The western Aleutian Trench remains one of the least characterized hadal environments with respect to trace metal biogeochemistry. Here we present the first dataset of mercury (Hg) concentrations, Hg fluxes, and sedimentation rates derived from excess 210Pb in three hadal sediment cores collected during a 2024 China-Russia collaborative expedition. Surface sediment Hg concentrations range from 59 to 172 ng g- 1, comparable to other Pacific trench systems. Down-core profiles reveal a strong coupling between Hg and total organic carbon (TOC) in two cores, indicating organic matter as the primary carrier phase, whereas a decoupled Hg-TOC relationship at the trench junction suggests enhanced lithogenic control under dynamic hydrodynamic conditions. Calculated Hg fluxes (15.8-22.7 mu g m-2 yr- 1) exceed global deep-sea averages, with substantially higher values (up to 208 mu g m- 2 yr- 1) at the trench junction, reflecting intensified sedimentation and material focusing. Thermodesorption analysis indicates that Hg predominantly occurs in sulfide-bound and mineral-associated forms. We infer that Hg accumulation in the western Aleutian Trench is governed by the interplay between organic matter deposition, tectonically enhanced sediment focusing, and additional inputs from cold seeps and regional volcanism. Despite elevated fluxes, Hg concentrations remain below thresholds associated with significant ecological risk.
Three-dimensional thermohaline reconstruction is crucial for understanding ocean dynamics; however, conventional methods encounter significant limitations. This study proposes a deep learning (DL) framework that integrates acoustic travel time measurements from Pressure-Recording Inverted Echo Sounders (PIES) with satellite-derived remote sensing data to overcome a key limitation of the classic Gravest Empirical Mode (GEM) method-its inability to reliably discriminate between distinct water masses that exhibit identical or highly similar acoustic travel times during thermohaline structure reconstruction. The multi-branch neural network synthesizes multi-source observational datasets to reconstruct profiles of temperature, salinity, and specific volume anomaly, effectively resolving uncertainties caused by the overlapping Kuroshio and South China Sea water masses in the Luzon Strait. The DL-based method achieved higher coefficient of determination (R2) and lower root-mean-square errors, demonstrating significantly improved performance compared to the GEM method. Comparisons with mooring time series and Argo/CTD profiles revealed strong consistency. Applied to PIES observations collected in the Luzon Strait during 2018-2019, the framework successfully reconstructed the three-dimensional thermohaline structures associated with Kuroshio intrusion pathways, including looping, leaking, and eddy-driven patterns. This method achieves a technical advancement of the inversion of PIES array over the traditional GEM approach, offering a cost-effective solution for monitoring complex oceanic regions.
Mesoscale eddies play a crucial role in oceanic variability, yet their impact on ocean bottom dynamics remains less understood. This study investigates such effects using a high-resolution, full-depth mooring system (similar to 680 m) on the continental slope of the northern South China Sea (October 2018-April 2019). Five distinct mesoscale features were identified from temperature anomaly theta' and bottom current variations: one anticyclonic warm eddy, two cyclonic cold eddies, one deep warm intrusion, and one subsurface cold eddy. Within the bottom layer similar to 100 m above the seafloor, eddy-induced temperature anomalies theta'' reached up to +0.3 degrees C in the warm eddy and down to -0.3 degrees C in the cold eddies.The subsurface cold eddy with a cold core at 250 m depth caused a significant temperature decrease (-1.2 degrees C), while its impact on bottom currents was weak. Bottom eddy kinetic energy (EKE) increased twofold to 15 cm(2)/s(2) in the eddies. Bottom turbulent diffusivity was enhanced by factors of 5 and 2 to approximately 1.7x10(-2) and 8x10(-3) m(2)/s in the warm and cold eddies, respectively. The bottom mixed layer (BML) thickness, averaging 23.5 m, generally thinned in the warm eddy and thickened in the cold eddies, likely associated with stronger and weaker stratification, respectively. The intrusion, centered at similar to 450 m, exhibited a temperature anomaly theta' of +0.8 degrees C and EKE of 36 cm(2)/s(2). This intrusion substantially compressed BML by 5 m and intensified the turbulent diffusivity to 1.8x10(-2) m(2)/s. The model reanalysis data suggest that the deep intrusion originated from the remote advection of the warm anticyclonic eddy, facilitated by interaction with the nearby cold cyclonic eddy.
