Subsidence occurring at geothermal fields requires monitoring, analysis, and understanding of the mechanisms in order to ensure that it does not affect field operations. This study utilised a broad range of techniques including spatial analysis, three-dimensional modelling, and the comparison of samples of the cover sequences to investigate the subsidence at Kawerau Geothermal Field. Subsidence at Kawerau is of concern because the Tasman Pulp and Paper Mill is located within the geothermal field and utilises machinery with small alignment tolerances that are sensitive to ground deformation. A probabilistic hazard analysis of Kawerau was completed and maps created indicating the potential for subsidence in the future. Spatial analysis of benchmark re-levelling surveys revealed two types of subsidence features: 1) field wide subsidence and 2) subsidence anomalies. Field wide subsidence, currently covering ~17 km2, is driven by thermal contraction of reservoir deep formations and/or compaction of the reservoir due to effective stress increases related to pressure drawdown. Four local subsidence anomalies each covering 150 – 400 m2 are likely driven by varying shallow processes. Two of these features, termed Bowls B and D, south of the Kawerau Geothermal Ltd. power station are the main focus of this thesis along with an assessment of the Tasman Mill site, its potential to develop an anomaly, and the mechanism of subsidence currently occurring across it. Three-dimensional modelling of the cover sequences to 750 m below relative level was completed in Leapfrog Geo using well logs from Kawerau. Modelling revealed an anomalous thickness of Tahuna Formation below Bowls B and D, and relatively uniform thicknesses across the mill site of other shallow formations. The anomalous thickness of Tahuna Formation was hypothesised as being responsible for the presence of the subsidence bowls by being more compressible than the overlying Caxton Formation which is thicker across the mill site while the Tahuna Formation is thinner. Alternative hypotheses were explored by mapping the relative level of the Matahina ignimbrite, thickness of the Caxton Formation, and distribution of brecciation. To test the main hypothesis, samples of Tahuna and Caxton formations were collected from the Kawerau Core Shed and tested for their physical properties and relative compressibility. XRD and thin section analysis was also completed on the samples. Tahuna Formation was found to have more than three times the porosity of the Caxton Formation and have smectite clays present. Using a method developed for testing the relative compressibility of weak rock the Tahuna Formation was found to generally be twice as compressible and elastic as the Caxton Formation when saturated. Samples of Recent alluvium from the mill site were also tested for their physical properties and found not to have the potential to contribute to subsidence across the mill site. However further investigation is required to confirm the mechanisms of Bowls B and D. A hazard analysis of Kawerau Geothermal Field found that the field has a low annual probability of being impacted by volcanic and volcanogenic, earthquake, and flooding events. Probabilities are calculated based on the reoccurrence intervals for each event. A hazard map for subsidence at Kawerau is also developed and outlines four zones of risk. Infrastructure at risk based on trends of subsidence is also analysed for its susceptibility to subsidence and mitigation methods discussed. The overall conclusion is that the geological conditions beneath the mill site are unlikely to form a local subsidence anomaly, and the mill site is largely unaffected by the field wide subsidence bowl. Ground tilt values are within mill machinery tolerances, and based on current trends the spatial extent of subsidence anomalies will remain approximately the same into the future.

Kawerau is located in the Bay of Plenty on the north-east coast of the North Island, New Zealand. Kawerau is an active geothermal field where fluids have been extracted for energy use since the 1950’s when a pulp and paper mill was constructed due to the close proximity to forestry areas and the geothermal energy source. Kawerau has seen significant development in the last 10 years with the commissioning of a 100 MW geothermal power station by Mighty River Power in 2005. Kawerau is located on the south-western edge of the Rangitaiki Plains; these plains have been modified considerably over the last 125 years since the 1886 Tarawera eruption by both natural and anthropogenic mechanisms. Processes at work in the Kawerau area include active volcanism, rifting, fluvial processes, shallow and deep water extraction, anthropogenic river modification and diversion, and construction of buildings and factories. Subsidence is an issue in geothermal and oil fields worldwide and Kawerau is no different. This research aims to determine the origin of localised subsidence features identified by levelling surveys within the Kawerau Geothermal Field. Ground subsidence surrounding the pulp and paper mill, geothermal power station and residential properties in Kawerau has been monitored with levelling surveys since the 1970’s. The potential effects of continued subsidence and tilt within this area could negatively affect the operation of the industry in the area, particularly the pulp and paper mill due to the sensitivity of the paper rollers to tilt. Subsidence in Kawerau occurs on two scales: the first is a large, field-wide subsidence feature, the second is a series of smaller, localised subsidence features which this thesis focuses on. First, identifying the location and characterising the properties of historic river channels, as well as their response to human demand, such as land and water use has been the primary approach in determining the origin of subsidence features. This helped build a picture of how the area appeared 125 years ago and add to our understanding of the history and landscape of the Rangitaiki Plains. Second, to determine the cause(s) and mechanism(s) of the subsidence in Kawerau, field and laboratory investigations were undertaken. Site investigations included geomorphological mapping, ground penetrating radar (GPR), electrical imaging, hand augering and face logging. Laboratory investigations included permeability testing, determination of Atterberg Limits, dispersion testing, grain size distributions, microscopy and allophane detection testing. Aerial photograph and LiDAR interpretation as well as a literature review has shown the approximate location of where the Tarawera River used to flow before it was diverted to aid in the draining of the Rangitaiki Plains. In the approximate location of the old Tarawera River, the geophysical survey identified an extension of the Onepu fault. This fault may have influenced the original location of the Tarawera River by creating low points in the topography as the result of seismic events. The Tarawera River path was diverted to its current path in the early 1900s following the large outbreak flood from Lake Tarawera. Basin wide subsidence at Kawerau has been attributed to geothermal fluid extraction and the resulting contraction and/or cooling of the reservoir. This has caused low rates of subsidence across the whole field. This subsidence is unlikely to cause any damage to surface features due to its low rate and low angles of tilt. Basin wide subsidence is not the focus of this thesis so is not covered in detail. This thesis focuses on two main sites of subsidence. Site 1 lies between the mill site and the Mighty River Power geothermal power station. Site 2 lies in farm land to the north of the mill and the old air strip. The mechanisms controlling subsidence at these sites is believed to be acting independently of each other. Primary mechanisms of subsidence at Site 1 include indirect seismic activity, direct disturbance by construction, vibration, apparent subsidence, the influence of drainage through the site, and wetting and drying sequences associated with rainfall and the soak ponds immediately adjacent to Site 1. Subsidence at Site 2 is likely to be caused by direct seismic activity, indirect seismic activity, consolidation of sediment due to changes in the groundwater table, and the influence of perched water tables.