Corinne A. Locke

Magma movements in Etna volcano associated with the major 1991–1993 lava eruption: evidence from gravity and deformation

The 1991–1993 eruption was probably the largest on Mt. Etna for 300 years. Since then the volcano has entered an unusually quiescent period. A comprehensive record of gravity and ground deformation changes presented here bracket this eruption and give valuable insight into magma movements before, during and after the eruption. The gravity and deformation changes observed before the eruption (1990–1991) record the intrusion of magma into the summit feeder and the SSE-trending fracture system which had recently been active in 1978, 1979, 1983 and 1989, creating the feeder dyke for the 1991–1993 eruption. In the summit region gravity changes between 1992 and 1993 (spanning the end of the eruption) reflect the withdrawal of magma from the conduit followed more recently (1993–1994) by the re-filling of magma in the conduit up to pre-eruption levels. In contrast, in the vicinity of the fracture zone, gravity has remained at the 1991–1992 level, indicating that no withdrawal has occurred here. Rather, magma has solidified in the fracture system and sealed it such that the 1993–1994 increase in magma level in the conduit was not accompanied by further intrusion into the flanks. Mass calculations suggest that a volume of at least 107 m3 of magma has solidified within the southeastern flank of the volcano.

Geophysical imaging of the Quaternary Wairoa North Fault, New Zealand: a case study

Abstract An integrated geophysical investigation, including gravity, vertical electric sounding (VES), 2D resistivity and seismic reflection/refraction techniques, was carried out along two profiles across the extensional Wairoa North Fault, New Zealand, prior to a trenching and paleoseismic study. The geophysical data clearly imaged the fault plane lying directly beneath a suspected fault scarp on one profile. The location of the intersection of the fault plane with the surface was deduced from 2.5D gravity and 2D resistivity modeling and subsequently corroborated by trenching to within 5 m. A 60° dip (±10–20°) for the fault plane along both profiles, similar to that revealed in the trench, was modeled to about 70 m depth at both locations, indicating a minimum vertical displacement. Seismic methods, together with a trial ground penetrating radar (GPR) profile, were less successful in imaging deformation above the fault plane, partly because of a lack of coherent reflectors within the upper Quaternary sequence, but mainly because of the deep weathering profiles on both sides of the fault. Despite this deeply weathered environment, geophysical methods (especially 2D resistivity) have proved very effective in confirming the significance of a suspected fault scarp, facilitating the accurate siting of a trench and providing valuable deeper information about fault morphology.

Geophysical evidence for temporal and structural relationships within the monogenetic basalt volcanoes of the Auckland volcanic field, northern New Zealand

Abstract Detailed gravity, aeromagnetic and geochemical studies have been used to investigate late Pleistocene monogenetic basaltic volcanoes in the Auckland volcanic field. The styles of eruption range from purely phreatomagmatic to purely magmatic, with some volcanoes displaying a range of eruptive characteristics. Appropriately scaled three-dimensional geophysical modelling methods have successfully delineated the subsurface structure of these small volcanoes and have been used to interpret the controls on their eruptive styles. This modelling has shown that a variety of structural styles exists in close proximity in the Auckland volcanoes, ranging from shallow effusive volcanoes to deeper-seated explosive volcanoes. The main control on the style developed appears to be the nature of the near-surface geology at the site of the eruption. Evidence for contemporaneous eruptions from multiple centres within this monogenetic field has been provided by both the geophysical and geochemical data. This information bears on the behaviour of monogenetic fields in general and in particular has important implications for modelling the past behaviour of the Auckland field and for the assessment of future activity which will form the basis of any contingency planning for volcanic hazard in the region.

