Document Actions

ijege-13_bs-blikra-et-alii.pdf

Editrice

307

DOI: 10.4408/IJEGE.2013-06.B-28

SUBSURFACE MONITORING OF LARGE ROCKSLIDES IN NORWAY:

A KEY REQUIREMENT FOR EARLY WARNING

ars

H. BLIKra

(*)

, L

ene

KrIsTensen

(*)

& M

arIo

LoVIsoLo

(**)

(*)

Åknes/Tafjord Beredskap, N-6200 Stranda, Norway;

(**)

Centro Servizi di Geoingegneria (CSG), Via Cazzulini 15A - 15010 - Ricaldone (AL) - ITALY

INTRODUCTION

The risk posed by large rockslides in Norway is

due to long run-out, the possibilities for river dam-
ming and the generation of disastrous tsunamis in
fjords (B

LIKra

et alii, 2006).

In order to have a good understanding of large and

complex landslides it is of vital importance to achieve
subsurface geological data, including the depth of the
instability and the related deformation. Borehole in-
strumentation is essential for both the investigations of
subsurface characteristics (sliding planes, depth, etc)
and for real-time operative early-warning. In Norway,
three large rockslides have been instrumented by the
multi-probe DMS system (Differential Monitoring of
Stability), which can be used both in the investigation
phase and later in the real-time operational monitoring.

Subsurface instrumentation have been established

at the Åknes rockslide (B

LIKra

, 2012), the Mannen

rockslide (K

rIsTensen

& B

LIKra

, 2011) and at the Jet-

tan rockslide in northern Norway. Examples will be
given from the Åknes and Mannen rockslides in west-
ern Norway (Fig. 1).

REQUIREMENTS AND CONCEPTS OF

SUBSURFACE MONITORING

Subsurface instrumentation and monitoring are

important both for the investigation stage and for suc-
cessive real-time early monitoring and early warn-
ing (B

LIKra

& K

rIsTensen

, 2011). No guidelines or

handbooks exist, which define specific requirements

ABSTRACT

The risk posed by large rockslides in Norway is

due to long run-out, the possibilities for river dam-
ming and the generation of disastrous tsunamis in
fjords. The investigations, monitoring and early-
warning that have been designed and implemented for
the Norwegian rockslides follows strong requirements
guided in national codes.

Extensive investigations and implementation of

relatively dense sensor network is needed in order to
achieve reliable and robust monitoring of large and
complex rockslides. The Norwegian codes have strong
and specific requirements for using early-warning to
reduce risk. The need of gaining sufficient knowledge
is especially important for the operative handling of
the total early-warning systems.

Subsurface borehole logging and instrumentation

is a mandatory part of the requirements. A real-time
instrumentation with a continuous coverage from the
surface and down to potential sliding planes of the
rockslides is implemented for the Norwegian examples.
These example shows that the multiparametric bore-
hole probe DMS used in these cases can be regarded as
a geotechnical lab within the rockslide, and gives in de-
tail crucial data about deformations, water-pressure and
temperatures at the active zone. This are considered to
be critical data during an acceleration phase.

words

: Early warning, rockslides, instrumentation, sub-

surface monitoring, displacements

L.H. BLIKRA, L KRISTENSEN & M. LOVISOLO

308

International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013

a daily and continuous scheduling.

- Demand of sufficient personnel and competence

for safe and reliable monitoring, warning and evacu-
ation systems.

- Continuous operation of the monitoring systems,

for example technical supervision of sensors, power
supply, communication systems and signal transmis-
sion etc.

-The warning time must be sufficiently long in

order to ensure a proper evacuation. The warning
time shall never be less than 72 hours. Evacuation of
people in the hazard zone must be ended in due time
before the expected rockslide event.

These restrictive and specific regulations and

terms require availability of a daily operative geologi-
cal competence. It is also clear that it puts extensive
demands on the level of knowledge which will be the
basis for geological interpretations. In our opinion,
an extensive investigation program and implementa-
tion of surface and subsurface monitoring systems are
needed in order to fulfill these requirements (see also
B

LIKra

& K

rIsTensen

, 2011).

