About the Taihape Landslide

Survey prism installed on the landslide.

The Taihape landslide first moved between 1,800 and 11,000 years ago. It is situated towards the west of Taihape township and includes approximately 45 hectares of land containing approximately 200 houses and a school. Several previous studies of it had been carried out, the earliest in 1971 when significant instability was first noticed.

The Taihape landslide is classified as a deep-seated translational block-slide. This classification refers to the characteristics of the landslide. Deep-seated refers to the depth of movement (depth to the landslide slip-plane); the slip plane of the Taihape landslide has been recorded at 25 m below ground level at the toe (bottom) and increasing to 35 m towards the back scarp (top). The term translational refers to the movement style of the landslide, where it moves as a relatively intact mass (raft) of material, along a planar zone of weakness. In this case, the zone of weakness corresponds to a thin clay layer thought to represent a bedding plane within the local Taihape sandstone. The term block-slide describes the structure within the landslide, which for the Taihape landslide is a series of large blocks (tens to hundreds of metres in size).

The robotic survey station.

Historical information from ground surveys (dating back to 1985) indicate that landslide movement is triggered by rainfall, and to a lesser extent erosion caused by Otaihape Valley Stream, located along the toe of the landslide. Periods of prolonged rainfall cause groundwater levels to rise, which lead to a reduction in strength of the materials forming the landslide slip plane (clay layer). It is this reduction in strength that leads to movement of the landslide. Movement of the landslide can therefore be linked to rainfall events. The intention of the study was to determine movement triggering thresholds, e.g. how much rainfall does it take to trigger movement and by how much does the landslide move. In the past it has been impossible to correlate the triggering factors to the different periods of landslide movement due to the large time intervals (typically greater than yearly) between the historical surveys. To overcome these problems GNS Science installed monitoring equipment on the landslide from June/July 2006 until March 2011.

Information recorded from the landslide can be used to calculate the landslide velocity, identify areas of movement and assess the movement patterns of the landslide to aid RDC (Rangitikei District Council) and EQC to plan potential mitigation strategies.

Typical rain gauge and piezometer installation.

Monitoring Equipment

The equipment installed on the landslide was designed to monitor two triggering factors (factors that trigger movement) and two response factors:

• Rainfall (trigger);
• Ground shaking intensity (trigger);
• Groundwater levels (response);
• Surface movement (response).

Rainfall was monitored by two rain gauges installed on the landslide. One rain gauge was installed near the back scarp of the landslide, the other near the toe of the landslide and so the variation in rainfall across the landslide could be monitored. These rain gauges recorded rainfall as and when it occured, therefore allowing the precise time of peak rainfall to be determined. Equipment to record ground motion in response to earthquakes was installed at the nearby Taihape Rural Hospital. Groundwater levels within the landslide were monitored using piezometers installed in 4 boreholes. The piezometers recorded groundwater levels acting along the slip plane of the landslide.

The surface movement of the landslide was monitored using a robotic survey station. The survey station automatically tracked the position of survey prisms installed on the landslide and the collected data were managed and displayed through custom software, which allowed the entire system to run remotely. The survey station was installed on the hillside opposite (south of) the landslide. The survey error of the total station installed at Taihape varies for each prism and is dependent on several factors, mainly temperature and pressure, caused by changes in the weather. The prisms installed on the landslide were strategically located based on engineering geological mapping and the results from historical surface movement surveys. Several prisms were also installed off the landslide (on stable ground) to act as back sites for the monitoring. These stable prisms were then used to calculate the movement of the prisms installed on the landslide. The distance from the survey station to each prism varies, and typically ranges from 700 to 1,000 m.

The monitoring network

The prisms installed on the landslide were monitored at hourly intervals, the piezometers recorded a groundwater measurement every five minutes and the rain gauges recorded rainfall as and when rainfall occured. The data from the instruments on the landslide were transferred to GNS Science via a communications network. The data were then checked and the results calculated. The photographs show what the different types of equipment actually look like. The process of retrieving the data took approximately one hour from site (Taihape) to the office (GNS Science, Lower Hutt).

Data Web services

All data are made available via web services.

Equipment Locations

The NZMG locations of prisms, tipping bucket rain gauges, and borehole piezometers.

• http://www.geonet.org.nz/services/landslide/taihape/locations/1.0

Processed Data

The processed data are available as XML and compressed (gzipped) XML web services. The compressed web services offer a considerable download bandwidth saving and should be used where possible.

• Compressed: http://www.geonet.org.nz/services/landslide/taihape/sax/1.0/compressed/
• Uncompressed: http://www.geonet.org.nz/services/landslide/taihape/sax/1.0 (drop .gz extension on names)