers since the beginning of the century.

Here is another apt formulation by Maillart from 1923 [8]:

“If we construct a tunnel lining so as to withstand the external rock pressure acting on
it, the strength of the rock mass is increased, thus enabling it to support itself.”

In a lecture held in 1956, i.e. 6 years before the NATM was proposed, Mohr [9] stated
that

“the forces acting on a lining will be smaller if the rock mass is allowed to deform a
little”

and he continues:

“The practical use of this knowledge requires that the rock mass should only be sup-
ported to the extent that it becomes able to support itself”.

In that the NATM claims exclusively to consider the supporting action of the ground, it
not only commits a logical error, but also ignores the achievements of those to whom
credit is due for recognizing and clearly formulating this fundamental law of tunnelling.
The NATM uses in its postulates without giving proper reference the terms and expres-
sions of leading tunnelling engineers and scientists of earlier times [6, 8].

1.2 The activation of a "ground ring" around an opening in rock and soil

Now we turn to the concept whereby, according to the official definition of the NATM,
the activation of a ring-like body of rock or soil must result. As mentioned above, we use
simply the term "ground ring". This requirement is particular to the NATM way of looking
at things and is not to be found in other literature on tunnel construction.

What is a ground ring? There are a number of answers to this question in NATM litera-
ture and in the brochure of the Austrian National Committee [1]. All these, however, dif-
fer fundamentally one from another. Here we summarize briefly the various, contradic-
tory ideas which are used by the authors of the NATM in their argumentations:

- the ground ring [10] is also called the protective zone (Fig. 5).
- in earlier tunnel constructions the ground rings apparently ran further from the open-
ing than is the case today with the NATM (Fig. 5).
- the ground rings are frequently represented as ellipses (Fig. 5).

5
 PROTECTIVE RING (GROUND RING) GROUND RING

TODAY EARLIER

Fig. 5 Protective rings (ground rings) according to Müller Fig. 6 Ground ring according to
and Fecker [10] Rabcewicz (1944, [11])

- according to Rabcewicz [11] such rings are in contact to the openings, whereas for
Müller and Fecker [10] they are not (Fig. 5 and 6).
- the ground rings allegedly have to be mobilized by means of admissible rock
deformations, otherwise they do not develop. If the rock deformation is insufficient,
these ground zones do not close to a ring [1].
- from the point of view of tunnel statics only the lining [1] and the ground ring (Fig. 7)
should count. The latter seems to be loaded one way or another. Prior to the NATM
concept, the ground ring apparently did not play a part.
- in the case of multiple adit excavation method (Fig. 8) a series of ground rings is
supposedly formed [12], which according to the understanding of the NATM, is detri-
mental to the rock mass. Thus the NATM propagates full face excavation as one of its
main principles. Plastic zones interrupt the ground rings [13].

TODAY EARLIER

P
v
Ph Ph

P
i CULVERT

LINING

GROUND RING

P
v

Fig. 7 Ground ring according to Müller and Fecker [10] Fig. 8 Ground rings around multiple adits
according to Müller [12]

6
 GROUND RING
INTACT RESIDUAL
ROCK ARCH ROCK ARCH

Fig. 9 Intact and residual rock arch according to Fig. 10 Ground ring above a cavern according to
Hagenhofer [3] Wisser [14]

- there are also reports [3] of intact and residual ground rings; the latter, however, do
not close (Fig. 9). What the difference is between the intact and the residual rock
rings, however, is not explained.
- in the vicinity of a cavern [14] the protecting zone, apparently, resembles the roof of a
hall (Fig. 10), i.e. it has the form of an arch. But it is not clear on what this arch rests.
It is also not clear why the ground ring is missing in the region of the invert.
- in the profile of the crossover point of the Channel Tunnel [15] the shape of the
ground ring is particularily strange (Fig. 11). At the top it is extremely thick, tapers in-
creasingly towards the sides and it does not close at the bottom.
- in NATM literature, descriptions of several concentric ground rings are to be found,
which are supposed to transform the rock mass to “an onion-skin-shell structure” [16]
(Fig. 12). Whether these rings develop simultaneously or successively is not ex-
plained.

