Northeast Rift Zone Research Articles

Rift Zone Architecture and Inflation‐Driven Seismicity of Mauna Loa Volcano

AbstractThe 2022 eruption at Mauna Loa, Hawai'i, marked the first extrusive activity from the volcano after 38 years of quiescence. The eruption was preceded by several years of seismic unrest in the vicinity of the volcano's summit. Characterizing the structure and dynamics of seismogenic features within Mauna Loa during this pre‐eruptive interval may provide insights into how pre‐ and co‐eruptive processes manifest seismically at the volcano. In particular, the extent to which seismicity may be used to forecast the location and timing of future eruptions is unclear. To address these questions, we construct a catalog of relocated seismicity on Mauna Loa spanning 2011–2023. Our earthquake locations image complex, sub‐kilometer‐scale seismogenic structures in the caldera and southwest rift zone. We additionally identify a set of streaks of seismicity in the volcano's northwest flank that are radially oriented about the summit. Using a rate‐and‐state friction model for earthquake occurrences, we demonstrate that the seismicity rate in this region can be modeled as a function of the stressing history caused by magma accumulation beneath the summit. Finally, we observe a mid‐2019 step change in the seismicity rate in the Ka'oiki region that may have altered the stress state of the northeast rift zone in the three years before the eruption. Our observations provide a framework for interpreting future seismic unrest at Mauna Loa.

Characterizing and Locating Seismic Tremor during the 2022 Eruption of Mauna Loa Volcano, Hawai’i, with Network Covariance

Abstract The 2022 eruption of Mauna Loa Volcano, Hawai’i, was accompanied by continuous seismic tremor that began about 30 min before and ended several days after the eruption. We characterize the amplitude history and frequency content of the tremor, and we use a network covariance-based method to estimate its source location. The tremor exhibits multiple narrow spectral peaks between 1 and 3 Hz, and its amplitude varies through time in a spasmodic manner. Our location results track a northeast migration of shallow sources through the summit region in the first few hours of the eruption. For the remainder of the eruption, source locations cluster in the vicinity of the erupting vent in the northeast rift zone. We attribute the tremor source to gas dynamics in the upper reaches of a basaltic dike. However, limitations in instrumentation and knowledge of the subsurface velocity structure may preclude an assessment of the source mechanism. Our results illustrate the value of characterizing and locating tremor for tracking magma movement, and demonstrate a use for dense and calibrated seismic instrumentation on active volcanoes. The location method we use requires substantial parameter testing, reflecting the potential benefit of developing more flexible approaches toward real-time automated assessment of tremor at volcanoes.

Monitoring the Mauna Loa (Hawaii) eruption of November–December 2022 from space: Results from GOES-R, Sentinel-2 and Landsat-8/9 observations

Mauna Loa, one of the most actives volcanoes on Earth, is a shield volcano, located on the Island of Hawaii (USA). On 27 November 2022, after about 38 years of quiescence, a new eruptive activity took place at the Moku‘āweoweocaldera, continuing in the following days (i.e. until 10 December) from the fissure vents opening on the Northeast Rift Zone.In this work, we investigate the Mauna Loa November − December 2022 eruption from space, integrating the information from different satellite sensors. The analysis of short-wave infrared (SWIR) data, at 10 min temporal resolution, from the Advanced Baseline Imager (ABI), aboard the Geostationary Operational Environmental Satellites − R series (GOES-R), performed through the Normalised Hotspot Indices (NHI), indicates that the Mauna Loa eruption started on 27 November in between 23:10–23:20 LT (28 November at 09:10–09:20 UTC). The same analysis shows the increase of thermal activity and its progressive reduction from the early morning of 28 November, in agreement with the eruption migration from the summit caldera to the Northeast Rift Zone. By analysing the second phase of eruption through SWIR data from the Multispectral Instrument (MSI) and Operational Land Imager (OLI), respectively aboard Sentinel-2 and Landsat 8/9 satellites, we estimated a maximum lava flow length of 17 km. Moreover, we retrieved values of the volcanic radiative power (VRP) up to 65 GW, and a time-averaged discharge rate (TADR) of ∼1000 (±500) m3/s. These results show that SWIR observations, at different spatial and temporal resolution, may give an important contribution to the monitoring, mapping and characterisation of intense lava effusions.

