Asymmetrical structure, hydrothermal system and edifice stability: The case of Ubinas volcano, Peru, revealed by geophysical surveys

Highlights

• The inner pit crater cuts the self-sealing layer and acts as a pressure-release “valve” for the underlying hydrothermal system.

• Ubinas volcano is a highly asymmetric edifice straddling a high plateau and the slope of the deep Ubinas valley.

• The extent of the hydrothermal system may be greater downslope on the eastern flank.

• The slope of the basement on which a volcano has grown plays a major role in the geometry of the hydrothermal systems.

Abstract

Ubinas volcano, the historically most active volcano in Peru straddles a low-relief high plateau and the flank of a steep valley. A multidisciplinary geophysical study has been performed to investigate the internal structure and the fluids flow within the edifice. We conducted 10 self-potential (SP) radial (from summit to base) profiles, 15 audio magnetotelluric (AMT) soundings on the west flank and a detailed survey of SP and soil temperature measurements on the summit caldera floor. The typical “V” shape of the SP radial profiles has been interpreted as the result of a hydrothermal zone superimposed on a hydrogeological zone in the upper parts of the edifice, and depicts a sub-circular SP positive anomaly, about 6 km in diameter. The latter is centred on the summit, and is characterised by a larger extension on the western flank located on the low-relief high plateau. The AMT resistivity model shows the presence of a conductive body beneath the summit at a depth comparable to that of the bottom of the inner south crater in the present-day caldera, where intense hydrothermal manifestations occur. The lack of SP and temperature anomalies on the present caldera floor suggests a self-sealed hydrothermal system, where the inner south crater acts as a pressure release valve. Although no resistivity data exists on the eastern flank, we presume, based on the asymmetry of the basement topography, and the amplitude of SP anomalies on the east flank, which are approximately five fold that on the west flank, that gravitational flow of hydrothermal fluids may occur towards the deep valley of Ubinas. This hypothesis, supported by the presence of hot springs and faults on the eastern foot of the edifice, reinforces the idea that a large part of the southeast flank of the Ubinas volcano may be altered by hydrothermal activity and will tend to be less stable. One of the major findings that stems from this study is that the slope of the basement on which a volcano has grown plays a major role in the geometry of the hydrothermal systems.

Another case of asymmetrical composite cone edifice, built on a steep topography, is observed on El Misti volcano (situated 70 km west of Ubinas), which exhibits a similar SP pattern. These types of edifices have a high potential of spreading and sliding along the slope owing to the thicker accumulation of low cohesion and hydrothermally altered volcanic products.

Introduction

Understanding the inner structure of a composite cone through geophysical survey can help to provide crucial insights into past eruptive history and the structural relationships between the edifice and regional tectonics. Among the major features that can influence the cone behaviour for future activity are the existence of faults and lithological discontinuities (such as those created by calderas or landslides) and the presence of a hydrothermal system that determines the fluid flows and the alteration of the edifice (Lopez and Williams, 1993, Reid et al., 2001). In this study of Ubinas volcano (south Peru), we used the self-potential (SP) technique and soil temperature measurements to outline the hydrothermal system and audio-magnetotelluric (AMT) measurements to investigate the internal structure through the distribution of resistivity. Moreover, the location and temperature of the water springs in a radius of 20 km all around Ubinas volcano have been taken. Information on the hydrothermal system is important because it is part of the plumbing system and as such, it plays a role on the eruptive activity. This is particularly true in the case of the last Ubinas eruption in 2006–2009 (Rivera et al., 2014).

The SP method has been used on active volcanoes for identifying and delineating anomalies associated with the presence of active hydrothermal systems (Lénat et al., 1998, Finizola et al., 2002, Finizola et al., 2003, Aizawa, 2004, Revil et al., 2004, Hase et al., 2005, Finizola et al., 2006; Aizawa et al., 2008; Revil et al., 2008, Barde-Cabusson et al., 2009a, Barde-Cabusson et al., 2009b, Finizola et al., 2009, Finizola et al., 2010, Bennati et al., 2011, Revil et al., 2011, Barde-Cabusson et al., 2012). This method can also be used to monitor the evolution of hydrothermal systems through time (Ishido et al., 1997, Yasukawa et al., 2005). SP anomalies due to the hydrothermal activity can reach hundreds to thousands of millivolts in amplitude (Finizola et al., 2004). These surface electric fields reflect streaming current effects occurring at depth. The main source of SP signal on volcanoes is thought to be electrokinetic coupling (Corwin and Hoover, 1979). Electrokinetic (or streaming) potentials are generated when a fluid flows through a porous medium (electro-osmosis) generating electric current and voltage difference in the double electrical layer (Corwin and Hoover, 1979, Avena and De Pauli, 1996, Lorne et al., 1999a, Lorne et al., 1999b, Revil and Leroy, 2001). The external layer, termed the electrical diffuse layer, is generally positively charged. The fluid flow drags positive charges from the diffuse layer, creating a macroscopic current density and an electrical field called the streaming potential. The current is therefore positive in the flow direction. This has been documented by laboratory experiments for silica and volcanic rocks (e.g. Ishido and Mizutani, 1981, Jouniaux et al., 2000), theoretical works (Lorne et al., 1999a, Lorne et al., 1999b, Revil et al., 1999a, Revil et al., 1999b, Revil and Leroy, 2001), and field data (e.g. Trique et al., 1999). As a result, the electrokinetic effect associated with the down flow of water in purely hydrogeological zones results in negative self-potential anomalies at the ground surface, whereas the uprising flow in hydrothermal systems will result in positive anomalies. In hydrogeological zones, the amplitude of the SP variation can be related to the distance between the topography and the water table (Jackson and Kauahikaua, 1987, Aubert et al., 1993, Aubert and Atangana, 1996, Boubekraoui et al., 1998, Revil et al., 2005). However, several cases of negatively charged electrical diffuse layer (positive zeta potential) have been reported for rocks and minerals located above hydrothermal areas and also for all minerals (with the exception of clays such as smectite) when the pH is below the isoelectric point of the mineral, typically for acidic solutions (Guichet and Zuddas, 2003, Hase et al., 2003, Guichet et al., 2006, Aizawa, 2008). This means that, under certain conditions (type of particles, bulk solution, temperature, pH), either SP maxima or minima can be measured with the same fluid circulation direction. But in most cases, SP profiles extending from the summit to the lower flanks of active volcanoes show two major SP domains (e.g. Sasai et al., 1997, Aubert et al., 2000, Finizola et al., 2004, Ishido, 2004). In the upper part of the edifice, the SP is generally dominated by hydrothermal flow with positive SP/elevation gradient, whereas in the lower flanks, hydrogeological flow is mostly expected with negative SP/elevation gradient.

This work has consisted in mapping extensively SP anomalies across the entire Ubinas composite cone about 10 km in diameter. In addition, 15 audio-magnetotelluric (AMT) soundings provided a resistivity cross-section of the western flank of the edifice. A detailed SP and soil subsurface temperature mapping was conducted on the floor of the summit caldera, covering an area of about 1 km in diameter. Finally, water springs, well known by the local inhabitants, where located and measured in temperature, in a radius of 20 km all around the volcanic edifice. Here we present results from each individual survey before drawing conclusions from their integrated interpretation. We then compare our findings with nearby Peruvian volcanoes (El Misti and Tiscani) where similar measurements have been conducted (Finizola et al., 2004, Byrdina et al., 2013). Finally, we discuss the implications of this study on hazard assessment at Ubinas volcano.