The Rio Grande Rise is a submerged plateau that has received increasing interest due to the presence of critical minerals within ferromanganese crusts that partially cover the plateau. The formation, distribution and preservation of the crusts are partially controlled by the hydrodynamic processes that are active around the region. This study was carried out to describe the erosional and depositional processes that have shaped the geomorphology of the Rio Grande Rise in an area where three scientific cruises have recently occurred. Particularly, we investigated the role of ferromanganese crusts in seafloor erosional processes and their interaction with deep ocean currents. For that, we combined the analysis of multibeam bathymetry, sub-bottom profiles and ROV imaging data. Multibeam bathymetry reveals the heterogeneity of submarine morphologies, composed of escarpments, slope scars, submarine canyons, subcircular depressions and erosive channels. Sub-bottom profiles show that deep currents have eroded the plateau flanks, forming contourite drifts only where ferromanganese crusts are absent. ROV images show that current directions are variable, and obstacle scouring occurs widespread. Overall, we show that the presence of ferromanganese crust is crucial in preventing erosion. Lastly, outcrops of basaltic escarpments expose the various characters of volcanic deposits in the Rio Grande Rise, while sediment mass movements including debris flows occur along the scarps. The integrated results shed light on the nature of the seabed erosional processes that affect the region and how the presence of ferromanganese crusts is shaping the morphology of the Rio Grande Rise.
Mesoscale eddies are energetic oceanic structures that contribute substantially to the ocean's kinetic energy. They frequently occur in pairs, known as eddy dipoles or mushroom-like eddies, and are potentially ubiquitous in the global ocean. These dipoles generate an intense surface current along their central axis, forming a narrow jet that is considerably stronger than the surrounding flow. We investigate the influence of the central jet of an eddy dipole on the properties and spectral shape of wind waves propagating in opposition to the current. To capture the complex modifications induced by wave-current interactions, a novel inversion method is applied to retrieve directional wave spectra from Synthetic Aperture Radar (SAR) data acquired along an Envisat ground track that crossed an eddy dipole on the Sao Paulo Plateau in the Southwestern Atlantic. The observational analysis is complemented by idealized numerical simulations using state-of-the-art wave and ocean circulation models. Furthermore, the positions and structures of the eddy dipoles and their associated jets are identified using absolute dynamic topography and derived absolute surface geostrophic velocities from satellite altimetry. The results demonstrate that, under favorable geometric conditions, eddy dipoles act as focal lenses for surface waves, refracting them toward the central jet and substantially enhancing wave height-and thus steepness-in the counter-current region. The presence of the dipole broadens the wave spectra in both frequency and directional space, with the magnitude of this broadening modulated by the initial directional spreading of the waves.