Therefore, the implementation of surface and sub-

surface monitoring systems related to large and com-
plex rockslides in Norway are mainly based on the
national technical building codes, but the guidelines
in the international standard Eurocode 7 have been
important. It has to be stressed however, that the final
design of the investigations and monitoring must be
based on the local conditions.

The design of the investigation program is impor-

tant for several critical issues. Firstly, the rockslide
scenarios need to be defined as the base for rockslide
and tsunami modeling. Secondly, the position and
distribution of the unstable area and the displacement
pattern are the most important sets of information for
the final design and implementation of a proper moni-
toring system. Thirdly, a reliable knowledge platform
is needed in order to perform reliable and real-time
operative monitoring and early warning. The under-
standing of the deformation dynamics is especially
important during critical events, when decisions re-
garding alarm levels and evacuation have to be taken
on relatively short notice. In order to reach a good un-
derstanding of the landslide it is of vital importance
to achieve subsurface geological data, including the
depth of the instability and the related deformation.
The investigations should provide data for a realistic

for the type and level of investigations needed in or-
der to perform reliable monitoring and early-warning
of large landslides. However, the European Standard
EN 1997-2:2007 (Eurocode 7) describes principles
and requirements related to geotechnical design and
ground investigations.

The regulations in the Norwegian technical build-

ing codes (TEK §7-4) from 2010 introduces some
major requirements that are needed to be fulfilled in
order to allow further construction and development
in tsunami hazard areas generated by large rockslides.
The population safety needs to be taken care of by
real -time monitoring, warning and evacuation. The
warning time shall not be shorter than 72 hours and
the evacuation time shall be maximum 12 hours. The
notes linked to the regulations also include specific
terms related to the monitoring and early-warning
systems:

- A system for daily continuous monitoring of the

conditions, e.g. by measurements of displacements.
Independent monitoring instruments and duplicate
data communication systems must be implemented.
The early-warning system needs to be based on real-
time monitoring, without long time delay.

- Need of sufficient competence for management

of monitoring network and interpretation of results on

Fig. 1 - The monitored rockslides in Norway (Åknes, Heg-

guraksla, Mannen and Jettan)

SUBSURFACE MONITORING OF LARGE ROCKSLIDES IN NORWAY: A KEY REQUIREMENT FOR EARLY WARNING

Editrice

309

inside the case and joints, so providing a complete
protection from external environment.

THE ÅKNES ROCKSLIDE

The Åknes rockslide is located on the northwest

flank of Sunnylvsfjorden in western Norway (Fig. 1). It
has an estimated volume of up to 54 Mm

and is mov-

ing at a velocity of up to 8 cm/year. The risk is associ-
ated with the generation of catastrophic tsunamis, hav-
ing run-up potential of up to 80 m in nearby villages
(B

LIKra

, 2012; G

LIMsdaL

& H

arBITz

, 2011). The rock-

slide is located in the Western Gneiss Region and is seat-
ed in medium-grained granitic and granodiorite gneiss
of Proterozoic age (B

raaTHen

et alii, 2004; G

anerød

alii, 2008). The gneiss has well developed foliation and
mineral banding (B

raaTHen

et alii, 2004) and numer-

ous centimetre to decametre-scale, close to tight folds.
At Åknes, biotite-rich layers up to 20 cm thick coincide
with zones of high fracture frequency (G

anerød

et alii,

2008). The foliation generally dips parallel to the slope
surface. Well defined, very steep, sharp folds are related
to the tension cracks at the top of the landslide (G

anerød

et alii, 2008; J

aBoyedoff

et alii, 2011).

Morphological investigations have revealed sev-

eral characteristic features of the landslide (Fig. 3),
including a prominent upper fracture system that can
be followed for more than 500 m.

The slope-parallel foliation and weak biotite-rich

layers control the large-scale displacement dynamics. A
large depression or graben has developed in the upper
west corner of the rockslide (Fig. 3), with a total ver-
tical displacement of 20-30 m. Tension cracks are also
present in the upper and the middle parts of the slope.

3D geometric model of the instability. Therefore some
selected borehole drillings combined with geophysi-
cal investigations and instrumentation are required in
order to investigate and monitor active displacements.

