Temporary stability of slopes

cut in London Clay

Dr Nesha Kovacevic

Geotechnical Consulting Group, London

Deep excavations at Terminal 5

• Bottom-up construction
• Open cut excavations for economy
• Maximum depth 22m
• Cut slopes to be as steep as possible
• Stand-up time up to 1/2 years

1
 End shunt

CTB

Launch
chamber

Launch chamber

2
 Temporary stability of deep cuts in
London Clay

• To predict stand-up time of deep cuts

using numerical analyses and so optimise
slope geometry – maximum steepness and
minimum excavation and backfill?

• Only after calibration against case

histories of temporary slope failures
(Bradwell, Prospect Park, Wraysbury)

Constitutive model for

London Clay

3
ϕ ', c'
ϕ'p , c'p G = G (p’, εd)
K’ = K’ (p’, εv)

ϕ'r , c'r

(εdp)p (εdp)r ε dp

Soil model used

3,000
Secant (3.G)/p' or K'/p'

modelled shear
2,500
recommended shear
modelled bulk
2,000

1,500

1,000

500
0
0.0001 0.001 0.01 0.1 1
Axial strain or volumetric strain (%)

Small strain stiffness curves in extension

4
 Displacement across a 0.5m thick layer, ∆ (m)
0 0.025 0.05 0.075 0.1
150
Shear stress, τ (kPa) σ'=166.7kPa
∆ τ
100
T γ=∆/T

50

0
0 5 10 15 20
Shear strain, γ (%)

Predicted behaviour in drained simple shear

Axial strain (%)

0 2 4 6 8 10
Shear stress (kPa)

-50

-100
measured
-150
predicted
PWP change

-50

50

Behaviour in undrained triaxial extension

5
 5

Volumetric strain (%) 4 Test (depth: 13.55m)

Predicted
3

0
0 50 100 150 200 250
Vertical effective stress (kPa)

Swelling in oedometer test

Horizontal permeability (m/s)

1E-012 1E-011 1E-010 1E-009 1E-008
0

surf ace
Near
Biii

Bi

10 Biii Biii
A3
A2 A3 A2
Biii
A3
A2 A2
Depth below ground level (m)

A3 Bii
Non-linear 20
A3
A3 A2 A2
Bii

permeability
e st

A2
A3 A2
dw

A2
A2 A2
East of basin
model:
l an

A2
tra

A3

k=k0.e-b.p’
c en

A3 A2 A2
30
A3

A2 A2
A2 central
A3 and west
A2

A2
A2
40
A2
A3

50
A2

6
 Short - term slope failure at
Bradwell

Su (kPa)
0 20 40 60 80 100
2.7m 3.5m

1
1: Clay Fill
Original ground level GWL
1 0.9m
1: Marsh Clay
11.3m

Weathered
7.2m

London Clay
:
0.5

Unweathered
London Clay

Excavated slope at Bradwell

7
 Tension zone

Clay Fill

Marsh Clay
11.3m

London Clay
εpD=1%
Rupture
surface

peak

Strength
Softening starts
residual
at εpD=5%
MoS=1.02 5% 20% εdp

Predicted rupture surface at Bradwell by FE analysis

1
1: Clay Fill

1
1: Marsh Clay

London Clay
11.3m

Back-analysis by LEM
1

(Skempton, 1965)
:
0.5

Predicted by FEM

Predicted rupture surfaces by LE and FE method of analysis

8
 Short-term (undrained) failures
at Bradwell
• Can be analysed in terms of effective stresses using
the same constitutive model and soil parameters
as for the delayed failures of cuttings in London Clay

• Role of progressive failure is small, and concentrated

in the area around the toe of the slope

• Soil stiffness is of importance in estimating

the short-term stability

M25 N
Slough
Prospect Park
M4
otorway
M4 M

Datchet
Windsor Reservoir
Heathrow
Wraysbury Airport
Reservoir
King George VI
Reservoir Staines
Reservoir
Scale M25
0 1 2 3 4km Staines

Site location

9
 Prospect Park failure 9 weeks
after excavation

Tectonic shear surface in London Clay

at Prospect Park

10
 Slip surface Cut-off wall

2 Terrace Gravels
11.6m

1
London Clay

Tectonic shear zone

London Clay

Typical cross section at Prospect Park

Scale
0 10m

0.5m

With shear No shear

Movement vectors during excavation

11
 Scale
0 10m

-50 0 -50
50 0
50
100 kPa
100 kPa
150 150

With shear No shear

Pore pressures at the end of excavation

Depth of excavation below top of

Horizontal movements (m)

London Clay (m)

0 2 4 6 8
0

0.1

0.2 shear
zone
with shear
no shear
with shear, no prior gravel removal
no shear, no prior gravel removal

Horizontal movements at the top of London Clay

during excavation

12
 Scale
0 10m
Rupture
Rupture
surface
surface
2%
2%

15% εDp=2% 15% εDp=2%

With shear No shear

Plastic shear strains just prior to collapse

Horizontal movements (m)

Time since excavation (yrs)

0 0.2 0.4 0.6 0.8 1
0

0.1

0.2

with shear
no shear
with shear, no prior gravel removal
no shear, no prior gravel removal

Horizontal movements at top of London Clay

after excavation – Zero suction

13
 Time since excavation (yrs)

Horizontal movements (m)

0 0.2 0.4 0.6 0.8 1
0
0.1
0.2 25
0 50
0.3
0.4 Surface suction in kPa
0.5

Horizontal movements at top of London Clay

after excavation – With shear zone

Key factors determining time to failure

• Presence of tectonic shear

• Previous site history
• Surface suction; 25kPa gave the best match
to observed time to failure
• Permeability profile and non-linearity
• Ko – if tectonic shear zone is absent
• Progressive failure – rate of drop from peak
to residual strength

14
Surface suction 25kPa

Predicted failure mode for

5m bermed 1:1 slopes at T5

Surface suction ‘zero’

Predicted failure mode

for 5m bermed 1:1 slope at T5

15
 Launch
chamber

Launch
chamber

16
 Lessons for the future

• Monitor movements to identify presence

of tectonic shears or development of basal
shear
• Take measures to maintain suctions at
slope surface
• Monitor suctions
• Differentiate between drying beds and
lagoons
• Check sensitivity to Ko, k and surface
suction

17