205

GEOTECHNICAL ASPECTS OF THE 22 FEBRUARY 2011

CHRISTCHURCH EARTHQUAKE

Misko Cubrinovski1, Brendon Bradley1, Liam Wotherspoon2,

Russell Green3, Jonathan Bray4, Clint Wood5, Michael Pender2,
John Allen6, Aaron Bradshaw7, Glenn Rix8, Merrick Taylor1,
Kelly Robinson1, Duncan Henderson1, Simona Giorgini1,
Kun Ma1, Anna Winkley1, Josh Zupan4, Thomas O’Rourke9,
Greg DePascale10, Donnald Wells11
SUMMARY
The 22 February 2011, Mw6.2-6.3 Christchurch earthquake is the most costly earthquake to affect New
Zealand, causing 181 fatalities and severely damaging thousands of residential and commercial
buildings, and most of the city lifelines and infrastructure. This manuscript presents an overview of
observed geotechnical aspects of this earthquake as well as some of the completed and on-going research
investigations. A unique aspect, which is particularly emphasized, is the severity and spatial extent of
liquefaction occurring in native soils. Overall, both the spatial extent and severity of liquefaction in the
city was greater than in the preceding 4th September 2010 Darfield earthquake, including numerous areas
that liquefied in both events. Liquefaction and lateral spreading, variable over both large and short
spatial scales, affected commercial structures in the Central Business District (CBD) in a variety of ways
including: total and differential settlements and tilting; punching settlements of structures with shallow
foundations; differential movements of components of complex structures; and interaction of adjacent
structures via common foundation soils. Liquefaction was most severe in residential areas located to the
east of the CBD as a result of stronger ground shaking due to the proximity to the causative fault, a high
water table approximately 1m from the surface, and soils with composition and states of high
susceptibility and potential for liquefaction. Total and differential settlements, and lateral movements,
due to liquefaction and lateral spreading is estimated to have severely compromised 15,000 residential
structures, the majority of which otherwise sustained only minor to moderate damage directly due to
inertial loading from ground shaking. Liquefaction also had a profound effect on lifelines and other
infrastructure, particularly bridge structures, and underground services. Minor damage was also
observed at flood stop banks to the north of the city, which were more severely impacted in the 4 th
September 2010 Darfield earthquake. Due to the large high-frequency ground motion in the Port hills
numerous rock falls and landslides also occurred, resulting in several fatalities and rendering some
residential areas uninhabitable.

1
Department of Civil and Natural Resources Engineering, University of Canterbury, Christchurch, New Zealand
2
Department of Civil and Environmental Engineering, University of Auckland, Auckland, New Zealand
3
Department of Civil and Environmental Engineering, Virginia Tech, Blacksburg, VA, USA
4
Department of Civil and Environmental Engineering, University of California Berkeley, Berkeley, CA, USA
5
Department of Civil and Environmental Engineering, University of Arkansas, Fayetteville, AR, USA
6
TRI Environmental, Inc., Austin, TX, USA
7
Department of Civil and Environmental Engineering, University of Rhode Island, Kingston, RI, USA
8
School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, USA
9
School of Civil and Environmental. Engineering., Cornell University., Ithaca, NY, USA
10
Fugro/William Lettis and Associates, USA
11
AMEC Geomatrix, Oakland, CA, USA

BULLETIN OF THE NEW ZEALAND SOCIETY FOR EARTHQUAKE ENGINEERING, Vol. 44, No. 4, December 2011
206

INTRODUCTION south-east dipping blind fault, which trends north-east to

south-west, with a reverse-oblique slip orientation [4, 5] and is
On 22 February 2011 at 12:51pm local time, a moment located to the south-east of the city centre (Figure 1).
magnitude 6.2-6.3 earthquake occurred beneath the city of
Christchurch, New Zealand, causing an unparalleled level of
GEOLOGY OF THE CHRISTCHURCH AREA
damage in the country’s history, and the largest number of
causalities since the 1931 Napier earthquake. Compared to Christchurch is located on the Canterbury Plains, a fan deposit
the preceding 4th September 2010 Darfield earthquake resulting from the numerous rivers flowing eastward from the
[1], which occurred approximately 30 km to the west of foothills of the Southern Alps [6]. In the vicinity of
Christchurch, the close proximity of the 22 February event led Christchurch, the Canterbury Plains are comprised of a
to ground motions of significantly higher amplitude in the complex sequence of gravels inter-bedded with silt, clay, peat,
densely populated regions of Christchurch. and shelly sands. The fine sediments form aquicludes and
aquitards between the gravel aquifers, and with the nearby
A defining feature of the 22 February 2011 earthquake, as well
coastline to the east, result in the majority of Christchurch
as other events which have produced strong ground shaking in
having a water table less than 3 m depth, with most of the area
Christchurch city, was the large severity and spatial extent of
to the east of the central business district having a water table
liquefaction that occurred in native soils. The severity of
of about 1 m from the surface [6]. The postglacial
strong motion also resulted in significant rock-falls in the Port
‘Christchurch formation’ created by estuarine, lagoonal, dune,
Hills, substantial damage to commercial and residential
and coastal swamp deposits (containing gravel, sand, silt, clay,
structures; and damage to infrastructure networks in the
shell and peat) is the predominant surface geology layer in the
eastern suburbs and central region of the city.
eastern Christchurch area which outcrops up to 11 km west of
This manuscript provides an overview of observations made the coast and has a thickness of approximately 40 m at the
during post-event reconnaissance as well as some of the present coastline [6]. At the southeast edge of Christchurch
associated research activities related to geotechnical aspects of lies the extinct Banks Peninsula volcanic complex.
this event. Firstly, the tectonic and geologic setting of
Christchurch is briefly discussed followed by presentation of OBSERVED GROUND MOTIONS
the salient features of the densely recorded ground motions
from the event. Observed liquefaction features are then Here a summary of the observed ground motions is given to
presented in an overarching context and subsequently the provide context for the observed response of geotechnical
impact of ground failure on the commercial structures, structures in the latter sections. Further details can be found in,
residential properties, and infrastructure is discussed. Finally, for example, Bradley and Cubrinovski (this issue), among
an overview of slope instability of the Port Hills is given. others.
Table 1 summarizes the intensity of ground motions in the
REGIONAL TECTONICS greater Christchurch region that were recorded within a
source-to-site distance of = 20 km of the causative fault,
New Zealand resides on the boundary of the Pacific and
Australian plates and its active tectonics are dominated by [2]: in terms of the geometric mean horizontal peak ground
(i) oblique subduction along the Hikurangi trough in the North acceleration ( ) and vertical peak ground acceleration,
Island; (ii) oblique subduction along the Puysegur trench in . Figure 1 illustrates the spatial distribution of fault-
the south west of the South Island; and (iii) oblique, right normal acceleration time histories recorded at the
lateral slip within the axial tectonic belt. There are numerous aforementioned strong motion stations. The inferred surface
identified faults in the Southern Alps and eastern foothills [3] projection of the causative fault [4] is also shown. It can be
and several significant earthquakes (i.e. ) have seen that the close proximity of the fault to the city led to large
occurred in this region in the past 150 years, most notably the ground motions in the horizontal and vertical directions, with
Darfield earthquake on 04/09/2010 [1]. The 6.2-6.3 a horizontal PGA of 1.41g recorded at Heathcote Valley, and
Christchurch earthquake occurred at 12:51pm on Tuesday seven ground motion records have horizontal PGA values
22/02/2011 beneath Christchurch and represents the most greater than 0.4g. In the central business district (CBD), PGA
significant earthquake in the unfolding seismic sequence in the values range from 0.37-0.52g, approximately 1.6 times higher
Canterbury region since the 04/09/2010 Darfield earthquake. seismic demand than that of the 4th September 2010
The 6.2-6.3 event occurred on a previously unrecognised earthquake in terms of liquefaction triggering [7].

Table 1: Summary of observed ground motions at strong motion stations.

Station Name Code (km) PGA (g) (g) Station Name Code (km) PGA(g) (g)

Canterbury Aeroclub CACS 12.8 0.21 0.19 Lyttelton Port Naval Point LPOC 6.6 0.34 0.39
Christchurch Botanic North New Brighton
CBGS 4.7 0.50 0.35 NNBS 3.8 0.67 0.80
Gardens School
Christchurch Cathedral
CCCC 2.8 0.43 0.79 Papanui High School PPHS 8.6 0.21 0.21
College
Christchurch Hospital CHHC 3.8 0.37 0.62 Pages Rd Pumping Station PRPC 2.5 0.63 1.88
Cashmere High School CMHS 1.4 0.37 0.85 Christchurch Resthaven REHS 4.7 0.52 0.51
Hulverstone Dr Pumping
HPSC 3.9 0.22 1.03 Riccarton High School RHSC 6.5 0.28 0.19
Station
Heathcote Valley School HVSC 4.0 1.41 2.21 Rolleston School ROLC 19.6 0.18 0.08
Kaipoi North School KPOC 17.4 0.20 0.06 Shirley Library SHLC 5.1 0.33 0.49
Lincon School LINC 13.6 0.12 0.09 Styx Mill Transfer Station SMTC 10.8 0.16 0.17
Lyttelton Port LPCC 7.1 0.92 0.51 Templeton School TPLC 12.5 0.11 0.16
 207

Figure 1: Observed fault-normal horizontal acceleration time histories at various locations in the Christchurch region from the
22 February earthquake with reference to the inferred surface projection of the causative fault which dips to the
south-east (Bradley and Cubrinovski, this issue).

