Rift Zone Of Iceland Research Articles

Magmatic evolution of the Heiðarsporður ridge, NE-Iceland

The Heiðarsporður ridge, located just south of the Krafla central volcano in the northern rift-zone of Iceland, was formed in several small eruptions in the early Holocene. The rock suite of Heiðarsporður is quite different from that of Krafla and other central volcanoes in the northern rift zone. It consists of an almost continuous series of olivine tholeiites, basaltic icelandites, icelandites and a dacite in addition to small amounts of Fe-rich tholeiites which are similar to rocks from Krafla. The occurrence of silicic rocks in the Heiðarsporður ridge is rather anomalous because usually silicic rocks in Iceland are associated with central volcanoes. This paper presents the petrography and mineral chemistry of the Heiðarsporður rock suite together with major and trace element geochemistry and oxygen isotope ratios. The intermediate and silicic rocks of Heiðarsporður are characterized by relatively high concentrations of CaO and Al 2O 3 coupled with relatively low concentrations of FeO, compared with evolved rocks elsewhere in the rift zones of Iceland. Furthermore, ∂ 18O values are higher in the evolved rocks of Heiðarsporður. The olivine tholeiites are of a very common regional type and represent mantle-derived magma. The Fe-rich tholeiites which only occur in the northern end of Heiðarsporður are derived from the Krafla central volcano. The basaltic icelandites represent hybrids of olivine tholeiites and icelandites. The icelandites and the dacite are proposed to have formed by near-solidus differentiation at a deeper level in the crust than silicic rocks in the nearby Krafla central volcano. This interpretation is supported by petrogenetic modelling using the MELTS software. The melts were extracted from the source by crustal deformation, most likely related to concurrent activity in the Krafla volcanic system.

Nitrogen isotopes of the mantle: Insights from mineral separates

We present the first nitrogen (N) isotope measurements determined by in‐vacuo crushing of mineral separates from arc lavas, OIBs (Ocean Island Basalts), and mantle xenoliths. Measured OIB δ15N values range from ∼−8‰ for the northern rift zone in Iceland to +3.1‰ for a dunite nodule from Hawaii. Most arc‐related olivines show distinctly positive values ‐ up to +6.2‰ (Cerro Negro, Nicaragua). The measured N isotope values in olivine separates are similar to gas samples collected at the same localities, suggesting that both media (olivines and gases) sample volatiles primarily derived from the magma. This observation also implies that N isotope fractionation does not occur during magma degassing, a notion supported by 4He/40Ar* data. Our results indicate a heterogeneous mantle source region, in terms of N isotopic composition, that may have resulted from surface recycling of N at some localities.

Azimuthal anisotropy and phase velocity beneath Iceland: implication for plume–ridge interaction

We have determined the pattern of azimuthal anisotropy beneath Iceland from shear-wave splitting and Rayleigh wave tomography using seismic data recorded during the ICEMELT and HOTSPOT experiments. The fast directions of shear-wave splitting are roughly N–S in western Iceland and NNW–SSE in eastern Iceland. In western Iceland azimuthal variations in Rayleigh wave phase velocity show that fast directions are close to the plate spreading direction at short periods of 25–40 s and roughly parallel to the Mid-Atlantic Ridge at longer periods of 50–67 s. Beneath the rift zones in central Iceland, we find a ridge-parallel fast direction at periods of 25–40 s and significantly weaker azimuthal anisotropy at periods of 50–67 s. The 2-D variation of isotropic phase velocity at periods of 33–67 s indicates that the lowest velocities are beneath the rift zones in central Iceland rather than above the plume conduit in southeast Iceland. While ridge-parallel alignment of melt films might contribute to anisotropy above 50 km depth beneath the rift zones, the overall observations are consistent with a model in which plume-influenced, hot, buoyant mantle rises beneath the Mid-Atlantic Ridge at depths greater than 50 km and is preferentially channeled along the ridge axis at the base of the lithosphere beneath Iceland. The shear-wave splitting results are attributed primarily to a N–S mantle flow at depths greater than 100 km.

