Description: Between 08.08.2003 and 02.09.2003, bathymetric data was acquired offshore Guatemala and Costa Rica during the R/V SONNE cruise SO173/2. The expedition comprised geophysical and biological research objectives. One aim was the acquisition of geophysical data for a better understanding of recent and long-term evolution of the Middle America Landbridge and mass flux into the subduction system. Moreover, the cruise was also dedicated to studying the sensory systems of mesopelagic fish, cephalopods, crustaceans and teleosts by using trawl gear and morphometric studies. Bathymetric mapping with the multibeam echosounder (MBES) SIMRAD EM120 was utilized to obtain a full coverage bathymetric map along the El Salvador and Guatemalan continental slope and to complete previous maps by filling gaps along the continental slope and oceanic plate of Nicaragua. Further geophysical instruments, such as the sub-bottom profiler PARASOUND, magnetometer, a dredge and seismic instrumentation, and biological equipment including trawling gear and lab instrumentation, complemented the research equipment. CI Citation: Paul Wintersteller (seafloor-imaging@marum.de) as responsible party for bathymetry raw data ingest and approval. Description of the data source: During the SO173/2 cruise, the hull-mounted multibeam echosounder (MBES) SIMRAD EM120 was utilized to perform bathymetric mapping. It allows to conduct surveys in water depths of up to 11,000 m. Two transducer arrays transmit successive frequency coded acoustic signals (11.25 to 12.6 kHz). Data acquisition is based on successive emission-reception cycles of the signal. While the emission beam has a dimension of 150° across and 2° along track, the reception is obtained from 191 overlapping beams with widths of 2° across and 20° along track. The beam footprint has a dimension of 2° by 2°. The beam spacing can be set to equidistant or equiangular. For further information on the system, consult: https://www.km.kongsberg.com/ Depth is estimated from each beam by using the two-way travel time and the beam angle known from each beam, and taking into account the ray bending due to refraction in the water column by sound speed variations. Combining phase and amplitude is used to provide measurement accuracy practically independent of the beam pointing angle. During the SO173 cruise, the EM120 was used continuously. At the beginning of the cruise, a sound velocity profile was measured to a depth of 2000 m. Responsible person during this cruise / PI: Wilhelm Weinrebe (wweinrebe@ifm-geomar.de) Chief Scientist: Wilhelm Weinrebe (wweinrebe@ifm-geomar.de) CR: http://oceanrep.geomar.de/13407/1/Geomar-Report-116.pdf CSR: https://www2.bsh.de/aktdat/dod/fahrtergebnis/2003/20040060.htm This dataset was published as part of: Geersen, Jacob (2019): Collated bathymetric data from convergent margins that experienced tsunami earthquakes. PANGAEA, https://doi.org/10.1594/PANGAEA.899049

Description: Active ridge propagation frequently occurs along spreading ridges and profoundly affects ridge crest segmentation over time. The mechanisms controlling ridge propagation, however, are poorly understood. At the slow spreading Mid-Atlantic Ridge at 21.5°N a seismic refraction and wide-angle reflection profile surveyed the crustal structure along a segment controlled by rapid ridge propagation. Tomographic traveltime inversion of seismic data suggests that the crustal structure along the ridge axis is controlled by melt supply; thus, crust is thickest, 8 km, at the domed segment center and decreases in thickness toward both segment ends. However, thicker crust is formed in the direction of ridge propagation, suggesting that melt is preferentially transferred toward the propagating ridge tip. Further, while seismic layer 2 remains constant along axis, seismic layer 3 shows profound changes in thickness, governing variations in total crustal thickness. This feature supports mantle upwelling at the segment center. Thus, fluid basaltic melt is redistributed easily laterally, while more viscose gabbroic melt tends to crystallize and accrete nearer to the locus of melt supply. The onset of propagation seems to have coincided with the formation of thicker crust, suggesting that propagation initiation might be due to changes in the melt supply. After a rapid initiation a continuous process of propagation was established. The propagation rate seems to be controlled by the amount of magma that reaches the segment ends. The strength of upwelling may govern the evolution of ridge segments and hence ultimately controls the propagation length.