Importance of nonlinear soil response Liquefaction observed in recorded ground motions

An illustration of the significant effects of non-linear soil As elaborated upon subsequently, one of the major causes of
response under strong ground motion can be seen by damage in the 6.2-6.3 Christchurch earthquake resulted
comparing the ground motions observed on rock and soil sites from the widespread and very severe liquefaction in
at Lyttelton Port (LPCC and LPOC, respectively). In addition residential, commercial and industrial areas. The horizontal
to a comparison of the acceleration time histories in Figure 1, components of acceleration depicted in Figure 1 show
Figure 2 illustrates the pseudo-acceleration response spectra of evidence of liquefaction in the central business district and
the geometric mean horizontal and vertical ground motion eastern suburbs which are located in the near-source region
components at the two sites. It can be seen that, compared to beyond the up-dip projection of the fault plane. For example,
LPCC, the observed ground motion at the LPOC soil site has in the central business district (e.g. CBGS), Cashmere
significantly lower amplitude of high frequency content, (CMHS) and Shirley (SHLC), evidence of liquefaction is
longer predominant period, and larger significant duration in inferred from the manifested reduction in high frequency
the horizontal direction. In contrast, it can be seen that there is content of ground motion following several seconds of S wave
relatively little difference between the vertical ground motions arrivals, and the subsequent acceleration ‘spikes’,
at the two sites, because of the relatively large compressive characteristic of strain hardening deformation during cyclic
stiffness of the sites, with peak vertical accelerations of 0.51 mobility.
and 0.39g, respectively.
Sedimentary basin generated surface waves and near-
Geometric Mean source directivity
Horizontal
0 Christchurch is located on a sedimentary fan deposit with the
10
Spectral acc, Sa (g)

volcanic rock of Banks Peninsula located to the south east.

Because of the location of the causative fault to the south of
the city, and the increasing depth of the volcanic rock-basin
interface moving in the north-west direction, it is likely that
-1 seismic waves emanating from the rupture entered the
10 Vertical
sedimentary basin through its thickening edge, leading to a
waveguide effect in which surface waves propagate across the
LPCC (Rock) basin resulting in enhanced long period ground motion
amplitudes and shaking duration. Because of the near-source
-2 LPOC (Soil)
10 -2 location of the city to the causative fault, directivity effects
-1 0 1
10 1010 10 associated with the rupture propagation were important
Period, T (s) features in the ground motions observed at specific locations.
Figure 2: Comparison of geometric mean horizontal and Directivity effects appear to be most significant in the eastern
vertical response spectra observed at two nearby suburbs of the city, due to its proximity to the rupture asperity.
strong motion stations in Lyttelton, one on For the central business district, near-source directivity effects
outcropping rock (LPCC), the other on soil are inferred as of secondary importance to the basin-generated
(LPOC) [8]. surface waves at long vibration periods.
208

(a) Geometric Mean (b)

Horizontal
70
0
10

Spectral disp, Sd (cm)

60
Spectral acc, Sa (g)

50

40

-1
CCCC 30
10 CHHC
Vertical
CBGS 20
REHS
NZS1170.5 10
-2 -1 0 1
10 10 10 10 0
4 6 8 010 2
Period, T (s) Period, T (s)
Figure 3: Comparison of response spectra from four strong motion stations located in the Christchurch central business district:
(a) horizontal and vertical pseudo-acceleration response spectra; and (b) horizontal displacement response spectra [8].
surface manifestation of liquefaction was evident. As
Seismic intensity in the CBD elaborated upon later, the suburbs to the east of the CBD along
The Christchurch earthquake caused significant damage to the Avon River (Avonside, Dallington, Avondale, Burwood
commercial structures in the CBD. Figure 3 illustrates the and Bexley) were most severely affected by liquefaction.
pseudo-acceleration and displacement response spectra of four
strong motion stations (CCCC, CHHC, CBGS, REHS) located Repeated liquefaction
in the CBD. Despite their geographic separation distances
(relative to their respective source-to-site distances) it can be Soil liquefaction repeatedly occurred at some sites during the
seen that the characteristics of the ground motion observed at earthquakes producing strong ground shaking in Christchurch,
these locations is relatively similar. This is particularly the and in particular during the 4 September 2010, 22 February
case for long-period ground motion amplitudes, which have 2011, and the 5.5 and 13 June 2011 earthquakes.
longer wavelengths, while there is more of a discrepancy in Figure 5 comparatively shows liquefied areas of Christchurch
seismic intensity at short periods due to lower wave coherency in these three events, as documented by field investigations.
and the importance of near-surface soil layers (which are The repeated liquefaction was often quite severe and many
highly variable as discussed subsequently). Figure 3, in residents reported that in some cases the severity increased in
particular, illustrates that the seismic demands were above the subsequent events.
475 year return period design ground motion for Christchurch
site class D as specified by the New Zealand loading standard, LIQUEFACTION AND ITS EFFECTS IN THE CBD
NZS1170.5 [9]. Figure 3b also illustrates that for structures Salient observations of the effects of soil liquefaction on
whose secant period at peak displacement is in the region of structures in the CBD of Christchurch are presented here
1.5 or 3.5 seconds, the displacement demands imposed by the including several important cases of buildings with varying
ground motion were in the order of two times the seismic foundation types that performed differently in liquefied
design level. ground, while further details can be found in Cubrinovski et
al. [7].
OVERVIEW OF OBSERVED LIQUEFACTION
CBD soil characteristics
Spatial extent of liquefaction in the 22 February 2011
earthquake The shallow alluvial soils vary substantially within short
distances, both horizontally and vertically within the CBD (as
A distinctive feature of the 22 February 2011 earthquake, as well as the greater Christchurch region in general). This
well as other recent events which have produced strong variation is depicted in Figure 6, where a simplified
ground shaking in Christchurch city, was the severity and stratification up to 30 m depth is shown for a cross section
spatial extent of liquefaction observed in native soils. Figure 4 through the CBD soils along Hereford Street [11]. To further
shows the extent of liquefaction caused by the 22 February illustrate the spatial variability of foundation soils, Figure 7
2011 earthquake in the greater Christchurch region based on a delineates several zones indicating the predominant soils in
drive-through reconnaissance [10]. Four areas of different the top 7 to 8 m of the CBD deposits [12]. In the south-west
liquefaction severity are indicated in the map: (a) moderate to part of the CBD, alluvial gravels are encountered at shallow
severe liquefaction (red zone, with very large areas covered by depths of 2.5 m to 3.5 m, while loose silts and peat comprise
sand ejecta, large cracks and fissures in the ground, and the top soils in the south-east part of the CBD. Relatively
significant liquefaction-induced impacts on buildings), (b) low clean and deep sands dominate the stretch along Avon River;
to moderate liquefaction (yellow zone, with generally similar this was the area most severely affected by liquefaction in the
features as for the severe liquefaction, but of lesser intensity 22 February earthquake. Further to the north of this zone
and extent), (c) liquefaction predominantly on roads with towards Bealey Avenue, loose silty soils and peat are
some on properties (magenta zone), and (d) traces of encountered in the top 7 to 8 m of the deposit.
liquefaction (red circular symbols, with clear signs of
liquefaction, but limited in extent and deemed not too
damaging for structures). Blue lines indicate areas where no
 209

Figure 4: Preliminary liquefaction map of Christchurch from drive-through reconnaissance [10]