Mauna Loa awakens

On 27 November 2022, Hawaii’s Mauna Loa volcano on the Big Island began erupting for the first time in 38 years. A day later, a flight over the eruption captured this photo. It clearly shows a line of fissures, at an elevation of 3 km, from which lava flowed across the volcano’s northeast rift zone. The eruption continued for about two weeks, with lava traveling some 20 km from the summit. The US Geological Survey observed an increase in the number of earthquakes beneath the volcano’s cauldron-shaped summit and to the northwest roughly two months before the eruption began. The island’s nearby Kīlauea volcano has been erupting since September 2021.PPT|High resolutionThe two volcanoes appear to be connected underground by a deep, widespread network of pathways, according to a recently published paper by researchers from Caltech. Although previous studies have hypothesized that such a network may exist, the large data set the Caltech team analyzed showed enough seismic activity to map out a detailed plumbing system that connects Mauna Loa to Kīlauea, 34 km away. Future monitoring of volcanic activity may benefit from considering both volcanoes as part of a larger, interconnected system. (J. D. Wilding et al., Science 379, 462, 2023, doi:10.1126/science.ade5755; image courtesy of the US Geological Survey.) Section:ChooseTop of page <<© 2023 American Institute of Physics.

The forgotten eruption: The basaltic scoria cone of Montaña Grande, Tenerife

We report on the Strombolian to Violent Strombolian eruption of Montaña Grande which occurred between 789 and 725 BCE in the Güimar Valley on the NE flank of Tenerife island. The eruption produced a ca. 180 m-high scoria cone, a thick fallout deposit mostly dispersed southwest and a vast lava flow field that extends east of the cone, towards the coast, for 3.3 × 2.2 km. The eruption occurred in an unusual geodynamic context, outside the North West Rift and North East Rift zones and out of Las Cañadas caldera which are the main geological structures of Tenerife, where the volcanic activity concentrated during the Holocene. The tephra and lava have a trachybasalt composition similar to products of the recent activity of Tenerife but characterized by a distinct trace element pattern (Ta depletion relative to Nb), that points to a distinct source for the magma feeding the eruption and an ascent history along the whole crust which is independent and different from the central feeding system of the Teide-Pico Viejo volcanic complex. The study of this eruption, which until now had been completely neglected, adds new significant data for the correct definition of volcanic risk in Tenerife.

Crustal versus source processes recorded in dykes from the Northeast volcanic rift zone of Tenerife, Canary Islands

The Miocene–Pliocene Northeast Rift Zone (NERZ) on Tenerife is a well exposed example of a feeder system to a major ocean island volcanic rift. We present elemental and O–Sr–Nd–Pb isotope data for dykes of the NERZ with the aim of unravelling the petrological evolution of the rift and ultimately defining the mantle source contributions. Fractional crystallisation is found to be the principal control on major and trace element variability in the dykes. Differing degrees of low temperature alteration and assimilation of hydrothermally altered island edifice and pre-island siliciclastic sediment elevated the δ18O and the 87Sr/86Sr ratio of many of the dykes, but had little to no discernible effect on Nd and Pb isotopes. Once the data are screened for alteration and shallow level contamination, the underlying source variations of the NERZ essentially reflect derivation from a young High-μ (HIMU, where μ=238U/204Pb)-type mantle component mixed with depleted mid-ocean ridge-type mantle (DMM). The Pb isotope data of the NERZ rocks (206Pb/204Pb and 207Pb/204Pb range from 19.591 to 19.838 and 15.603 to 15.635, respectively) support a model of initiation and growth of the rift from the Central Shield volcano (Roque del Conde), consistent with latest geochronology results. The similar isotope signature of the NERZ to both the Miocene Central Shield and the Pliocene Las Cañadas central volcano suggests that the central part of Tenerife Island was supplied from a mantle source that remained of similar composition through the Miocene to the Pliocene. This can be explained by the presence of a discrete column of young HIMU-like plume material, ≤100km in vertical extent, occupying the melting zone beneath central Tenerife throughout this period. The most recent central magmatism on Tenerife appears to reflect greater entrainment of DMM material, perhaps due to waning of the HIMU-like “blob” with time.

Dykes and structures of the NE rift of Tenerife, Canary Islands: a record of stabilisation and destabilisation of ocean island rift zones

Many oceanic island rift zones are associated with lateral sector collapses, and several models have been proposed to explain this link. The North–East Rift Zone (NERZ) of Tenerife Island, Spain of ...