Aristeid shrimps are valuable deep-sea resources along the Brazilian Meridional Margin (BMM). Between 2002 and 2009, three species were heavily exploited by bottom trawlers within a narrow depth range of the continental slope (700-800 m), resulting in important stock reductions and ecosystem impacts. Limited scientific records indicated, however, that these species occurred in much deeper zones (up to 1800 m) mostly unavailable to the trawl fisheries. This study aimed to identify and map suitable habitats of Aristaeopsis edwardsiana in the northern BMM and to evaluate the potential role of submarine canyon systems in shaping its spatial distribution. The study area encompassed continental slopes of the Campos and Espirito Santo basins, characterized by numerous cross-slope canyon systems. Species distribution was modelled using occurrence data from commercial trawl surveys and environmental predictors describing geomorphology and water properties derived from highresolution datasets. An ensemble habitat suitability modelling framework was implemented by combining three modelling approaches: Generalized Linear Models, Maxent, and Random Forest. Water column properties (bottom temperature and salinity), current speed and seafloor aspect (easterness) tended to exhibit a high explanatory role in all models. In general, suitable habitats included water masses mixture zones with reduced bottom flow dynamics. These habitats were most likely to occur between 200 and 800 m depths throughout the entire study area, and particularly within canyon systems where suitable habitats are likely to extend to over 1800 m depths. These results indicate the existence of shrimp distribution areas not previously identified in the Brazilian Meridional Margin at this level of spatial resolution, which may function either as refugia from trawl fisheries or as spatial connectors between deep and mid-slope populations. Given that fisheries catches were dominated by mature individuals, canyon systems may contribute to maintaining population connectivity between spawning and pre-reproductive habitats in the northern BMM.
Understanding trophic sources and nutritional frameworks is fundamental for deciphering the structure and function of ecological communities. In cold seep ecosystems, chemoautotrophic bacteria constitute the primary producers, sustaining the food web. However, distinct geochemical processes associated with these bacteria may lead to significant divergence in seep food web dynamics compared to typical marine environments. While stable carbon and nitrogen isotopes (S13C and S15N) combined with rare earth elements (REEs) are established tracers for predator-prey relationships and trophic positions in conventional marine ecosystems, their application in cold seep systems remains limited. Here, we systematically analyzed S13C and S15N signatures (202 measurements across 42 species) and REE patterns in diverse organisms from the Haima cold seep (South China Sea). Our findings revealed that S13C and S15N effectively trace trophic linkages, exhibiting progressive enrichment with increasing trophic levels (R2 = 0.68). Notably, sulfur-oxidizing taxa display exceptionally broad S15N values (-2.6%o to 9.8%o), potentially attributable to variable nitrogen sources and sediment burial depths. Conversely, REEs lack systematic trophic transfer patterns observed in typical marine settings, likely due to: (1) heterogeneous nutritional sources among seep macrofauna, and (2) preferential assimilation of light REEs (LREEs) by aerobic methane-oxidizing bacteria. This study evaluates the applicability of S13C-S15N-REEs frameworks in chemosynthetic food webs, establishing a methodological foundation for future research on trophic interactions and elemental cycling in chemosynthesis-driven ecosystems.
The strong thermohaline Subpolar Front divides the Japan/East Sea (JES) into subtropical and subarctic regions, making it one of the most eddy-rich seas in the world. However, eddies in the northern JES, outside the frontal zone, have been far less studied than those in the south. Using the AMEDA algorithm and altimetry data from January 1, 1993, to January 1, 2025, we detected mesoscale anticyclonic eddies (AEs) and cyclonic eddies (CEs), and identified hotspots for ‘subarctic’, ‘subtropical’, and frontal eddies. The regions, where eddy splitting and merging occur frequently, have also been identified. The kinematic characteristics of eddies in both parts of the sea were calculated and compared, and the mechanisms behind their regular generation in different regions were investigated. We focused on the previously unstudied northeastern part of JES, where several hotspots for the regular formation of both anticyclonic eddies (AEs) and cyclonic eddies (CEs) have been identified. Notably, vortex structures form recurrently within confined areas in the region of bifurcation of the Tsushima Warm Current. A clear spatial segregation of AEs and CEs, persisting in both cold and warm seasons, was observed; the eddies are largely mutually exclusive, with CEs being practically absent where AEs regularly form, and vice versa. CEs (AEs) are predominantly generated on the offshore (onshore) side of the boundary currents. Various diagnostics, such as eddy kinetic energy and its gradient, the vorticity field and its zonal gradient, and the horizontal strain tensor, demonstrate that eddies are generated in the same regions where the observation frequency is high.