Figure 5: Preliminary liquefaction maps documenting areas of observed liquefaction in the 4 September 2010 (white contours),
22 February 2011 (red, yellow, magenta areas), and 13 June 2011 (black contours) earthquakes [12].
damage in the CBD was limited in the 4th September 2010
Spatial distribution of liquefaction in the CBD earthquake.
Figure 7 shows the resulting liquefaction documentation map Even though the map shown in Figure 7 distinguishes the zone
for the CBD. The principal zone of liquefaction stretches west most significantly affected by liquefaction, the severity of
to east through the CBD, from Hagley Park in the west, along liquefaction within this zone was not uniform. The
the Avon River to the northeast boundary of the CBD at the manifestation of liquefaction was primarily of moderate
Fitzgerald Avenue Bridge. This zone is of particular interest intensity with relatively extensive areas and volumes of
because many high-rise buildings on shallow foundations and sediment ejecta (Figure 8). There were also areas of low
deep foundations were affected by the liquefaction in different manifestation or only traces of liquefaction, as well as pockets
ways. Note that this zone consists mostly of sandy soils and of severe liquefaction with very pronounced ground distortion,
largely coincides with the path of the Avon River and network fissures, large settlements and substantial lateral ground
of old streams shown in 1850’s survey maps [12]. The movements. This non-uniformity in liquefaction manifestation
performance of the surficial soils in the 22 February 2011 reflects the complex and highly variable soil conditions even
earthquake is also significant because liquefaction-induced within the CBD principal liquefaction zone.
210

fissures and a distorted pavement surface marked this feature,

which runs continuously through properties and affected a
number of buildings causing cracks in both the foundations
and superstructures. Liquefaction and associated ground
deformation were pronounced and extensive on the down
slope side between the identified geomorphic feature and the
Avon River, but noticeably absent on the slightly higher
elevation to the north (upslope side away from the river). This
feature is thought to delineate the extent of a geologically
recent river meander loop characterized by deposition of loose
sand deposits under low velocity conditions. A similar
geomorphic feature was observed delineating the boundary
between liquefaction damage and unaffected ground within a
current meander loop of the river to the east of this area
(Oxford Terrace between Barbadoes Street and Fitzgerald
Avenue).
Figure 6: Representative subsurface cross section of
Christchurch CBD along Hereford Street [11]. Ground Failure Effects on Nearly Identical Structures –
East Salisbury Area
The northern extent of the zone, which is shown by the thick
solid line in Figure 7, is a clearly defined geomorphic feature A mini-complex of three nearly identical buildings (with one
running east-west that was delineated by a slight change in small but important difference) is shown in Figure 9. The
ground elevation of about 1 m to 1.5 m over approximately 2 buildings are three-storey structures with a garage at the
m to 10 m wide zone. After the 22 February event, it was ground floor, constructed on shallow foundations. This case
further characterized by ground fissuring or distortion clearly illustrates the impact of liquefaction, with nearly
associated with localized spreading, as well as gentle slumping identical structures built across the east-west trending
of the ground surface on the down slope side. Ground cracks, geomorphic feature identified previously in Figure 7, one

Figure 7: Preliminary CBD liquefaction map for the 22 February earthquake [10]; predominant soils in the top part of the
deposits are also indicated [12].

(a) (b)

Figure 8. Representative areas of: (a) moderate liquefaction; and (b) severe liquefaction within the CBD principal liquefaction
zone [7].
 211

(a) (b)

Figure 9: Apartment complex: (a) looking south from northern building showing tilt of southern building, and (b) looking north
at liquefaction feature at edge of southern building [7].

Apartment buildings
shown in Figure 9

Duplex homes; centre

structure is shown in
Tilted structure
Figure 11
shown in Figure 9b

Figure 10: Location of geomorphic feature in area of apartment and duplex complexes north of Salisbury Street in CBD [7].
(a) (b)

Figure 11: Duplex housing complex: (a) looking north at centre building, and (b) close-up of ground settlement next to centre
building [7].
212

building located on the higher level to the north suffering no

damage, and the buildings located below the crest suffering
progressively higher amounts of damage. This geomorphic
feature, which is expressed here by a significant change in
grade of the pavement between the northern and middle
buildings, is shown in Figure 10. The northern building that
sits on the higher ground showed no evidence of cracking and
distortion of the pavement surface. Conversely, large sediment
ejecta were found along the perimeter of the southern building
indicating severe liquefaction in its foundation soils (Figure
9b). Liquefaction features were also observed near the middle
building, but the resulting distress of this building was
significantly less than that of the southern building. The
southern building had a shortened end wall with a column at
its southwest corner, which appeared to produce additional
settlement at the location of the column’s concentrated load. It
suffered differential settlement of about 40 cm and over 3
degrees of tilt towards the west-southwest, which is visible in Figure 12: Two-story building that underwent liquefaction-
Figure 9a. This building was uneconomical to repair and was induced punching movements [7].
demolished after the 22 February 2011 earthquake. Adjacent
to these buildings is another complex of three identical but Differential Settlement and Sliding - Armagh-Madras
structurally different buildings from the former set. Their Area
locations relative to the abovementioned geomorphic feature is
Further to the south at the intersection of Madras and Armagh
identical, but these buildings are two-storey duplexes. Figure
Streets, several buildings were affected by severe liquefaction
11a shows the middle building with clear evidence of
that induced significant differential settlements or lateral
pavement distortion, cracking and settlement of the
movements. At this location, the liquefaction was manifested
surrounding ground. The settlement of the building was likely
by a well-defined, narrow zone of surface cracks, fissures, and
not significant, but the ground settled about 20 cm exposing
depression of the ground surface about 50 m wide, as well as
the top of the foundation at the southwest corner (Figure 11b).
water and sand ejecta (Figure 13a, and the wide black zone to
Another apartment complex that was constructed on a single the south of the Avon River in Figure 7).
level basement that extends almost the full length of the
This zone stretches from the Avon River to the north towards
complex and provides off-street parking for the development
the buildings, which were affected by this liquefaction feature.
lies to the west of the two case histories discussed previously.
Traces of liquefaction were evident further to the south of
It also crosses the geomorphic feature. Noticeable settlement
these buildings. Figure 13b and Figure 13c illustrate two
of the ground at the southern end of the complex of the order
buildings, founded on isolated shallow foundations, that were
of 15-20 cm occurred and compression features in the
located on the edges of the well-defined liquefaction zone in
pavement suggest that it displaced laterally towards the street.
Figure 13a. It can be seen that for both structures lateral
The concrete basement floor and structure appeared to have
displacements, differential settlements, and consequent tilting
undergone negligible distortion, which indicates an overall
were observed. Both buildings were considered uneconomical
rigid response despite the differential ground movements
to repair and will be demolished.
across the site.
Performance of Adjacent Structures - Town Hall Area
Punching Settlement - Madras-Salisbury-Peterborough
Area The Christchurch Town Hall for Performing Arts is located
within the northwest quadrant of the CBD, with the
Several buildings with shallow foundations located within the
meandering Avon River to its immediate south. It is a complex
liquefied zone underwent punching settlements with some
facility comprising a main auditorium (seating 2,500) with
undergoing significant differential settlements. An example of
adjoining entrance lobby, ticketing, and café areas. Further
such performance is shown in Figure 12 for a two-storey
extensions provide a second, smaller auditorium James Hay
industrial building located 200 m south-west of the buildings
Theatre (seating 1,000) and a variety of function rooms and a
discussed previously. There was clear evidence of the mud-
restaurant. The structures are supported on shallow
water ejecta on the walls of the building indicating about 25
foundations, except the kitchen facility that was added later.
cm thick layer of water and ejected soils due to the severe
Air bridges connect the complex to the Crowne Plaza, a major
liquefaction. Note the continuous sand ejecta around the
hotel, and to the Christchurch Convention Centre (opened
perimeter of the footing and signs of punching shear failure
1997) to the north. Tiled paved steps lead from the southern
mechanism in Figure 12. At the front entrance of the building
side of the complex down to the river’s edge.
large ground distortion and sinkholes were created due to
excessive pore water pressure and upward flow of water. The facility suffered extensive damage primarily caused by
Settlement of the building around its perimeter was evident liquefaction-induced ground failure. Differential settlements
and appeared substantially larger than that of the surrounding caused by punching shear beneath the building’s main internal
soil that was unaffected by the building. The building settled columns that surround the auditorium and carry the largest
approximately 25 cm relative to a fence at its south-east corner dead loads to shallow foundations and a second ring of
and settled 10-20 cm relative to the ground at its north-west exterior columns (Fig. 14a) that are connected to the inner ring
corner. The ground floor at the entrance was uplifted and via beams (Fig. 14b) caused distortion to the structure. The
blistered which is consistent with the pronounced settlement cracked beam shown in Fig. 14b underwent an angular
beneath the walls or along the perimeter of the building. distortion of 1/70 across its span. The seating for the
auditorium has been tilted; dragged backward by the
settlement of the columns, leaving a large bulge or doming to
the floor of the auditorium itself. The air bridge connecting the
main auditorium to the Christchurch Convention Centre to the
north (away from river) has separated from the building.
 213

(a)
Structure shown in
Figure 13c

Structure shown in
Figure 13b

(b) (c)

