Earth and Planetary Science Letters最新文献

Pulsing volcanotectonic cycles characterized by short-lived (10 s Myr) switches in magmatic and tectonic patterns have been broadly identified in active margins. However, the specific mechanism that causes these switches remains ambiguous, i.e., whether the subduction continuity and/or terrane arrival (accretion, underthrusting, or subduction of buoyant continental/oceanic blocks with a thicker crust than their surrounding oceanic plate) plays a crucial role in controlling the observed volcanotectonic cycles remains controversial. Here, by modeling subduction processes involving the sequential arrival of buoyant terranes, we show that 1) in scenarios where the oceanic plate is weakly coupled with terranes, the entraining of the terrane into the subduction induces slab breakup to occur between the partially subducted terrane and its adjacent oceanic slab, with the terrane then rebounding and moving away from the trench. This evolution causes switches in the magmatic and tectonic patterns within the overriding plate. 2) In these models, the exposed terrane materials can preserve characteristic pressure-temperature-time trajectories, i.e., nearly isothermal compression to isothermal decompression and/or isobaric heating after partial exhumation. 3) slab breakup does not guarantee the occurrence of a trench jump; only when the terrane has a medium scale (∼300 km) will a new trench tend to develop prominently behind the rebounded terrane. 4) in scenarios where the composite slab (terrane and oceanic portions) resists yielding deformation and the terrane density is close to that of the oceanic plate (<0.6% density contrast), continuous subduction will occur. In this latter scenario, slab deformation (revealed by subduction angle and slab curvature), instead of slab breakup, will control the magmatic and tectonic patterns in the overriding plate. By further comparing model results with observations, we demonstrate that intermittent subduction interrupted by the subduction of terranes can be a tectonic driver for episodes of compression-to-extension transformations and magmatism (or piston-like volcanotectonic cycles) in two representative accretionary belts — with or without trench jumps (exemplars in south-central-Tibet and eastern-Mediterranean). In contrast, continuous subduction with strongly coupled upper and subducting plates could have contributed to similar cycles in the example without accretion (e.g., the Altiplano in the Central Andes). Therefore, over 10 s of Myr, the scale and frequency of terrane arrivals could essentially control the specific motion pattern of the subducting plate, creating the observed short-lived volcanotectonic switches at these three subtypes of active margins.

This article has been retracted: please see Elsevier Policy on Article Withdrawal (https://www.elsevier.com/about/policies/article-withdrawal).

This article has been retracted at the request of Publisher/Editor for the below details:

Authors submitted a revised paper due to some errors appearing in the paper. Rather than a corrigendum being posted for this original paper, a second version was published: Earth Planet. Sci. Lett. 628 (2024) 118587, https://doi.org/10.1016/j.epsl.2024.118587. We need to retract at least one version of this paper.

The knowledge of dynamics and retreat patterns of marine-based ice streams under multiple stressors are of foremost importance for predicting Antarctic Ice Sheet response to climate changes. The Holocene palaeoglaciological record of former ice streams draining the northeast Antarctic Peninsula can elucidate the influences of changes in atmospheric and oceanic circulation and sea-level oscillations on the ice thinning and grounding line retreat. Here, terrestrial cosmogenic nuclide (TCN) dating of erratic boulders across the James Ross Island group sheds light on the pattern and timing of the ice recession along the two main arteries of the palaeo-ice drainage: Croft Trough and Prince Gustav Channel. The approach of using paired ¹⁰Be-²⁶Al nuclides enabled an assessment of cosmogenic isotope inheritance and complex burial-exposure history, notably on the high-altitude volcanic mesas. The TCN ages suggest that the Prince Gustav Channel Ice Stream was thinning from at least ∼12 ka, with subsequent separation of the Antarctic Peninsula and James Ross Island ice masses by 10–8 ka. The transition from grounded ice to open marine conditions in the Croft Trough occurred rapidly at 8.6–7.2 ka, following the Early Holocene Warm Period, concomitant with eustatic and relative sea-level rise and incursions of warmer circumpolar waters. Grounding line retreat was possibly further accelerated by buoyancy response of thinning ice stream to low-gradient bed topography. The lessons of rapid deglaciation of James Ross Island palaeo-ice streams may provide analogues for recent or future intensification of pressures on Antarctic glaciers.

Halokinetic deformations synchronous with salt deposition are processes already suggested in several salt giants including the Lower Cretaceous salt deposits of the Santos Basin in Brazil. However, the dynamics of syn-depositional deformation has never been studied in a coherent depositional and structural framework. This study investigates well data and high-resolution 3D seismic images in the evaporitic Ariri Formation of the Santos Basin to understand the intra-salt deformations, estimate their timing and establish a link with the depositional setting of the evaporites. The seismic data shows numerous examples of intra-salt onlaps, erosive surfaces and sedimentary wedges highlighting the occurrence of deformations during salt deposition. Two dominant processes have initiated salt creep: the density contrasts between a dense upper anhydrite-rich unit and a less dense lower halite-dominated unit and the gravitational system promoted by the late-rift subsidence. Updip, extension prevailed during salt deposition and favored a boudinage of the Ariri Formation. Downdip, early contraction favored the development of well-developed intra-salt minibasins. Deformation occurring during salt deposition and induced topographic relief likely influenced the hydrological conditions. It probably led to intra-salt dissolution/reprecipitation processes and hyper-saline conditions favoring the precipitation of tachyhydrite deposits in localized intra-salt depocenters. Eventually, post-salt minibasins systematically developed above syn-salt depocenters, indicating that syn-salt deposition deformation had a strong impact on post-salt evolution. This new understanding of the intra-salt deformation in the Santos Basin paves the way for new interpretations of halokinetic deformations reported elsewhere in the South Atlantic or in other salt-giant provinces.