Effect of stress fields on magma chamber stability and the formation of collapse calderas

The summits of many of the Earth’s and other planets’ larger volcanoes are occupied by calderas that formed by collapse into an evacuating, underlying magma chamber. These collapse calderas are typically several tens of square kilometers in area and are commonly elliptical in shape. We show that the long axes of late Quaternary collapse calderas in the Kenya rift valley, the western Basin and Range province, the Snake River‐Yellowstone Plateau, and the Iceland rift zone are parallel to the upper crustal minimum horizontal stress direction (Sh) as determined by independent criteria. We suggest that circular magma chambers beneath these volcanoes became elliptical by stress‐induced spalling of their chamber walls, by a mechanism that is analogous to the formation of breakouts in boreholes and tunnels. In breakouts, the hole becomes elongate parallel to the far‐field minimum stress. In the Kenya rift, Late Pleistocene caldera collapse was accompanied by a 45° rotation of Sh and an increase in the magnitude of the maximum horizontal stress (SH). The breakout model predicts increasingly unstable caldera walls under these conditions, a possible explanation for the sudden appearance of so many collapse events in a volcanic setting that had never experienced them before. This mechanism of stress change‐induced collapse may have played a role in other caldera settings.

Platinum-group elements in the Icelandic rift system: melting processes and mantle sources beneath Iceland

Thirty new platinum-group element (PGE) analyses of various basalt types from Iceland are presented in this study. The analysed samples are divided into three groups based on their Mg-contents, high-Mg tholeiites (10–14 wt.% MgO), composed of primitive olivine tholeiites and picrites, evolved olivine tholeiites (7–10 wt.% MgO) and evolved basalts (4–7 wt.% MgO) consisting of FeTi basalts, quartz tholeiites and alkaline basalts. The high-Mg tholeiites have a range of compositions between a relatively Cu-rich and Pd-poor (120 ppm Cu and 6 ppb Pd) end-member and a relatively Cu-poor and Pd-rich (74 ppm Cu and 17 ppb Pd) end-member. There is a positive correlation between the highly siderophile elements, whereas Cu and Pd correlate negatively between the Cu-rich primitive olivine tholeiite and the Cu-poor picrite end-member. Negative correlation between Cu and the PGEs cannot be reconciled in a model where the two end-members form by varying degrees of partial melting of a common source. The sub-primitive mantle Cu/Pd ratio of the picrite end-member (∼4300) could indicate a strongly depleted mantle source, where Pd was efficiently retained in mantle sulphide, relative to Cu, during previous melt extraction episodes. In mantle melting models presented here, the picrite composition can be approximated by 25% melting of a mantle source that previously lost a normal MORB component ( F∼15%). The olivine tholeiite end-member (Cu/Pd∼19000) can be approximated in mantle melting models by assuming derivation from a separate source, one slightly Cu-enriched source undergoing melting beneath a rift. Based on Cu and Pd variations, the majority of high-Mg tholeiites are interpreted to be mixtures between melts derived from highly depleted to slightly enriched sources, which is consistent with previous studies of Sr–Nd isotopic variations. The similar range in Cu/Pd ratios of the high-Mg tholeiites and the evolved olivine tholeiites could infer that the latter evolved from parental liquids similar to the high-Mg tholeiites under S-undersaturated conditions. Relatively high Pd and low Ir concentrations in the evolved olivine tholeiite samples (∼7 wt.% MgO) indicate, too, that these magmas underwent S-undersaturated fractionation during which Pd accumulated in the melt whereas Ir was incorporated into the fractionating assemblage. This is in contrast to the group of evolved basalts with less than 7 wt.% MgO with low PGE contents and high Cu/Pd ratios that strongly suggest that these samples have experienced S-saturation. We propose a scenario for melt generation beneath the Icelandic rift zone, where picritic magma is generated from a highly depleted mantle, whereas primitive olivine tholeiite magma forms from the depleted as well as a Cu-enriched source. It is proposed that Cu-enriched mantle domains were exhausted at depth since their melting products are absent in the most depleted picrite derived from the uppermost part of the central melting region. The olivine tholeiites can be approximated by around 11–12% partial melting in a triangular melting regime beneath the Icelandic rift zone. The olivine tholeiites are interpreted to reflect efficient melt collection from the deepest and most distal parts of the melting triangle. In contrast, the picrite represent a high degree of melting ( F∼25%) without incorporation of melt batches from the Cu-enriched mantle source.