1.8 deg

6 5 4
9
15 cm 18
29 cm

Figure 13: Relatively narrow liquefaction-induced feature and induced differential settlement and sliding of buildings [7].
With no significant deformations of the ground as the obvious
source of this lengthening between the two buildings, the Contrasting Performance of a Pile-Supported Structure -
explanation appears to be that distortions to the auditorium Kilmore Area
structure have pulled the outer walls in towards the building, Several pile-supported structures were identified in areas of
creating this separation. The entire complex appears to have severe liquefaction. Although significant ground failure
moved laterally towards the river (albeit by a barely occurred and the ground surrounding the structures settled, the
perceptible amount on the northern side) with parts of the buildings supported on piles typically suffered less damage.
complex closest to the river undergoing increasingly larger However, there are cases where pile-supported structures were
movements (Fig. 14c). These sections have settled and moved damaged in areas that underwent lateral spreading near the
laterally towards the river more than the remainder of the Avon River. In other cases, such as the building shown in
building leading to significant structural deformations where Figure 16, located approximately 200 m to the east from the
the extension and original structures are joined. Town Hall, the ground floor garage pavement was heavily
Contrary to the liquefaction-induced punching settlement of damaged in combination with surrounding ground
buildings into the surrounding ground that was observed at the deformation and disruption of buried utilities. The settlement
Town Hall and in other parts of the CBD, the seven-storey of the surrounding soils was substantial with about 30 cm of
building on shallow foundations shown in Figure 15a did not ground settlement on the north side of the building and up to
punch significantly into the liquefied ground nor undergo 17 cm on its south side. The first storey structural frame of the
significant differential settlement. As shown in Figure 15b building that was supported by the pile foundation with strong
there were significant amounts of sand ejecta observed in this tie-beams did not show significant damage from these
area. However, there was no obvious evidence of significant liquefaction-induced ground settlements. Following the 13
differential ground or building movement. The differential June 2011 earthquakes, the settlement of the surrounding soil
settlement measured between adjacent columns was typically at the north side of the building reached about 50 cm.
negligible, but differential settlements of up to 3.5 cm were Across from this building to the north, is a seven-storey
measured at a few locations. This building is across the street reinforced concrete building on shallow spread footing
and slightly to the west of the Town Hall. It is a case of foundations that suffered damage to the columns at the ground
liquefaction without significant differential settlement and level. This building tilted towards south-east as a result of
building damage. approximately 10 cm differential settlement caused by the
more severe and extensive liquefaction at the south, south-east
part of the site. Hence, these two buildings provide invaluable
information on the performance of shallow foundations and
pile foundations in an area of moderate to severe liquefaction
214

that induced uneven ground settlements. At this site, extensive (a)

field investigations were conducted including shear wave
velocity profiling and a dense array of CPTs and Gel-Push
sampling of undisturbed samples of sandy and silty soils from
2 m to 13 m depth.

(a)

(b)

(b)

Figure 15: Building in area of significant liquefaction that

displays negligible to minor differential
settlement or punching settlement [7].
(a)
(c)

17 cm
30 cm
Figure 14: Town Hall auditorium and adjacent dining (b)
facility undergoing significant liquefaction-
induced differential settlement and lateral
movements [7].

30 cm
Foundation beam

Figure 16: Building on pile foundations in area of severe

liquefaction showing large settlement of the surrounding
soils relative to the foundation beams [7].
 215

LIQUEFACTION AND ITS EFFECTS IN RESIDENTAL

Presence of Shallow Gravelly Soils - Victoria Square AREAS
Near Victoria Square, the liquefied zone was predominantly
composed of relatively loose sand deposits that transitioned Soil characteristics of residential areas
relatively sharply into a zone where gravelly soil layers reach
close to the ground surface. Shallow foundations (spread As previously noted, the near-surface geology of Christchurch
footings and rafts) for many of the high-rise buildings in this is dominated by fluvial processes and, as such, has highly
latter area are supported on these competent gravelly soils. variable soil properties. Despite this variability, gross features
However, the ground conditions are quite complex in the of the near-surface soil characteristics can be used to explain
transition zone, which resulted in permanent lateral the observed ground response, particularly in suburban areas,
movements, settlements, and tilt of buildings either on shallow as is the focus of this section.
foundations or hybrid foundation systems (with both shallow Figure 18 provides a schematic illustration of an east-west
and pile foundation elements), as illustrated in Figure 17. cross-section of the near surface geology of Christchurch
Immediately to the north of these buildings, the liquefaction taken along Bealey Avenue (location shown in Fig. 5).
was severe with massive sand ejecta; however, approximately Important features of this figure include a water table with
100 m further to the south where the gravels predominate, depth of approximately only 1 m below the surface in almost
there was neither evidence of liquefaction on the ground the entire eastern side of the city (with the exception of those
surface nor visible distress of the pavement surface. Again, it colluvium areas at the base of the Port Hills to the south); and
appears that the ground and foundation conditions have played an increasing depth of Riccarton gravel horizon, the upmost
a key role in the performance of these buildings, with these aquifer beneath the city. Although not shown in Figure 18 it is
buildings, accordingly, being selected for further in-depth also noteworthy that the Springston formation (alluvial
inspections and field investigations. gravels, sands and silts) is the dominant surface layer in the
west of Christchurch, and the Christchurch formation
(estuarine, lagoon, beach, dune, and coastal swamp deposits of
sand, silt, clay and peat) to the east of Christchurch. Hence, it
can be argued that the significant liquefaction observed in the
eastern suburbs of the city, and the absence in the west of the
city can be attributed to several contributing factors: (i) a
reduction in the amplitude of ground shaking moving from
east to west (i.e. Figure 1 and Table 1); (ii) a gradual change in
surficial soil characteristics; and (iii) an increase in water table
depth.
In both (4 September 2010 and 22 February 2011)
earthquakes, widespread liquefaction occurred in the urban
areas of Christchurch and Kaiapoi causing extensive damage
to residential properties. The liquefaction was manifested by
massive sand boils and large amount of sand/silt ejecta and
water littering streets, residential properties and recreation
grounds of Christchurch and Kaiapoi (town north of
Christchurch shown in Fig. 21). Nearly 15,000 residential
houses and properties were severely damaged due to
liquefaction and lateral spreading, more than half of those
beyond economical repair.
The distribution of liquefied areas shown in Figure 5 reflects
the combined effects of two important factors for liquefaction
Figure 17: Buildings on shallow and hybrid foundations in triggering: the soil resistance to liquefaction (a measure for the
transition area from moderate liquefaction to capacity of soils to sustain cyclic loading) and the severity of
low/no liquefaction; arrows indicate direction of ground motions (measure for the seismic load or demand)
tilt of the buildings [7]. produced by the two quakes. The suburbs most severely
affected by liquefaction in Christchurch were along the Avon
Lateral Spreading – Avon River within CBD River to the east and northeast of CBD (Avonside, Dallington,
Avondale, Burwood and Bexley). The soils in these suburbs
Liquefaction-induced lateral spreading was evident within the are predominantly loose fluvial deposits of liquefiable clean
CBD along the Avon River in the liquefied zone, and the fine sands and sands with non-plastic silts. The top 5-6 m are
horizontal stretching of the ground adversely affected several in a very loose state, with a CPT cone resistance, qc, of about
buildings. Detailed measurements by ground surveying 2-4 MPa. The resistance typically increases to 7-12 MPa at
conducted at about 10 transects on Avon River within the depths between 6 and 10 m, however lower resistances are
CBD after the 22 February earthquake indicated that at several often encountered in areas close to wetlands. The more
locations the maximum spreading displacements at the banks extensive liquefaction observed in these areas during the
of Avon River reached about 50-70 cm, whereas at most of the February 2011 earthquake is consistent with the fact that the
other locations the spreading was on the order of 10 cm to 20 seismic demand specific to liquefaction was about 1.5 to 2.0
cm. There were many smaller buildings suffering serious times higher during the February event as compared to the
damage to the foundations due to spreading as well as clear September 2010 earthquake. Similarly, at the southwest end of
signs of effects of spreading on some larger buildings both at the city in Hoon Hay and Halswell, more extensive
the foundations and through the superstructure. liquefaction occurred during the 2010 Darfield earthquake.
216

Figure 18: East-west cross-section of surface elevation, water table and Riccarton gravel horizon at Bealey Avenue in the North-
South direction [12].
(a) (b)

(c) (d)

Figure 19: Typical manifestation of liquefaction in residential areas.