Body size is one of the most significant functional trait of an organism, while the size spectrum defines the functioning of a multispecies community. Body size affects energetics, which is why changes in body size have implications for resilience to climate change at both individual and community scales. To investigate the influence of changing environmental conditions on body size, secondary production, and carbon demand of deep-sea benthos in the Fram Strait, the passage between NE Greenland and the Svalbard archipelago, we analyzed meiofauna and macrofauna samples collected in 2000, 2010, and 2017 at station depths ranging from 973 m to 5561 m. Surface sediments were collected using a box corer to assess the influence of environmental changes on the structure (biomass size spectra) and functioning (secondary production and carbon demand) of benthic communities. At depths between 900 and 1500 m (RIDGE), macrofauna individual biomass decreased from 2000 to 2017. However, at depths below 2000 m, it remained at a similar level across the studied years. In all bathymetric zones, macrofauna secondary production and carbon demand remained stable across the three years. In contrast, nematodes individual biomass, total biomass and secondary production decreased from 2000 to 2017. Results of DISTLM model confirmed that nematodes did not rely on the bacterial activity (indicated by FDA). Responses of benthic biomass size spectra to the environmental changes varied among bathymetric zones. The decreasing nematodes standing stocks and magnitude of its functioning may be due to an increasing interference competition with macrofauna.
Due to a lack of direct observations of global ocean surface currents (OSC), the spatial distributions and characteristics of its ageostrophic component are unclear. This study used surface drifters to analyze the global spatial distribution and contribution of ageostrophic currents. The ratio of the ageostrophic current to the OSC is 63% (56%) on the global average, derived from hourly (6-hourly) drifters, with a high ageostrophic current ratio in the eastern tropical and mid-latitude subtropical regions (>70%) and a low ratio in western boundary current systems and Antarctic Circumpolar Current regions (<50%). Furthermore, this study quantitatively calculates the global distribution of the ratio contributed by different types of ageostrophic components to OSC. The decomposition shows that the hourly drifter-derived ageostrophic current includes effects from near-inertial oscillation (16% in the OSC), Ekman current (9%), tide (6%), Stokes drift (2%), and unclassified motions (30%), which include inertial gravity waves and submesoscale motions. The latter term indicates the presence of high-frequency currents with intensified horizontal speeds, which can potentially impact material transport and energy cascade dynamics.
We present new biogeochemical and mineralogical analyses of ocean waters, suspended particulate matter (SPM) and deep seafloor sediments (DSS) across three different sites of the eastern slope of the Atacama Trench in front of northern Chile. These analyses have revealed a singular chemistry of upwelling deep waters from the Humboldt Current with respect to the overlying ocean water. Scanning electron microscopy (SEM) on SPM (including marine snow) collected at different depths in the water column (20 m to 4500 m) show a high diversity of solids, including biological debris (settling phytoplankton biomass dominated by diatoms), detrital minerals (e.g. quartz, K-feldspar, plagioclase, oxides, clay minerals), authigenic minerals formed in the water column (e.g., calcite, barite), and microplastic fibers. We observed abundant colonization of settling biological particles by rod-shaped bacteria in the sharpest part of the oxycline (at around 50 m depth) and intense heterotrophy (including different groups of sulfate-reducing bacteria) in the oxygen minimum zone (OMZ). The fast degradation of biological matter in the upper layer results in an extremely limited vertical export of organic carbon to the bathyal and abyssal zones. However, the detection of abundant phytoplankton remains (e.g. diatom frustules), lipid biomarkers from the phototrophic active zone (e.g. phytol, fatty acids and sterols) and certain pigments (e.g. carotenoids) in the upper cm of the bottom sediments suggest that a small fraction of organic matter still reaches the deep ocean floor, allowing for microbial activity like sulfate reduction which favors the formation of pyrite below the water/sediment interface.
