Background & Summary

In the southeast Brazilian Margin at latitude 20°S is located the Vitória-Trindade Ridge (VTR), a group of volcanic islands and carbonate banks that splits the deep Brazilian margin (>1000 m) in two oceanographic domains, north and south of the VTR1. Formed by submarine volcanism since the Oligocene, the VTR seamount chain extends 1200 km offshore the Brazilian coast and is composed by numerous volcanic seamounts and carbonate banks2,3,4,5 (Fig. 1). The seamounts of the VTR rise up to 5 km from the oceanic crust seafloor, with multiple of the seamount’s seafloor reaching the photic zone (Besnard, Jaseur, Columbia and Davis Bank). Consequently, the VTR represents an important barrier for ocean currents flowing along the Brazilian margin, which resulted in a complex and heterogenous environment regarding its physical, geological and biological constituency and evolution6,7,8. Although rich and complex, little is still known about the seafloor and subsurface morphology of the region around the VTR.

Fig. 1
figure 1

(a) Study area and the track (white line) of the CORE-VTRCC Cruise across the Vitoria-Trindade Ridge (VTR) and the Columbia Channel (white dashed line)26. (b) Current configuration in the region includes the south-flowing surficial Brazilian Current (BC), the north-flowing Antarctic Intermediate Water (AAIW) above 1000 m, the south-flowing North Atlantic Deep Water (NADW) between 1000–3500 m and the Antarctic Bottom Water (AABW) below 3500 m13. (c) Topographic profile along the VTR show the presence of numerous seamounts and carbonate banks/platforms. Nomenclatures are from the Brazilian Navy40 and bathymetry from GRMT41.

Full size image

The depositional systems around the VTR are marked by the interaction between ocean currents, volcanic seamounts and carbonate factories9,10,11. In the deep (>3000 m), the steep volcanic slopes block the northward passage of the Antarctic Bottom Water Current (AABW) along the margin, while gravitational flows slip down the ridge to form a mixed contourite-turbidite depositional system11,12. Such conditions may have influenced the formation of the turbidite-contourite Columbia Channel (CC) at the southern region of the VTR, along the displacement of the AABW11,12,13. At water depths between 1000–3000 m, the southward flowing North Atlantic Deep Water (NADW) passes through the bathymetric gaps along the VTR (Fig. 1)13. Within the photic zone (<100 m), reef communities thrived atop the volcanic seamounts, which formed carbonate platforms observed in the bathymetry as top planar surfaces10,14,15. Red algae are the main constituents of the benthic fauna atop the VTR, but a rich and heterogenous habitat developed14,15. The action of waves and surface currents over the carbonate factories triggered the formation of rhodolits, carbonate nodules formed by coralline red algae that move across the platform’ surface under the action waves and currents14. Rhodolits thrive within the VTR, under the influence of the Brazilian Current14,15.

In order to better characterize the marine environment around the VTR, the Cruise CORE-VTRCC was organized in late 2021 by the Oceanographic Institute of the University of São Paulo in partnership with the Brazilian Navy (Directorate of Hydrography and Navigation – DHN), to acquire geophysical data (seismic data, bathymetry, magnetometry) and seafloor samples in the region along the VTR, including the continental slope, the volcanic seamounts, the carbonate factories atop the seamounts, and the deepwater channels of the region. The research vessel NHo Cruzeiro do Sul was assigned for that cruise due to its geophysical and sediment sampling capabilities. The cruise departed the city port of Niterói on October 18th of 2021 and returned on November 6th of 2021, to a total of 18 days of cruise activities. The scientific group was composed by 13 students and researchers from multiple Brazilian universities. In total, the cruise completed 1873.70 nautical miles, surveying 1513.50 nautical miles with multibeam bathymetry and sub-bottom profiler, 14 multichannel seismic lines, 31 XBT (Expendable Bathythermograph) profiles, and 9 sediment samples (Fig. 2). Seafloor depth across the survey area ranged from ~80 m atop the seamounts down to 5200 m near Trindade Island. Survey boundaries were 23° S, 19.65 ° N, 29.20° E and 40.81° W.

