Proceedings World Geothermal Congress 2015 Melbourne, Australia, 19-25 April 2015 1 Hydrothermal Fluid Flow in a Structurally-controlled Basin, Ngakuru Graben, Taupo Rift, New Zealand Warwick Kissling, Andrew Rae, Pilar Villamor and Susan Ellis GNS Science, 1 Fairway Drive, Avalon 5010, New Zealand w.kissling@gns.cri.nz Keywords: Ngakuru Graben, geothermal systems, silica sinter, hydrothermal alteration, fluid flow, modelling, TOUGH2 ABSTRACT The Ngakuru Graben lies in the northern Taupo Rift, New Zealand and hosts a number of major geothermal areas on its western and eastern margins – including Te Kopia, Waikite, Orakeikorako, Atiamuri and Horohoro. As well as these active geothermal systems, areas of extinct hydrothermal activity (i.e. silica sinter, hydrothermal eruption deposits and hydrothermally-altered tephras and lake sediments) are also found at the margins of the graben, including its northern and southern terminations. The graben contains four major active faults (or fault systems) – the Ngakuru, Maleme, Whirinaki and Paeroa. These not only define the graben structurally but play host to its hydrological system which, because of the deep interconnectedness of the faults, can behave in a complex way when there are significant, sudden changes in the fault permeabilities. Such changes are expected to have implications for the location and longevity of past geothermal activity in the graben, as interpreted from fossil sinter formations and other manifestations. A numerical model of this system has been developed using the fluid flow code TOUGH2. The model represents the major geological features of the area and a recently inferred deep heat source to the east of the Paeroa fault. We explore models where there are major perturbations to the permeabilities of the faults. An increase in fault permeability is assumed to be associated with a fault rupture, while a permeability reduction to ‘background’ values can occur as a result of (unmodelled) chemical or geolog ical processes. The key result from the modelling is that the flows in an inter-connected system of faults are complicated, and fluid pathways can be altered by changes in the fault permeabilities. More specifically, despite significant changes in the flows immediately following a rupture, even large increases in fault permeabilities do not lead to major changes such as the cutoff or reversal of flow in the faults on ~100 year timescales. Conversely, we identify a model where a reduction in fault permeability leads to drastic changes in the flows, and these have important consequences for the fossil sinter deposits found at various locations within the Ngakuru graben. 1. INTRODUCTION The Taupo Rift is geologically one of the “hottest” rifts in the world, with high crustal heat flow ( ~ 700 mW/m 2 ; Bibby et al. 1995), high extension rates (up to 12 mm/yr, Wallace et al., 2004), and high silicic eruption rates (e.g., 12.8 km 3 Ka -1 ; Wilson et al., 2009) and volumes (>35 km 3 ; Wilson et al., 2009). In this geological environment, numerous active geological processes are contributing to defining the characteristics and evolution of geothermal systems located within the rift. The Ngakuru Graben in the northern Taupo Rift, has areas of current and fossil geothermal activity located along the basin margins. Geothermal systems occur on the eastern and western margins at Te Kopia, Waikite, Ngapouri, Orakeikorako, Atiamuri and Horohoro. Fossil geothermal areas, in the form of deposits of silica sinter and hydrothermal eruption breccia, as well as hydrothermally-altered tephras and lake sediments are located elsewhere on the graben margins. The coexistence of active and extinct hydrothermal activity among a dense and complex active fault network, and the proximity to large active and extinct silicic volcanic centres, provides a unique opportunity to gain insights into the evolution of geothermal activity in a structurally controlled basin. Both the active and fossil geothermal areas and their spatial relationship to fault structures are the subject of this paper. In particular, we are concerned with understanding the role of faults providing the permeability that gives rise to these geothermal areas in the Ngakuru graben. We address important questions as to what controls the flow of fluid to the surface, why are surface expressions mainly restricted to the basin margins, and how can we explain their apparent ephemeral nature. A hydrological model of the graben has been developed using TOUGH2, which will address these questions and test the influence of both fault architecture and the rate of tectonic activity on hydrothermal activity. The model represents the major geological features of the area on a 30 km wide 2-D northwest-southeast cross-section of the graben that extends to a depth of 7 km. This depth corresponds to the depth where ductile processes are expected to reduce the permeability to a level too low to support convective fluid flow (Bryan et al., 1999; Bibby et al., 1995). The model includes four major faults for which good estimates of average slip rates and other properties are available – the Ngakuru, Maleme, Whirinaki and Paeroa faults. The permeabilities of these faults are estimated from their slip rates, and permeabilities elsewhere in the model are drawn from our experience with regional-scale models of the area (Kissling and Weir, 2005).