Evolution of ocean-island rifts: The northeast rift zone of Tenerife, Canary Islands

The northeast rift zone of Tenerife presents a superb opportunity to study the entire cycle of activity of an oceanic rift zone. Field geology, isotopic dating, and magnetic stratigraphy provide a reliable temporal and spatial framework for the evolution of the NE rift zone, which includes a period of very fast growth toward instability (between ca. 1.1 and 0.83 Ma) followed by three successive large landslides: the Micheque and Guimar collapses, which occurred approximately contemporaneously at ca. 830 ka and on either side of the rift, and the La Orotava landslide (between 690 +/- 10 and 566 +/- 13 ka). Our observations suggest that Canarian rift zones show similar patterns of development, which often includes overgrowth, instability, and lateral collapses. Collapses of the rift flanks disrupt established fissural feeding systems, favoring magma ascent and shallow emplacement, which in turn leads to magma differentiation and intermediate to felsic nested eruptions. Rifts and their collapses may therefore act as an important factor in providing architectural and petrological variability to oceanic volcanoes. Conversely, the presence of substantial felsic volcanism in rift settings may indicate the presence of earlier landslide scars, even if concealed by postcollapse volcanism. Comparative analysis of the main rifts in the Canary Islands outlines this general evolutionary pattern: (1) growth of an increasingly high and steep ridge by concentrated basaltic fissure eruptions; (2) flank collapse and catastrophic disruption of the established feeder system of the rift; (3) postcollapse centralized nested volcanism, commonly evolving from initially ultramafic-mafic to terminal felsic compositions (trachytes, phonolites); and (4) progressive decline of nested eruptive activity.

Vertical axis rotation of the upper portions of the north-east rift of Tenerife Island inferred from paleomagnetic data

Paleomagnetic sampling sites were established in 82 dykes along an 8 km long section of the north-east rift-zone (NERZ) of Tenerife, Canary Islands, Spain. Of the 70 interpretable sites, 16 are of normal polarity and 54 of reversed polarity. Four normal polarity sites and fifteen reverse polarity sites were excluded from the grand mean calculation for statistical reasons. After inverting the reverse polarity sites through the origin, the in-situ grand mean yields a declination ( D) = 023.8°, an inclination ( I) = 42.3°, α 95 = 3.2°, ĸ = 39.0, N = 51 that is discordant to the expected late Miocene to Pleistocene field direction ( D = 357.6°, I = 38.8°, α 95 = 4.7°). This discordance can be explained as either a 26° clockwise vertical axis rotation or a 28° WNW-side-down-tilt about an average 009° horizontal tilt axis. The sampled section is composed of numerous semi-vertical dykes cutting mainly lava flow units that are sub-horizontal and cross-cut by steeply dipping faults (70°–90°). Field evidence is therefore more compatible with a vertical-axis rotation rather than a horizontal axis tilt of the drilled units. We argue that this clockwise vertical-axis rotation is likely related to strike-slip movements that occurred along the edges of the collapse scars and accommodate the emplacement and growth of the underlying intrusive core and associated dykes. Six new 40Ar/ 39Ar age determinations constrain the main interval of dyke emplacement within the NERZ between 0.99 Ma and 0.56 Ma. The intrusive activity in the sampled section of the rift appears to have been almost continuous, with several intrusion pulses that are probably related to flank destabilisation event(s) during the mid Pleistocene. Our study thus demonstrates a long-lived, multi-faceted history that shaped the NERZ.

Volcano‐earthquake interaction at Mauna Loa volcano, Hawaii

The activity at Mauna Loa volcano, Hawaii, is characterized by eruptive fissures that propagate into the Southwest Rift Zone (SWRZ) or into the Northeast Rift Zone (NERZ) and by large earthquakes at the basal decollement fault. In this paper we examine the historic eruption and earthquake catalogues, and we test the hypothesis that the events are interconnected in time and space. Earthquakes in the Kaoiki area occur in sequence with eruptions from the NERZ, and earthquakes in the Kona and Hilea areas occur in sequence with eruptions from the SWRZ. Using three‐dimensional numerical models, we demonstrate that elastic stress transfer can explain the observed volcano‐earthquake interaction. We examine stress changes due to typical intrusions and earthquakes. We find that intrusions change the Coulomb failure stress along the decollement fault so that NERZ intrusions encourage Kaoiki earthquakes and SWRZ intrusions encourage Kona and Hilea earthquakes. On the other hand, earthquakes decompress the magma chamber and unclamp part of the Mauna Loa rift zone, i.e., Kaoiki earthquakes encourage NERZ intrusions, whereas Kona and Hilea earthquakes encourage SWRZ intrusions. We discuss how changes of the static stress field affect the occurrence of earthquakes as well as the occurrence, location, and volume of dikes and of associated eruptions and also the lava composition and fumarolic activity.