Fig. 2
figure 2

Data acquired during the CORE-VTRCC Cruise. (a) Multibeam bathymetry survey occurred throughout the vessel track with the exception of shallow depths (<100 m). Seafloor samples were acquired atop the Columbia Seamount (V01 and V02), atop the Dogaressa Bank (V03 and V04), atop the Davis Bank (V05 and V06) and atop the Jaseur Seamount. (b) Sub-bottom and sound velocity profiles were also acquired throughout the cruise track but not at <100 m depths. (c) Multichannel Sparker Seismic data was acquired along the Columbia Channel (L01-L03), along the VTR (L07-L15) and at the Rio de Janeiro continental slope (L17). (d) Marine magnetometry was acquired throughout the survey track when sea conditions were safe.

Full size image

Methods

High-resolution multichannel seismic

Multichannel seismic data was acquired using a sparker seismic source with 800 tips model Geo-Source 800, up to 7 kJ energy power supply (GeoSpark 2000X) and a 48-channel streamer, all from the Geo Marine Survey Systems. Seismic acquisition and SEG-Y conversion were performed with the software GeoRecorder, also from Geo Marine Survey Systems. The Geo-Source 800 uses a negative discharge for seismic pulse generation which results in a more stable seismic signature with a frequency range between 200–3000 Hz, centered at ~1 kHz. The streamer is composed of three hydrophones per channel, with a total active length of 150 m and 2 × 25 m cable stretches, to a total of 200 m of streamer deployment. Channel distance in the streamer is 3.125 m and channel sampling occurred at 10 kHz. Geographic geometry was supplied by a DGPS antenna placed atop the seismic source configured with the WGS-84 ellipsoid and UTM projection zone 25S. Geometry distance averaged 5 m between source and streamer (dx) and ~20 m between source and first channel (dy; Table 1). The source was placed at 1.5 m water depth while the streamer was at 0.5 m water depth. Energy levels and trigger interval for pulse generation varied across the survey depending on the bathymetric profile (Table 1). Vessel velocity during seismic acquisition averaged 4 knots (Table 1).

Table 1 Main parameters used during the acquisition and processing of multichannel seismic data.
Full size table

Seismic processing was performed with the software RadExPro v2022.1. Basic brutestack seismic processing was made for all seismic lines, following the steps: quality control (data filtering), preprocessing (noise filtering), geometry evaluation, normal moveout correction and stacking. For Lines L01 and L17 was performed an improved seismic processing, including the following additional processing steps: static correction, deconvolution, multiple suppression and migration. Bandpass filtering was performed between 200–1500 Hz. For stacking, common mid points (CMD) gathers were established according to Table 1.

Sub-bottom profiler

High resolution sub-bottom profiler data were acquired employing an SBP-120 from Kongsberg installed in the hull of the ship. Survey geometry came from a SeaPath 330 positioning system with two GNSS antennas without differential positioning. Acquisition occurred in the Kongsberg SIS software with the ellipsoid WGS-84 and a UTM projection with zone 25S. The equipment was operated with one main channel, in a frequency of 2.5–6.5 kHz, with chirp up linear pulse, which results in a maximum vertical resolution ~0.3 ms. The data was logged in Topas raw format and converted to SEG-Y format. The power of the SBP pulse and trigger interval varied along the survey depending on ocean conditions and seafloor depth. Survey velocities ranged from 4–7 knots.

Multibeam bathymetry and sound velocity profiling

Multibeam bathymetry was performed with the deep echosounder Kongsberg EM-122 with 12 kHz of operating frequency. The system was coupled with the SeaPath 330 positioning system with two GNSS antennas. Acquisition occurred in the Kongsberg SIS software with the ellipsoid WGS-84 and UTM projection. The swath width was set to automatic up to 55° with a ping frequency of 40 Hz. Due to the deep characteristics of the EM-122 equipment, multibeam bathymetry was not performed in water depths shallower than 200 m (Fig. 2). Survey velocities ranged between 4–7 knots.