Mineralized Microbiota: A Possible Indicator of Hydrocarbon Flows in the Rift Zone of Iceland

Relationships of mineralized microbiota with the content of polycyclic aromatic hydrocarbons in hydrothermal systems are considered. It has been established that the mineralized microbiota can serve as an indicator of hydrothermal hydrocarbon flows in present-day and ancient deposits.

Oxygen isotope variations in primitive tholeiites of Iceland: evidence from a SIMS study of glass inclusions, olivine phenocrysts and pillow rim glasses

Oxygen-isotope compositions are reported for glass inclusions, their host olivines (Fo 87–90) and tholeiitic pillow rim glasses from the neovolcanic rift zone of Iceland. The inclusion compositions range from enriched ([La/Sm] n=1.4–4.8) to very depleted ([La/Sm] n=0.08–0.35). The whole rocks and matrix glasses display little variation ([La/Sm] n=0.41–0.51). Glass inclusions vary somewhat stronger in 18O/ 16O ratios (δ 18O=4.0–6.2±0.6‰) than the matrix glasses (δ 18O=4.6–5.6±0.6‰). The range of olivine phenocryst composition is even wider (δ 18O=2.4–6.9±0.5‰), and in some cases the variations within individual crystals reach 1.0–2.5‰. Most glass inclusions are in O-isotopic equilibrium with the surrounding host olivine but not with the central parts of the host crystals; similarly, most olivines are out of equilibrium with their host matrix glasses. This implies that the glass inclusions and olivine phenocrysts can preserve O-isotopic compositions inherited from contrasting parental mantle-derived magmas, which might have mixed in a common volcanic system. The isotopic heterogeneity of individual olivine crystals allows one to estimate their residence times in the magma reservoir using the rate of 18O diffusion in olivine. The time required to achieve isotopic homogeneity over 90–520 μm ranges from 25 to 1030 yr. The δ 18O values found in depleted glass inclusions and their host olivines are unlikely to have resulted from magma contamination by upper crustal rocks. We suggest two possibilities, implying either an O-isotope heterogeneity of the Icelandic mantle or a mixing of two end-members: a refractory component with ‘normal-mantle’ or somewhat elevated δ 18O values, high Ca/Na and strong light rare earth element depletion mixed with a component having lower δ 18O and Ca/Na but also depleted in incompatible elements. Based on δ 18O–La/Sm and δ 18O–δ 11B relationships, the variations of δ 18O between 4.6 and 6.2‰ in enriched and depleted glass inclusions are interpreted to reflect those of primary magmas and, consequently, of their source region.

The link between volcanism and deglaciation in Iceland

Temporal variation in the eruption rate and lava composition in the rift zones of Iceland is associated with deglaciation. Average eruption rates after the end of the last glacial period, ∼12 kyr BP, were up to 100 times higher than those from both the glacial period and recent times (<5 kyr BP). This peak in volcanic activity finished less than 2 kyr after the end of deglaciation. New geochemical data from ∼80 basalt and picrite samples from the Theistareykir and Krafla volcanic systems show that there is a temporal variation in both the major and trace element composition of the eruptions. Early postglacial eruptions show a greater range in MgO contents than eruptions from other times, and at a fixed MgO content, the concentration of incompatible elements in subglacial eruptions is higher than that in early postglacial eruptions. Recent eruptions from the Krafla system have similar compositions to subglacial eruptions. The high eruption rates and low rare earth element (REE) concentrations in the lava from early postglacial times can be accounted for by increased melt generation rates in the shallow mantle caused by unloading of an ice sheet. Magma chamber processes such as crystallization and assimilation can produce the temporal variation in REE contents if garnet is present. However, garnet is not observed as a phenocryst or xenocryst phase and is not required to match the variation in major element contents observed at Krafla and Theistareykir. If the increase in eruption rates reflects increased melt production rates in the mantle, then the relative timing of deglaciation and the burst in eruption rates can be used to estimate the rate of melt transport in the mantle. The observed duration of enhanced eruption rates after deglaciation can be reproduced if the vertical melt extraction velocity is >50 m yr−1.