Description: The subduction of partially serpentinized oceanic mantle may potentially be the key geologic process leading to the regassing of Earth’s mantle and also has important consequences for subduction zone processes such as element cycling, slab deformation, and intermediate-depth seismicity. Little is known about the quantity of water that is retained in the slab during mantle serpentinization. Recent studies using thermodynamical and/or experimental models of subduction zone processes have assumed that the mantle is uniformly serpentinized to a depth determined from the equilibrium stability of serpentine minerals in P-T space. This approach yields an incomplete picture of the pattern of serpentinization that may occur during bending-related faulting; an initial state that is essential for quantifying subsequent dehydration processes. In order to provide further constraints on the pattern of hydration and the amount of water trapped in the subducting mantle, we build a 2-D reactive-flow model incorporating the kinetic rate-dependence of serpentinization based on experimental results. After simulating hydration processes at the trench outer-rise, we find that the water content in serpentinized mantle strongly depends on the age of the subducting lithosphere and subduction rate, with values ranging between 1.8x105 and 4.0x106 kgm-2 reactive water uptake into the subducting mantle column. Serpentinization also results in a reduction in surface heat flux towards the trench caused by advective downflow of seawater into the reaction region. Observed heat flow reductions are larger than the reduction due to the minimum-water downflow needed for partial serpentinization, predicting that active hydrothermal vents and chemosynthetic communities should also be associated with bend-fault serpentinization. Model results agree with previous studies that the lower plane of double Benioff zones can be generated due to dehydration of serpentinized mantle at depth. The depth-dependent pattern of serpentinization including reaction kinetics predicts a separation between the two Benioff planes consistent with seismic observations.

Description: Hydration of the oceanic lithosphere is an important and ubiquitous process which alters both the chemical and physical properties of the affected lithologies. One of the most important reactions that affect the mantle is serpentinization. The process of serpentinization results in a drastic decrease in the density (up to 40%), seismic velocity and brittle strength as well as water uptake of up to 13 wt.% of the ultramafic rock. In this paper, we use numerical models to study the amount and extent of serpentinization that may occur at mid-ocean ridges and its effects on fluid flow within the lithosphere. The two dimensional, FEM model solves three coupled, time-dependent equations: (i) mass-conserving Darcy flow equation, (ii) energy conserving heat transport equation and (iii) serpentinization rate of olivine with feedbacks to temperature (exothermic reaction), fluid consumption and variations in porosity and permeability (volume changes). The thermal structure of the ridge is strongly influenced by rock permeability in addition to the spreading velocity of the ridge. Increased rock permeability enhances hydrothermal convection and results in efficient heat mining from the lithosphere whereas higher spreading velocities result in a higher thermal gradient. Serpentinization of the oceanic mantle, in turn, depends on the aforementioned, competing processes. However, serpentinization of mantle rocks is itself likely to result in strong variations of rock porosity and permeability. Here we explore the coupled feedbacks. Increasing rates of serpentinization lead to large volume changes and therefore, rock fracturing thereby increasing rock porosity/permeability while as serpentinization reaches completion, the open pore space in the rock is reduced due to the relative dominance of mineral precipitation. Although, variations in the relation between porosity and permeability and serpentinization before the reaction reaches completion do not significantly affect the degree of serpentinization, we find that unreasonably large portions of the mantle would be serpentinized if rock closure does not occur at the final reaction stage. The amount of water trapped as hydrous phases within the mantle shows a strong dependency on the spreading velocity of the ridge with water content ranging from 0.18 × 105 kg/m2 to 2.52 × 105 kg/m2. Additionally, two distinct trends are observed where the water content in the mantle at slow-spreading ridges drops dramatically with an increase in spreading velocity. The amount of water trapped in the mantle at fast-spreading ridges, on the other hand, is lower and does not significantly depend on spreading velocity.

Description: We present 2D and 3D numerical model calculations that focus on the physics of compositionally buoyant diapirs rising within a mantle wedge corner flow. Compositional buoyancy is assumed to arise from slab dehydration during which water-rich volatiles enter the mantle wedge and form a wet, less dense boundary layer on top of the slab. Slab dehydration is prescribed to occur in the 80–180 km deep slab interval, and the water transport is treated as a diffusion-like process. In this study, the mantle's rheology is modeled as being isoviscous for the benefit of easier-to-interpret feedbacks between water migration and buoyant viscous flow of the mantle. We use a simple subduction geometry that does not change during the numerical calculation. In a large set of 2D calculations we have identified that five different flow regimes can form, in which the position, number, and formation time of the diapirs vary as a function of four parameters: subduction angle, subduction rate, water diffusivity (mobility), and mantle viscosity. Using the same numerical method and numerical resolution we also conducted a suite of 3D calculations for 16 selected parameter combinations. Comparing the 2D and 3D results for the same model parameters reveals that the 2D models can only give limited insights into the inherently 3D problem of mantle wedge diapirism. While often correctly predicting the position and onset time of the first diapir(s), the 2D models fail to capture the dynamics of diapir ascent as well as the formation of secondary diapirs that result from boundary layer perturbations caused by previous diapirs. Of greatest importance for physically correct results is the numerical resolution in the region where diapirs nucleate, which must be high enough to accurately capture the growth of the thin wet boundary layer on top of the slab and, subsequently, the formation, morphology, and ascent of diapirs. Here 2D models can be very useful to quantify the required resolution, which we find for a 1019 Pa · s mantle wedge to be about 1 km node spacing for quadratic-order velocity elements.