Innovative management of geological data as a tool for safe tunnelling

Abstract

The construction of a tunnel in hard or soft ground requires adaptation to the material provided by the prevailing geological conditions, whose history of origin, properties, three-dimensional structure and groundwater conditions all have to be taken into account. Geologists at the site work within a team of experts. Their job is to observe and to evaluate the encountered geological, hydrogeological and geotechnical conditions and record all the key parameters and influential factors, which were identified during the geotechnical design phase. Based on the field observations, the geologists give a prognosis of the further geological conditions for the future tunnel advance. Unknown and unfavourable geological conditions during tunnel advance carry extremely high risks for the whole excavation progress. Therefore, the prognosis must be given as accurately as possible to avoid dangers. Geological observations in underground constructions, such as tunnels, are almost exclusively spatial information and therefore must be recorded and analysed in three dimensions. In addition to the types and properties of ground encountered, the positional relationships of the geological bodies to each other and the discontinuity sets of the rock mass have a great influence on the stability of the underground structure. Therefore, powerful software tools were developed, which help the expert on the site to deal with the high amounts of data produced during face mapping and record the geological information in a 3 dimensional system. With this data basis it is now possible to generate a detailed prognosis of the geometry of geological structures with geostatistical methods. In the following paper, such a system is presented.

Full text

1
Innovative management of geological data
as a tool for safe tunnelling
Dr. Giorgio Höfer-Öllinger and Dr. Franz Peter Weichenberger
GEOCONSULT, Hölzlstraße 5, 5071 Wals bei Salzburg,
AUSTRIA,giorgio.hoefer- oellinger@geoconsult.eu
GEOCONSULT, Hölzlstraße 5, 5071 Wals bei Salzburg, AUSTRIA,
franz.weichenberger@geoconsult.eu
ABSTRACT
The construction of a tunnel in hard or soft ground requires adaptation to the
material provided by the prevailing geological conditions, whose history of origin,
properties, three-dimensional structure and groundwater conditions all have to be
taken into account. Geologists at the site work within a team of experts. Their job
is to observe and to evaluate the encountered geological, hydrogeological and
geotechnical conditions and record all the key parameters and inuential factors,
which were identied during the geotechnical design phase. Based on the eld
observations, the geologists give a prognosis of the further geological conditions
for the future tunnel advance. Unknown and unfavourable geological conditions
during tunnel advance carry extremely high risks for the whole excavation
progress. Therefore, the prognosis must be given as accurately as possible to
avoid dangers. Geological observations in underground constructions, such as
tunnels, are almost exclusively spatial information and therefore must be recorded
and analysed in three dimensions. In addition to the types and properties of ground
encountered, the positional relationships of the geological bodies to each other
and the discontinuity sets of the rock mass have a great inuence on the stability
of the underground structure. Therefore, powerful software tools were developed,
which help the expert on the site to deal with the high amounts of data produced
during face mapping and record the geological information in a 3 dimensional
system. With this data basis it is now possible to generate a detailed prognosis of
the geometry of geological structures with geostatistical methods. In the following
paper, such a system is presented.
1. Introduction
The construction of underground structures poses a special challenge for the
project team. The goal of geological documentation is to record the conditions as
precisely as possible and to provide this information to the whole team. To this
end, the geological situation is recorded during the tunnel advance. At the tunnel
face and/or wall geological sketches are drawn manually, digital photos are taken
and structural measurements are made. The geologist evaluates on the site which
kinds of rock or ground occur, which properties they have and their geometrical
relationship to each other. Rock mass is typically fractured by faults, fault zones
or simply by sets of joints. Groundwater maybe owing into the heading at certain
positions, like out of installed anchors or bolts or directly out of discontinuities,
which have some aperture. For obtaining structural geological data, faults or joints
are measured with a geological compass or - more recently - by three dimensional
image analysis. Samples of rocks or groundwater inows are also taken at regular
intervals. It has become standard practice now to store and evaluate the observed
data digitally. Before the introduction of information technology (IT), documentation
was done on “paper” and the results were shown in the form of sketches, drawings
and reports. With the spread of IT, signicant limitations, which occur especially on
larger projects, have been removed.
Due to the sheer amount of information gained, it was formerly difcult or
much too time-consuming to compile the signicant ndings quickly to answer the