Crustal accretion under northern Iceland

Successful models of oceanic crustal accretion should be consistent with both geochemical and geophysical observations from active mid-ocean ridges as well as those from ophiolite sections. The major element concentrations of a set of basalt and picrite samples from the Krafla and Theistareykir volcanic systems of the Northern Volcanic Zone of Iceland show two distinct trends. The compositional variation in samples with MgO>9.5 wt% can be explained by addition/removal of a wehrlitic cumulate, while the major element variability in samples with 5–9.5 wt% MgO is dominated by gabbro removal. The results of thermobarometry based on the composition of the samples and the crystals found within them show that crystallisation took place at a range of temperatures (1160–1350°C) and pressures (<0.3–0.9 GPa) in the crust and uppermost mantle under Krafla and Theistareykir. The geochemical results are consistent with crustal accretion models where crystallisation takes place over a range of depths in the crust and uppermost mantle (<10–30 km). The geochemical observations allow estimates of the composition, mineralogy, pressure and temperature of material in the crust and shallow mantle under northern Iceland to be made and these estimates can be used to predict the seismic velocity of the material at the ridge axis. These P-wave velocity estimates are in agreement with the results of a seismic survey of the ridge axis at Krafla. The presence of temperatures of over 1000°C and magma chambers at depths greater than 10 km in the Icelandic crust cannot be ruled out using the available geophysical data from the Icelandic rift zones. Therefore both the geochemical and geophysical observations are consistent with models where crustal accretion takes place at a range of depths under northern Iceland.

Explosive silicic volcanism in Iceland and the Jan Mayen area during the last 6 Ma: sources and timing of major eruptions

Fifty-three major explosive eruptions on Iceland and Jan Mayen island were identified in 0–6-Ma-old sediments of the North Atlantic and Arctic oceans by the age and the chemical composition of silicic tephra. The depositional age of the tephra was estimated using the continuous record in sediment of paleomagnetic reversals for the last 6Ma and paleoclimatic proxies (δ18O, ice-rafted debris) for the last 1Ma. Major element and normative compositions of glasses were used to assign the sources of the tephra to the rift and off-rift volcanic zones in Iceland, and to the Jan Mayen volcanic system. The tholeiitic central volcanoes along the Iceland rift zones were steadily active with the longest interruption in activity recorded between 4 and 4.9Ma. They were the source of at least 26 eruptions of dominant rhyolitic magma composition, including the late Pleistocene explosive eruption of Krafla volcano of the Eastern Rift Zone at about 201ka. The central volcanoes along the off-rift volcanic zones in Iceland were the source of at least 19 eruptions of dominant alkali rhyolitic composition, with three distinct episodes recorded at 4.6–5.3, 3.5–3.6, and 0–1.8Ma. The longest and last episode recorded 11 Pleistocene major events including the two explosive eruptions of Tindfjallajökull volcano (Thórsmörk, ca. 54.5ka) and Katla volcano (Sólheimar, ca. 11.9ka) of the Southeastern Transgressive Zone. Eight major explosive eruptions from the Jan Mayen volcanic system are recorded in terms of the distinctive grain-size, mineralogy and chemistry of the tephra. The tephra contain K-rich glasses (K2O/SiO2>0.06) ranging from trachytic to alkali rhyolitic composition. Their normative trends (Ab–Q–Or) and their depleted concentrations of Ba, Eu and heavy-REE reflect fractional crystallisation of K-feldspar, biotite and hornblende. In contrast, their enrichment in highly incompatible and water-mobile trace elements such as Rb, Th, Nb and Ta most likely reflect crustal contamination. One late Pleistocene tephra from Jan Mayen was recorded in the marine sequence. Its age, estimated between 617 and 620ka, and its composition support a common source with the Borga pumice formation at Sør Jan in the south of the island.