To estimate sound velocity profiles for bathymetric processing, temperature profiles were first obtained for the water column throughout the survey every 50 nautical miles, to a total of 31 profiles. For that we used Expendable Bathythermograph (XBT) Lockheed Martin model T5 to the deeper portion of the survey (down to 1860 m water depth) and the model T7 over the shallower regions (down to 760 m). A salinity profile is then calculated by linearly interpolating the measured temperature profile to the nearest WOA/RTOFS salinity profile. Sound speed values are then calculated from the salinity updated profile using the UNESCO equation16. Vertical extrapolation of sound velocity profiles below 1800 m was performed with the UNESCO equation within the SVP editor using WOA09 CTD profiles down to 5500 m and Challenger Deep CTD profiles down to 12000 m depth17,18,19. Sound velocities were then incorporated during the multibeam bathymetry acquisition directly into the SIS software.

Multibeam bathymetry preprocessing was performed on the software CARIS HIPS from Teledyne during the expedition. A Patch Test calibration of the echosounder was done atop the Columbia Seamount on October 26th, where roll, pitch, yaw, heading and the time delay were calibrated. The corrected values from the Patch Test were then inserted into the processing procedure. Finally, spike removal and surface equalization processing were performed with the software EIVA NaviModel Producer after the expedition.

Marine magnetometry

The SeaSPY magnetometer from Marine Magnetics was used to acquire the magnetic field data. The sensor (torpedo) was towed 250 m behind the vessel and 300 m from the vessel’s GPS. Survey velocities ranged between 4–7 knots. The SeaSPY magnetometer is a Proton Precession (Overhauser) sensor of absolute accuracy, and the resolution of the system is 0.1 nT and 0.001 nT, respectively. The data was acquired in a 10 Hz frequency, from the continental slope to the furthest point from the continent, near Trindade Island. The frequency resulted in measurements for every meter traveled, on average, with the sensor about 30 to 35 m below the sea level. The survey also occurred in parallel to the seismic survey, without any perceptible interference.

In the processing step, the data had low-quality/high-noise measurements removed and was filtered using a non-linear filter of 10 fiducials to eliminate spikes from the data. The quality approved field was reduced using the International Geomagnetic Reference Field (IGRF) to correct long-wavelength effects non-related to crustal features. The resulting Total Magnetic Field was finally filtered using a first-order polynomial filter to highlight medium- to short-wavelength anomalies related to the VTR, producing the Residual Magnetic Field surveyed by the cruise.

Seafloor sediment sampling

During the expedition nine seafloor sediment samples were collected within the Vitória-Trindade Ridge with a Van-Veen grab sampler (3600 cm²) on the top of the volcanic seamounts (Fig. 2; Table 2). Sample V7 collected only rhodoliths, without additional sediments. Grain sizes larger than 40 mm diameter (pebble size) were separated for rhodolith’ measurements. Sediment samples were washed and dried to remove salt grains and other impurities. Sediment samples were bathed in a water container for 48 h, and the water medium was removed with a syringe. Finally, the decanted sediments were then oven-dried at 45 °C for 72 hours. For rhodolith samples, we dried samples at 35 °C for 48 hours. The morphometry of the rhodoliths was classified in spheirodal, discoidal and ellipsoidal based on the measurement of the long (L), intermediate (I) and short (S) axis with a Vernier Caliper (Tables 3, 4)20.

Table 2 Sediment Sampling Sites. Geographic Coordinates were represented in Decimal Degrees (DD).
Full size table
Table 3 Morphometric measurements for Rhodolith samples V1-V7.
Full size table
Table 4 Morphometric measurements for Rhodolith samples V8.
Full size table

Subsequently, samples had their weight measured and carried out the sieving in 13 phi fractions (−1.5, −1.0, −0.5, 0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0 and >4.0) for 10 minutes. Samples sieved were descripted in phi fractions, then were grouped in the following orders, according to Wentworth classification21: granules (−2 to −1 phi), very coarse sand (−1 to 0 phi), coarse sand (0 to 1 phi), medium sand (1 to 2 phi), fine sand (2 to 3 phi), very fine sand (3 to 4 phi) and silt (>4 phi). Mean grain sizes and sorting were analyzed on Gradistat v9.1 software22, using the parameters described by Folk23,24. Sediment samples were also placed on a Petri dish for microscope analysis and photography using the Fiji software25. Microscopic photographs were taken with a Zeiss Stemi DV4 stereomicroscope, at zoom scale of 8x and a total diameter of 26 mm. Grains were then classified in carbonate debris, foraminifers, bryozoans, sponge’s spicules, bivalves, gastropods, crustaceans, echinoderms, and annelida, and its abundance was quantified based on 300 random point counts per sample.