Influence of volcanic activity at Mauna Loa, Hawaii, on earthquake occurrence in the Kaoiki Seismic Zone

The Kaoiki Seismic Zone at Mauna Loa's east flank is the locus of some of the largest earthquakes on Hawaii. In this paper we examine the link of those earthquakes to volcanic activity at Mauna Loa. We calculate the changes of Coulomb failure stress along the Kaoiki faults for typical volcanic events at Mauna Loa volcano. We simulate the dislocation associated with the 1950 eruption at the Southwest Rift Zone, the 1984 eruption at the Northeast Rift Zone, and the subsequent replenishment of the shallow magma chamber. Our model calculations suggest that the earthquake occurrence in the Kaoiki Seismic Zone strongly depends on the type of volcanic activity. Kaoiki earthquakes are encouraged by dike intrusions into the Southwest Rift Zone of Mauna Loa, but discouraged by dike intrusion into the Northeast Rift Zone. Inflation of a shallow magma chamber may again encourage earthquakes.

An Exercise in Forecasting the Next Mauna Loa Eruption

Mauna Loa, on the island of Hawaii, is one of the most studied volcanoes on Earth. Based on Mauna Loa's long historic record and detailed geologic mapping, volcanologists have published forecasts for future activity at the volcano. The mean recurrence interval for eruptions at Mauna Loa is 1,459 days. The probability of the next eruption of Mauna Loa during or before the year 2009 exceeds 50%. The eruption should send lava flows away from the town of Hilo (assuming the active vent is in the northeast rift zone) and produce about 323×l06 m3 of lava. This inquiry-based exercise requires students to think like scientists while interpreting data about Mauna Loa's past activity. Students will forecast the next eruption of Mauna Loa, predict the location of the vent, and estimate the amount of lava that will be produced.

Geochemical stratigraphy of lava flows sampled by the Hawaii Scientific Drilling Project

Geochemical discriminants are used to place the boundary between Mauna Loa flows and underlying Mauna Kea flows at a depth of about 280 m. At a given MgO content the Mauna Kea flows are lower in SiO2 and total iron and higher in total alkali, TiO2, and incompatible elements than the Mauna Loa lavas. The uppermost Mauna Kea lavas (280 to 340 m) contain alkali basalts interlayered with tholeiites and correlate with the postshield Hamakua Volcanics. In addition to total alkalis, the alkali basalts have higher TiO2, P2O5, Sr, Ba, Ce, La, Zr, Nb, Y, and V relative to the tholeiites and lower Zr/Nb and Sr/Nb ratios. Some of the alkali basalts are extensively differentiated. Below 340 m all the flows are tholeiitic, with compositions broadly similar to the few “fresh” subaerial shield‐building Mauna Kea tholeiites studied to date. High‐MgO lavas are unusually abundant, although there is a wide range (7–28%) in MgO content reflecting olivine control. FeO/MgO relationships are used to infer parental picritic magmas with about 15 wt % MgO. Lavas with more MgO than this have accumulated olivine. The Mauna Loa lavas have compositional trends that are controlled by olivine crystallization and accumulation. They compare closely with trends for historical (1843–1984) flows, tending toward the depleted end of the spectrum. They are, though, much more MgO‐rich (9–30%) than is typical for most historical and young (&lt;30 ka) prehistoric lavas. The unusual abundance of high‐MgO and picritic lavas is attributed to the likelihood that only large‐volume, hot, mobile flows will reach Hilo Bay from the northeast rift zone. FeO/MgO relationships are used to infer parental picritic magmas with about 17 wt % MgO. Again, lavas with more MgO than this have accumulated olivine. Systematic changes in incompatible element ratios are used to argue that the magma supply rate has diminished over time. On the other hand, the relatively constant Zr/Nb and Sr/Nb ratios that compare closely with historical and young (&lt;30 kyr) prehistoric flows are used to argue that the source components for these lavas in the Hawaiian plume have remained relatively uniform over the last 100 kyr.