Interaction and linkage of extension fractures and normal faults: examples from the rift zone of Iceland

Field and photogeological studies were made of 90 zones of interacting fracture segments along the rift zone of Iceland. Each zone consists of a pair of extension fractures or a pair of normal faults, with lengths from metres to kilometres. These zones evolve from an underlapping stage, through an overlapping stage (the most common configuration) and, finally, to a linkage stage. Of all the zones, only 7% are underlapping, whereas 93% are overlapping, with hook-shaped fracture pairs. The length/width ratios of the overlapping zones have a mean value of 3.5. The preferred geometry of the overlapping zones depends upon the initial configuration of the interacting fractures (length, overstep) and the development of the interaction. In the overlapping zones, most fracture pairs show moderate shear (strike-slip) components, related to local variations in the extension (opening) directions. Vertical displacements on normal faults decrease as the overstep and length of overlapping zones increase; both, in turn, are proportional to the total lengths of the faults forming the pair. The geometrical and kinematic features of overlapping spreading centres at mid-ocean ridges show close similarities to those reported here. These similarities indicate that the architecture and evolution of overlapping zones are scale independent.

Fracture dimensions, displacements and fluid transport

There is commonly a linear relationship between the lengths of rock fractures and their displacements, but for many fracture populations there is a very large scatter in the data. In the lava flows of the rift zone in Iceland, the displacements on a fracture or fault of a given length may vary by a factor of 2–10. Similar scatter is obtained for the aperture (width)/length ratios of several hundred mineral-filled veins in a major fault zone. I propose that the displacement on a fracture depends mostly on the smaller of its dip and strike dimensions, referred to as the controlling dimension. Thus, in a horizontal outcrop, fractures with the same strike dimension (outcrop length) can have widely different displacements depending on whether the displacements of individual fractures are controlled by strike or dip dimensions. During growth of a fracture, its controlling dimension may alternate between the dip dimension and the strike dimension. The volumetric rate of flow of fluid through a rock fracture with smooth, parallel walls depends on the cube of the fracture aperture. This cubic law implies that when the aperture of a fracture of a given length in a single set or population can vary by a factor of 2–10, the corresponding volumetric rate of fluid flow through that fracture can vary by a factor of 8–1000. A single, wide fracture in a set of as many as several hundred fractures may thus largely dominate the fluid transport through that set. Fracture aperture depends not only on the associated stress field, but also on its controlling dimension.

Plagioclase ultraphyric basalts in Iceland: the mush of the rift

Glassy, plagioclase ultraphyric basalts from six locations in Iceland have bimodal phenocryst size distributions where microphenocrysts (ol+plg±cpx±mt) are in equilibrium with the matrix glass, but macrophenocrysts (ol+plg±cpx) are too primitive to be so. Matrix glass compositions are similar to those of other rift zone glasses from Iceland, and oxygen isotope variations suggest they interacted with the Icelandic crust. A lack of negative Eu-anomalies in matrix glasses precludes large amounts of plagioclase crystallisation from their parental liquids. Compositions of glass inclusions in plagioclase and olivine macrophenocrysts indicate that parental magma compositions of the macrophenocryst assemblage are similar to those of primitive, Icelandic rift zone glasses. Temperatures for plagioclase macrophenocryst crystallisation obtained from Linkam® heating stage experiments, and from glass inclusion compositions corrected for post-entrapment crystallisation, give temperatures up to 1260°C, corresponding to crystallisation at middle to deep crustal levels. Temperature differences of less than 100°C between plagioclase-hosted glass inclusions before and after post-entrapment plagioclase crystallisation show that the macrophenocrysts must have been kept at elevated temperatures prior to incorporation in their present host magmas. We suggest that the macrophenocrysts of the plagioclase ultraphyric basalts accumulated in crystal mush bodies underneath the rift zone and were picked up by their present hosts during a rifting event with increased magma supply from the mantle.

