1

GEO 4180The Åknes rock slopeDr. Vidar KveldsvikNGI

Content• Background on large rock slides• Triggers• Stability analysis• Risk mitigation (risk reduction)• The Åknes rock slope

2

Background

Landslides due to massive rock slope failures represent a major geological hazard in many parts of the world.

Volumes 105 to 1010 m3.

Low-frequency /high magnitude events which occur mainly in areas of high relief

Rock avalanching is extremely rapid (>30m/s) flow-likemovements of large volumes (>106 m3) which are beeingincreasingly fragmented during the runout process.

High mobility and long runout make rock avalanches some ofthe most destructive geological processes on the Earth’ssurface.

Background

10

100

1000

10000

0.1 1 10 100 1000 10000 100000

Volume (106 m3)

Swas

h he

ight

(m)

vmax = (2ghsw)0.5

vmax = maximum velocity g = gravitational acceleration hsw = swash height

vmax = 20 m s-1

vmax = 130 m s-1

3

Background

• Failure may often be preceded by observableslope deformation:– Growth and widening of

tension cracks– Increased rock fall activity– Increased disaggregation of

the intial failure mass on theslope

Background

Secondary processes from massive rock slope failuresinclude:

Landslide dams

Landslide generated waves (tsunamis)

4

Usoi dam and LakeLake Sarez

The volume of the landslide was 2.2 km3

Scarp of the landslide

BackgroundThe 600m high Usoi Dam on Lake Sarez in Tajikistan

Artist’s depiction of tsunami at Geiranger

Background

5

Alaska 1958

540 moh.

Background

Background

From Blikra et al. (2006)

Historical: ~1700 - present

6

Background

Åknes

Large rockslide

Background

7

Loen, 1905

Tafjord, 1934

Western Norway 1900’s:

3 rockslides causing tsunami

Caused 175 fatalities

Loen, 1936

Background

Triggers

• Earthquake (>M6?)• Water pressure and its fluctuations• Erosion• Permafrost thaw• Weathering and breakdown of fracture asperities• Progressive failure in intact rock bridges

8

Stability analysis

Stability analysis

9

Stability analysis

Stability analysis

10

Stability analysis

• Material properties– Intact rock properties (σc, E, ν, σt)

– Often more important: shear strength parameters of fractures

⎥⎦

⎤⎢⎣

⎡+⎟⎟⎠

⎞⎜⎜⎝

⎛×= r

nn

JCSJRC ϕσ

στ'

logtan' 10

Where

τ: peak shear strength

σ’n: effective normal stress

JRC: joint roughness coefficient

JCS: joint wall compressive strength

φr: residual friction angle.

φr = φb (basic friction angle) for fresh unweatheredjoints

r: Schmidt hammer rebound on a joint surface

R: Schmidt hammer rebound on intact rock

⎟⎠⎞

⎜⎝⎛ ×+−=

Rr

b 2020r ϕϕand

Stability analysis

11

⎥⎦

⎤⎢⎣

⎡+⎟⎟

⎠

⎞⎜⎜⎝

⎛××=×+= r

nnini

JCSJRCc ϕσ

σϕστ'

logtan'tan' 10

φa

Stability analysis

Stability analysis

12

Stability analysis

Stability analysis

13

⎥⎦

⎤⎢⎣

⎡+⎟⎟⎠

⎞⎜⎜⎝

⎛×= r

nn

JCSJRC ϕσ

στ'

logtan' 10 ⎟⎠⎞

⎜⎝⎛ ×+−=

Rr

b 2020r ϕϕand

Stability analysis

Risk reduction: What is an Early Warning System?

• In common usage, an EWS is a component of a risk management system for detecting and dealing with an anticipated natural or man-made hazard.

• Early warning systems are not restricted to natural hazards and disasters. They are applicable to any activity or situation that may create a problem that must be dealt with.

14

Risk reduction: Elements of an EWS

• Knowledge of and means of forecasting the danger faced• Information from technical monitoring and visual

observations• A response plan• Dissemination of meaningful warnings to population at risk• Public awareness and preparedness to respond to the

warning.

An early warning system will normally have a minimum of 5 components:

Risk reduction: Available Technology for EWSs

• Sensors and sensing technology• Communication technology• Data collection systems as well as data

processing, reporting and analyzing software• Forecasting methods and modelling tools

15

Risk reduction: Principal activities in an EWS

Risk reduction: The key to a successful EWS

• The key to a successful EWS is to be able to identify and measure the relevant precursors to the event.