V GROUND RINGS

1 2 3

Vt
Vr

GROUND RING

Fig. 11 Specific form of the ground ring with the Channel Fig. 12 Several ground rings according to
Tunnel (Myers et al, [15]) Müller, Sauer and Vardar [16]
forming an "onion-skin-shell
structure"

7
 1 PROTECTIVE ZONE
(WIESMANN) c, M 
2
1 2 GROUND RINGS
G
d
H

3 PLASTIC ZONE p
3
4
4 ANCHORED ARCH

Fig. 13 The varieties of concept of the ground ring Fig. 14 Shape, thickness and material
properties of the ground ring
remain undetermined

Essentially the NATM works selectively with four basically different hypotheses of a
supporting ground ring. According to the NATM the ring can signify (Fig. 13):

1. the protective zone according to Wiesmann (natural structural action) extending far
from the opening,
2. areas in the rock mass with maximum circumferential stress (protective zones),
3. the plastic zone and
4. the rock zone defined by systematic anchoring (grid of anchors).

In NATM literature, authors shift arbitrarily between these four rather divergent hypothe-
ses and the reader may find that even in the same article the meaning of the ground
ring can undergo changes.

It is therefore understandable that neither the authors of the NATM nor those who sup-
port it could ever give information on how to determine the shape and thickness of such
a ground ring. Even the material properties are not characterized (Fig. 14). There are
indications according to which the thickness of the ground ring is chosen arbitrarily [17].
Fig. 14 shows the quantities, which should all be known in order that one might handle
such a system from an engineering point of view. But has a ground ring ever been sub-
jected to a quality control according to Fig. 14? Müller [13] maintains that

"the closed ring-shaped protective zones are for most tunnel engineers experienced
reality".

The NATM places conditions upon the establishment of the ground ring. One has to
mobilize or activate it, else it does not form or only gradually forms or the ring closure is
incomplete [1]. Under activation one author means systematic anchoring, the other
controlled rock deformations, and yet another waiting for a period of time to elapse, etc.
However, the ground ring may also be initiated by a shotcrete lining. One of the most

8
widespread formulations asserts that the NATM "permits" ground deformations and
"allows" time for the rock to support itself. Since in tunnelling, rock deformations cannot
be completely prevented by reasonable means and the installation of support measures
inevitably takes time, this postulate is unacceptable and particularly misleading.

We summarize the results of our investigation of the official definition of the NATM as
follows:

- The ground represents of necessity in the whole of tunnelling, independent of the

selected tunnelling method, a structural component. The recognition of this phe-
nomenon is emphatically not due to the authors of the NATM.

- The specific requirement of the NATM to activate the ground ring cannot be ac-
complished. The words “ground ring” and “activation” are so ambiguous that they are
useless from the scientific standpoint. The definition of the NATM has proved to be a
murky definition, since it explains one unknown (the NATM) with the aid of another
(structural activation of the ground ring).

Thus we have an explanation for the observation made by Müller and Fecker [10] con-
cerning the NATM, namely

“that practically everyone who applies this method of construction has a different
conception of it in his mind”.

Nothing has changed regarding the correctness of this statement since 1978. Thus, the
question "does the NATM really exist ?" can be answered with an emphatic no.

2. Minimizing rock pressure

We now go a step further and investigate the central idea of the NATM about
minimizing rock pressure acting on the tunnel lining.

Since 1972 the hypothesis of Pacher [18] published in 1964 concerning the trough-
shaped ground response curve and the minimalization of the rock pressure and the lin-
ing thickness based upon it has become more and more central to the concept of the
NATM [19]. According to Müller [12],

"the main concept of the NATM is based on Pacher's ground response curves".
What are we dealing with here? It is simply a matter of choosing [1, 10] the position and
shape of the lining characteristics according to the NATM in such a way as to intersect
the ground response curve of the rock at its lowest point (Fig. 15). Rabcewicz [19] is of
the opinion that

"with the aid of measurement one is in the position to keep the forces under control
and the lining resistance pi can be chosen accordingly, until an optimum value is
achieved".