Volcán Ecuador, Galapágos Islands: erosion as a possible mechanism for the generation of steep-sided basaltic volcanoes

Volcan Ecuador (0°02′S, 91°35′W) consists of two strongly contrasting components: the eroded and vegetated remnant of a once-circular main volcano with a probable caldera, and a prominent rift zone extending to the northeast that is neither strongly eroded nor weathered. There are about 20 young-looking flows and vents on this caldera floor but only one on the higher remnant of the main volcano. The southwest half of the main volcano is faulted into the ocean. The main part of Volcan Ecuador possesses steep erosional slopes (average 30–40°) that cut into a sequence of flows that dip radially outward at <10°. In contrast, the northeast rift zone consists entirely of young flows and vents. The upper 10 km of the rift zone forms a peninsula about 7.5 km wide that connects Volcan Ecuador to Volcan Wolf. The rift zone bends to the southeast and the lower 8 km is tangential to the coast of Volcan Wolf. The rift zone axis dips away from the northeast edge of the main volcano, and its flanks slope roughly northwest and southeast at <4°. The rift zone is the Galapagos structure that most closely resembles a Hawaiian rift zone because it is constructed of lavas from subparallel linear vents, shows evidence of a deep feeder conduit, and has changed its direction to avoid a direct intersection with neighboring Volcan Wolf. The steep erosional slopes extending around the perimeter of the main volcano (except to the southwest where slumping occurred) were probably generated by marine erosion during a prolonged period of eruptive inactivity (perhaps 20 000–30 000 years). Only a few post-erosional eruptions have taken place at the main volcano in and near what was once the caldera. The entire rift zone postdates the period of prolonged erosion. Using the evidence for prolonged inactivity at Volcan Ecuador, we propose that erosion may have helped to produce steep slopes on the other western Galapagos volcanoes. On these more active volcanoes, however, numerous subsequent eruptions have completely mantled the erosional slopes with lava. The mechanism by which the volcanoes may shut off for long periods of time is unknown, but the fact that the Galapagos hotspot is presently supplying nine active volcanoes suggests that the magma supply at an individual volcano could vary greatly over periods of (tens of?) thousands of years.

Implications of historical eruptive-vent migration on the northeast rift zone of Mauna Loa Volcano, Hawaii

Implications of historical eruptive-vent migration on the northeast rift zone of Mauna Loa Volcano, Hawaii

Implications of historical eruptive-vent migration on the northeast rift zone of Mauna Loa Volcano, Hawaii

Research Article| July 01, 1990 Implications of historical eruptive-vent migration on the northeast rift zone of Mauna Loa Volcano, Hawaii John P. Lockwood John P. Lockwood 1U.S. Geological Survey, P.O. Box 51, Hawaii Volcanoes National Park, Hawaii 96718 Search for other works by this author on: GSW Google Scholar Author and Article Information John P. Lockwood 1U.S. Geological Survey, P.O. Box 51, Hawaii Volcanoes National Park, Hawaii 96718 Publisher: Geological Society of America First Online: 02 Jun 2017 Online ISSN: 1943-2682 Print ISSN: 0091-7613 Geological Society of America Geology (1990) 18 (7): 611–613. https://doi.org/10.1130/0091-7613(1990)018<0611:IOHEVM>2.3.CO;2 Article history First Online: 02 Jun 2017 Cite View This Citation Add to Citation Manager Share Icon Share Facebook Twitter LinkedIn MailTo Tools Icon Tools Get Permissions Search Site Citation John P. Lockwood; Implications of historical eruptive-vent migration on the northeast rift zone of Mauna Loa Volcano, Hawaii. Geology 1990;; 18 (7): 611–613. doi: https://doi.org/10.1130/0091-7613(1990)018<0611:IOHEVM>2.3.CO;2 Download citation file: Ris (Zotero) Refmanager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search Search Dropdown Menu toolbar search search input Search input auto suggest filter your search All ContentBy SocietyGeology Search Advanced Search Abstract Five times within the past 138 yr (1852, 1855-1856, 1880-1881, 1942, and 1984), lava flows from vents on the northeast rift zone of Mauna Loa Volcano have reached within a few kilometres of Hilo (the largest city on the Island of Hawaii). Most lavas erupted on this rift zone in historical time have traveled northeastward (toward Hilo), because their eruptive vents have been concentrated north of the rift zone's broad topographic axis. However, with few exceptions each successive historical eruption on the northeast rift zone has occurred farther southeast than the preceding one. Had the 1984 eruptive vents (the most southeasterly yet) opened less than 200 m farther southeast, the bulk of the 1984 lavas would have flowed away from Hilo. If this historical vent-migration pattern continues, the next eruption on the northeast rift zone could send lavas to the southeast, toward less populated areas.The historical Mauna Loa vent-migration patterns mimic the southeastern "younging" of the Hawaiian-Emperor volcanic chain and may be cryptically related to northwestward movement of the Pacific plate. Systematic temporal-spatial vent-migration patterns may characterize eruptive activity at other volcanoes with flank activity and should be considered as an aid to long-term prediction of eruption sites. This content is PDF only. Please click on the PDF icon to access. First Page Preview Close Modal You do not have access to this content, please speak to your institutional administrator if you feel you should have access.

