Shallow Magma Plumbing System Research Articles

Inflating Shallow Plumbing System of Bezymianny Volcano, Kamchatka, Studied by InSAR and Seismicity Data Prior to the 20 December 2017 Eruption

Explosive eruptions at steep-sided volcanoes may develop with complex precursor activity occurring in a poorly-understood magma plumbing system so that timelines and possible interactions with the geologic surrounding are often unresolved. Here we investigate the episode prior to the energetic December 20, 2017 eruption at Bezymianny volcano, Kamchatka. We compare degassing activity inferred from time-lapse camera images, seismicity and real-time seismic amplitude (RSAM) data derived from a temporary station network, as well as high-resolution InSAR displacement maps. Results show that the first changes can be identified in low-frequency seismicity and degassing at least 90 days before the eruption, while the first volcano-tectonic (VT) seismicity occurred 50 days before the eruption. Coinciding with significant changes of the RSAM, surface displacements affect the volcanic flanks at least 9 days prior to the eruption. Inversion modeling of the pre-eruptive surface deformation as well as deflation-type, co-eruptive surface changes indicate the presence of a shallow and transient reservoir. We develop a conceptual model for Bezymianny volcano initiating with deep seismicity, followed by shallow events, rockfalls, steaming and an inflating reservoir. The eruption is then associated with subsidence, caused by deflation of the same reservoir. This sequence and conceivable causality of these observations are providing a valuable contribution to our understanding of the shallow magma plumbing system beneath Bezymianny and may have relevance for volcano monitoring and early warning strategies at similar volcanoes elsewhere.

Volcano morphology as an indicator of stress orientation in the Java Volcanic Arc, Indonesia

Stress fields in the Earth's crust are important indicators leading to better understanding of geologic processes. One way to acquire stress data is through morphometric analysis of volcanoes. Here we evaluate 141 Quaternary-age monogenetic and polygenetic volcanoes and volcanic fields in Java, Indonesia, using a 30-m SRTM DEM, an 8-m DEMNAS, aerial photos, published geological maps, and Landsat and Google Earth satellite imagery. We analyze parameters that imply the geometry of the shallow magma plumbing system and ascent pathways: crater ellipticity and elongation axis, and the orientation of crater-breaching bisectors. The dominant trend of all of these parameters should indicate the direction of the maximum horizontal compressive stress (σHmax) while the minimum horizontal stress (σHmin) should be perpendicular to it. The goal here is to estimate the principal stress axes at the time of dike emplacement, to recognize any spatial variations in stress orientations, and assess factors controlling the emplacement. The resulting orientation of σHmax in the western part of Java is NE-SW, and the trend gradually rotates to NW-SE moving eastward. Abrupt changes in stress orientations and anomalies are also observed in the vicinity of prominent geologic structures, highlighting the influence of preexisting structures on dike emplacement and eruption characteristics. Results of the morphometric analysis of volcanoes and volcanic fields conducted in this study help to estimate the past and present stress orientations in Java, and provide valuable understanding regarding the relationship between volcanic emplacement and crustal structures, complementing other types of analysis.

Cascading Partial Rupture of the Flores Thrust during the 2018 Lombok Earthquake Sequence, Indonesia

AbstractA series of four Mw>6 earthquakes struck the northern region of Lombok, eastern Indonesia, in a span of three weeks from late July to mid-August 2018. The series was thought to be associated with the Flores thrust, but the exact mechanism causing the unusual earthquake series has remained elusive. Our Interferometric Synthetic Aperture Radar analysis, combined with insights from seismology, indicates that the events originated at different hypocenter depths with differing fault geometries, which may explain the cascading behavior of the events, and indicates that better imaging of active fault geometry might provide some insight into future rupture behavior on other similar thrust systems. Our static stress change calculations suggest that the earlier events in the sequence played a role in promoting the later events. In addition, the second event brought the most significant impact on a nearby volcano, by causing volumetric expansion at its shallow magma plumbing system and unclamping its magma ascent zone, which may potentially have an impact on its future eruptive activity. However, no volcanic activity has so far occurred after the earthquakes. Finally, our damage proxy maps suggest that the second event caused the greatest damage to buildings.