Building upon previous work (Yang et al., 2022) which provided the first depth-resolved Net Primary Production (NPP) estimates for the open-ocean Gulf of Mexico (GoM) using profiling float data, this study presented three major advancements. First, the performance of the Visible Infrared Imaging Radiometer Suite (VIIRS) satellite product was tested for float data correction. The results showed it can effectively replace the retiring Moderate Resolution Imaging Spectroradiometer (MODIS) product used previously, with cumulative differences in water column-integrated NPP within 2% over a three-year period. Second, the annual NPP remained relatively stable between 1.5 & times; 105 and 1.8 & times; 105 mg C m- 2 yr- 1, consistent with the previous estimate of 1.6 & times; 105 mg C m- 2 yr- 1. The new data (2020-2024) corroborate the distinct seasonal controls identified earlier (2011-2015): surface NPP peaks in winter and is likely driven by nutrient availability, while subsurface NPP is more controlled by light availability and peaks from spring to early fall. Third, the persistence of a subsurface NPP maximum was confirmed. While this phenomenon can occur throughout the year, it is most prevalent during the spring and summer. Furthermore, the contribution of NPP from below the mixed layer to the entire water column was found to be significant, ranging from 0% in winter to as high as 70% in the spring and summer.
Internal tides contribute significantly to ocean mixing and circulation, yet their dynamics and energetics in semi-enclosed seas, including the Mediterranean, remain poorly quantified. This study investigates internal tides in the Central Mediterranean Sea using a high-resolution (1 degrees/ 60) ROMS model, that agrees well with available observations. Total barotropic-to-baroclinic energy conversion is estimated at 147 MW with similar to 72% occurring in the Sicily Strait. The diurnal K-1 signal dominates from northwestern Sicily to the Adventure Bank, where slope-trapped Kelvin waves induce strong baroclinic shear, producing low Richardson numbers (Ri <025) and elevated bottom turbulent kinetic energy dissipation consistent with thermistor observations. Northeast of Pantelleria, the K-1 tide splits into two energy flux branches, while additional K-1 generation near thermistor station AT propagates energy northeastward toward the western Sicilian shelf. Although contributing only similar to 6% of total conversion, dominated by M-2 tides, overtides modulate conversion variability, while turbulent mixing remains elevated. Over the Malta Plateau, baroclinic conversion is predominantly diurnal, with flux pathways agreeing with recent observational studies. A theoretical framework decomposes baroclinic energy flux divergence into pressure work, advection, and diffusion; neglecting advection and diffusion leads to misrepresentation of local energy budgets. While these terms compensate in the Sicily Strait and across the domain, the balance breaks down over the Sicily subregion, the Malta Plateau and Messina Strait, resulting in enhanced mixing and residual circulation. By providing the first systematic energetic characterization of internal tides in this region, this study identifies key generation sites, flux pathways, and dissipation hotspots, offering a framework for understanding internal tide-driven mixing in other marginal seas worldwide.
Eddies can retain water for several months to over a year, but their water properties may change as they move. In the potential spicity-density (pi-sigma) space, variations in water mass properties are decomposed into a diapycnal component Delta sigma and an isopycnal component Delta pi. This study quantifies the change rate V_lambda of water properties of global eddies in this space, where V_lambda = d[sqrt(Delta sigma<^>2+Delta pi<^>2)]/dt, with a mean value ranging from 0 to 0.03 kg/m<^>3/day. Using Argo floats and altimetry data, more than 150,000 eddies were sampled by Argo floats, providing over 600,000 profiles from 2000 to 2022. Results show water properties near eddy centers are more conserved than at edges, with higher V_lambda during eddy generation and extinction and V_lambda decreasing with depth. The 3-D structure of V_lambda can be approximated by a unified solution by solving the spicity equation on isopycnal layers. Weaker V_lambda values are observed in Lagrangian eddies than in Eulerian eddies, and the Eulerian view overestimates V_lambda by similar to 50% due to the misattribution of property changes to trapped water that is actually being exchanged. Analogous to decomposing kinetic energy into components, the contribution of isopycnal and diapycnal components to V_lambda is given by the fraction of total variance. It is found that the isopycnal component accounts for 50%-70% of V_lambda in 0-1000 m, and a larger V_lambda is observed in the eddy flanks adjacent to strong currents (e.g., Kuroshio Extension, Gulf Stream). This pi-sigma decomposition approach may be a useful tool for the eddy research community.