9
 pr pi / Vr [%] ROCK LOOSENING
100
GROUND RESPONSE CURVE
(PACHER, 1964) 80 70 %
p
i
R 60
'R
p 40
r
LINING CHARACTERISTICS 20

'R / R 0
5 15 100 1000 ' R [mm]

Fig. 15 Optimization of the rock pressure after Fig. 16 Bifurcation of the ground response
Müller and Fecker [10] curve due to strain softening in the
rock mass after Müller [12]

In Fig. 15 we have used a representation following Müller and Fecker. The abscissa is
the relative radial displacement 'R/R and the ordinate the lining resistance pr. The axial
symmetry of both the geometry and the stresses should be noted in this representation.

In Fig. 16 a bifurcation of the curve may be observed [12]. The lower curve, which ap-
proaches the horizontal axis, is given by plasticity theory. NATM postulates a curve
turning upwards, which is supposedly caused by rock loosening and strain softening. In
this diagram presented by Müller [12] the units of radial displacement are given in
millimeters and the lining resistance in percent of the overburden pressure.

Müller justifies the dramatic influence of loosening due to deformation as follows [12]:

“The higher the rock pressure, the greater the loosening of the rock; this increases in
turn the rock pressure phenomena”.

Such a process is similar to a chain reaction (a reaction which, once induced, causes
further reactions of the same type to occur). If one considers more closely the diagram
of Müller shown in Fig. 16 then the curve turning upward seems to reflect the result of a
chain reaction. One sees that the rise in rock pressure produced by rock loosening can
reach about 70% of the overburden pressure. For a tunnel situated at a depth of 1000
m the rock pressure according to Fig. 16 would correspond to the weight of a column of
rock of 700 m. That this is impossible was clear even to engineers in the middle of the
last century. What is the reason for this extraordinary contradiction? It lies in the false
assumption of a chain reaction in the rock mass. Indeed, there is no evidence theoreti-
cally or empirically for the existence of the ground response curve with a shape postu-
lated by Pacher.

10
 p PLASTIC ZONE
V1
i V1
ASSUMED BY NATM
pi (NOT APPROVED) PEAK STRENGTH
V3
'R

RESIDUAL STRENGTH

PLASTICITY THEORY

'R LOSS OF STRENGTH

H1


Fig. 17 Ground response curve according to Fig. 18 Loss of strength in shear and
plasticity theory in comparison with triaxial tests
that of the NATM

Only the ground response curve that is resulting from the theory of plasticity is theoreti-
cally founded, see Fig. 17. Whether or not a loss of strength (strain softening) is taken
into account - as may be observed in a shear or triaxial test - the curve does not turn
upwards.

In a detailed research report [20] on the use of the ground response curves as a design
basis for the NATM, the possibility of a trough-like Pacher curve is not even mentioned.
Also, in the ITA Guidelines [21] there is no mention of trough-like characteristic curves.
In the publications "Finite element analysis of the NATM" [22] and "NATM and finite
elements" [23] nothing is said about either activated rock ring structures or a Pacher
curve.

p [MPa]
v G
W
2.5 (H = 100 m)

H
2.0

1.5
INFLUENCE OF LOOSENING
D = 10 m F GF W
H= 5m
'p = 0.1 MPa
F
1.0
D
pv
0.5
GF [cm]

10 20 30 40

Fig. 19 Influence of loosening in the roof area Fig. 20 Increase of pressure on the
on the ground response curve roof lining due to rock loosening
(erratic,unpredictable event
which does not lend itself to
numerical computation)

11
There is only one conceivable possibility of a deviation of the ground response curve
from its downwards trend, which is also pointed out in NATM literature. It is the case
when in the roof area of the tunnel, due to unfavourable jointing or the development of
slip surfaces, a body of rock in a state of failure detaches itself partially or completely
from the parent rock and due to its self-weight increases the pressure on the roof lining
(Figs. 19 and 20). This possibility is also mentioned in the ITA guidelines [21]. At what
point along the length of a tunnel, to what extent and at what time such sudden occur-
rences are to be expected cannot be predicted from calculations. For us here it is
important to note, that no chain reaction is thereby induced. Taking for instance the dia-
gram shown in Fig. 19 one recognizes how small an influence the detaching of a body
of rock of about 5 m height (Fig. 20) has on the ground response curve in the roof re-
gion. This is confirmed by Rabcewicz [11] when he states that

“for the experienced tunnelling engineer rock loosening is mostly harmless”.