Magma Oscillations in a Conduit‐Reservoir System, Application to Very Long Period (VLP) Seismicity at Basaltic Volcanoes: 2. Data Inversion and Interpretation at Kīlauea Volcano

AbstractVery long period (VLP) seismic events (with dominant periods of 15 to 40 s), observed from 2007 to 2018 at the summit of Kīlauea Volcano, Hawai‘i, arise from resonant oscillations in the shallow magma plumbing system. Utilizing an oscillation model developed in the companion paper (Liang et al., 2020), we perform Bayesian inversions on seismic data from four representative VLP events separately for the parameters of the shallow conduit‐reservoir system, exploring both sphere and crack reservoir geometries. Both sphere and crack geometries are preferentially located ∼1–2 km beneath the northeast edge of Halema‘uma‘u crater and produce similar fits to the data. Considering a reasonable range for reservoir storativity, magma density, and density contrast between the top and bottom of the conduit, we favor a spherical reservoir with a radius of 0.8 to 1.2 km and a short conduit of less than a few hundred meters. For this geometry, buoyancy from density stratification in the conduit provides the dominant restoring force for the VLP oscillation. Viscosity is constrained within an order of magnitude for each event (e.g., approximately 2 to 23 Pa s for one event versus 27 to 513 Pa s for another). Changes in VLP period T and quality factor Q can be explained by changes in viscosity, density stratification, and/or conduit/reservoir geometry. In particular, observed fluctuations in Q over short time intervals (e.g., hours) with minimal changes in T apparently require rapid changes of magma viscosity by over an order of magnitude, assuming geometry remains unchanged, possibly reflecting changes in volatile content, bubble concentration, or conduit flow regime.

Timescales of magma ascent and degassing and the role of crustal assimilation at Merapi volcano (2006–2010), Indonesia: Constraints from uranium-series and radiogenic isotopic compositions

We present new 238U-230Th-226Ra-210Pb-210Po, 87Sr/86Sr and 143Nd/144Nd isotopic data of whole-rock samples and plagioclase separates from volcanic deposits of the 2006 and 2010 eruptions at Merapi volcano, Java, Indonesia. These data are combined with available eruption monitoring, petrographic, mineralogical and Pb isotopic data to assess current theories on the cause of a recent transition from effusive dome-building (2006) to explosive (2010) activity at the volcano, as well as to further investigate the petrogenetic components involved in magma genesis and evolution. Despite the significant difference in eruption style, the 2006 and 2010 volcanic rocks show no significant difference in (238U/232Th), (230Th/232Th) and (226Ra/230Th) activity ratios, with all samples displaying U and Ra excesses. The 226Ra and 210Pb excesses observed in plagioclase separates from the 2006 and 2010 eruptions indicate that a proportion of the plagioclase grew within the decades preceding eruption. The 2006 and 2010 samples were depleted in 210Po relative to 210Pb ((210Po/210Pb)i < 1) at the time of eruption but were variably degassed (69%–100%), with the degree of 210Pb degassing strongly related to sample texture and eruption phase. In good agreement with several activity monitoring parameters, 210Po ingrowth calculations suggest that initial intrusion into the shallow magma plumbing system occurred several weeks to a few months prior to the initial 2010 eruption. The 2006 and 2010 samples show a wide range in (210Pb/226Ra) activity ratio within a single eruption at Merapi and are largely characterised by 210Pb deficits ((210Pb/226Ra) < 1). Assuming a model of complete radon degassing, the 210Pb deficits in the 2006 volcanic rocks indicate relatively longer degassing timescales of ∼2–4 years than those given by the 2010 samples of ∼0–3 years. The uranium-series and radiogenic isotopic data do not support greater crustal assimilation of carbonate material as the explanation for the more explosive behaviour of Merapi in 2010 (as has been previously suggested) and instead indicate that relatively rapid ascent of a more undegassed magma was the primary difference responsible for the transition in explosive behaviour. This interpretation is in good agreement with gas monitoring data, previous petrological studies (mineral, microlite and melt inclusion work) and maximum calculated timescale estimates using Fe-Mg compositional gradients in clinopyroxene, that also suggest more rapid movement of relatively undegassed magma in 2010 relative to 2006.