(a) (b)

Figure 20: Illustration of a house in Kaiapoi which sustained liquefaction in both the (a) 4 th September 2010 Darfield; and (b) 22
February 2011 Christchurch earthquakes.
 217

down to 6.1-9.1 m below the surface. The experimental

Typical damage in residential areas surface waves dispersion curves obtained from the SASW
Total and differential settlements, lateral movements, and testing were used to determine a best-fitting 1D shear wave
flooding due to liquefaction and lateral spreading is estimated velocity ( ) profile. In total, 30 DCP and 36 SASW tests were
to have severely affected 15,000 residential performed across Christchurch and its environs after the
properties/buildings. Particularly, damage due to liquefaction Darfield and Christchurch earthquakes as depicted in Figure
related phenomena was widespread in the suburbs to the east 21.
of the CBD along the Avon River (Avonside, Dallington,
Avondale, Burwood and Bexley). In these areas only moderate
damage was directly due to inertial loading from ground
shaking. About 5,000 residential properties in such suburbs
will be abandoned due to the infeasibility of repair (New
Zealand Government, 2011). Examples of damage as a result
of liquefaction in the residential areas are presented in Figure
19, with the volume of ejected material in residential
properties indicated by the piles of sand in Figure 19a, a
typical view in many streets following the Christchurch
earthquake. Figure 19b provides a good indication of the
flooding and ejected material in the streets themselves
immediately following the earthquake. The large sand boils in
Figure 19c, about 20-30 m long and 10-15 m wide, indicate
both a large severity and extent of liquefaction throughout the
depth of the deposit. Figure 19d shows a typical differential
settlement and damage to the building due to separation of
walls as a result of loss of bearing capacity of the liquefied
foundation materials.
Figure 20 indicates damage to a residence in Kaiapoi after the
Darfield and Christchurch earthquakes. Following the Darfield
earthquake there was large settlement of the ground and
house, and approximately 40 cm of ejected material covering
the ground surface. Site investigations performed following
the Darfield earthquake indicated loose/soft soils up to depths
of 9 m. Ground motions were largest in Kaiapoi during the Figure 21: Locations of SASW (“+” symbol) and DCPT (“”
Darfield earthquake and lesser in the Christchurch earthquake symbol) tests performed post-event. Blue sites
(PGA’s of approximately 0.33g and 0.21g, respectively). are field tests post-Darfield earthquake, and
Despite this, Figure 20b illustrates that the volume of ejected red post-Christchurch earthquake.
material following the Christchurch earthquake was again
significant, and highlights the potential for repeated For use in liquefaction assessment, the obtained (converted)
liquefaction during multiple earthquakes of the typical soil SPT N-values and profile from the DCP and SASW tests
deposits in the region. A smaller volume of ejected material were normalized for effective confining stress and the cyclic
was again evident at this site following the 13 June 2011 resistance ratio for a M7.5 event ( ) computed
earthquakes. following Youd et al. [14]. Figure 22 provides a comparison
of the computed ground motion severity in terms of the cyclic
Field investigations of soil characteristics stress ratio (CSRM7.5) for both the Darfield and Christchurch
earthquakes and CRRM7.5 for a test site in the eastern
Following the 22 February 2011 earthquake field Christchurch suburb of Bexley. As shown in this figure,
investigations were performed in concert with documentation liquefaction is predicted to have occurred during both
of observed damage. Here a summary of some of the methods earthquakes (i.e., CSRM7.5 > CRRM7.5). However, the factor of
and obtained data are given. Readers are referred to Green et safety against liquefaction (FS) is lower for the Christchurch
al. [13] for further details. Other field investigations including earthquake than the Darfield earthquake; where FS =
comprehensive CPT, SPT and Gel-Push sampling of CRRM7.5/CSRM7.5. The lower factor of safety indicates
undisturbed soils are still in the phase of processing and increased severity of liquefaction. These predictions are
interpretation. consistent with field observations in Bexley made after the
two earthquakes.
The in-situ testing procedures discussed herein are: spectral
analysis of surface waves (SASW), and dynamic cone Figure 23 illustrates the resulting data at all of the sites where
penetrometer (DCP), which were used to estimate the shallow DCP and SASW tests were performed following both the 4
soil shear wave velocity and strength (via corrected SPT N September 2010 and 22 February 2011 earthquakes, based on
values), respectively. The DCP used in the field tests utilizes a the normalised and convert SPT N-value (N1,60cs) and CSRM7.5
6.8 kg mass on an E-rod slide drive to penetrate an oversized of the liquefaction-susceptible layer, as well as the surface
45° apex angle cone. The cone is oversized to reduce rod evidence of liquefaction. For comparison, the liquefaction
friction behind the tip. Experience has shown that the DCP can triggering relationship proposed for clean sand by Youd et al.
be used effectively in augered holes to depths up to 6.1 m. The [14] is also shown, for which it can be seen there is a good
DCP tests consists of counting the number of drops of the 6.8 correlation with the obtained field data.
kg mass that is required to advance the cone ~4.5 cm, with the
number of drops referred to as the DCP N-value or NDCPT, and
can be correlated to SPT N value. The SASW field
measurements in this study were made using three 4.5 Hz
geophones, a ‘pocket-portable’ dynamic signal analyzer, and a
sledge hammer. Each tests took less than 45 minutes per
location and typically enabled VS profiles to be generated
218

(a) (a)

(b)

(b)

Figure 23: Summary of the: (a) DCP tests; and (b) SASW
tests at all sites illustrating the computed cyclic
stress ratio (CSRM7.5) of the inferred liquefiable
layer as well as whether surface manifestation
of liquefaction was evident. Test data includes
both the 4th September 2010 and 22nd February
2011 earthquakes [13].

Figure 22: Comparison of CSRM7.5 for the Darfield (DF EQ)

and Christchurch (CH EQ) earthquakes with
CRRM7.5 for a site in Bexley (FC = 9%): (a)
profiles for DCP test; and (b) profiles for SASW
test [13].

Lateral spreading
Along the Avon River, particularly to the east of CBD, lateral
spreading occurred, causing horizontal displacements at the
river bank on the order of several tens of centimeters to more Figure 24: Lateral spreading toward the Avon River.
than two meters (Figure 24). At ten locations along the Avon
River, where lateral spreading measurements were conducted 100-200 meters from the river, while other aerial observation
after the 4 September earthquake, measurements of lateral methods suggest that the effects of spreading might have been
spreading displacements were carried out again after the even beyond these distances. Further more detailed
February earthquake. It was found that the permanent lateral evaluation/analysis of permanent ground displacements is
displacements were two to three times the displacement currently in progress.
measured after the September earthquake indicating increased
spreading movement which is in agreement with the more Avon River temporary stop banks
severe liquefaction observed in these areas during the
The significant subsidence of large areas of Christchurch as a
February event. Ground surveying indicated that ground
result of the widespread liquefaction and associated lateral
cracks associated with lateral spreading extended as far as
spreading increased the risk of flooding from both tidal and
 219

local rainfall events in areas of the city mostly coinciding with areas with limited space, reinforced earth and diamond block
the residential red zone. In some areas, settlements of over a 1 walls were used. An example of stop bank construction in
metre were measured, a significant drop in ground elevation Figure 25c shows the height of the stop banks relative to the
given the low lying nature of the city, even prior to the roadway. Flood levels along the Avon River are
Darfield earthquake. approximately 3 m above MSL, meaning that stop banks with
crests 1.4 m above the current ground levels would be required
Emergency stop banks were initially constructed due to the in places. Hence, seawalls and treatment of the foundation
expected spring tides in April 2011. These were built up to 1.8 soils may be required to provide adequate flood protection in
m above mean sea level (MSL), with 11 km of stop banks the medium to long term.
built over four days along the Avon River. Silty gravel was
used for the construction material, as it was readily available
and reasonably impermeable. Lateral spreading cracks were LIQUEFACTION EFFECTS ON INFRASTRUCTURE
filled prior to stop bank construction, but otherwise there were
no improvements to the foundation material. An example of Bridge structures
the stop bank construction along the Bexley Wetland is shown The city of Christchurch and the surrounding districts contain
in Figure 25a, with a geogrid used under stop banks if the over 800 road, rail and pedestrian bridges. Most bridges are
foundation material was weak. Large settlements in this area reinforced concrete, symmetric, and have small to moderate
meant that many houses that were approximately a metre spans (15 – 25 m). Even though bridges were subjected to
above the water level now sit below the crest of the stop banks ground motions at or above their design levels throughout the
(Figure 25b). Overall this emergency system performed well central and eastern city, the majority sustained minimal
during the spring tides. damage as a result of inertial loading from shaking alone. The
majority of bridge damage during the Christchurch earthquake
(a)
was a result of kinematic loads imposed by lateral spreading
of river banks along the Avon and Heathcote Rivers. Bridges
along these rivers suffered varying levels of lateral spreading-
induced damage, with ground conditions and distance from the
fault rupture controlling this response. Most of the bridge
damage was located in the central and eastern parts of the city,
where ground shaking was the strongest and soil conditions
weakest. Although liquefaction was widespread, only five
bridges within the city suffered major damage and ten
developed moderate damage. This compares to only two
bridges with moderate damage in the city following the
Darfield earthquake. All bridges affected by lateral spreading
were open to traffic within a week of the Christchurch
earthquake. Only four bridges in the city had appreciable
damage on sites that did not experience liquefaction, two with
(b) major damage, and two with moderate damage. A more
complete summary of the bridge damage can be found in
Wotherspoon et al. [15] and Palermo et al. [16].
The majority of the significant bridge damage occurred along
the Avon River downstream of the Christchurch CBD. Of the
nine bridges along this stretch of river, two had major damage
and five were moderately damaged. The remaining two had
only minor approach damage. The damage patterns along the
Avon were fairly consistent: settlement and lateral spreading
of approaches, back rotation and cracking of the abutments,
and pier damage [15]. In most cases the bridge decks
restrained the movement of the top of the abutment, resulting
(c)
in their back rotation (e.g., Figure 26a). Some of the damaged
bridges had pile foundations, with lateral spreading placing
large demands on the abutment piles, and likely resulting in
plastic hinging below grade. Settlement and spreading of the
approaches impaired bridge serviceability, and was the main
reason for bridge closure. The most severe case of settlement
was the northern abutment of Gayhurst Rd Bridge, with large
settlements of the surrounding area occurring after the
Darfield, Christchurch, and June 13 2011 earthquakes. The
Figure 25: (a) Fill material used for construction and combined effect of these events is clearly shown in Figure
geogrid/liner for temporary stop banks; (b) 26b, with over 1 metre of settlement. The southern abutment
position of residences below temporary stop developed only minor settlements of the order of a few
banks in Bexley wetland; and (c) stop bank centimetres, highlighting the variability of damage from one
construction along New Brighton Rd. bridge to another, but also at each individual site.
Following the construction of emergency stop banks, a more In the Christchurch CBD, bridges crossing the Avon River
comprehensive, but still temporary, stop bank network was performed well, with only one of the 14 bridges in this area
constructed along the Avon River. A total of 17 km of stop suffering moderate damage, 10 developing minor damage, and
banks were constructed from the mouth of the Avon River, three experiencing no damage at all. All were single span
upstream to the suburb of Avonside. Where space was bridges, with the most common damage being minor lateral
available, a trapezoidal stop bank cross section with 3:1 or 4:1 spreading, compression or slight slumping of approach
horizontal to vertical slope, and a 2.5 m crest was used. In material, and minor cracking in abutments. Colombo St
220