Müller [12] says also elsewhere that

“in tunnelling we have in general to reckon with zones of rock loosening of 0.5 to
5 m".

NATM literature, however, warns of the damaging consequences of rock loosening. Ac-
cording to Rabcewicz [24],

“the prevention of inadmissible rock loosening (is) an integral demand of the NATM".

p p
r r

TOO EARLY TOO LATE

p
r

R
'R/R 'R/R
pr pr

TOO RIGID TOO FLEXIBLE

'R/R 'R/R

Fig. 21 Sixth basic principle of NATM: “Construct the lining not

too early or too late, and not too rigid or too flexible”
(Müller and Fecker [10]).

12
Rabcewicz creates the term “admissible loosening”. This observation stands in contra-
diction to the above statement (loosening pressure is harmless). In addition, the
concept of admissible loosening was never defined. Müller [12] in 1978 correctly
observed that

“we have unfortunately no experimental evidence as to how much strain softening re-
sults from a certain amount of loosening".

Regardless of what is understood by a ground ring there should be a relation between

it and the ground response curve. In the NATM literature, however, there is no relation
whatever between the postulate of the activation of a ground ring and that of a ground
response curve as proposed by Pacher.

We summarize:

The minimization of the lining resistance in the sense of the NATM is not possible at all,
because its prerequisite of a trough-shaped ground response curve according to Fig. 21
cannot be explained theoretically and has never been verified empirically. Thus the
principle of the NATM “Construct the lining neither too early nor too late, and neither too
rigid nor too flexible" [10] is without meaning. The optimum choice of the strength- and
deformation-properties of the lining, as well as the time at which it is placed, must be
based on other criteria. It represents one of the most difficult problems in tunnelling.
Thus, its treatment by the NATM amounts to an attempt to trivialise the problem, and
thus tunnel design itself.

If a theory contains a gross error, it opens the door to even greater errors. The "Theory
and Practice of the NATM" [25] works, according to Sauer, with further types of charac-
teristic curves:

"The observations, experiences and measurements made thus far require in summary
an extension of the Fenner-Pacher curve to account for an additional maximum and
minimum in the excavation zone".

The justification for this claim is an altered Pacher curve, which is supposed to show the
relationship between six variables. Each coordinate axis represent not just one, but
three variables. To complete the total confusion, two of the six variables are not defined.
The effort put into an unsound mathematical formalism might give the impression of it
being scientific. It is incomprehensible that the Austrian National Committee for Under-
ground Construction of the ITA has not taken steps to stop such claims, although it acts
in the manner of an institution against critics of the NATM [26].

13
3. The NATM edifice of thought and its inferences

We close our evaluation of the NATM with a general assessment of its edifice of thought
[10].

We all know that ideas and concepts form the backbone of our thinking and conse-
quently of our judgements [27]. The basis of our scientific knowledge presupposes
clearly defined concepts and sound ways of arriving at true conclusions. Knowledge
without valid arguments or based on false premisses cannot be true.

If we consider the method by which the NATM develops its concepts then it may be
seen that it works on the whole with “nominal” definitions. By "nominal definition" one
understands simply "words with no clearly defined meaning". Nominal definitions, in
contrast to “real definitions”, are only loosely related to concepts. The latter have a clear
content. Thus only real definitions are appropriate in a scientific field.

The following examples taken from NATM literature clearly show the pitfalls of merely
creating words without clearly defined concepts.