Seismo-acoustic wavefield of strombolian explosions at Yasur volcano, Vanuatu, using a broadband seismo-acoustic network, infrasound arrays, and infrasonic sensors on tethered balloons

Seismo-acoustic wavefields at volcanoes contain rich information on shallow magma transport and subaerial eruption processes and inform our understanding of how volcanoes work. Acoustic wavefields from eruptions are predicted to be directional, but sampling this wavefield directivity is challenging because infrasound sensors are usually deployed on the ground surface. We attempt to overcome this observational limitation using a novel deployment of infrasound sensors on tethered balloons in tandem with a suite of dense ground-based seismo-acoustic, geochemical, and eruption imaging instrumentation. We conducted a collaborative multiparametric field experiment at the active Yasur volcano, Tanna Island, Vanuatu, from 26 July to 2 August 2016. Our observations include data from a temporary network of 11 broadband seismometers, 6 single infrasonic microphones, 7 small-aperture 3-element infrasound arrays, 2 infrasound sensor packages on tethered balloons, an FTIR, a FLIR, 2 scanning Flyspecs, and various visual imaging data; scoria and ash samples were collected for petrological analyses. This unprecedented dataset should provide a unique window into processes operating in the shallow magma plumbing system and their relation to subaerial eruption dynamics.

Transient deformation in the Asal‐Ghoubbet Rift (Djibouti) since the 1978 diking event: Is deformation controlled by magma supply rates?

AbstractThe Asal‐Ghoubbet Rift (AG Rift) in Djibouti lies in the subaerial continuation of the Aden ridge system, thereby constituting a unique location to study rifting processes and mechanisms involved in continental breakup and oceanic spreading. Continually upgraded and expanded geodetic technology has been used to record the 1978 Asal rifting event and postdiking deformation. In light of recent results obtained for the Manda Hararo‐Dabbahu rifting event (2005–2010), we propose that the horizontal and vertical geodetic data can be modeled with a double source, involving a dike‐like inflation component aligned along the rift axis and a spherical pressure source located at midsegment below the Fieale caldera. By revisiting the codiking data, we propose that the reservoir below Fieale could have fed, at least partially, the 1978 injection and the contemporaneous Ardoukôba eruption and potentially induced local subsidence due to magma draining out of the central reservoir. As an alternative to previously proposed viscoelastic relaxation models, we reinterpret postdiking observations using a purely elastic rheology. We determine the relative contribution of a midsegment reservoir inflation and a dike‐like opening component, together with their respective time evolutions. Our results suggest that interactions between steadily accumulating tectonic strain and temporal variations in melt supply to the shallow magma plumbing system below the AG Rift may entirely explain the geodetic observations and that viscoelastic deformation processes played a minor role in the 30 years following the 1978 rifting event.

Transitions between explosive and effusive phases during the cataclysmic 2010 eruption of Merapi volcano, Java, Indonesia.

Transitions between explosive and effusive activity are commonly observed during dome-forming eruptions and may be linked to factors such as magma influx, ascent rate and degassing. However, the interplay between these factors is complex and the resulting eruptive behaviour often unpredictable. This paper focuses on the driving forces behind the explosive and effusive activity during the well-documented 2010 eruption of Merapi, the volcano’s largest eruption since 1872. Time-controlled samples were collected from the 2010 deposits, linked to eruption stage and style of activity. These include scoria and pumice from the initial explosions, dense and scoriaceous dome samples formed via effusive activity, as well as scoria and pumice samples deposited during subplinian column collapse. Quantitative textural analysis of groundmass feldspar microlites, including measurements of areal number density, mean microlite size, crystal aspect ratio, groundmass crystallinity and crystal size distribution analysis, reveal that shallow pre- and syn-eruptive magmatic processes acted to govern the changing behaviour during the eruption. High-An (up to ∼80 mol% An) microlites from early erupted samples reveal that the eruption was likely preceded by an influx of hotter or more mafic magma. Transitions between explosive and effusive activity in 2010 were driven primarily by the dynamics of magma ascent in the conduit, with degassing and crystallisation acting via feedback mechanisms, resulting in cycles of effusive and explosive activity. Explosivity during the 2010 eruption was enhanced by the presence of a ‘plug’ of cooled magma within the shallow magma plumbing system, which acted to hinder degassing, leading to overpressure prior to initial explosive activity.Electronic supplementary materialThe online version of this article (doi:10.1007/s00445-016-1046-z) contains supplementary material, which is available to authorized users.

Degassing during quiescence as a trigger of magma ascent and volcanic eruptions.