2
questions that arose.
The display and analysis of data together with data from other professions
was often only possible to a very limited extent. But exactly this interdisciplinary
view can guarantee a successful tunnel project. If complicated and complex
geological conditions occurred, extrapolations could only be done manually and
by simplifying the model to a high degree.
In the last two decades, specialist geological information systems have been
developed to gain new possibilities in geological tunnel documentation. The use of
these systems is now standard practice on many sites. Based on experience with
the implementation of the system TUGIS.NET [1] with more than 25 deployments
until today and extensive feedback from various engineering geologists at the sites
we present:
• A requirement prole
• The state of the technology
• Examples showing how to improve tunnel safety by
generating appropriate extrapolations
• Three-dimensional modelling technique of important geological
ndings to further tunnel advance stages
Figure 1 shows how the geological documentation system is integrated in the
general project environment.
Figure 1: Workow of geological tunnel documentation
2. Digital tunnel documentation requirements
The base of all further work is a proper and accurate database derived from the
geological observations in the tunnel. Only well-structured data can be used for
further analysis.
Therefore, during the digitalization procedure the information must be
standardized and categorized. The following points have to be considered for the
deployment of a suitable digital geological documentation system:
• All geological and hydrological information have to be recorded digitally
into one dataset to provide in the future a rapid availability of the data.
• Since geological observations in a tunnel are nearly always three-
dimensional information, coordinates in three dimensions have to be used
in the system.

3
• Particularly for geological and geotechnical parameters dened at the
design phase, it is necessary to standardise data inputs. The values used
as geological attributes are to be categorised, classied and as far as
possible quantied. Additionally textual descriptions are also necessary.
• Geological documentation during tunnelling often happens in a time-
critical environment, therefore a graphical user interface is necessary for
data input and it should be designed so that observations can be entered
extremely quickly and with as few errors as possible.
• Faulty data must be eliminated by manual correction or better by automatic
data validation or plausibility checking before data analysis or visualisation.
• Because of the high amount of data generated, a capable data storage
system has to be used, which can also deal with multiple users. Lost
datasets can only be recovered laboriously, so a high degree of data
security should be provided. Modern database systems are most useful
for these aims.
• Analysis functions, visualisations and three-dimensional modelling
represent the added value of the digital processing method. There must
be integrated tools in the software therefore to analyse, visualize and
extrapolate geological structures quickly. Complicated geological settings
require even more sophisticated data processing for example by using
geostatistical methods, which can be done with modern GIS applications.
Interfaces to these applications must be implemented.
3. Data structures
Geological data are mostly three-dimensional. During the development process
over many years, it turned out that it is best to classify geological structures similar
to data structures in geographic information systems (GIS). There the themes are
categorized by their “type” of geometry. Point data have only one coordinate; an
example would be a dip-measurement of a fault plane at a certain position. The
intersection of a joint surface with the tunnel face is geometrical equivalent to a
polyline. A geological body - a dyke for example - intersects with the tunnel face in
the form of a polygon. Volume bodies, like the model of a whole lithological layer
need data structures, which normally cannot be found in standard GIS applications.
GIS datasets don’t only store the geometry but also the so called attributes. In
geological documentation this attribute data corresponds to the properties of the
geological ndings. An example would be the type of the rock, its uniaxial strength
or its grade of weathering. Using GIS compatible data formats have the advantage
of an easy export in such systems and the data is as “BIM ready” as possible at
this time. Building Information Modelling (BIM) is a method of achieving holistic
consideration of the entire lifecycle of a tunnel. The essential objectives are to
improve communication and the efciency and quality of a construction project
by creating a comprehensive, integrative digital model [2]. This is undertaken in
a 3D environment, in which the construction works are already created virtually
during the design stage. Regarding the specialist geological application of BIM,
the situation at the moment is that the structure and the geological conditions
are maintained in different models. If the geological ndings, which are already
available in three-dimensional form, are integrated into the BIM, then the interaction
of the structure with the geological-hydrogeological model can be represented.
This would offer advantages regarding the design of suitable support measures,
tunnelling parameters or more accurate cost estimation. In order to make a
specialist geological information system compatible with a BIM system, appropriate