Data Records

The final data and its descriptions are available at the Marine Geoscience Data System repository under the Creative Commons 4.0 International Public License (CC-BY 4.0). Navigation data is provided as an ASCII file at https://doi.org/10.26022/IEDA/331475 with data, hour and UTM and geographic coordinates26. Raw and high-resolution multi-channel seismic reflection are available in SEG-Y format (https://doi.org/10.26022/IEDA/331494 and https://doi.org/10.26022/IEDA/331485)27,28. Name structure follows expedition name VTRCC, method (MCS – Multichannel Seismic), line number, processing step (original data – RAW; stack – STK; and post stack time migration – PostSTM). Subbottom profiler data is available in SEG files (https://doi.org/10.26022/IEDA/331484)29. Name structure follows VTRCC, method (Subbottom Profiler – SBP) and line number.

Raw multibeam swath bathymetry data is available in format Kongsberg.all (https://doi.org/10.26022/IEDA/331477)30 while processed multibeam swath bathymetry is available in MB-System-compatible GSF format (https://doi.org/10.26022/IEDA/331478)31. Name structure follows line number, date and hour, and research vessel. Bathymetry grids are also available in format ESRI ASCII (https://doi.org/10.26022/IEDA/331479)32. XBT-derived Sound velocity profiles are available bundled in format.svp (https://doi.org/10.26022/IEDA/331480)33. Data for each profile includes date and time, position, depth and sound velocities (m/s). Magnetometry data is available as an ASCII file at https://doi.org/10.26022/IEDA/331476 with the headers geographic and UTM coordinates, magnetic field, residual magnetic field, date and time34.

For the seafloor sediment samples, both macro and microscopic photographs are available in JPEG format at https://doi.org/10.26022/IEDA/33148335. The abundance counts of biofacies, as well as grain-size data and morphometric measurements of rhodolits are available in CSV format at https://doi.org/10.26022/IEDA/33148136 and in Excel format at https://doi.org/10.26022/IEDA/33148237.

Technical Validation

In order to avoid magnetic interference during magnetometric acquisition (i.e., vessel interference), a distance greater than three times the ship length was established between the towed fish and the vessel. For multichannel seismic data, quality control procedures were implemented during the acquisition. The seismic streamer was towed alongside the vessel to ensure attenuation of acoustic noise generated by the vessel. Weights were added to the streamer within a specific geometry to maintain streamer depth stable throughout the acquisition, which enhances the seismic signal in the far offsets and prevents the occurrence of burst noise. The geometry of acquisition (Table 1) was tested through the modeling and comparison of the direct wave arrival in the data, during acquisition. Bin sizes were then determined to optimize the fold number in the stacked section. Further processing procedures (i.e., pre-stack migration, noise modeling) may be performed to further improve data results, depending on the investigation goals. Even so, technical limitations of using a short streamer result in a difficulty of properly imaging surfaces/areas with a high gradient, particularly along the continental slope and seamounts, where the seismic traces are often distorted and do not represent the regional stratigraphy.

Bathymetric and SBP surveys were conducted by the Brazilian Navy, which follows the bathymetric technical standards of the NORMAM 50138 and IHO S-4439. A multibeam patch test was conducted prior to the scientific expedition, where pitch, roll, and heave corrections were applied. Since the bathymetric survey covered deep regions, tide correction was deemed unnecessary as tide variations represent less than 1% of the minimum seafloor depth observed. The vertical extrapolation of sound velocity profiles for depths below 1800 m is a major cause of depth uncertainty in the deep part of the survey. Although sound velocity values below the thermocline (>1000 m depths) tend to vary similarly across ocean regions, for each 1 m/s of error between calculated and real sound velocities, we can expect a bathymetric uncertainty of ~1 m at 2000 m depth and at ~3 m at 5000 m depths. For sediment analysis, sedimentological standards of Folk23,24 were followed. The initial and final masses were measured and when those values had a >10% difference, the analysis was repeated.