Understanding the mechanisms that control the start-up of volcanic unrest is crucial to improve the forecasting of eruptions at active volcanoes. Among the most active volcanoes in the world are the so-called persistently degassing ones (e.g., Etna, Italy; Merapi, Indonesia), which emit massive amounts of gas during quiescence (several kilotonnes per day) and erupt every few months or years. The hyperactivity of these volcanoes results from frequent pressurizations of the shallow magma plumbing system, which in most cases are thought to occur by the ascent of magma from deep to shallow reservoirs. However, the driving force that causes magma ascent from depth remains unknown. Here we demonstrate that magma ascent can be triggered by the passive release of gas during quiescence, which induces the opening of pathways connecting deep and shallow magma reservoirs. This top-down mechanism for volcanic eruptions contrasts with the more common bottom-up mechanisms in which magma ascent is only driven by processes occurring at depth. A cause-effect relationship between passive degassing and magma ascent can explain the fact that repose times are typically much longer than unrest times preceding eruptions, and may account for the so frequent unrest episodes of persistently degassing volcanoes.

Magma storage, differentiation, and interaction at Lake City caldera, Colorado, USA

Rocks from the 23 Ma Lake City caldera show diverse chemical affinities attesting to a complex magmatic system beneath the caldera. Field and geochemical data from ignimbrites and intrusions constrain magma storage and magma interactions during the formation of the caldera. Two geochemically distinct magma batches erupted during caldera formation: batch A, consisting of rhyolites and trachytes, and batch B, consisting of dacites and trachyandesites. The ignimbrites of the Lower, Middle, and Upper Sunshine Peak Tuff represent the bulk of erupted batch A magma, with an increasing proportion of trachyte to rhyolite as the eruption progressed. Overall, the observed trends of major and trace elements are consistent with the sequential eruption of a magmatic system with a rhyolitic upper portion and trachytic lower portion. The Middle Sunshine Peak Tuff contains two distinct types of pumice clast, while the Upper Sunshine Peak Tuff contains four distinct pumice clast types, with one type chemically related to batch B magma. The link between the rhyolite and trachyte of batch A is supported by major- and trace-element geochemical modeling of an initially trachytic magma that fractionated and was subjected to crystal/melt segregation following 50%–60% crystallization. Compositional gaps and chemical heterogeneity in the bulk ignimbrite composition show that the proportions of these different magma types varied significantly during eruption. We propose that the fractionating batch A and B magmas formed distinct magma pods, some containing residual magma mush, that were tapped during different phases of caldera formation. After collapse, dacite lavas of batch B were erupted concurrent with resurgent uplift from shallow intrusion of both residual mingled batch A and batch B magma. In summary, our observations suggest (1) a complex magma chamber geometry from two fractionating magma batches, and (2) magma replenishment and accelerated periods of magma reorganization in the shallow magma plumbing system during a single caldera cycle at Lake City.

P37 地殻変動・火山ガス観測に基づく浅間山地下のマグマ収支見積(ポスターセッション)

P37 地殻変動・火山ガス観測に基づく浅間山地下のマグマ収支見積(ポスターセッション)

Budget of shallow magma plumbing system at Asama Volcano, Japan, revealed by ground deformation and volcanic gas studies

AbstractMultiple cycles of the intensive volcanic gas discharge and ground deformation (inflation and deflation) were observed at Asama Volcano, Japan, from 2000 to 2011. Magma budget of the shallow magma plumbing system was estimated on the basis of the volcanic gas emission rates and ground deformation data. Recent inflations observed in 2004 and 2008 were modeled as a dike intrusion to 2–3 km west of Asama Volcano. Previous studies proposed that magma ascends from a midcrustal magma reservoir to the dike and reaches the surface via a sinuous conduit which connects the dike to the summit. The intensive volcanic sulfur dioxide discharge of up to 4600 t/d at the volcano was modeled by magma convective degassing through this magma pathway. The volcano deflates as shrinkage of the magma in a reservoir by volcanic gas discharge. We estimated the volume change of the dike modeled based on the GPS observations, the volume decrease of the magma by the volcanic gas discharge, and the amount of degassed magma produced to calculate the magma budget. The results show that the volume decrease of the magma by the volcanic gas discharge was larger than the volume change of the dike during the inflation periods. This indicates that a significant volume of magma at least more than 2 times larger than the volume change of the dike was supplied from the midcrustal magma reservoir to the dike. The volume decrease of the dike was comparable with the volume decrease of the magma by the volcanic gas discharge during the deflation periods. The long‐term deflation trend of the dike and the volume of degassed magma (108–9 m3) suggest that the degassed magma produced is not stored in the dike and the magma is mainly supplied from the midcrustal magma reservoir. In both periods, the volume of degassed magma produced was 1 order of magnitude larger than the volume change of the dike. This indicates that the actual volume of the magma supplied from the midcrustal magma reservoir is up to 1 order of magnitude larger than the volume change of the dike. These results strongly suggest that an amount of magma moved through a magma reservoir is possible to be significantly larger than volume change of the magma reservoir estimated by the geodetic observations.