SUBSURFACE MONITORING TECHNI-

QUE

Borehole instrumentation is essential for both the

investigation of subsurface characteristics (sliding
planes, depths, etc) and for real-time operative early-
warning. The Norwegian site characteristics with steep
slopes and harsh environment lead to very costly drill
holes, since all transportation is by air lift and the water
supply for core-drilling is often complicated and costly.
It is thus important that the expensive drill holes can be
used for multipurpose in terms of instrumentation. The
concepts that have been chosen for the Norwegian bore-
hole instrumentation have the following requirements:
- be used both for investigation and later real-time mon-

itoring. This means that it must be a flexible system
to be used in several portions of the boreholes;

- allow to measure several parameters (displacements,

water pressure, temperature);

- be easy to install, robust and capable to survive as

long as possible inside the borehole;

- provide real-time data and as much as possible re-

mote controlled.
At Åknes and Mannen, four boreholes have been

instrumented by the DMS system (Differential Moni-
toring of Stability, L

oVIsoLo

et alii, 2003). This system

meets the requirements and specifications listed above.
The DMS column is like a sensorized spinal cord or
long, thin, bundle of hard tubular modules connected
to each other by specially designed joints (Fig. 2).

The adopted DMS systems are 100 to 120 m long

columns measuring the movement in 2D. The 120 m
long columns consists of a total of 245 sensors. The
sensors include biaxial inclinometers, temperature
sensors as well as piezometers and digital compasses
in selected modules. The DMS system has a very ro-
bust structure in order to support high pressure, trac-
tion and deformations. Each module contains all the
electronic devices for measurement, control and dig-
ital transmission. The special 2D/3D strong and flex-
ible joints allow continuous adaptability to bending
and twisting of the drilling hole, whilst maintaining
rigorously the orientation with respect to a reference
system defined during installation. All the cables are

Fig. 2 - The DMS borehole installation of a 120 m long

column at Mannen (left) and an overview of the

system (right)

L.H. BLIKRA, L KRISTENSEN & M. LOVISOLO

310

International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013

The surface monitoring system is based on exten-

someters/crackmeters, tiltmeters, single lasers, GPS, a
total station and a microseismic network, in addition
to a meteo station (Fig. 3). Large efforts have been
paid to the establishment of subsurface investigations
and monitoring systems in deep boreholes. Deep core
drillings down to 200 m depth were completed at 3
localities (Fig. 3), with geophysical logging and core
logging (G

anerød

et alii, 2007). In addition to the

geophysical logging (e.g. temperature, conductivity,
resistivity, velocity), several methods were applied
for identification and characterization of groundwater
flow in the boreholes (r

ønnInG

et alii, 2006; T

Höny

2008).Three boreholes have been instrumented with
100 to 120 m long DMS instrumentation (see location
in figure 3 and depth data in Figs 4 and 5).

The DMS data has documented a well-defined up-

per sliding zone in two boreholes at depth of 35 and 50
m depth. This is above the water-level measured in open
standpipes. The displacements from the last two years
in upper borehole are shown in figure 4. The largest de-
formation is seen at 49-50 m depth, and the differential
displacements along the axis of movement direction
(SW, 225°), show a total annual displacement of nearly
3 cm/year. Some less-pronounced sliding planes are
also seen at 9-10 m, 28-30 m and 40-42 m depth (Fig.
4). At the moment the deeper sectors seem to be stable.

The mode of deformation varies from continuous

Prominent slide scarps characterize the east side of the
deep canyon that defines the western landslide bound-
ary. Small slide scars also characterize the lower part
of the rockslide. Springs discharge water at the lower-
most part of the slope at about 100 m asl, and smaller
springs are located in the middle part of the landslide
(f

reI

, 2008).