Nominaldefinitionen Nominal definitions 

Entfestigungsempfindlichkeit sensitivity to strain softening

Entspannungsbereitschaft capacity for stress relief
Entspannungsgeschwindigkeit rate of stress relief
Halbsteife Schale semi stiff lining
Kräftevakuum force vacuum
Momentenfilter bending moment filter
Nachdrängende Gewichtskomponente pushing weight components
Nachdrängender Belastungscharakter pushing character of loading
Selbstschutzwirkung self-protection effect
Setzungsbeeinflusste Tragreserve structural reserves influenced by settlements
Spannungsschatten der Umlagerungen stress shadows of the redistributions
Spannungsumlagerungsgeschehen stress redistribution event
Spannungsgeschehen stress event
Spezifischer Zeitfaktor specific time factor
Umlagerungsnotwendigkeit necessity for stress redistribution
Zulässige Auflockerung admissible rock loosening

(Aus dem Schrifttum der NÖT) (taken from NATM literature)


As the meaning of these expressions is frequently not quite clear, even in the original German version, a
literal translation is adopted in the English version.

14
How does language react to such a system of ideas? It begins to proliferate. A series of
expressions related to the term “ground ring” are here assembled from NATM literature.

Dem Begriff "Gebirgstragring" Terms related to "Ground ring" 

sinnverwandte Wörter

Aussenschale outside lining

Aussengewölbe outside arch
Aussenring outside ring
Druckring pressure ring
Entspannungszone stress relief zone
Felstragring rock support ring
Fliesszone zone of plastic flow
Gebirgstragkörper rock support structure
Gewölbering arch ring
Gewölbeschale arch lining
Gleitbruchzone rupture zone
Ringgewölbe ring arch
Schutzgewölbe protective arch
Schutzhülle protective zone
Schutzzone ground ring
Stützring support ring
Traggewölbe structural arch
Tragring structural ring
Tragschalengebilde structural shell
Zwiebelschalen onion-shell-shaped shell
Dickwandiges Rohr thick-walled tube
Intaktes Felsgewölbe intact rock arch
In sich verspannende Zone self-spanning zone
Kastnersche Ohrenzonen Kastner “ear-shaped” zones
Massgebender Gebirgstragring critical rock ring structure
Mittragende Zone auxiliary rock support zone
Natürliches Gewölbe natural arch
Ohrenförmige Scherzonen ear-shaped shear zone
Plastische Zone plastic zone
Pseudoplastische Fliesszone pseudo-plastic flow zone
Residueller Gebirgstragring residual rock ring structure
Ringflächenförmige Fliesszone ring-surface-shaped flow zone
Schalenförmige Tragzonen shell-shaped structural zone
Selbsttragendes Gewölbe self-supporting arch
Spannungsverminderte Schutzschalen stress reducing protective shells
Stabilisierter Tragring stabilized structural ring
Taschenförmige Fliesszonen pocket-shaped flow zone
Virtuelles Traggewölbe virtual structural arch
Wiesmannsche Zone Wiesmann zone
Zungenförmige Zone tongue-shaped zone

(Aus dem Schrifttum der NÖT) (taken from NATM literature)


As the meaning of these expressions is frequently not quite clear, even in the original German version, a
literal translation is adopted in the English version.

15
This variety of terms reminds us of Goethe’s Faust:

“It's exactly where a thought is lacking

That, just in time, a word shows up instead.
With words you can argue beautifully
With words you can make up a system".
(Translation by Randall Jarrell)

If the definitions are imprecise then the judgements based on them must be either un-
true or at least not binding. In NATM literature, wrong or incomplete conclusions are to
be found. A wrong inference arises when the same word is used but with a different
meaning, so that in fact several definitions are involved in the same conclusion [27].
Thus, the one who makes the judgement can always find a statement to fit the circum-
stances and allow inconsistent statements to stand side by side. A good example of a
wrong conclusion is the description of how the ring-like rock support structure is acti-
vated.

An incomplete conclusion results if one or more premisses in arriving at an evaluation

are not stated. The claim of the NATM that the ground requires time for the
development of a new state of equilibrium is an incomplete conclusion. Here one would
have to add that this statement is only of practical significance in certain types of rock.
A bad error of judgement also occurs, if a name is applied to ideas, different from their
common use [28]. Such an arbitrary "abuse of words" is met with Sauer's [25]
representation of the ground response curve with additional Minima and Maxima.