4
interfaces have to be dened. Standardisation efforts are currently underway for
BIM for underground construction.
4. Data acquisition and digitalization
The methodological basis for digital recording is the working procedure produced
for the construction of the Tauern Tunnel (1st tube) [3]. The geologist draws a sketch
at the exposed surface in the tunnel and notes all the signicant parameters. An
example is shown in Figure 2.
Figure 2: Field sketch of a tunnel face and digital image
Besides sketching the exposed area with pen and paper digital images are taken.
Additionally, more advanced systems like ShapeMetriX3D [4] are available today.
The photos can, after being scaled and georeferenced, serve as the basis for
geological mapping. ShapeMetriX3D allows the derivation of three-dimensional
information from two referenced images. This data can be imported in TUGIS.
In order to place the geological observations into a spatial context, they have to
be georeferenced. In TUGIS.NET, the axes of the tunnel headings are entered
as part of the initial conguration. The boundaries of the recorded areas –
such as regular tunnel face cross-sections– are congured in advance. Both of
these can be updated later on site. With this data basis, any geological record
can be unambiguously referenced. When the geometry of geological ndings is
drawn (Figure 3), coordinates are lled into the database and the properties are
documented.

5
Figure 3: Graphical User Interface TUGIS.NET, digitizing and
attributing of the lithologies
Figure 4 shows as an example a graphical representation of geological objects. The
intersections of lithological bodies with the face are shown as transparent coloured
polygons. Fault plane intersections (red) and joint intersections (green) are shown
as lines. Measurements of the geological structure (dip-direction, dip-angle) are
shown as uniform squares with their insertion at the point of measurement. The
view of the model can be rotated as required.
Figure 4: GIS model of the geological situation of a tunnel section with several
tunnel faces, the tunnel wall visible is derived from external
3D measurement data

6
Another aspect is the quantication, categorisation and classication of the
data entries. The objective is a common designation of the geological ndings.
Data evaluations can be carried out for comparisons of the predicted and actual
situations or to answer specic questions. When the quantity of data is large, an
automatic analysis is preferred, for which purpose the ndings have to be available
in an appropriately structured form. For the attributes of geological ndings,
the individual categories are made available in a selection list. When digitising
the geological conditions, the geologist has to dene a clear boundary in each
case. Without clear boundaries statistical evaluations, such as distributions of
different lithologies in a certain section of a tunnel heading, would be impossible.
Implemented classication systems like RMR [5] prot a lot by proper and validated
input data.
5. Extrapolations, geometrical modelling and prognosis for future headings
In the design phase of a tunnel project a geological and hydrogeological model is
established, which should include all the available data from literature, the geological
surface mapping, from boreholes and from groundwater measurements. The goal is
to describe the geological conditions as precisely as possible to avoid unexpected
situations and risks when the tunnel advances. Possible dangers are for example
the existence of fractured fault zones with unfavourable material properties such as
fault gouges or disadvantageous discontinuity orientations to certain heading axes,
open water bearing joint systems especially where a high groundwater level above
the tunnel exists or the occurrence of minerals like asbestos which are hazardous
to the human health.
Observations and investigations at the design phase are always limited by many
factors. The geological conditions, which are observed on the surface, can deviate
signicantly from those found at tunnel level, caused for example by weathering or
gravitational movement processes on slopes.
Boreholes are pinpricks in comparison to the dimension of a typical tunnel project.
Therefore, a geological model always includes an overriding interpretation of the
ndings as well. Today much effort, time and expert knowledge is invested in
investigation campaigns in advance of the construction phase but some details of
the whole geological framework often remain uncertain.
Therefore, it is a very important aspect to receive a continuous updating of the
geological prognosis during the construction phase. The data, which is obtained
as the tunnel is excavated, is much more detailed than that from the exploration
or design phase and is thus suitable for the production of extrapolations and
accurate predictions. Another aspect is that the observations in the tunnel are
nearby to the position where an extrapolation is required and the diameter of the
tunnel prole allows - in contrast to boreholes – to observe geological structures
in the dimensions of some metres of length. Fault surfaces for example, which are
striking subparallel to the tunnel axis can be followed over a long distance and at
several tunnel faces. So the input data obtained in the construction phase is much
more suitable for accurate modelling and extrapolation. As the tunnel advances,
the preliminary geological model can therefore be rened with every documented
tunnel face or wall.