In this study, a Pressure Core Processing and Analysis System (PCPAS) was developed to facilitate the integrated handling of intact pressure cores. The system enables the transfer of cores up to 4 m in length from pressureretaining coring tools, followed by non-destructive characterization and precise sectioning into subsamples for storage-all while strictly maintaining in-situ temperature and pressure conditions. The integrated nondestructive analysis, featuring X-ray scanning and P-wave velocity measurements, allows for the rapid identification of hydrate-bearing zones, which is critical for optimizing subsample selection. Storage modules range from standard pressure-holding cylinders to advanced instrumented chambers for subsequent physical and mechanical characterization. The system's pressure-handling integrity was validated through laboratory tests at 35 MPa. Furthermore, multiple sea trials demonstrated the system's operational stability, maintaining core temperatures between 3.0 degrees C and 4.5 degrees C and restricting pressure fluctuations to within +/- 1.5% of the target values.
Near-inertial waves (NIWs) in the Bay of Bengal (BoB) were investigated using data from five deep moorings deployed between 2018 and 2019. The paper describes the kinematic and some dynamical properties of vertically propagating NIWs generated by monsoon events and summer and winter pre-monsoon cyclones, in the presence of background mean flows and mesoscale eddies. The pre-monsoon cyclones rapidly developed nearinertial currents in the mixed layer, while propagation of NIWs into the thermocline occurred more gradually. NIWs carried energy downward to depths of up to 200 m, with maximum vertical propagation speeds of 20 md- 1 and vertical and horizontal wavelengths ranging from 300 to 1000 m and 300-1400 km, respectively. On average, NIWs contributed about 50% of the total current shear, increasing to 50%-80% or more during cyclone events. In the southern BoB, NIWs interacted with monsoon-driven currents, the cyclonic Sri Lanka Dome (SLD), and a large anticyclonic eddy (ACE). In the south-central BoB, interactions occurred with intraseasonal baroclinic currents. Energy and shear penetrated deep in the thermocline within the ACE where vertical wavenumber exhibited linear growth with rates similar to(1-8) x 10-4 cpm d- 1 in relatively uniform background shears similar to the theoretical prediction of Young and Jelloul (1997). In contrast, NIW energy and shear within the SLD remained confined to shallower depths. Collectively, these observations gathered under a range of background conditions describe various possible spatial and vertical structures of NIWs in the BoB.
The impact of tides on the thermohaline function of the North Aegean Sea is hereby investigated. Two similar hindcasts were performed using a high-resolution model, covering the period 1986-2013. In the first simulation, no tidal forcing was included. In the second, eight (8) harmonic components M2, K2, N2, S2, K1, O1, P1, Q1 from the Oregon State University Tidal Atlas were used. The model results showed significant differences between the two simulations. The most interesting was the production of more dense deep waters (sigma(theta)> 29.3 kg/m(3)) during the Eastern Mediterranean Transient (EMT) when the tidal signal was included compared to when it was not. This led to a better representation of the long-term variability of the North Aegean deep basins, provided insight for the role of tides in the lateral exchanges between the south and north part of the Aegean Sea and revealed an unexpected mechanism that redistributes vertical stratification rather than decreasing it.