Mapping out the conduit of the Iceland mantle plume with helium isotopes

In order to investigate the lateral dimensions and the shallow dispersion of the Iceland mantle plume we have measured the 3He/ 4He ratios in closely spaced primitive (Mg#: 59–70) basaltic subglacial pillow lavas erupted along the Northern Rift Zone (NRZ) of Iceland. This detailed sampling has revealed for the first time a clear gradient in 3He/ 4He ratios along the NRZ, increasing from 8.0 times atmospheric ( R/ R a) in the northernmost samples to 20.4 R/ R a in central Iceland. In combination with previously published 3He/ 4He data, the new results from subglacial basaltic glasses from the neovolcanic rift zones of Iceland reveal a ‘plateau’ of high values ( 3He/ 4He ∼20 R/ R a), which is approximately 100 km in diameter and located over the southeastern part of the island. The high 3He/ 4He plateau is well-defined and coincides with a regional gravity minimum and a seismic low velocity anomaly within the upper mantle beneath Iceland. We suggest that this zone of elevated 3He/ 4He ratios (>20 R/ R a) outlines the width of the Iceland plume conduit at depth. 3He/ 4He ratios are well correlated with Nb/Th and Nb/Rb, but not well correlated with other isotope ratios (as indicated by Pb isotopes). The correlations between helium isotopes, the geophysical variations, and some extremely incompatible elements, such as Rb and Th, and the lack of correlation between He and Pb isotopes, are most likely due to preferential concentration of He into volatile-rich incipient melts, and the subsequent lateral dispersion of the plume at shallower levels.

Englacial vs lacustrine origin of volcanic table mountains: evidence from iceland

Detailed facies analysis of hyaloclastites and associated lavas from eight table mountains and similar "hyaloclastite volcanoes" in the Icelandic rift zone contradict a rapid and continuous, "monogenetic", entirely subglacial evolution of most volcanoes studied. The majority of the exposed hyaloclastite deposits formed in large, stable lakes as indicated by widespread, up to 300-m-thick, continuous sections of deep water, shallow water and emergent facies. Salient features include extensively layered or bedded successions comprising mainly debris flow deposits, turbidites, base surge and fallout deposits consisting of texturally and compositionally variable, slightly altered hyaloclastites, as well as sheet and pillow lavas. In contrast, chaotic assemblages of coarser-grained, more poorly sorted and more strongly palagonitized hyaloclastite tuffs and breccias, as well as scoria and lava are interpreted to have formed under sub- or englacial conditions in small, chimney-like ice cavities or ice-bound lakes. Irregularly shaped and erratically arranged hyaloclastite bodies produced at variable water levels appear to have resulted mainly from rapid changes of the eruptive environment due to repeated build-up and drainage of ice-bound lakes as well as the restricted space between the ice walls. We distinguish a "deep water" facies formed during high water levels of the lake, a hydroclastic shallow water and emergent facies (leakage of the lake or growth of the volcano above the water surface). Our model implies the temporary existence of large, stable lakes in Iceland probably formed by climatically induced ice melting. The highly complex edifices of many table mountains and similar volcanoes were constructed during several eruptive periods in changing environments characterized by contrasting volcanic and sedimentary processes.