7
As an example, if a fault zone causing failures or bearing risks is encountered on
the excavation of one tube already, the project team is well prepared for an equal
situation in a not yet started nearby second tube or cross passage. The support
of the tunnel can be adapted and optimized in the prognosed area and additional
investigations - for example, the drilling of horizontal boreholes - can be planned
and carried out. This reduces the potential risk of failure of structures and helps to
save costs. Examples of the extrapolations of geological structures are shown in
Figure 5 and Figure 6.
Figure 5: Simple planar extrapolations of fault planes from an already
excavated tube to a not yet started one
Boundaries can also be approximated as curved surfaces, with geostatistical
methods being used as shown in Figure 6.
Figure 6: Curved extrapolation of a lithological boundary to a not yet
excavated tube

8
For safe tunnelling, it is important not only to nd potential risks and to model them
but to communicate the results of this geological research at the site to the whole
project team. Three-dimensional visualizations of the same give a comprehensive
and comprehensible view to everybody involved.
6. Conclusion
By using digital geological documentation applications, a pre-existing geological
model can be updated continuously during the excavation works. Using the data
from this model, a prognosis for not yet excavated parts of the same heading or
other headings of the tunnel project can be carried out. In order to achieve this,
a proper data structure is necessary. Using this system, in combination with GIS
applications, is already state of the art. Geological information can be visualised
and analysed together with external data (deformation measurements, TBM data or
groundwater measurements on the surface for example) in geographic information
systems. Using GIS means 2 ½ dimensional modelling for geological ndings. A
more sophisticated approach takes volume bodies into account as well. Therefore,
some specialized applications exist and they are widespread in mining already.
There are some main advantages of real three-dimensional modelling, because
then cross sections can be derived at every position in the model. For advanced
hydrogeological analysis, volume bodies are necessary as well. The target of
recent developments is it to provide the geological documentation system with
functions for a data export and interfaces to fully 3D-capable applications.
Further developments will enable the integration of geological ndings and models
into BIM compatible data structures, the work therefore can be done as soon as
the interfaces for BIM in this area are dened. Standardisation efforts are currently
underway for BIM for underground construction.
7. References
[1] Weichenberger, F.P.: Konzeption und Implementierung eines digitalen
Systems zur ingenieurgeologischen und hydrogeologischen Bearbeitung von
Tunnelbauprojekten – Dokumentation, Analyse, Prognose. Dissertation, Universität
Salzburg, 2008.
[2] Pläsken, R., Flandera, T., Keuschnig, M.: Workshop Building Information
Modeling – (R)Evolution? Georesearch Forschungsgesellschaft mbH, Salzburg
2017.
[3] Fehleisen, F., Halbmayr H., Kunz F., Kaiser J.: Tunneldokumentation Geologie
und Ausbruch, Tauerntunnel. Bericht im Auftrag der ÖSAG, Salzburg, 2008.
[4] Gaich, A. und Pischinger, G.: 3D images for digital geological mapping.
Geomechanics and Tunnelling 9 (2016), No. 1, pp. 45–51.
[5] Bieniawski, Z.T.: Engineering classication of jointed rock masses, Trans S. Afr.
Inst. Civ. Engrs. 15, pp. 335–344, 1973.