Mapping surface liquefaction caused by the September 2010 and February 2011 Canterbury earthquakes: a digital dataset

Figures & data

Figure 1. Overview of the Christchurch area, showing the epicentres of the Darfield and Christchurch earthquakes (stars), and the areas of interest (AOI) for which we have mapped the effects of liquefaction. Note that the central city area (dashed outline) was mapped on the ground by The University of Canterbury and Tonkin & Taylor, not from aerial image interpretation. Surface trace of the Greendale Fault is from Langridge et al. (2016) and sub-surface fault (green dashed line, location of dipping fault plane as projected to within c. 1 km of the ground surface) is from http://www.gns.cri.nz/Home/Our-Science/Natural-Hazards/Recent-Events/Canterbury-quake/Hidden-fault. Earthquake epicentres are from GeoNet and background map is LINZ Topo250.

Figure 2. Areas covered by various images used in the analysis. See Table 1 for image details. Background map is LINZ Topo250.

Table 1. Details of images used in this analysis.

Image (originator)	Image acquisition date^a	Spatial resolution (cm)	Band information
SEPTEMBER 2010 DARFIELD EARTHQUAKE IMAGES
GeoEye	4/09/2010	50	RGB-NIR
NZAM aerial photography	5/09/2010	25	RGB
NZAM aerial photography	6/09/2010	25	RGB
WorldView-2	12/09/2010	50–66^b	RGB
WorldView-1	22/09/2010	50	Panchromatic
WorldView-1	26/09/2010	50	Panchromatic
Lincoln University aerial photography	22/09/2010	54–66^c	RGB
FEBRUARY 2011 CHRISTCHURCH EARTHQUAKE IMAGES
WorldView-2	24/02/2011	50	RGB-NIR
NZAM aerial photography	24/02/2011	10	RGB
WorldView-2	29/03/2011	50	RGB-NIR

^aImage acquisition date is corrected to corresponding NZ date.

^bImages captured at different off-nadir angle resulted in spatial resolution varying from 50 to 66 cm.

^cSpatial resolution varied due to change in flying heights.

See Figure 2 for area covered by each image.

Table 2. Attributes assigned to polygons at the time of digitising.

Attribute	Description
FEATURE	Feature being mapped, e.g. ‘liquefaction’, ‘flooding – water’ (see Appendix 1 for details)
EVIDENCE	Note or supporting evidence of reason for digitising feature
ORIGINATOR	Initials of person/company who mapped the feature (see Appendix 2 for details)
IMAGE	Name/source of the image that the feature was digitised from
IMAGE_RES	Ground pixel resolution of the image used for digitising (in metres)

See Appendix 1 for a list of features in the FEATURE attribute and their descriptions.

Figure 3. Liquefaction (sand boils, cracking/lateral spread and fissures) east of Kaiapoi, September 2010. The alignment of sand boils and fissures (grey sediment/ejecta) relates to an in-filled former channel of the Waimakariri River (Wotherspoon et al. 2012). Image is NZAM 05/09/2010; centre of view approximately 1573200, 5195800 (NZTM). Rectangle indicates location of detail shown in Figure 4.

Figure 4. Part of Figure 3 (boxed area) viewed at approximately full image resolution reveals details of the features being mapped.

Figure 5. Liquefaction at Rawhiti, south of Halswell, September 2010. Damage includes cracking/lateral spread of the road, and sand boils and flooding (including sedimentation) in the paddocks (image GeoEye 04/09/2010; centre of view approximately 1564800, 5173000).

Figure 6. Sand boils and surface flooding across Carters Road, southwest of Taitapu, September 2010 (image WorldView 12/09/2010; centre of view approximately 1560000, 5164000). Note the en-echelon pattern of short fissure segments that merge to form longer contiguous features (e.g. arrows).

Figure 7. Liquefaction adjacent to the Styx River, west of Brooklands, February 2011. Note the alignment of sand boils parallel to the river margin (arrowed). This area was also affected in September 2010 (image NZAM 24/02/2011; centre of view approximately 1574970, 5194550).