Seismotectonics of the central part of the South Iceland Seismic Zone

The largest earthquakes in Iceland are associated with strike-slip faults in the South Iceland Seismic Zone (SISZ). Major destructive earthquake sequences occur in this zone at an average interval of 80 years, the last two sequences being in 1784 and 1896. The last major single earthquake in the zone was an earthquake of magnitude 7 ( M) in 1912. The seismicity is mostly associated with a conjugate system of NNE-trending (mainly) right-lateral faults and ENE-trending (mainly) left-lateral faults, but in addition there are many NW-trending faults in the SISZ. Particular attention is paid to the comparison between present-day fault activity (earthquake mechanisms) and Quaternary faulting (fault-slip data). This comparison reveals a general similarity despite the temporal difference between these subsets independently collected. The fault populations are well exposed in the hyaloclastite mountain Vördufell, near the central part of the South Iceland Seismic Zone, where considerable fault slip occurred during the earthquake sequences of 1784 and 1896. The cumulative displacement on most of these faults is 1–6 m. In addition to the large-scale faults, we measured 79 minor faults (displacements of the order of centimetres) in the Vördufell Mountain. Most of the minor faults are left-lateral. Analysis of these minor faults recording palaeostresses in rocks ranging in age from 3.1 to 0.7 Ma, as well as 50 focal mechanisms from a swarm of present-day microearthquakes occurring beneath the Vördufell Mountain, gave consistent results in terms of seismotectonic stress (orientations of stress axes and ratios of principal stress differences). This shows that the tectonic regime remained unchanged through recent times. The results indicate, however, that there are two contrasting stress regimes in this part of the SISZ. The primary subset indicates NW–SE tension and NE–SW compression, whereas the secondary subset indicates NW–SE compression and NE–SW tension. We propose that the primary subset is consistent with the dominating time-averaged regional stress field in the SISZ. This dominating stress field can be largely explained as a consequence of simple plate pull parallel with the spreading vector in South Iceland. The secondary subset and the associated stress field, however, may be partly related to stress release and rebound and partly to dike injections in the nearby segments of the rift zone in South Iceland.

Generation of Icelandic rhyolites: silicic lavas from the Torfajökull central volcano

The Torfajökull central volcano in south-central Iceland contains the largest volume of exposed silicic extrusives in Iceland (∼225 km 3). Within SW-Torfajökull, postglacial mildly alkalic to peralkalic silicic lavas and lava domes (67–74 wt.% SiO 2) have erupted from a family of fissures 1–2.5 km apart within or just outside a large caldera (12×18 km). The silicic lavas show a fissure-dependent variation in composition, and form five chemically distinct units. The lavas are of low crystallinity (0–7 vol.%) and contain phenocrysts in the following order of decreasing abundance: plagioclase (An 10-40), Na-rich anorthoclase (<Or 23), clinopyroxene (Fs 37-20), FeTi oxides (Usp 32-60; Ilm 93-88), hornblende (edenitic–ferroedenitic) and olivine (Fo 22-37), with apatite, pyrrhotite and zircon as accessory phases. The phenocryst assemblage (0.2–4.0 mm) consistently exhibits pervasive disequilibrium with the host melt (glass). Xenoliths include sparse, disaggregated, and partially fused leucocratic fragments as well as amphibole-bearing rocks of broadly intermediate composition. The δ 18 O values of the silicic lavas are in the range 3.6–4.4, and these are lower than the δ 18 O values of comagmatic, contemporaneous basaltic extrusives within SW-Torfajökull, implying that the former can not be derived from the latter by simple fractional crystallization. FeTi-oxide geothermometry reveals temperatures as low as 750–800°C. To explain the fissure-dependent chemical variations, 18 O depletions, low FeTi-oxide temperatures and pervasive crystal-melt disequilibrium, we propose the extraction and collection of small parcels of silicic melt from originally heterogeneous basaltic crustal rock through heterogeneous melting and wall rock collapse (solidification front instability, SFI). The original compositional heterogeneity of the source rock is due to (1) silicic segregations, in the form of pods and lenses characteristically formed in the upper parts of gabbroic intrusives, and (2) extreme isostatic subsidence of the earlier, less differentiated lavas of the Torfajökull central volcano. Ridge migration into older crustal terranes, coupled with establishment of concentrated volcanism at central volcanoes like Torfajökull due to propagating regional fissure swarms, supplies the heat source for this overall process. Continued magmatism in these fissures promotes extensive prograde heating of older crust and the progressive vitality and rise of the central volcano magmatic system that leads to, respectively, SFI and subsidence melting. The ensuing silicic melts (with relict crystals) are extracted, collected and extruded before reaching complete internal equilibrium. Chemically, this appears as a two-stage process of crystal fractionation. In general, the accumulation of high-temperature basaltic magmas at shallow depths beneath the Icelandic rift zones and major central volcanoes, coupled with unique tectonic conditions, allows large-scale reprocessing and recycling of the low- δ 18 O , hydrothermally altered Icelandic crust. The end result is a compositionally bimodal proto-continental crust.