Fig. 3 The Åknes rockslide with the main scenarios in

different colors, geomorphological features and

the instrumentation (modified from B

likra

, 2012)

Fig. 4 - An example of recent DMS data from the upper borehole at Åknes measured in the period 22nd of November 2010

until the 14th of December 2012 (see location in figure 2). a) total displacements, b) total displacements along the

main sliding direction towards SW , c) the differential displacements and d) the differential displacement in the main

sliding direction (right). It clearly documents the sliding zone at 50 m depth, in accordance with core and borehole

data. Note also several sliding zones on shallower depths

SUBSURFACE MONITORING OF LARGE ROCKSLIDES IN NORWAY: A KEY REQUIREMENT FOR EARLY WARNING

Editrice

311

A and B in Fig. 5 (d

aHLe

et alii, 2011). Runout mod-

elling shows that a rockslide would destroy the road
and railroad at the valley bottom as well as some
farms. Furthermore, the rock debris may dam the
river Rauma, with a subsequent risk of dam collapse
and flooding downstream of the river.

The bedrock consists of Proterozoic gneiss (s

aIn

ToT

et alii 2011) with structural weaknesses from the

tectonic deformation. At the top, where the main back-
crack developed, foliation is near vertical. In the upper
part of the moving area, a borehole showed foliation
dipping about 30° to the north and downslope, but fur-
ther down the pattern is more complex, as the gneiss is
intensely folded. The bedrock within the instability is
extremely fractured. The morphology is characterized
by the well-defined backscarp and numerous smaller
slide scars (Fig. 6). These smaller scars indicate multi-
ple recent rock falls or smaller rock avalanches.

A 137 m deep borehole was drilled in the upper

part of the instability, see Fig. 6. Core logging demon-

displacements through time to suddenly and abrupt
steps in movement rates. The advantages with real-time
monitoring along the entire depth of the rockslide are
the possibility to observe in detail at a certain time and
depth windows, what happen along the sliding surfaces.
This is exemplified with the DMS data for sensor 77 at
49-50 m (see Fig. 4) depth together with the water-level
fluctuations and temperatures measured in the same
borehole (Fig. 5). An increase in deformation along the
sliding plane occurred in late March 2011, followed by
a sudden increase in the water level (7 m) at the 2

April 2011. The temperature also dropped by 1°C at the
same time, indicating inflow of water from snowmelt.
After the peak the water-level dropped again some me-
ters and stabilized nearly 5 m higher than before the
event. The displacement continued at the same rate for a
long time. This event can be explained with an internal
change in fractures opening due to the deformations,
which has led to a change in the hydrological condi-
tions. The stabilized higher water level led to a general
increase in displacement of about 50% compared to the
years before, and demonstrates that the stability con-
ditions can be changed drastically due to the internal
dynamics of the rockslide. This highly unstable area
of the rockslide is presently carefully followed, and
we are especially concerned about the situation during
the snowmelt periods. A new borehole drilled in 2012,
close to the upper site, documented a 40 cm thick brec-
cia at 63 m below ground surface that seem to be cor-
related to the 49-50 m zone. This indicates that a large
collapse of the south-western flank could also occur as
one large event (Fig. 3). The DMS instrumentation is
like a geotechnical lab that gives us the possibility to
study and follow in detail the internal deformations,
water-pressure and temperature conditions.

THE MANNEN ROCKSLIDE

Mannen is a 1295 m high mountain which is

a part of a plateau above the steep, glacial eroded
Romsdalen valley in western Norway. A part of the
plateau (100 mill. m

) towards the valley is bounded

by deep clefts indicating past movement, but most
of this seems not to be active today. A portion of
the outer edge, of possibly 15-20 million m

, moved

downslope 15-20 m from the plateau, and a smaller
part of this (2-3 mill. m

) shows an annual displace-

ment of 2-5 cm dipping 45-50 against ENE (Fig. 6).
These blocks are the basis of the proposed scenarios

Fig. 5 - Time line for sensor 77 at 49-50 m depth at the up-

per borehole at Åknes, see location in figure 2. The

upper curve shows the displacement and water lev-

el in 2011, while the lower diagrams show displace-

ment, water level and water temperature for the se-

lected time period between the 1st of March until

of May 2011. Note that the increased displace-

ments start about one week before the increased

water level and lowering of water temperature

L.H. BLIKRA, L KRISTENSEN & M. LOVISOLO

312

International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013

strated that the bedrock is highly fractured to a depth of
113 m, with several levels of breccia as well as bedrock
crushed to silt fraction. It was instrumented with a 120
m long DMS column, measuring displacement or tilt
every meter (Fig. 7).