Geochemical evolution of a shallow magma plumbing system during the last 500 years, Miyakejima volcano, Japan: Constraints from 238U– 230Th– 226Ra systematics

In order to unravel magma processes and the geochemical evolution of shallow plumbing systems beneath active volcanoes, we investigated U-series disequilibria of rocks erupted over the past 500 years (1469–2000 AD) from Miyakejima volcano, Izu arc, Japan. Miyakejima volcanic rocks show 238U– 230Th– 226Ra disequilibria with excess 238U and 226Ra, due to the addition of slab-derived fluids to the mantle wedge. Basaltic bombs of the 2000 AD eruption have the lowest ( 230Th/ 232Th) ratio compared to older Miyakejima eruptives, yielding the youngest 238U– 230Th model age of 2 kyr. This reinforces our previous model that fluid release from the slab and subsequent magma generation in the mantle wedge beneath Miyakejima occur episodically on a several-kyr timescale. In the last 500 years, Miyakejima eruptives show: (1) a vertical trend in a ( 230Th/ 232Th)–( 238U/ 232Th) diagram and (2) a positive linear correlation in a ( 226Ra/ 230Th) 0 − 1/ 230Th diagram, which is also observed in lavas from some of the single eruptions (e.g., 1940, 1962, and 1983 AD). The variations cannot be produced by simple fractional crystallization in a magma chamber with radioactive decay of 230Th and 226Ra, but it is possibly produced by synchronous generation of melts in the mantle wedge with different upwelling rate or addition of multiple slab-derived fluids. A much more favorable scenario is that some basaltic magmas were intermittently supplied from deep in the mantle and injected into the crust, subsequently modifying the original magma composition and producing variations in ( 230Th/ 232Th) and ( 226Ra/ 230Th) 0 ratios via assimilation and fractional crystallization (AFC). The assimilant of the AFC process would be a volcanic edifice of previous Miyakejima magmatism. Due to the relatively short timescales involved, the interaction between the assimilant and recent Miyakejima magmatism has not been recorded by the Sr–Nd–Pb isotopic systems. In such cases, Th isotopes and ( 226Ra/ 230Th) ratio are excellent geochemical tracers of magmatic evolution.

Along‐strike variation of the sheeted dike complex in the Oman Ophiolite: Insights into subaxial ridge segment structures and the magma plumbing system

Along‐strike variations of the 60‐km‐long sheeted dike complex in the northern Oman Ophiolite were studied in order to understand the shallow magma plumbing system beneath the fossil fast spreading ridge. The presence of numerous dikes intruding into the layered gabbro defines the northern end of the paleoridge segment at Wadi Fizh. The postulated segment center is located at Wadi Thuqbah, which has a thicker Mono transition zone than elsewhere along the fossil segment. Aphyric dikes predominate in the sheeted dikes, of which 99% are simple, while multiple and composite dikes are few. The thickness of 1511 dikes ranges from &lt;1 cm to &gt;13 m, with an average thickness of 71.3 cm. Restored dike trends in the 30‐km‐long northern half of the dike complex display the NS trending north domain and NNE‐NS trending south domain bordered at the south of Wadi Bani Umar al Gharbi. In the domain boundary, dikes gradually change strikes or are mutually intrusive. Dike thickens northward with the largest peak along Wadi Fizh at the northern end of the paleoridge segment and a small peak at Wadi Bani Umar al Gharbi. Most dikes have a bulk Mg# of 55–66, which overlaps the majority of MORB. Less common, highly evolved dikes with Mg# 34–40 characterize the north domain. Thicker dikes (&gt;3 m) tend to have high Mg#, while thinner dikes (&lt;2 m) are highly variable in Mg#. The regional variations of the dike trends and the whole rock compositions can be explained by the secular variation in the structures of the paleoridge segment comparable to a third‐order segment of the present mid‐ocean ridge system spanning a few tens of thousands of years. Initially, the segment‐long melt lens developed along the paleoridge axis, which fed long and thick dikes with high Mg#. With decreasing supply of magma, the melt lens split up into the north and south smaller lenses bordered by a DEVAL that fed thinner dikes intruding into the former thick dikes. Cut‐off of the northern melt lens from the magma source changed the melt composition to highly evolved, low‐Mg# magmas, which subsequently intruded as short evolved dikes. Meanwhile, the main melt lens in the segment center continued feeding the high‐Mg# dikes which were maintained by the larger melt lens size and intermittent magma supply from deep magma chambers.