(a) (b)

(c) (d)

Figure 26: Summary of bridge damage: (a) back-rotation and settlement of the western abutment of South Brighton bridge; (b)
settlement of the northern approach of Gayhurst Road bridge following the June 13 2011 earthquake; (c) abutment
back-rotation and buckling of the steel arches of Colombo Street bridge; and (d) movement of temporary construction
platform for Ferrymead bridge due to lateral spreading.
Bridge was one of the few bridges in the CBD to develop
noticeable back-rotation of its abutments, being also one of the
only bridges in the city with shallow foundations. This back- made up of a gravel core with 1-m thick silt cap, which
rotation resulted in the buckling of the arched steel girders and extends from the river side and across the top, as shown in
handrails at the edge of the bridge, as shown in Figure 26c. Figure 27a . The stop banks typically sit on sandy soils at or
near the ground water level. A toe filter was also constructed
Of the 14 bridges along the Heathcote River, one had major on the land side of the levee to prevent piping of sand during a
damage, two were moderately damaged, and the remainder high water event. The majority of the damage to stop banks
were either undamaged or suffered minor approach damage, during the Christchurch earthquake was a consequence of
despite being near the fault rupture. This lack of damage is liquefaction in the foundation soils that resulted in lateral
inferred as a result of soils in this region having larger spreading, slumping, and/or settlement. As illustrated in
resistance to liquefaction and lateral spreading than those in Figure 27b, longitudinal cracks were observed along the crest
the vicinity of the Avon River. The most severely damaged of the stop banks. Although not desirable, moderate
bridge was the Ferrymead Bridge, at the mouth of the longitudinal crack widths are not as critical to the functionality
Heathcote River and entrance of the Avon-Heathcote Estuary. of the stop banks as transverse cracks because they do not
This structure was undergoing a deck widening upgrade at the provide a direct seepage path through the stop bank. However,
time of the Christchurch earthquake, with the original there is the potential for such longitudinal cracks to connect
structure, partially completed upgrade, and the temporary undetected transverse cracks or flaws that only penetrate
construction platforms all damaged by lateral spreading of the partway through opposite sides of the stop bank. The resulting
river banks (Figure 26d). seepage path could potentially rapidly enlarge due to internal
erosion and piping at high river levels.
Flood stop banks
Transverse cracks in the stop banks were less commonly
In addition to liquefaction and lateral spreading along the observed than longitudinal cracks and were often associated
Avon River and, to a much lesser extent, Heathcote River with sharp bends along the length of the stop banks and/or
flowing through Christchurch, liquefaction also impacted the slumping of the embankment. Because these cracks provide a
stop banks along the eastern extent of the Waimakariri and direct seepage path from one side of the stop bank to the other,
Kaiapoi rivers to the north [17]. they can severely impact the functionality of the stop banks.
Stop banks in the Canterbury region were often constructed by Even transverse cracks having minor widths could potentially
pushing up river gravels and silts. A typical cross section is rapidly enlarge due to internal erosion and piping at high river
levels and lead to the failure of that section of the stop bank.
 221

Settlement of stop bank sections resulted from both post- area (e.g. PGA’s of 0.33 and 0.21g were recorded at the
liquefaction consolidation in the foundation soils and bearing nearby Kaiapoi seismograph in these respective events). Note
capacity failures due to the reduced strength of the liquefied that some portions of the stop banks were already under repair
foundation soil. In addition to the degradation of stop bank by the time of the authors reconnaissance inspection following
functionality due settlement-induced cracking associated with the Christchurch earthquake. In these cases, the authors
the settlement (similar to that discussed above), settlement supplemented their field observations, to the extent possible,
also reduces the amount of freeboard at high river levels. The with both observations from high resolution aerial images
significance of this loss depends on settlement magnitude, but taken the day after the Christchurch earthquake and field
in general it is not thought to be a significant issue with the observations made by ECan consultants [18].
stop bank system.
To examine the relationship between the severity of the
(a) induced damage to the stop banks and the liquefaction
response of the foundation soil, a stretch of stop banks along
the Kaiapoi River was examined in more depth. As shown in
Figure 28, these stop banks sustained damage ranging from
No Damage to Severe Damage (Table 2). Following the
Darfield earthquake, the New Zealand Earthquake
Commission (EQC) contracted a local firm to perform a series
of cone penetration tests (CPT), among other in-situ tests,
throughout North and South Kaiapoi [19].
From interpretation of CPT logs and available borehole data
[19], the soil profile along the north bank of the Kaiapoi River
east of the Williams Street Bridge is characterized by
(b) approximately 4 m of very loose to loose sand overlying
approximately 4 m of loose to medium dense gravelly sand.
Below approximately 8 m, the sand and gravel layers tend to
be significantly denser than the overlying layers. The depth to
the ground water table varies, but is approximately 1.5 m
deep. On the south bank of the Kaiapoi River east of the
Williams Street Bridge, the soil profile is characterized by
approximately 6 m of very loose to loose silty sand/sand
overlying an approximately 2-m thick layer of loose to
medium dense sand/gravelly sand. Samples of the liquefiable
soils taken adjacent to the stop banks on the north bank had
fines contents around 15%, with the fines being non-plastic.
Using the CPT soundings, the liquefaction susceptibility of the
foundation soils was analyzed following the procedures
Figure 27: (a) Typical geometry and soil composition of stop outlined in Youd et al. [14]. The horizontal PGA recorded at
bank cross section; and (b) typical longitudinal the strong motion seismograph station at Kaipoi (KPOC) was
cracks running along the crest [17]. used for determining CSRM7.5, with PGAs of 0.33 and 0.21g
The majority of stop bank damage from the Christchurch for the Darfield and Christchurch events, respectively. Figure
earthquake occurred east of State Highway 1 (SH1) as shown 29 shows the results from the liquefaction evaluation for the
in Figure 28. In this figure, damage severity is categorized two representative CPT soundings mentioned above. In these
using the scale given in Table 2. The scale has five grades that figures, the cyclic resistance ratio (CRR) for each profile and
range from No Damage to Severe Damage. The damage the CSRs for both events are plotted together, where both the
patterns shown in Figure 28 are very similar to those from the CRR and CSR are adjusted to a Mw7.5 earthquake. For
Darfield earthquake, but are in general less severe for the liquefiable soils (i.e., gravels, sands and cohesionless silts),
Christchurch earthquake. The principal reason for this is that liquefaction is predicted to have occurred at depths where the
the ground motion severity is inferred to be less for the CSRM7.5 > CRRM7.5. Accordingly, for both profiles,
Christchurch earthquake than the Darfield earthquake in this

Table 2: Damage severity categories

Category Description
No Damage No observed damage
Cracks up to 5 mm wide and/or 300
Minor
mm deep. Negligible settlement of
Damage
crest.
Moderate Cracks up to 1 m deep. Some
Damage settlement of crest.
Cracks greater than 1 m deep.
Major
Evidence of deep seated movement
Damage
and/or settlement.
Severe damage or collapse. Gross
Severe lateral spread and/or settlement, cracks
Figure 28: Observed damage to stop banks following the Damage showing deformation of 500 mm or
Christchurch earthquake [20], [19] more.
222

(a) (b)