A bad error of judgement occurs, however, when knowledge is propagated without justi-
fying it. An example of this is to be found in Sauer [25] “Theory and Practice of the
NATM” in his diagram shown in Fig. 22, which attempts to find a relation between the
following quantities:

- shear displacement ('S) and shear stress (W) in a direct shear test,
- convergence of displacements ('R) and distance (L) of the ring closure from the
excavation face.

Sauer reports [25]:

"Analogously the costs for rock support can be read off from this curve. They are
given by the difference to an assumed need of minimum support measure at a
maximum exploitation of the primary shear strength".

16
 V W
$ [x1000] W[KN / m2 ]
W L 0

min
V 2
1000
'S
COSTS FOR 800
SUPPORT MEASURES 4
COSTS
6 600

' S (' R) 400

8

L 200
10
W 1 2 3 ' S [mm]
'R
10 20 L [m]

Fig. 22 Unfounded relation between shear test, Fig. 23 Costs for the support measures in a
convergence, distance from the excavation supposed function of the shear strength
face and costs for support measures following the NATM theory
(Sauer, [25])

Thus, in this diagram we should have according to the NATM a means at our disposal
to determine the costs of support measures for a tunnel based upon a few laboratory
shear tests. This curve could be of the form shown in Fig. 23. On the left vertical axis
one could have, for instance, the costs for support measures per meter of tunnel. De-
pending on the shape of the shear strength diagram, the costs would rapidly rise at a
certain distance to the excavation face.

This theory of the NATM after Sauer has not been validated. Furthermore it cannot be
validated. In fact, it can easily be shown that such relations simply cannot exist. Such
misleading curves give the reader the impression, however, that under the NATM
method one obtains information which cannot be gained from any other tunnelling
methods.

As a last example of mistaken judgement in NATM literature we consider the following

basic statement of the NATM [1]:

“Adaptation of the methods of construction and operation to changing rock prop-

erties, to the stand-up time as well as to the stability of the excavation face by the
right choice of area and depth of attack".

On careful perusal of the above it is evident that in this basic statement two expressions
along with their synonyms occur. In tunnelling by method of construction and operation
is understood the procedure followed in excavating the cross-section, i.e. the choice of
the area of attack - e.g. full or partial cross-section, and under method of operation the
depth of this attack. Thus this statement of the NATM merely says that:

17
 “Adaptation of the methods of construction and operation using the right choice of
the methods of construction and operation”.

It is a tautology to repeat the same things using synonymous terms. How one should
make the right choice, the heart of the matter, is not answered.

4. Conclusions regarding the NATM edifice of thought

A critical discussion of the NATM within its own framework of ideas is not possible. Its
terms are so ambiguous that they defy close examination. If one considers the NATM as
a whole, however, not only is it not free from criticism, it is simply groundless.