The Qiongdongnan Basin (QNDB) in the northern South China Sea is a key region for hydrocarbon exploration, yet research on its Quaternary source-to-sink system remains limited. This study aims to reconstruct the provenance and sediment-routing pathways of the Quaternary Ledong Formation in the QNDB, integrating geochemistry, heavy mineral assemblages, and seismic geomorphology. Our analysis identifies three primary sediment sources: the Red River, the Kontum Uplift, and the Hainan Uplift, with clear spatial differentiation across the basin. The Yacheng area receives mixed inputs from all three sources, while the Songtao area is dominated by Hainan-derived sediments. The Lingshui area is primarily influenced by the Kontum Uplift with additional Red River contributions, and the Baodao area is primarily impacted by Red River-derived fine-grained sediments. Seismic data reveal a west-high-east-low paleotopographic gradient that governs sediment routing, with the western basin serving as the dominant sediment-entry corridor. The sedimentary contributions from the Kontum Uplift formed large submarine fans in the west, while fine-grained mud from the Red River accumulated widely across the basin. This study provides a refined source-to-sink model for the QNDB, highlighting the dominant role of western sediment influx and the strong geomorphic control exerted by the basin’s paleotopography, offering valuable insights for deep-water reservoir prediction and exploration in similar passive-margin basins.
From a Lagrangian perspective, a three-dimensional particle tracking approach is employed to investigate the dynamic processes governing the spatiotemporal evolution of typical chlorophyll structures, as regulated by mesoscale eddies through advective transport and vertical motion. The results demonstrate that the formation and maintenance of typical chlorophyll structures are driven by an eddy-mediated cyclic exchange of water masses, which facilitates persistent vertical turnover within a three-dimensional flow field, rather than resulting solely from passive advection. During this process, water masses sequentially enter the chlorophyll structures in discrete batches. Each batch can be classified into ascending, quasi-stationary, and descending states, with their relative dominance shifting over time. A comprehensive analysis reveals distinct stage-dependent dynamics throughout the lifecycle of chlorophyll structures, which can be divided into developing, maintaining, and decaying stages. Statistical analysis of 560 chlorophyll structures shows that the maintaining stage typically lasts about 7 days, during which a marked increase in chlorophyll concentration occurs around the third day. The decaying stage is characterized by predominant downward particle movement accompanied by a decline in chlorophyll concentration. From a Lagrangian viewpoint, this study elucidates the mechanisms governing the formation and evolution of typical chlorophyll structures under the influence of mesoscale eddies, providing new insights into oceanic energy-material coupling processes.
The U.S. North Pacific harbors some of the densest and most diverse cold-water coral and sponge communities globally, yet quantitative data for depths below 900 m in the Gulf of Alaska and Aleutian Islands remain scarce. Most records originate from <300 m, despite similar to 80% of Alaska's seafloor exceeding 200 m depth. Using Remotely Operated Vehicle video imagery from two NOAA Ocean Exploration expeditions in 2023 (Seascape Alaska 3 and 5), we conducted a quantitative, image-based assessment of deep-sea coral and sponge communities across 15 previously unvisited sites spanning 380-3200 m depth. From 15,531 observations, we documented 164 distinct morphotypes-90 Porifera and 74 Cnidaria-substantially extending known distributions for multiple taxa, including five sponge genera new to Alaska and the northernmost Pacific record of the coral genus Umbellapathes. Density and diversity peaked along the margins of the Oxygen Minimum Zone (similar to 500-1800 m; O-2 <= 1.43 mg L-1). Additionally, eight high-density aggregations were identified using quantitative spatial criteria, reaching densities up to 20.68 individuals m(-2). Community composition was structured by a hierarchy of drivers, with oceanographic setting explaining most variation. These findings highlight the importance of oceanographic context in structuring deep-sea biodiversity and provide a baseline for future ecological, taxonomic, and conservation research in North Pacific deep-sea ecosystems. Understanding the distribution and drivers of cold-water coral and sponge communities is vital for anticipating ecosystem responses to future ocean scenarios and for designing effective management strategies to safeguard these vulnerable deep-sea habitats.