Figure 29: Liquefaction evaluation for representative sites on the: (a) north bank; and (b) south stopbank of the Kaiapoi river in
the Darfield and Christchurch earthquakes.
liquefaction is predicted to have occurred during the Darfield the 1766 km long waste water network, 142 km (8%) were out
earthquake for almost the entire depth from the ground water of service; and 542 km (31%) with limited service on 16
table to the top of the dense gravel/sand layer (i.e., to ~7.5 m March 2011 (i.e. three weeks after the February earthquake),
and ~11 m for the north and south river banks, respectively). as shown in Figure 31. A significant part of the network was
However, during the Christchurch earthquake, marginal still out of service even three months after the earthquake, and
liquefaction is predicted to occur at a few isolated depths it is estimated that it will take two to three years to fully
within both profiles. recover the system.
Loss of grade, joint failures, cracks in pipes and failure of
Impacts of liquefaction on pipe networks laterals were the most commonly observed types of failures.
The large ground movements and deformation (in extension, Loss of critical facilities such as pump stations also
compression, shear, and combined modes) including ground contributed to the overall poor performance of the system.
distortion, cracks, fissures and venting sink-holes, resulting Buoyancy of concrete vaults at potable water and wastewater
from the severe liquefaction and lateral spreading caused pump stations, compounded by liquefaction-induced
severe damage to underground pipe networks such as the settlement, caused pipeline breaks at their connections with
potable water, wastewater and storm water systems. These the vaults. Approximately 1 m of settlement at the Bexley
systems have different characteristics and they were affected Pump Station ruptured the well, flooding the surrounding
and performed quite differently in the 22 February earthquake. neighborhood at 140 m3/hr. Silt and sand from liquefaction
washed into the Bromley sewage treatment plant from broken
The potable water system is a system of relatively shallow wastewater pipelines, causing damage in the primary settling
pipe network (mostly in the top 60 cm of the ground). It is a tanks. Nearly all facilities at the sewage treatment plant were
pressurized system composed of mains and sub-mains. Figure affected by liquefaction, which caused differential settlement
30 shows a summary of the water mains network in of the clarifiers, thereby seriously impairing secondary
Christchurch indicating pipe materials (solid lines) and the treatment capabilities.
location of breaks (faults; red solid symbols) caused by the 22
Note that the waste water system includes both pressurised
February earthquake. Superimposed in this figure is the
and gravity components, and the network consists of pipes of
liquefaction map (shown in Figure 4) indicating the areas
different sizes and materials including concrete, ceramic, cast
affected with different severity of liquefaction. It is apparent
iron and plastic (PVC and PE) pipes. This system is much
in this figure that most of the breaks were located in the area
deeper, typically at 3-4 m depth from the ground surface,
affected by liquefaction. A more rigorous preliminary analysis
making it more vulnerable to liquefaction effects. For both
indicates that about 4.6 % of the pipes (pipe segments) were
potable water and waste water systems, the most severe
damaged, or about 78 km out of 1676 km total pipe length.
damage was inflicted by lateral spreading.
About 80% of the damaged pipes were in areas that
manifested either moderate-to-severe or low-to-moderate
liquefaction. Similar observations and preliminary findings Impact on electrical infrastructure
were obtained for the sub-mains system which is dominated There was serious damage to the underground electric power
by polyethylene (PE) pipes. Despite the relatively large system, with failure of all major 66 kV underground cables
number of breaks, the potable water supply was quickly supplying the Dallington and Brighton areas caused by
restored within several days of the earthquake. liquefaction-induced ground movements. Over 50% of all 66
The waste water system was hit particularly hard in the areas kV cables suffered damage at multiple locations.
severely affected by liquefaction and lateral spreading. Out of
 223

Figure 30: Water mains pipe network and location of breaks (faults) caused by the 22 February 2011 earthquake; coloured lines
indicate pipe materials; coloured areas indicate liquefaction severity as defined and mapped in Figure 4.

Figure 31: Waste water service status on 16 March 2011. (M. Christison, pers. comm.)
spread cracks were noted on the north side of the building
PERFORMANCE OF IMPROVED GROUND between the water and the building. The crack closest to the
water had a maximum width of 13 cm and the crack closest to
Waterside Apartment Building the building had a maximum width of about 4 cm. The larger
The Waterside apartment building area was investigated in crack extended along the top of the embankment north of
detail with regard to the performance of improved ground to Tidal View Rd, and additional cracking was present between
resist liquefaction. The building is indicated by the dashed the road and building.
white line in Figure 32a, is situated between the Avon- There was evidence of movement of the building following
Heathcote estuary to the north and Ferry Rd to the south, and the Christchurch earthquake, however these were relatively
immediately west of the Ferrymead Bridge. As shown in minor given the severity of liquefaction in the surrounding
Figure 32b, the structure consists of a 6-story precast concrete area. The building settled between 4 and 8 cm and had a slight
panel building with a single basement level carpark. The tilt toward the water. On the north side of the building,
building is supported on shallow foundations overlying stone separation walls on the ground surface showed differential
columns. Large volumes of ejecta were evident in the movement as shown in Figure 32c. The separation walls
unimproved areas surrounding the structure, with sand boils sloped downward towards the building at an angle of about 0.4
still present in the estuary adjacent to the structure. Two lateral to 0.5 degrees. This caused the caulk in the expansion joint to
224

(a) (b)

(c) (d)

Figure 32: (a) Aerial view of Ferrymead apartment building indicating surrounding ejected material; and (b) south-west looking
view of the apartments; (c) separation of walls on the north side of the building, including sloping of the
separation walls and shattered glass panels, and compression of caulk at top of expansion joint; and (d)
the basement parking garage. Note the high water mark on the walls.
be compressed at the top of the wall. Three of the glass panels Deans Stand were 600 mm in diameter and were installed on
that were mounted on top of the separation walls were approximately 2.5 m spacing in an arc pattern away from the
shattered, likely due to the compression of the glass against centre of the stadium. Both were damaged during the
the deck above, and cracking of the wall connection beneath earthquake, however it is likely that the stone column ground
this was also evident. The separation walls on the south side of improvements prevented more severe damage.
the building also sloped downward toward the building at an
angle of about 0.8 degrees. Cracking along concrete walkways The Paul Kelly stand settled by up to 400 mm, with settlement
extending out from the structure also indicated differential variations of approximately 70 mm. The thick slab beneath
settlement of the building relative to the ground to the north. the structure prevented any ejected material from coming up
Significant flooding was observed in the basement as shown in within the structural footprint. The Deans stand developed
Figure 32d, and sand had flowed up through the drains in the similar overall settlements, but with much larger variations in
basement slab. In the absence of ground improvement, this settlement across the structure of up to 300 mm. No
structure would most likely have suffered much more liquefaction was present beneath the northern part of the
significant damage. Deans Stand, however, there was a large area with surface
evidence of liquefaction beneath the southern part of this
stand. Both stands suffered structural damage from both the
AMI Stadium differential settlements and the ground shaking, which was
AMI Stadium is located in an area that experienced extensive approximately 30% larger than design levels.
liquefaction and ground damage, with the ejected material Significant liquefaction occurred on the field and manifested
within and surrounding the stadium shown in Figure 33a. All in the form of sand boils and surface deformations (Figure
four stands suffered varying levels of damage during the 33b). Due to mesh below the turf, significant undulations of
earthquake, and all were founded on shallow foundations. The up to 70 cm high occurred across the field because the ejected
Hadlee Stand had no ground improvement measures, suffered material could not vent to the surface.
severe structural damage and has been recommended for
demolition. The Tui Stand was constructed on a fill platform
to raise its level, and suffered less severe structural damage
during the earthquake.
Both the Paul Kelly and Deans Stands were constructed on
widely-spaced stone columns installed within their footprint.
The Deans Stand has shallow foundations connected by grade
beams built upon 8 m deep stone columns, while the Paul
Kelly Stand has a slab foundation up to a metre thick founded
on 9 m deep stone columns. The stone columns beneath the
 225

Both natural and modified (quarry) volcanic rock faces were

sources of rock fall and block collapse, forming large talus
slopes at the base of cliffs, or rock fall run out on some slopes.
The volcanic rocks exposed across the northern part of the
Banks Peninsula are part of the Lyttelton Volcanic Group, and
include dominantly basaltic to trachytic lava flows interbedded
with breccia and tuff, and lava domes. More than 20
residential and commercial buildings downslope of the cliffs
in Redcliffs and Sumner were destroyed by rock fall debris
(e.g., Figure 34).
Several types of rock fall protective measures were observed
at the base of the quarry wall in the Redcliffs. These included
a gabion, rock fall fences, and a rock berm. The gabion
performed well in stopping the block collapse of the cliff from
impacting the house below the gabion. Two rock fall fences
adjacent to the gabion were less successful, as both were filled
and overtopped by the large volume of the block failures. A
rock berm was constructed along the schoolyard border at the
base of the quarry wall, possibly using debris from a more
(b) limited rock fall that may have been generated by the 2010
Darfield earthquake (the berm is not present on the 2009 pre-
earthquake imagery). This berm was successful in protecting
the schoolyard, as no rocks were observed in the area beyond
the rock berm.