During the process of differentiation, the magmas in convergent margins undergo an increase of oxidized nature, accompanied by a decreased Fe content and concentration of heavy Fe isotopes. Garnet and amphibole are both Fe-rich minerals, which can be responsible for this phenomenon through fractional crystallization. One prevailing hypothesis suggests that Fe²⁺-rich garnet cumulates in the arc root as a "crustal redox filter." However, the stability of garnets is highly dependent on pressure conditions. In contrast, amphibole can crystallize under a broader range of temperature and pressure conditions and is a more common mineral phase in magmas. As such, the contribution of amphibole might have been underappreciated. Here, we conducted elemental composition, zircon trace element, and high-precision Fe isotope analyses on Miocene magmatic rocks from the Gangdese arc to trace the evolution of magmatic oxidation. The results indicate that the enrichment of heavy Fe isotopes in these magmas is primarily controlled by amphibole-dominated fractional crystallization rather than garnet. This also implies that amphibole fractional crystallization may play a role in enhancing the oxygen fugacity of the magmas. Taking a global perspective, we found a pervasive correlation between amphibole fractional crystallization and Fe isotope fractionation in magmatism at convergent plate margins, indicating its widely applicable influence on oxidation. The influence of garnet cannot be entirely neglected in some specific scenarios, such as within thickened continental arcs, but its impact is generally limited. Continuous amphibole fractional crystallization increases oxidation, facilitating the mobilization and concentration of Cu within the magma, thereby enhancing the potential for porphyry deposit formation. This impact is especially notable in spatiotemporally related magmatic events and could be decisive in determining the magmatic mineralization potential.

Seismic swarms represent clusters of seismicity without large mainshocks. While they occur naturally, they can also be induced by human activities, particularly during reservoir hydraulic stimulations. A striking feature of seismic swarms is the migration of their hypocenters. The seismic front, initially attributed to fluid diffusion, has more recently been understood as the result of the propagation of a fluid-induced aseismic slip. Close to the center of the swarm, a seismic back-front is commonly admitted after the injection end, but a low density of events is also observed during the injection period. In our investigation, based on a compilation of 22 swarms of both natural or anthropogenic origin, we aim to explore the existence and origin of a seismic back-front. Interestingly, we observe a post-injection back-front only in rare cases, where a rapid fluid pressure decrease is imposed at the injection point. Conversely, a back-front during the injection period is always observed in both types of swarms. Consequently, the back-front cannot be reliably used to infer the end of injection, as commonly done for natural swarms. Moreover, the occurrence of this back-front during injection is linked to an increase in the minimum magnitude of seismic events. We interpret the vanishing of the seismicity close to the injection point as a consequence of the increase in earthquake nucleation length with increasing fluid pressure. With a substantially enhanced capability for detecting small events, it may become feasible to use this back-front as a means of monitoring injection pressure, even in the context of natural swarms.

The Midcontinent Rift (MCR), hosting several world-class ore deposits, is the fossil remnant of a massive Mesoproterozoic rifting event (1.1 Ga) that did not lead to the formation of an ocean basin. To better understand the lithospheric processes associated with the rifting stage and its subsequent failure, we developed a novel full-waveform joint inversion method using ambient noise data and teleseismic P waves for this seismically inactive region. We apply this approach to three years (2011-2013) of seismic recordings from the Superior Province Rifting EarthScope Experiment (SPREE) (~12 km average station spacing) and the USArray Transportable Array (~70 km average station spacing), and obtain a new 3D high-resolution Vs model down to 100 km depth, as well as Vp and density models down to 60 km depth. The model shows major velocity anomalies in agreement with previous seismic studies for the western arm of the MCR. In particular, we observe high density (2.8-3.0 g/cm³), Vp (6.3-6.5 km/s), and Vs (3.6-3.7 km/s) structures in the shallow upper crust within the rift, likely associated with volcanic rocks. Similar to a previously identified underplated layer, we also observe extensive normal-to-high Vs (3.8-4.2 km/s) along the whole rift axis and Vp (6.8-7.5 km/s) beneath the northern segment of the rift within the lower crust. However, the Vs and Vp values are lower than average for typical underplated materials. We suggest that this underplated layer may represent a combination of different intrusive rock types (e.g., gabbro, anorthosite) developed during magma differentiation processes, or contamination of the mafic magma by surrounding crustal material, or intrusions of sills.

Elucidating the structure and composition of Mercury is important for understanding its interior dynamics and evolution. The planet is characterised by unusual chemical characteristics and a weak magnetic field generated in a large metallic core, and its early evolution was also marked by the presence of a magnetic field, widespread volcanism and global contraction. Here we develop a parameterised model of coupled core-mantle thermal and magnetic evolution considering a layered Fe-Si(-S) core structure with chemical and physical properties of the mantle and the core based on previous laboratory studies. We seek successful solutions that are consistent with observations of Mercury's long-lived dynamo, total global contraction, present-day crustal thickness, and present-day interior structure. Successful solutions have a mantle reference viscosity $>{10}^{21}$ Pa s (corresponding to a present-day bulk mantle viscosity $>2\times {10}^{20}$ Pa s), a silicon concentration in the core >13 wt%, a present inner core radius of $\sim 1000-1200$ km and a thermally stable layer ∼ $500-800$ km thick below the core-mantle boundary. Our results show that if present, a molten FeS layer atop the core has minimal effect on Mercury's long-term thermal and magnetic evolution. Predictions from our models can be tested with upcoming Bepi-Colombo observations.