Thermal and mechanical aspects of magma emplacement in giant dike swarms

We consider the thermal history and dynamics of magma emplacement in giant feeder dikes associated with continental flood basalts. For driving pressure gradients inferred for giant dike swarms, thicknesses of &lt;10 m would enable dikes to transport magma laterally over the distances observed in the field (up to thousands of kilometers) without suffering thermal lock‐up. Using time‐dependent numerical solutions for the thermal evolution of a dike channel under laminar and turbulent flow conditions in the presence of phase transitions, we investigate the possibility that the observed dike thicknesses (of the order of 100 m) result from thermal erosion of the country rocks during dike emplacement. This implies that the observed range of dike widths in giant dike swarms may reflect variations in the source volume and not the excess magma pressure. It is found that the total volume of intruded magma required to produce an order of magnitude increase in dike width via wall rock melting broadly agrees with the estimated volumes of individual flows in continental flood basalts. The presence of chilled margins and apparently low crustal contamination characteristics of some giant dikes may be consistent with turbulent magma flow and extensive melt back during dike emplacement. In this case, measurements of the anisotropy of magnetic susceptibility most likely indicate magma flow directions during the final stages of dike intrusion. Shear stresses generated at the dike wall when the dike starts to freeze strongly decrease with increasing dike width, which implies that thicker dikes may have less tendency to produce consistent fabric alignment. Our results suggest that if the dike was propagating downslope off a plume‐related topographic swell, the mechanism responsible for flow termination could possibly have been related to underpressurization and collapse (implosion) of the shallow magma plumbing system feeding the intrusion. Radial dikes that erupted at the periphery of the topographic uplift might have increased (rather than decreased) extensional stresses in the crust within the topographic uplift upon their solidification.

Gondwana dispersion and Asian accretion: IGCP 321 final results volume : I. Metcalfe, ed I. Metcalfe, (A.A. Balkema), Gondwana dispersion and Asian accretion. IGCP 321 final results volume, ISBN 9054104465, 1999, (361p)

We present new 238U-230Th-226Ra-210Pb-210Po, 87Sr/86Sr and 143Nd/144Nd isotopic data of whole-rock samples and plagioclase separates from volcanic deposits of the 2006 and 2010 eruptions at Merapi volcano, Java, Indonesia. These data are combined with available eruption monitoring, petrographic, mineralogical and Pb isotopic data to assess current theories on the cause of a recent transition from effusive dome-building (2006) to explosive (2010) activity at the volcano, as well as to further investigate the petrogenetic components involved in magma genesis and evolution. Despite the significant difference in eruption style, the 2006 and 2010 volcanic rocks show no significant difference in (238U/232Th), (230Th/232Th) and (226Ra/230Th) activity ratios, with all samples displaying U and Ra excesses. The 226Ra and 210Pb excesses observed in plagioclase separates from the 2006 and 2010 eruptions indicate that a proportion of the plagioclase grew within the decades preceding eruption. The 2006 and 2010 samples were depleted in 210Po relative to 210Pb ((210Po/210Pb)i < 1) at the time of eruption but were variably degassed (69%–100%), with the degree of 210Pb degassing strongly related to sample texture and eruption phase. In good agreement with several activity monitoring parameters, 210Po ingrowth calculations suggest that initial intrusion into the shallow magma plumbing system occurred several weeks to a few months prior to the initial 2010 eruption. The 2006 and 2010 samples show a wide range in (210Pb/226Ra) activity ratio within a single eruption at Merapi and are largely characterised by 210Pb deficits ((210Pb/226Ra) < 1). Assuming a model of complete radon degassing, the 210Pb deficits in the 2006 volcanic rocks indicate relatively longer degassing timescales of ∼2–4 years than those given by the 2010 samples of ∼0–3 years. The uranium-series and radiogenic isotopic data do not support greater crustal assimilation of carbonate material as the explanation for the more explosive behaviour of Merapi in 2010 (as has been previously suggested) and instead indicate that relatively rapid ascent of a more undegassed magma was the primary difference responsible for the transition in explosive behaviour. This interpretation is in good agreement with gas monitoring data, previous petrological studies (mineral, microlite and melt inclusion work) and maximum calculated timescale estimates using Fe-Mg compositional gradients in clinopyroxene, that also suggest more rapid movement of relatively undegassed magma in 2010 relative to 2006.