CONCLUSIONS
This manuscript has provided documentation of some of the
salient geotechnical features of the 22 February 2011
earthquake, as evident during extensive post-event
reconnaissance. The geotechnical aspects of this earthquake
Figure 33: (a) Aerial photo of AMI Stadium indicating are exceptional from several viewpoints. The earthquake-
liquefaction damage; and (b) Liquefaction induced ground shaking caused very widespread and severe
damage to the stadium field. liquefaction in native soils of Christchurch (including its
central business district), and numerous rock falls and slope
failures in the Port Hills, all of which resulted in fatalities,
huge damage to the city infrastructure and/or large economic
SLOPE INSTABILITY IN THE PORT HILLS
losses. Tens of thousands residential properties were
Rock falls, block failures, and other forms of landslides were profoundly affected by the liquefaction and rock/slope
widespread in the near-fault region around the Port Hills instabilities; the central business district still remains cordoned
(Sumner, Redcliffs, Lyttelton, Cass Bay, and Rapaki). off for general use (at the time of writing), with at least one
Landslides were significant on the northern side of Lyttelton third of its building stock going to be demolished; and the
Harbour, with relatively less rock falls on the southern side of lifelines of Christchurch suffered extensive damage. The 22
the Harbour (e.g. Diamond Harbour). These slope failures February 2011 earthquake was the second strong earthquake
resulted in five deaths and damaged or destroyed many roads, to hit Christchurch in a period of less than six months, and was
tracks, and structures. Almost every cliff face in the Port Hills subsequently followed by a two other notable strong
generated a rock fall, while over-steepened road cuts and earthquakes on 13 June 2011. The observations of re-
quarry walls were subjected to block collapse or large volumes liquefaction and cumulative effects from this series of strong
of rock fall. Rock falls were the most widespread earthquakes are also unprecedented. More in-depth studies on
manifestation of slope failure, causing the five deaths and the the phenomena and damage features presented herein are
most structural damage. Deep-seated landslides were found currently under way, with the hope to advance the research
only at a few locations, most of which were at the top of findings in this area and provide strong support to the rebuild
coastal headlands. Numerous failures occurred in retaining and recovery of Christchurch.
walls and fill slopes, resulting in damage to roads, property,
and commercial and residential structures.
(a) (b)

Figure 34: Examples of rock falls on the Ports hills.

226

[9] NZS 1170.5. (2004) "Structural design actions, Part 5:

ACKNOWLEDGEMENTS Earthquake actions - New Zealand". Standards New Zealand:
Wellington, New Zealand, 82.
The primary support in the reconnaissance effort for the NZ
members was provided by the Earthquake Commission New [10] Cubrinovski, M. and Taylor, M. (2011) "Liquefaction
Zealand (EQC) and University of Canterbury. Liam map of Christchurch based on drivethrough reconnaissance
Wotherspoon’s position at the University of Auckland is after the 22 February 2011 earthquake". University of
funded by the Earthquake Commission. The primary support Canterbury.
for the US GEER Team members was provided by grants
from the U.S. National Science Foundation (NSF) as part of [11] Elder, D. and McCahon, I. (1990) "Near surface
the Geo-engineering Extreme Event Reconnaissance (GEER) groundwater hydrology and excavation dewatering in
Association activity through CMMI-00323914 and CMMI- Christchurch", in Groundwater and Seepage Symposium:
1137977. Rod Cameron and Mark Christison, Christchurch Auckland, New Zealand.
City Council, provided invaluable information on the potable [12] Cubrinovski, M. and McCahon, I. (2011) "Foundations
water and waste water systems. The first author would also on deep alluvial soils", University of Canterbury,
like to acknowledge the funding support from the Natural Christchurch. 40.
Hazards Research Platform (NHRP) through the short-term
recovery projects. Any opinions, findings, and conclusions or [13] Green, R.A., Wood, C., Cox, B., Cubrinovski, M.,
recommendations expressed in this material are those of the Wotherspoon, L., Bradley, B.A., Algie, T., Allen, J.,
authors and do not necessarily reflect the views of the Bradshaw, A. and Rix, G. (2011) "Use of DCP and SASW
National Science Foundation, EQC, or the host institutions of tests to evaluate liquefaction potential: Prediction vs.
the authors. observations during the recent New Zealand earthquakes".
Seismological Research Letters, Focused Issue on the 2011
Finally, we would also like to acknowledge the assistance of Christchurch New Zealand Earthquake.
all NZ and US GEER team members who participated in the
reconnaissance of these events. Their contributions are noted [14] Youd, T.L, Member C, ASCE, Idriss, I.M., Fellow C-C,
at the GEER website at: http://www.geerassociation.org/. Andrus, R.D., Arango, I., Castro, G., Christian, J.T., Dobry,
R., Finn, W.D.L., Jr. LFH, Hynes, M.E., Ishihara, K., Koester,
REFERENCES J.P., Liao, S.S.C., III WFM, Martin, G.R., Mitchell, J.K.,
Moriwaki, Y., Power, M.S., Robertson, P.K., Seed, R.B. and II
[1] NZSEE. (2010) "Special Issue: Preliminary observations KHS. (2001) "Liquefaction Resistance of Soils: Summary
of the 2010 Darfield (Canterbury) Earthquakes". Bulletin of Report from the 1996 NCEER and 1998 NCEER/NSF
the New Zealand Society for Earthquake Engineering; 43(4): Workshops on Evaluation of Liquefaction Resistance of
215-439. Soils". Journal of Geotechnical and Geoenvironmental
Engineering; 127(10): 817-833.
[2] Gledhill, K., Ristau, J., Reyners, M., Fry, B. and Holden,
C. (2011) "The Darfield (Canterbury, New Zealand) Mw 7.1 [15] Wotherspoon, L., Bradshaw, A., Green, A.G., Wood, C.,
Earthquake of September 2010: A Preliminary Seismological Palmero, A., Cubrinovski, M. and Bradley, B.A. (2011)
Report". Seismological Research Letters; 82(3): 378-386. "Bridge performance during the 2011 Christchurch
earthquake". Seismological Research Letters, Focused Issue
[3] Stirling, M.W., Gerstenberger, M., Litchfield, N.,
on the 2011 Christchurch New Zealand Earthquake; 82(6):
McVerry, G.H., Smith, W.D., Pettinga, J.R. and Barnes, P.
950-964.
(2007) "Updated probabilistic seismic hazard assessment for
the Canterbury region". 58. [16] Palmero, A. et al. (2011) "Bridge performance 22 Feb
earthquake". Bulletin of the New Zealand Society for
[4] Beavan, J., Fielding, E.J., Motagh, M., Samsonov, S. and
Earthquake Engineering.
Donnelly, B.S. (2011) "Fault location and slip distribution of
22 February 2011 Mw 6.3 Christchurch, New Zealand, [17] Green, R.A., Allen, J., Wotherspoon, L., Cubrinovski, M.,
earthquake for geodetic data". Seismological Research Letters, Bradley, B., Bradshaw, A., Cox, B. and Algie, T. (2011)
Focused Issue on the 2011 Christchurch New Zealand "Performance of Levees (Stopbanks) during the 4 september
Earthquake. 2010 Mw 7.1 Darfield and 22 February 2011 Mw 6.2
Christchurch, New Zealand, Earthquakes". Seismological
[5] Holden, C. (2011) "Kinematic source model of the
Research Letters; 82(6): 939-949.
February 22nd 2011 Mw6.3 Christchurch earthquake using
strong motion data". Seismological Research Letters, Focused [18] Riley Consultants. (2011) "Waimakariri and Kaiapoi
Issue on the 2011 Christchurch New Zealand Earthquake. River Stopbanks Findings of Condition Assessment Post 22
February 2011 Earthquake. Letter Report 10820/2-A from
[6] Brown, L.J. and Weeber, J.H. (1992) "Geology of the
Riley Consultants to Environment Canterbury (ECan)":
Christchurch urban area", Geological and Nuclear Sciences
Christchurch, New Zealand.
110.
[19] Tonkin and Taylor. (2011) "Darfield Earthquake
[7] Cubrinovski, M., Bray, J.D., Taylor, M.L., Giorgini, S.,
Recovery Geotechnical Factual Report - Kaiapoi North",
Bradley, B.A., Wotherspoon, L. and Zupan, J. (2011) "Soil
Tonkin and Taylor: Christchurch, NZ.
liquefaction effects in the Central Business District during the
February 2011 Christchurch earthquake". Seismological [20] Green, R.A., Allen, J., Wotherspoon, L., Cubrinovski, M.,
Research Letters, Focused Issue on the 2011 Christchurch Bradley, B.A., Bradshaw, A., Cox, B. and Algie, T. (2011)
New Zealand Earthquake. "Performance of levees (Stopbanks) during the 4 September
2010 Mw7.0 Darfield and 22 February 2011 Mw6.1
[8] Bradley, B.A. and Cubrinovski, M. (2011) "Near-source
Christchurch, New Zealand Earthquakes". Seismological
strong ground motions observed in the 22 Febraury 2011
Research Letters, Focused Issue on the 2011 Christchurch
Christchurch earthquake". Seismological Research Letters,
New Zealand Earthquake.
Focused Issue on the 2011 Christchurch New Zealand
Earthquake.