Figure 8. New (darker grey/wet; post February 2011 earthquake) and old (paler grey/dry; pre-February 2011) sand boils west of Lansdowne, February 2011 (image NZAM 24/02/2011; centre of view approximately 1565000, 5171530). Note: there were other earthquakes/aftershocks that caused localised liquefaction (Quigley et al. 2013) that was not mapped separately. The various episodes of liquefaction presented in this figure may represent some of these aftershocks, but by far the most damage occurred during the September 2010 and February 2011 earthquakes. Image capture was not frequent enough to determine a higher temporal resolution of damage.

Figure 9. Main area affected by liquefaction from the September 2010 Darfield earthquake. Grey dashed box outlines the area shown on Figure 10. Note that this map does not display the liquefaction mapping by Tonkin & Taylor – see text for details. See Brackley et al. (2012) and Datasets S1–S2 for full coverage. Epicentre location (black star) is from GeoNet. Background map is LINZ Topo250. Surface geological data from Forsyth et al. (2008) and Heron (2014). This figure includes data extracted from the Canterbury Geotechnical Database (https://canterburygeotechnicaldatabase.projectorbit.com/), which were prepared and/or compiled for the Earthquake Commission (EQC) to assist in assessing insurance claims made under the Earthquake Commission Act 1993 and/or for the Canterbury Earthquake Recovery Authority (CERA). The source maps and data were not intended for any other purpose. EQC, CERA, their data suppliers and their engineers, Tonkin & Taylor, have no liability for any use of the maps and data or for the consequences of any person relying on them in any way.

Figure 10. Detailed example of liquefaction mapping in the Kaiapoi area for the September 2010 Darfield earthquake. This map does not display liquefaction mapped by Tonkin & Taylor – see text for details. Note the distribution of observed liquefaction along abandoned river channels and in swampy ground near the coast. See Figure 9 for location. Background map is LINZ Topo50. This figure includes data extracted from the Canterbury Geotechnical Database (https://canterburygeotechnicaldatabase.projectorbit.com/), which were prepared and/or compiled for the Earthquake Commission (EQC) to assist in assessing insurance claims made under the Earthquake Commission Act 1993 and/or for the Canterbury Earthquake Recovery Authority (CERA). The source maps and data were not intended for any other purpose. EQC, CERA, their data suppliers and their engineers, Tonkin & Taylor, have no liability for any use of the maps and data or for the consequences of any person relying on them in any way.

Figure 11. Total mapped area affected by liquefaction from the February 2011 Christchurch earthquake. Comparison with Figure 9 shows much more liquefaction in Christchurch City area and eastern suburbs, reflecting the closer epicentre and therefore locally higher shaking intensities for this event. Some areas were affected by both events. Green dotted line shows projection of dipping subsurface fault plane at c. 1 km depth. Epicentre location (black star) is from GeoNet. Background map is LINZ Topo250. Surface geological data from Forsyth et al. (2008) and Heron (2014). This figure includes data extracted from the Canterbury Geotechnical Database (https://canterburygeotechnicaldatabase.projectorbit.com/), which were prepared and/or compiled for the Earthquake Commission (EQC) to assist in assessing insurance claims made under the Earthquake Commission Act 1993 and/or for the Canterbury Earthquake Recovery Authority (CERA). The source maps and data were not intended for any other purpose. EQC, CERA, their data suppliers and their engineers, Tonkin & Taylor, have no liability for any use of the maps and data or for the consequences of any person relying on them in any way.

Figure 12. Furthest distance from the epicentre (

,

) and fault (

,

) of known liquefaction occurrences plotted against magnitude for the Darfield and Christchurch earthquakes. Data for the Darfield earthquake falls within the bounds of global observations of earthquake data (

,

), but for the Christchurch earthquake sits slightly beyond the Ambraseys (1988) bound for both epicentral and fault distance. Adapted from Maurer et al. (2015).