Post-eruptive gravity changes from 1990 to 1996 at Krafla volcano, Iceland

The 1975–1984 Krafla rifting episode was a major lava- and dyke-producing event associated with the release of extensional strain accumulated over more than 200 years at the divergent plate boundary in North Iceland. The present work provides a unique example of gravity decreases and increases sustained over a long period following a major eruptive episode at a rift volcano. After height correction, persistent net gravity decreases over the source of observed Mogi-type deflation occur with gravity increases occurring further away from this centre of deformation. Gravity decreases are interpreted in terms of drainage from a shallow magma chamber. The net gravity decreases require that at least 4×1010 kg of magma must have been drained during the 6-year observation period. Assuming a density of 2700 kg m−3, this magma would occupy 1.5×107 m3 and by analogy with results obtained for Kilauea, this implies a magma chamber volume change of 4.1×106 m3. This is consistent with the chamber volume change deduced from ground deformation data assuming a Poissons ratio of 0.25 and a Mogi source. Net gravity increases are more spatially extensive and are most likely caused by migration of the steam–water interface and/or closure of micro-fractures in lavas above the magma chamber during post-eruptive cooling and contraction. We present a model for the Krafla magma chamber in which a cooling, contracting and draining magma body causes subsidence at the surface. These results contrast with observations from the Askja caldera, Iceland, where post-eruptive deflation has been shown to be accompanied by negligible net gravity changes above the Mogi-type source in the caldera. Long-term post-eruptive deflation and magma drainage have not been observed at subduction-related volcanoes; this may be a function of a difference in magma viscosity.

Magma transfer processes at persistently active volcanoes: insights from gravity observations

Magma transfer processes at persistently active volcanoes are distinguished by the large magma flux required to sustain the prodigious quantities of heat and gas emitted at the surface. Although the resulting degassed magma has been conjectured to accumulate either deep within the volcanic edifice or in the upper levels of the sub-edifice system, no direct evidence for such active accumulation has been reported. Temporal gravity data are unique in being able to quantify mass changes and have been successfully used to model shallow magma movements on different temporal scales, but have not generally been applied to the investigation of postulated long-term accumulation of magma at greater spatial scales within volcanic systems. Here, we model the critical data acquisition parameters required to detect mass flux at volcanoes, we review existing data from a number of volcanoes that exemplify the measurement of shallow mass changes and present new data from Poas and Telica volcanoes. We show that if a substantial proportion of degassed magma lodges within the sub-edifice region, it would result in measurable annual to decadal gravity increases occurring over spatial scales of tens of kilometres and propose that existing microgravity data from Sakurajima and, possibly, Etna volcanoes could be interpreted in these terms. Furthermore, such repeat microgravity data could be used to determine whether the accumulation rate is in equilibrium with the rate of production of degassed magma as calculated from the surface gas flux and hence identify the build-up of gas-rich magma at depth that may be significant in terms of eruption potential. We also argue that large magma bodies, both molten and frozen, modelled beneath volcanoes from seismic and gravity data, could represent endogenous or cryptic intrusions of degassed magma based on order of magnitude calculations using present-day emission rates and typical volcano lifetimes.

The Auckland volcanic field, New Zealand: geophysical evidence for its eruption history

Abstract The Late Quaternary monogenetic basalt volcanoes of the Auckland volcanic field exhibit styles of eruption ranging from phreatomagmatic to magmatic. New detailed aeromagnetic and other geophysical data from the southern half of the field provide constraints on the style and relative timing of eruptions. Concealed basalt bodies are shown to be common beneath maars indicating deep excavation by phreatomagmatic events and subsequent filling by magma. Depth of excavation is unrelated to the presence of surficial and potentially saturated Plio-Pleistocene sediments but commonly involves magma-water interaction in widespread aquifers within the underlying Miocene sediments. Coincident anomalous bulk magnetization directions show that at least three volcanoes were active contemporaneously and suggest a very short duration of activity which is probably typical of other centres in the field. These results emphasize the spasmodic nature of activity in the field over the last 140 ka culminating in the most recent centre, Rangitoto, which erupted after a long quiescent period and represents a large spasm of activity, confined to a single centre. The locations of contemporaneous centres are tentatively correlated with the regional NNW-SSE structural trend.

