Page 6 - gt1503-04

GSA TODAY | MARCH/APRIL 2015 avulsions. This natural sediment replenishment entails the rapid
advance of coarse alluvium along relict and newly excavated chan-
nels, driven by high river flows and accompanied by extensive
flooding. Such avulsions pose a severe physical threat to the built
environment. Extensive flood protection works, including gravel
extraction, were first established in 1928, with three subsequent
flood events breaching the primary stopbank (levee) system,
resulting in floodplain inundation. Throughout the majority of
European settlement, the city has been spared from major floods
from the Waimakariri, although stopbank failure remains a hazard.
Christchurch has also long been vulnerable to localized flooding
from its urban rivers, exacerbated by low-lying, relatively flat terrain of 0.44 to 0.55 m, with a vertical accuracy of ±0.15 m for the 2003
with low gradients and high groundwater levels, extreme tides, and
storm surge. Urban expansion since the 1880s imparted distinct
anthropogenic signatures on local hydrology. Widespread drainage
works undertaken for urban development caused ground surface
subsidence due to reduction of the groundwater levels, leading to
historical surface flooding and ponding in low-lying areas. In
parallel, separate underground storm water and waste water systems
were established, with the latter long recognized as being “leaky”
—that is, allowing infiltration into pipes with associated draining of
groundwater and suppression of local water tables (Wilson, 1989).
The storm water system, originally integrating open channels and
buried pipes and then incorporating roadside gutters, was developed
to manage overland flow runoff exacerbated by expansion of imper-
meable surfaces through suburban development.
THE CES AND URBAN LANDSCAPE EVOLUTION
Between September 2010 and December 2011, Christchurch was
damaged by six earthquakes: 4 Sept. 2010 (MW = 7.1); 22 Feb. 2011
(MW = 6.2, 185 fatalities); 13 June 2011 (two earthquakes:
MW = 5.3 at 1 p.m. and MW = 6.0 at 2:20 p.m.) and 23 Dec. 2011
(two earthquakes: MW = 5.8 at 1:58 p.m. and MW = 5.9 at 3:18 p.m.)
(Fig. 1; for detailed reviews of the geologic and seismic aspects of
the CES, see Beavan et al., 2010, 2011, 2012a, 2012b; Duffy et al.,
2013; Quigley et al., 2012; Bradley et al., 2014). The close proximity
of causative faults to Christchurch generated strong ground motions
(Bradley and Cubrinovski, 2011; Bradley, 2012) that caused exten-
sive damage to residential and commercial properties (Bech et al.,
2014; Fleischman et al., 2014; Moon et al., 2014) and infrastruc-
ture lifelines, particularly potable water, waste water, and road
networks (Cubrinovski, et al., 2014a, 2014b, 2014c; O’Rourke et
al., 2014). Much of the damage to the city’s built environment was
caused by widespread soil liquefaction that occurred predomi-
nantly in saturated, unconsolidated alluvial and marine fine sedi-
ments in east Christchurch, in the region of late Holocene coastal
progradation. In susceptible soils with high water tables (e.g.,
suburbs adjacent to the Avon River), liquefaction was manifested
at the ground surface in earthquakes as low as MW 5.0 and PGAs
as low as 0.08 g (Quigley et al., 2013). Less-susceptible soils
required higher shaking intensities for liquefaction initiation
(Tonkin & Taylor, 2013; van Ballegooy et al., 2014b). Liquefaction
caused significant ground deformations, ejection of groundwater
and sediments on to the ground surface, and lateral spread around
rivers (Cubrinovski et al., 2014c; Quigley et al., 2013; Green et al.,
2014; van Ballegooy et al., 2014b). In some areas, loadings from
structures and preferential ejecta pathways through roads and
buried infrastructure imparted distinct anthropogenic signatures
on surface ejecta patterns.
6
In 2003, the Christchurch City Council commissioned an aerial
LiDAR survey for hydrological modeling purposes. Following the
4 Sept. 2010 Darfield earthquake, another LiDAR survey was
commissioned and flown on 5 Sept. 2010 by the New Zealand
Ministry of Civil Defense and Emergency Management to quantify
property subsidence and to facilitate insurance assessments and
reconstruction work. Further LiDAR campaigns were flown typi-
cally one month after each subsequent major CES earthquake to
provide time for ejected sand and silt to be removed from most
properties and streets, so that measurements recorded ground
surface level. LiDAR capture equipment had a horizontal accuracy
survey and ±0.07 m for the post-earthquake surveys. These errors
exclude Global Positioning System network error and approxima-
tions within the New Zealand Quasigeoid 2009 reference surface,
which has an expected vertical accuracy of ±0.07 m. From each
LiDAR dataset a bare-earth 5-m-resolution Digital Elevation
Model (DEM) was generated; the 5-m-resolution was determined
to be optimal for interpolation of pre- and post-earthquake
LiDAR ground returns in the urban environment. The accuracy
of LiDAR data and bare-earth DEMs were assessed against refer-
ence geodetic survey control benchmarks and topographic surveys
conducted pre-CES on roads and subdivisions at suburb-level in
August 2011 and on residential properties in January 2012. These
assessments showed reasonable accuracy as a whole, with hard
surfaces providing smaller standard deviations of errors for roads
than for residential properties, reflecting the differing roughness
of the two types of terrain. Here we show total vertical elevation
changes (∆ETot), elevation changes due to liquefaction (∆ELiq),
lateral ground movements due to liquefaction (∆XLiq), and vertical
tectonic changes (∆ETec) (Fig. 2). Tectonic movements were deter-
mined using satellite interferometry synthetic aperture radar data
(see Beavan et al., 2011, 2012b), which we subtracted from ∆ETot as
determined by LiDAR-derived DEMs to produce ∆ELiq.
We also present pre-/post-earthquake differential elevation anal-
ysis (∆ETot) for the Avon-Heathcote Estuary, based on 1-m-resolution
DEMs interpolated from LiDAR data (area of bed exposed above
water surface during survey), supplemented by ground survey and
depth-sounder survey data for areas covered by estuarine waters
during LiDAR surveys (Measures et al., 2011; Measures and Bind,
2013). Pre-/post-earthquake ground surveys and echo-sounder
surveys were conducted using Real-Time Kinetic Global Navigation
Satellite System positioning, on foot or with a boat-mounted depth
sounder, and calibrated to local benchmarks.
The 4 September 2010 Darfield earthquake caused 74% of
central and eastern Christchurch to subside; 60% of this area
subsided up to 0.2 m (Fig. 2A). Vertical tectonic displacements
of 0.8 to 1.8 m along the associated surface rupture ~50 km west
of Christchurch caused partial river avulsion and flooding
(Duffy et al., 2013). The 22 February 2011 Christchurch earth-
quake caused 83% of eastern and central Christchurch to
subside further; 78% subsided up to 0.3 m, with localized areas
exceeding 1 m. This event also caused a clear signature of
tectonic uplift (~0.45 m) around the Avon-Heathcote Estuary
caused by blind faults (Fig. 2A and 2E). Compared to pre-
earthquake elevations, 86% of central and eastern Christchurch
subsided through the CES; 10% subsided more than 0.5 m, with
some localized locations exceeding 1 m. Cumulative tectonic
subsidence through the CES reached 0.18 m (Fig. 2E). Both