References
[1] Neue Österreichische Tunnelbaumethode, Definition und Grundsätze, Selbstverlag der Forschungs-
gesellschaft für das Strassenwesen im ÖIAV, Wien, 1980.
[2] Lauffer, H.: Forderungen der NÖT an maschinelle Vortriebssysteme, Felsbau 6, Nr. 4, 1988.
[3] Hagenhofer, F.: NATM for tunnels with high overburden, Tunnels and Tunnelling, May, 1990.
[4] Ritter, W.: Statik der Tunnelgewölbe, Berlin, 1879.
[5] Engesser, F.: Über den Erddruck gegen innere Stützwände (Tunnelwände), Deutsche Bauzeitung,
1882.
[6] Wiesmann, E.: Ein Beitrag zur Frage der Gebirgs- und Gesteinsfestigkeit, Schweiz. Bauzeitung, Band
53, 1909.
Wiesmann, E.: Über Gebirgsdruck, Schweiz. Bauzeitung, Band 60, Nr. 7, 1912.
[7] Rabcewicz, L.: Aus der Praxis des Tunnelbaus, Einige Erfahrungen über echten Gebirgsdruck, Ge-
ologie und Bauwesen, Jg. 27, Heft 3-4, 1962.
[8] Maillart, R.: Über Gebirgsdruck, Schweizerische Bauzeitung, Band 81, Nr. 14, 1923.
[9] Mohr, F.: Kraft und Verformung in der Gebirgsmechanik untertage, Deutsche Baugrundtagung, Köln,
W. Ernst Verlag, 1957.
[10] Müller, L., Fecker, E.: Grundgedanken und Grundsätze der "Neuen Österreichischen Tunnelbauwei-
se", Felsmechanik Kolloquium Karlsruhe, Trans Tech Publ., Claustal, 1978.
[11] Rabcewicz, L.: Gebirgsdruck und Tunnelbau, Springer-Verlag, Wien, 1944.
[12] Müller, L.: Der Felsbau, Dritter Band: Tunnelbau, Enke Verlag Stuttgart, 1978.
[13] Müller, L.: Der Einfluss von Klüftung und Schichtung auf die Trompeter-Wiesmannsche Zone, 10.
Ländertreffen, Int. Büro für Gebirgsmechanik Leipzig, Akad. Verlag, Berlin, 1970.
[14] Wisser, E.: Die Gestaltung von Krafthauskavernen nach felsmechanischen Gesichtspunkten, Felsbau
8, Nr. 2, 1990.
[15] Myers, A., John, M., Fugeman, I., Lafford, G., Purrer, W.: Planung und Ausführung der britischen
Überleitstelle im Kanaltunnel, Felsbau 9, Nr. 1, 1991.
[16] Müller, L., Sauer, G., Vardar, M.: Dreidimensionale Spannungsumlagerungsprozesse im Bereich der
Ortsbrust, Rock Mechanics, Suppl. 7, 1978.
[17] Duddek, H.: Zu den Berechnungsmodellen für die Neue Österreichische Tunnelbauweise (NÖT), Rock
Mechanics, Suppl. 8, 1979.
[18] Pacher, F.: Deformationsmessungen im Versuchsstollen als Mittel zur Erforschung des Gebirgsver-
haltens und zur Bemessung des Ausbaues, Felsmech. und Ing. Geol., Suppl. I, 1964.
[19] Rabcewicz, L., Golser, J., Hackl, E.: Die Bedeutung der Messung im Hohlraumbau, Teil I, Der Bau-
ingenieur 47, Heft 7, 1972.

18
[20] Seeber, G., Keller, S., Enzenberg, A., Tagwerker, J., Schletterer, R., Schreyer, F., Coleselli, A.: Be-
messungsverfahren für die Sicherungsmassnahmen und die Auskleidung von Strassentunneln bei
Anwendung der neuen Österreichischen Tunnelbauweise, Strassenforschung, Heft 133, Wien, 1980.
[21] ITA-Richtlinien für den konstruktiven Entwurf von Tunneln, Taschenbuch für den Tunnelbau, Verlag
Glückauf, 1990.
[22] Swoboda, G.: Finite element analysis of the New Austrian Tunnelling Method, 3rd Int. Conf. on Num.
Methods in Geomechanics, Aachen, 1979.
[23] Wanninger, R.: New Austrian Tunnelling Method and finite elements, 3rd Int. Conf. on Num. Methods
in Geomechanics, Aachen, 1979.
[24] Rabcewicz, L.: Die Neue Österreichische Tunnelbauweise, Entstehung, Ausführungen und Erfah-
rungen, Der Bauingenieur, 40. Jg., Heft 8, 1965.
[25] Sauer, G.: Theorie und Praxis der NÖT, Tunnel 4, 1986.
[26] Austrian National Committee of the ITA: In defence of NATM (Communication), Tunnels & Tunnelling,
June 1986.
[27] Hessen, J.: Wissenschaftslehre, Lehrbuch der Philosophie, Erasmus-Verlag, München, 1947.
[28] Locke J., An Essay concerning Human Understanding, Clarendon Press, Oxford, 1975.

19
Adress of the author:

Dr. K. Kovári
Professor of tunnelling
Swiss Federal Institute of Technology
CH-8093 Zurich
Switzerland

20