Three-dimensional structure of relict stratovolcanoes in Taranaki, New Zealand: evidence from gravity data

Abstract The andesite stratovolcanoes of Taranaki occur in a spatially distinct age progression, being eroded to different levels and situated within a large well-known sedimentary basin, thus providing particularly favourable circumstances for geophysical investigation. New gravity data presented here show that the Kaitake and Pouakai volcanoes, two of the older relict centres, are associated with large positive residual gravity anomalies (240 g.u. and 160 g.u., respectively). Detailed three-dimensional gravity models define confined high-density dyke/stock complexes below the volcanoes which constrain the likely locus of eruptive activity and extend to at least basement depths (about 5 km). The older Kaitake edifice is modelled as solid andesite whereas the younger Pouakai edifice comprises a solid andesite core mantled by lower-density volcaniclastics. These andesite bodies are interpreted as regions of extensive dyke injection, a process clearly important in cone development and possibly also in cone stability.

Te Pouhawaiki Volcano and pre-volcanic topography in Central Auckland: Volcanological and hydrogeological implications

Abstract Te Pouhawaiki Volcano in the Auckland Volcanic Field was identified on the basis of a small scoria cone, but whether this cone marked the location of a significant eruption centre has been unknown. Volcanic stratigraphy in the central Auckland isthmus is complex, with older deposits (possibly entire volcanic centres) obscured by younger deposits. The distribution of lava flows in the central Auckland isthmus was strongly influenced by the pre‐volcanic topography, and is a major control on present‐day groundwater flow regimes. Detailed gravity data from the central Auckland isthmus are used here to model the thicknesses of volcanic deposits and hence determine the pre‐volcanic topography. The site of the former Te Pouhawaiki scoria cone is shown to correlate with a distinct positive gravity anomaly (c. 6 μN.kg‐1) interpreted in terms of a lava‐filled depression in the Waitemata surface, surrounded by a tuff ring. This inferred explosive eruption centre is similar in both size and eruption style to a number of others in the Auckland Volcanic Field and suggests that the Te Pouhawaiki scoria cone may have been the surface manifestation of a substantial eruption centre which also produced phreatomagmatic deposits and lavas. The gravity model also defines the location and geometry of the paleotopographic divide between the ancestral Waitemata and Manukau River systems, showing it to be a complex ridge system. These buried ridges peak at c. 10–20 m depth (60–70 m a.s.l.) with a saddle in an eastern limb of the ridge which may have allowed lava from One Tree Hill Volcano to flow north of this divide. The configuration of the pre‐volcanic Waitemata surface indicates that the present‐day groundwater flow regime is likely to be complex and divergent away from the ridge system, controlled in some areas by narrow paleovalleys. Within the ridge complex, an area in which groundwater flow is likely to be convergent has been defined which correlates with the location of occasional surface flooding.

Magma plumbing processes for persistent activity at Poás volcano, Costa Rica

New microgravity data from the active crater of Poas volcano, Costa Rica, collected in 2002-2004 extends the existing dataset to provide a unique 20-year time series. These data show that gravity has decreased monotonically in the north and east of the crater over the last 5 years, whilst it has increased to the west and remained approximately constant in the south. These changes are interpreted in terms of convective recharge within dendritic intrusions beneath the crater, with overall down-welling in the north and up-welling in the west. The data reveal a 5-10 year periodicity in sub-crater mass movement, but overall, the upper part of the conduit system appears to have maintained a state of mass equilibrium.

Egmont Volcano, New Zealand: three-dimensional structure and its implications for evolution

Abstract Egmont Volcano (Mt. Taranaki), a large active andesite stratovolcano, is characterised by a well-defined large positive residual gravity anomaly of 350 g.u. Detailed three-dimensional modelling of extensive gravity data delineates a large subedifice intrusion of andesite density extending to at least basement depths (6 km) and a core of similar density within the edifice. These andesite bodies are attributed mainly to repeated magma injection from deeper magma chambers into both the underlying sediments and the volcanic edifice. The dense edifice core of Egmont Volcano probably represents significant dyke intrusion; such dykes may have played a major role not only in edifice construction but also in edifice collapse. Egmont Volcano appears to share a common evolution with the three older relict centres in the Taranaki succession since all four Taranaki volcanoes are shown to have large subedifice intrusions and similar dense edifice cores. However, intrusive volumes are somewhat smaller below the older centres which, given their ages, suggests that magma production rates may have increased with time; the total volume of magma involved in the formation of the Taranaki volcanoes is estimated to be at least 1500 km 3 .