• For example, typical precursors for an impending landslide event are:– Intense rainfall– Earthquakes and ground vibrations– High rate of slope movement– Rapid increases in pore water pressure– Erosion at the toe of the slope

16

Risk reduction: Early Warning Systems

Risk reduction: Early Warning Systems

Ground water level From Blikra (2008)

Laser

From Blikra (2008)

Extensometers

17

Risk reduction: Early Warning Systems

Risk reduction: Other measures than EWS, evacuation, etc.?

Protection barriers may be applicable for ”small large” rock slides.

Drainage may stabilize even very large unstablerock slopes

18

The Åknes rock slope

Artist’s depiction of tsunami at Geiranger

19

Flooding in Hellesylt with run-up 25 – 35m

2D experimental setup

• Surface elevation measured at gauges 1-3• Velocity field measured

PIV PIV

20

Ongoing 3D laboratory experiments• Coast and Harbour

Research Laboratory at SINTEF, Trondheim

• Instrumentation based on numerical simulations and 2D experiments

Results from initial numerical modelling – assumed worst case scenario

Run-up

Surface elevation

21

Overview of the Åknes rock slope

~20m

~1m

8-15cm/year

2-3cm/year

~600m

~180masl

Dip≈35deg~650,000m2

Assumed max. slide area

N

Monitoring systems• At the slope surface:

• Permanent GPS network with 8 antennas

• Total station with 30 prisms

• Ground-based radar with 8 reflectors (radar located accross the fjord)

• Five surface rod extensometers

• Surface crackmeters

• Surface tiltmeters

• Two single lasers measuring distances across the upper tension crack

• 8 geophones: micro-seismic network

22

Monitoring systems

• Climate station:

• Temperature• Precipitation• Two snow-depth sensors• Wind speed• Ground temperature

Monitoring systems

• In boreholes:

• Two 50 m long DMS systems with 50 inclinometers• One 100m long DMS (not installed yet)• Pietzometers, conductivity and temperature sensors in 3

boreholes

23

Monitoring: overview

Early warning centre: now in operation 24hrs a day

Alarm tresholdscriteria based on:• Total displacements

• Velocity inn defined time periods

• Acceleration

• Treshold values need to be defined and updated

Sirens in all the villages located in the tsunami hazardzone

Phone messages

Evacuation procedures and routes

The police responsible for the evacuation

24

Displacements across the upper tension fracture: 1993-2007

Displacements across the upper tension fracture: 2004-2007

25

Displacements per year - horizontal componentshown on a possible block model

2004 -2006: GPS, Tot. stat., Extensometers

1961-1983: Photogrammetry

1983-2004: Photogrammetry

More displacements in the NW part from 1961 to 1983 than later

LISA Radar

26

LISA Radar results

Displacements and block boundaries – Displacements across the upper tension fracture (4)

Ground water level From Blikra (2008)

Laser

From Blikra (2008)

Extensometers

27

Displacements in the upper borehole

1st interval ofmeasurements

2nd interval ofmeasurements

Geological and geotechnicalinvestigations

Geophysicalsurveys: resisitivity, georadar and seismic

2D Resistivity

28

2D resistivity: Interpretation of depth of unstable rock mass

Boreholes

• Core logging• Samples for lab testing• Optic televiewer and borehole

logging• Instrumentation

29

Geological model – Core logging (1)

Crushed and core loss

Results of field mapping: fracture orientationFol. frac. downslope the upper tensionfracture: mean dip 32deg.

Fractures non-parallel with the foliation

30

Geological model

Block model: DDA analysis on possible blockboundaries based on all three displacement data sets

Profile for stabilityanalyses

The area of Block 11 is 201,000m2 and a major part moved insignificantly from 2004 to 2007

Borehole location

Big question: does Block10 move????

31

Stability analyses: static

UDEC model of the whole slope

Stability analyses: static•One major conclusion from the numericalmodelling: Instability at great depth agrees withthe back-calculated limiting friction angle ofthe unstable area

•Instability at 120m later indicated by borholemeasurements

32

Stability analyses: dynamicUDEC model

Earthquakes with returnperiods of 100 and 1000 years

The analyses indicate that an earthquake with a return period of 1000 years is likely to trigger a slide to great depth at the present ground water conditions and that the slope will remain stable if it is drained.

An earthquake with a return period of 100 years is not likely to trigger a slide at the present ground water conditions.

Drainage

33

References??????