The displacement pattern during a 15 months pe-

riod between 2011 and 2012 along the DMS column
in the upper borehole is shown (Fig. 7). Sliding planes
are clearly defined at 24 and 28 m depth, where also the
recovered borehole core and the televiewer logging re-
vealed well-defined breccias (Fig. 7, middle). However,
at the moment we cannot exclude the presence of other
sliding planes at greater depths. The movement along
the sliding planes is towards NE, and this is in accord-
ance with the surface GPS displacement data (K

rIs

Tensen

& B

LIKra

, 2011). The piezometers installed in

the column show that the entire borehole is dry, and the
temperature conditions during the winter is close to 0.
This indicates that the temperature regime at the local-
ity is close to permafrost conditions.

The time-line data (Fig. 7, upper right) from the

sliding at 24-25 m depth shows increased displacement
starting during the snowmelt period from late May to
middle June. In general, the displacement rate is lim-
ited during the winter time, and the increased velocity
in late spring is interpreted as an effect of water seeping
into fractures and percolating along the sliding planes
reducing the shear strength even if the main landslide
body remains prevalently dry.

CONCLUSIONS

Extensive investigations and implementation of

relatively dense sensor network is needed in order to
achieve reliable and robust monitoring of large and
complex rockslides. The Norwegian codes have strong
and specific requirements for using early-warning sys-
tem to reduce risk. The need of sufficient knowledge is
especially important for the operative handling of the
total early-warning system.

Subsurface borehole data and instrumentation is a

Fig. 6 - The Mannen rockslide located on the south-western side of the valley Romsdalen. The two scenarios, A & B are indi-

cated in yellow as well as the position of the GB InSAR system in the valley

SUBSURFACE MONITORING OF LARGE ROCKSLIDES IN NORWAY: A KEY REQUIREMENT FOR EARLY WARNING

Editrice

313

ACKNOWLEDGEMENTS

The investigations and monitoring at the Åknes

and Mannen rockslide have been performed in cor-
poration with a number of national and international
partners, including the Geological Survey of Norway,
Norwegian Geotechnical Institute, NORSAR, Norwe-
gian University for Science and Technology, University
Centre of Svalbard, University of Bergen, University
of Oslo, University of Milano-Bicocca, University of
Lausanne, ETH, and the companies Ellegi srl, Cautus
Geo, and Fugro.

vital part of the requirements. A real-time instrumen-
tation with a continuous coverage from the surface
and below potential sliding planes of the rockslides is
implemented for the Norwegian examples. The sub-
surface monitoring has given critical data for the eval-
uation and revision of scenarios (depth and volume),
and for the understanding of deformation mechanisms
and seasonal changes. These example shows that the
DMS instrumentation used in these cases can be re-
garded as a geotechnical lab within the rockslide, and
gives in detail crucial data about deformations, water-
pressure and temperatures at the active zone. This will
be critical data during an acceleration phase.

Fig. 7 Data from the upper borehole at the Mannen rockslide. DMS depth data from the 2nd of September 2011 until the 14th

of December 2012 (left), televiewer images (middle), a time history from the sensor at 24-25 m depth (upper right)

and a photo of the clay-rich breccia at 28 m depth (lower right). Note the documented displacements at 24 and 28 m

depth. The displacement spike at 41 m might be due to a problem with one of the sensors. The televiewer images are

oriented and show also the dip and direction of foliations, frac

tures and breccias. The images show that the sliding

planes dips towards N-NE

L.H. BLIKRA, L KRISTENSEN & M. LOVISOLO

314

International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013

REFERENCES

LIKra

. (2008) - The Åknes rockslide; monitoring, threshold values and early-warning. In: C

Hen

z., z

HanG

J.-M., H

K., W

f.-Q. & L

z.-K. (

eds

.). Landslides and engineered slopes. Proceedings of the 10

International Symposium on Landslides

and Engineered Slopes, 30 June - 4 July 2008, Xi'an, China. Taylor and Francis. ISBN: 978-0-415-41196-7.

LIKra

L.H. (2012) - The Åknes rockslide, Norway. In: C

LaGue

J.J. & s

Tead

d. (

eds

.). Landslides: types, mechanisms and

modeling. Cambridge, 323-335.