Figure 13. Comparison of the distribution of liquefaction for the Darfield and Christchurch earthquakes in the rural and residential areas (mapped separately), as a proxy for comparing liquefaction severity between the two events. A, The Darfield earthquake affected c. 2.5% of the total area assessed. B, Because of its more distant source, only the most susceptible materials in the city (5% of mapped urban area) were affected. C, The Christchurch earthquake affected c. 3% of the same total area as in A. D, The proximity of the epicentre directly beneath the liquefaction-prone soils of the city and the consequently higher local PGA values resulted in more severe liquefaction, affecting both highly and moderately susceptible soils (c. 30% of mapped urban area). Epicentres are shown by black star and sub-surface fault (at 1 km depth) by green dashed line.

A1.1 Names and description of fields in the accompanying GIS datasets

Fields	Description
FEATURE	Feature being mapped, e.g. ‘liquefaction’, ‘flooding – water’ (see below for details)
EVIDENCE	Note or supporting evidence of reason for digitising feature
ORIGINATOR	Initials of person/company who mapped the feature (see Appendix 2 for details)
IMAGE	Name of the image supplier that the feature was digitised from
IMAGE_RES	Ground pixel resolution of the image used for digitising (in metres)
SITE_VISIT	Denotes whether the mapped feature was checked on the ground
GNS_QA	Initials of GNS personnel who performed the internal QA
GNSQA_NOTE	Comments about whether the GNS QA agreed (or otherwise) with the Originator
GNS_CERT	Combined certainty from Originator and GNS QA
EXTERNAL_QA	Comments from LandCare during visual assessment of mapped feature (see Belliss & Lynn 2012)
CERTAINTY	Overall certainty of the mapped feature, accounting for all QA processes
local_img	File name of the stored image(s) or tile that the feature was digitised from
image_date	Acquisition date of image

Langridge RM, Ries WF, Litchfield NJ, Villamor P, Van Dissen RJ, Barrell DJA, Rattenbury MS, Heron DW, Townsend DB, Lee JM, et al. 2016. The New Zealand active faults database: NZAFD250. New Zeal J Geol Geophy. 59. doi: 10.1080/00288306.2015.1112818

 Web of Science ®Google Scholar

Wotherspoon LM, Pender MJ, Orense RP. 2012. Relationship between observed liquefaction at Kaiapoi following the 2010 Darfield earthquake and former channels of the Waimakariri River. Eng Geol. 125:45–55.

 Google Scholar

Quigley MC, Bastin S, Bradley BA. 2013. Recurrent liquefaction in Christchurch, New Zealand, during the Canterbury earthquake sequence. Geology. 41:419–422. doi:10.1130/G33944.1

 Web of Science ®Google Scholar

Brackley HL comp., Almond P, Barrell DJA, Begg JG, Berryman KR, Christensen S, Dellow GD, Fraser J, Grant H, Harwood N, et al. 2012. Review of liquefaction hazard information in eastern Canterbury, including Christchurch City and parts of Selwyn, Waimakariri and Hurunui Districts. GNS Science consultancy report 2012/218; Environment Canterbury report R12/83. 99 p. + 1 CD.

 Google Scholar

Forsyth PJ, Barrell DJA, Jongens R comps. 2008. Geology of the Christchurch area: scale 1:250,000. GNS Science, Lower Hutt. Institute of Geological & Nuclear Sciences 1:250,000 geological map 16. 67 p. + 1 folded map.

 Google Scholar

Heron DW (custodian). 2014. Geological map of New Zealand 1:250,000. Lower Hutt: GNS Science. GNS Science geological map 1. 1 CD.

 Google Scholar

Ambraseys NN. 1988. Engineering seismology: part I. Earthq Eng Struct Dyn. 17:1–50. doi: 10.1002/eqe.4290170101

 Web of Science ®Google Scholar

Maurer BW, Green RA, Quigley M, Bastin S. 2015. Development of magnitude-bound relations for paleoliquefaction analyses: New Zealand case study. Eng Geol. 197:253–266. doi: 10.1016/j.enggeo.2015.08.023

 Web of Science ®Google Scholar

Belliss SE, Lynn IH. 2012. Quality assurance on GNS-compiled liquefaction mapping after September 2010 and February 2011 seismic events. Lincoln: Landcare Research. Landcare Research Contract Report LC 926.

 Google Scholar