Magma chambers modeled as cavities explain the formation of rift zone central volcanoes and their eruption and intrusion statistics

The volcanic rift zone in Iceland is characterized by elongate volcanic systems, consisting of tension fractures, normal faults and volcanic fissures, where most of the volcanotectonic activity takes place. Most volcanic systems develop central volcanoes, the formation of which is still poorly understood. There is great difference in the eruption and intrusion statistics of the volcanic systems inside and outside the central volcanoes. In the central volcanoes, eruptions are frequent (typically one every several hundred years), of small volume (normally less than 0.1 km3) and fed mainly by thin (average thickness of 0.5 m), inclined sheets. Outside central volcanoes the eruptions are rare (typically one every several thousand years) but of large volume (typically more than 1 km3) and fed mainly by thick (average thickness of 4–5 m), subvertical dikes. Using the results of a boundary element study of magma chambers modeled as a cavities or holes in a semi‐infinite plate, these empirical relations, as well as the formation of specific central volcanoes in the volcanic systems, can be explained. Once a cavity‐like magma chamber has formed, its existence in the rift zone concentrates tensile stress. This stress concentration causes the segment containing the magma chamber to rupture much more frequently than the other parts of the volcanic system, which partly explains the formation of central volcanoes and their high eruption frequencies. This concentration also gives rise to a local stress field that encourages injection of small‐volume sheets in all directions and at various dips from their source magma chamber. The magma chamber acts as a trap for upward propagating dikes from mantle reservoirs and channels magma, through inclined sheets, toward a limited area at the surface where the central volcano gradually forms. The inverse relationship between eruption frequency and eruption volume, when the central volcano is compared with other parts of the volcanic system, is also partly due to the trap‐like nature of the magma chamber. The chamber is normally much smaller than its source mantle reservoir, so that a single magma flow (through a dike) from the reservoir (lasting perhaps many years) may trigger tens of magma flows (through sheets) from the chamber, many of which would reach the surface in the central volcano.

Oxygen isotopic fluxes associated with high-temperature processes in the rift zones of Iceland

High-temperature reactions in the rift zones in Iceland, involving magma, crust and hydrothermal fluids, cause the lowering of the 18O/ 16O ratio of the crust relative to the mantle. A significant flux of 18O to the hydrosphere results from these reactions. We use the available oxygen isotope data to estimate the isotopic composition of the crust produced in this environment. This enables us to place limits on the flux of 18O from the mantle to the hydrosphere. The rift zones in Iceland are efficient in transferring 18O to the hydrosphere; we find that they release up to five times the amount of 18O as does an equivalent length of mid-ocean ridge. In spite of this large flux the net depletion of the Icelandic crust is relatively small, the data show that the δ 18O of the crust in the rift zone, has been lowered by about 1 to 3‰. Our findings have implications for the 18O-history of seawater. We argue that the rift zones in Iceland are a model analog for a global rift system above sea level, a proposed tectonic configuration that perhaps was prevalent in the geological past. Contrary to the proposed termination of the 18O-flux from the mantle to the hydrosphere, we demonstrate that a significant 18O-flux still prevails. This particular tectonic configuration, therefore, does not allow depleted seawater to persist.

Hafnium and Strontium Isotope Measurements in High-MgO Basalts from Theistareykir, Northern Iceland

We report Hf and Sr isotope compositions from 31 samples from Theistareykir volcano, northern Iceland. The samples are high-MgO olivine tholeiites and picrites with the majority having MgO contents higher than 9 wt.%. The total range of MgO contents is from 5.5 to 22 wt.%. All samples are younger than 10,000 years and were collected from different postglacial lava flows of the Theistareykir volcano in the northern Icelandic rift zone.