LIKra

L.H. & K

rIsTensen

L. (2011) - Monitoring concepts and requirements for large landslides in Norway. Proceedings of the

Second World Landslide Forum Abstracts. 3-7 October 2011, Rome, Italy.

LIKra

L.H., L

onGVa

o., B

raaTHen

a., a

nda

e., d

eHLs

J. & s

TaLsBerG

K. (2006) - Rock-slope failures in Norwegian fjord areas:

examples, spatial distribution and temporal pattern. In: e

Vans

s.G., s

CarasCIa

uGnozza

G., s

TroM

a. & H

erManns

r.L.

(

eds

.). Landslides from massive rock slope failure. NATO Science Series: IV: Earth and Environmental Sciences, 59.

raaTHen

a., B

LIKra

L.H., B

erG

s.s. & K

arLsen

f. (2004) - Rock-slope failures of Norway; type, geometry, deformation

mechanisms and stability. Norwegian Journal of Geology, 84: 67-88.

aHLe

H., s

aInToT

s., B

LIKra

L.H. & a

nda

e. (2010) - Mannen i Romsdalen - geofaglig oppfølging 2009. Geological Survey of

Norway, Report 2010.022. (In Norwegian)

reI

C. (2008) - Groundwater Flow at the Åknes Rockslide Site (Norway). Results of multi-tracer test. M.Sc. thesis, ETH Zurich,

Zurich, Switzerland.

anerød

G.V., G

rønenG

G., a

ardaL

I.B. & K

VeLdsVIK

V. (2007) - Core logging of seven boreholes from Åknes, Stranda

municipality, Møre and Romsdal County. Geological Survey of Norway report no. 2007.020, 222 pp.

anerød

G., G

rønenG

G., r

ønnInG

J.s., d

aLseGG

e., e

LVeBaKK

H., T

ønnesen

J.f., K

VeLdsVIK

V., e

IKen

T., B

LIKra

L.H. &

raaTHen

a. (2008) - Geological model of the Åknes rockslide, western Norway. Engineering Geology, 102: 1-18.

LIMsdaL

s. & H

arBITz

, C.B. (2011) - Numerical simulations of tsunamis from potential and historical rock slides in Storfjorden:

hazard zoning and comparison with 3D laboratory experiments. Norwegian Geotechnical Institute Report 20051018.

aBoyedoff

M., o

ppIKofer

T., d

erron

M-H., B

öHMe

M., B

LIKra

L.H. & s

aInToT

a. (2011) - Complex landslide behaviour and

structural control: a 3D conceptual model of Åknes rockslide, Norway. In: J

aBoyedoff

M. (

.). Slope Tectonics. Geological

Society of London Special Publication, 351: 147-161.

rIsTensen

L. & B

LIKra

L.H. (2011) - Monitoring displacement on the Mannen rockslide in Western Norway. Proceedings of the

World Landslide Forum Abstracts. 3-7 October 2011, Rome, Italy.

oVIsoL

, M., G

HIroTTo

s., s

CardIa

G. & B

aTTaGLIo

M. (2003) - The use of Differential Monitoring Stability (D.M.S.) for remote

monitoring of excavation and landslide movements. In: M

yrVoLL

(

.) - Proceedings of the 6

International Symposium on

Field Measurements in Geomechanics, 519-524, Oslo, Balkema.

ønnInG

J.s., d

aLseGG

e., e

LVeBaKK

H., G

anerød

G. & T

ønnesen

J.f. (2006) - Geofysiske målinger Åknes og Tafjord, Stranda

og Nordal kommuner, Møre og Romsdal. Geological Survey of Norway Report 2006.002 (in Norwegain).

aInToT

a., e

LVeBaKK

H., G

anerød

G., o

ppIKofer

T. & f

arsund

T. (2011) - Mannen unstable rock slope (Møre & Romsdal):

logging of borehole and drill core KH-01-10, geomorphologic digital elevation model interpretation and displacement
analysis by terrestrial laser scanning. NGU report no. 2011.026.


	Content © 2018 Italian Journal of Engineering Geology and Environment This is the new scientific journal of the Research Center CERI