Radiocarbon Dates for Lava Flows from Northeast Rift Zone of Mauna Loa Volcano, Hilo 7 1/2’ Quadrangle, Island of Hawaii

Twenty-eight 14C analyses are reported for carbonized roots and other plant material collected from beneath 15 prehistoric lava flows erupted from the northeast rift zone (NERZ) of Mauna Loa Volcano (ML) utilizing the recovery techniques of Lockwood and Lipman (1980). Most samples were collected from the Hilo 7 1/2’ quadrangle during field work for a geologic map of that quadrangle (Buchanan-Banks, unpub data); a few sample sites are located in adjacent quadrangles: Piihonua to the west and Mountain View to the south. Altitudes are given in English units as well as metric to facilitate locating sites on USGS topographic maps.

Dikes and the petrology of Waianae Volcano, Oahu

We summarize the distribution, orientation, and composition of the dikes of Waianae Volcano and attempt to relate this summary to the evolution of the volcano. Dikes are located in the caldera, in a well‐defined northwest rift zone, and in less well‐defined south and northeast rift zones. The active axis of the south rift zone may have shifted with time. The dikes generally are parallel to their rift zones and are oriented in many directions in the caldera area. They commonly dip about 70° to 80° and are generally less than a meter thick. Most are virtually identical in texture, mineralogy, and composition to nearby extrusive rocks. A few have unusual low‐magnesium silicic compositions not yet observed in flows; these may be classified as icelandites. Trends defined by the compositions of tholeiitic dikes are similar to, but extend further than, crystal fractionation trends of tholeiitic flows of Waianae and other Hawaiian volcanoes. The basalts can be divided into a more silica saturated strongly tholeiitic subgroup and a less saturated mildly tholeiitic subgroup. The icelandites lie on trend with the second group. Most of the analyzed rift zone dikes are strongly tholeiitic; most of the analyzed caldera dikes are mildly tholeiitic; all of the differentiated high‐silica low‐magnesium dikes found are inside the caldera near the eruptive center.

The 1984 eruption of Mauna Loa Volcano, Hawaii

When Mauna Loa showed signs of unrest in the spring of 1974, this volcano, the largest on earth, had not erupted since 1950. This unrest, marked by increased seismic activity and geodetic changes, led to a small summit eruption in July 1975, a type of eruption which in the past had typically been followed by flank activity within a few years. Based on this historic pattern and on the continuing inflation, a forecast was published that called for a major northeast rift zone eruption “between 2,800 and 3,000 meters elevation… sometime before the summer of 1978” [Lockwood et al, 1976]. The forecast proved correct in all details except for timing, which proved in error by almost 6 years!

Degassing-induced crystallization of basaltic magma and effects on lava rheology

During the north-east rift eruption of Mauna Loa volcano, Hawaii, on 25 March–14 April 1984 (Fig. 1), microphenocryst contents of erupted lava increased from 0.5 to 30% without concurrent change in either bulk magma composition or eruption temperature (1,140 ± 3 °C). The crystallization of the microphenocrysts is interpreted here as being due to undercooling of the magma 20–30 °C below its liquidas; the undercooling probably resulted from separation and release of volatiles as the magma migrated 12 km from the primary summit reservoir to the eruption site on the north-east rift zone. Such crystallization of magma during an eruption has not been documented previously. The undercooling and crystallization increased the effective viscosity of the magma, leading to decreased eruption rates and stagnation of the lava flow.