Comparable deep-water benthos datasets collected by epibenthic sledges (EBS) with an epibenthic and a suprabenthic netsampler in the Atlantic Ocean have been gathered since 2006. They were collected during the international research expeditions: ANDEEP-SYSTCO II, BIOPEARL I, DIVA1-3, IceAGE1-3&RR, IceDIVA1,2, JR275 and Vema-TRANSIT. While EBS diversity data at high taxon level were published for ANDEEP_SYSTCO II, DIVA1-3 and Vema-TRANSIT, unpublished diversity data for BIOPEARL I, IceAGE1-3&RR, IceDIVA1, 2 and JR275 came from sample databases at DZMB Senckenberg and British Antarctic Survey, and are published here for the first time. In total, diversity data for 143 EBS deployments from 13 expeditions were available for analyses based on identification on 50 taxon levels, including phyla, subphyla, classes and orders. During all 13 expeditions EBS with an epibenthic and a suprabenthic netsampler following sampler sizes and height , enabling comparability of samples. This type of EBS was a suitable device for sampling small benthic fauna on and above the seabed, including macrofauna and small-sized megafauna. We analyse pan-Atlantic benthic data from a range (119m - 8338m) of depths. For the pan-Atlantic analyses we defined seven regions to pool EBS locations based on their position North and South of the Equator and to the mid-Atlantic Ridge (MAR): East and West of the MAR, the Vema Fracture Zone as a gap in the MAR, the Southern Ocean south of the MAR and the Puerto Rico Trench as a deep-sea trench. In this study we included data for 41 higher taxa of the initially separated 50 taxa ranging from phyla to orders. The environmental parameters for this study were provided by Bio-ORACLE, which identifies mean values for different physical and chemical variables over a 14 year time period through a combination of satellite and in-situ measurements (2000 - 2014), at a resolution of 5 arcmin. 4 multivariate analyses (principal components analysis, analysis of similarities, similarity of percentages and BioEnv BEST) were carried out on standardised abundances using PRIMER software, the results and parameters of which are presented in this dataset. Funding over the years for the sample collection and analyses was provided by multiple NERC grants and international grants. Katrin Linse, Peter Enderlein and Huw J. Griffiths were part of the British Antarctic Survey Polar Science for Planet Earth Programme funded by The Natural Environment Research Council (NERC) [NC-Science] and included the funding for the RSS James Clark Ross expeditions BIOPEARL I and JR275. This study was directly funded by the IceAGE_RR and IceDIVA grants by the German Science Foundation (DFG) and Bundesministerium fur Bildung und Forschung (BMBF) under grant numbers MSM75 (MerMet17-5), SO280 and SO286 to PIs Saskia Brix, James Taylor and Katrin Linse. Funding for previous expeditions that provided data were: IceAGE1-3, BR3843-3-1& 4-1, & SO276 (MerMet17-6). James Taylor and Karlotta Kurzel were supported via DFG grant GPF 20-3_087 as part of the IceDiva project 2021 - 2022 by DFG. Anne-Nina Lorz was funded by the German Science Foundation Project IceAGE Amphipoda, LO2543/1-1. Additionally, Angelika Brandt was granted funding (SO 237, Forderziffer 03G0237A) by the Bauer Foundation for the VEMA-Transit project. Inmaculada Frutos was supported through the junior research group''''Vema TRANSIT. Puerto Rico Trench, Vema Fracture Zone and Abyssal Atlantic Biodiversity Study'''' as part of the project ''''Biodiversitatnachhaltige Ressourcennutzun'''' (Aktenzeichen T237/25054/).

The Antarctic food webs represent two entire above-belowground communities from Signy Island Reference Sites on Signy Island, one of the South Orkney Islands in the Maritime Antarctic. The two communities are a dry moss community (Antarctic dry tundra) and a wet moss community (Antarctic wet tundra). These two communities were the focus of intensive biological study by personnel from the British Antarctic Survey over the course of a decade in the 1970''s, of which the results were finally compiled into a meta-analysis by Davis (1981). The Jacobians of the dry and wet tundra were calculated by Neutel and Thorne (2014) using the empirical biomass and flow data of the respective systems from Davis'' analysis. The Jacobians represent the interaction strengths of the species in the two food webs, evaluated at equilibrium.

The soil food webs in this collection represent seven belowground communities from native and agricultural soils. The seven communities are from experimental research sites in the USA, Sweden and the Netherlands. The Jacobians of the seven food webs were calculated by de Ruiter et al. (1995) using the empirical biomass data of the respective systems, and inferring steady-state biomass flow data using a procedure described by Hunt et al. (1987), see further references below. The Jacobians represent the interaction strengths of the species in the two food webs, evaluated at equilibrium.

The soil food webs in this collection represent a total of 32 belowground communities studied by Neutel et al. (2007), from two natural successions in sandy dune soils: one on the Waddensea Island of Schiermonnikoog in the north of the Netherlands and the other at Hulshorsterzand, on the Veluwe, in the central Netherlands. The study sites, which constitute the two gradients, represent four consecutive stages in chronosequences of early primary vegetation succession, increasing in aboveground and below-ground productivity. The Jacobians of the 32 food webs (two series, four stages with four replicates per stage) were calculated by Neutel et al. (2007) from observed average biomass data of the respective systems, and inferring steady-state biomass flow data using a procedure described by Hunt et al. (1987). The Jacobians represent the interaction strengths of the species in the two food webs, evaluated at equilibrium.

In 2018 RRS James Clark Ross investigated the marine benthic biodiversity of the Prince Gustav Channel area and the macrobenthic molluscan fauna collected by epibenthic sledge (EBS) has been assessed for species richness, abundance and assemblage composition as well as for functional traits. In total 20,307 mollusc specimens assigned to 50 morphospecies and 4 classes (Solenogastres, Bivalvia, Gastropoda and Scaphopoda) were identified. Assemblage analyses across the Prince Gustav Channel area did not show apparent pattern or separation across depth, taxon or station. To set the bivalve dataset into a wider context, unpublished bivalve species richness and abundance data from EBS collected stations in the area influenced by the Weddell Gyre were added. This doi dataset provides data for 1) PGC EBS locations, 2) PGC EBS molluscan abundances, 3) PGC molluscan functional traits, 4) Weddell Gyre EBS stations (300 - 2000 m depth), 5) Weddell Gyre EBS bivalve standardised 1000 m trawl length abundances (300 - 2000 m depth). Funding was provided by NERC urgency grant NE/R012296/1 ''Benthic biodiversity under Antarctic ice-shelves - baseline assessment of the seabed exposed by the 2017 calving of the Larsen-C Ice Shelf''.

Benthic Macrofaunal abundance and biomass data collected during a series of ship-board sampling campaigns in the Barents Sea. As part of the NERC funded Changing Arctic Ocean programme, in July of 2017, 2018 and 2019, the RSS James Clark Ross followed a North-South transect at 30 degrees Longitude. In 2017 six stations were sampled (B3, B13, B14, B15, B16, B17); in 2018 seven stations were sampled (B3, B13, B14, Xs, B15, B16, B17); and in 2019 five stations were sampled (B3, B13, B14, B15, B16). Macrofaunal samples were collected using a USNL corer (surface area 0.1m<sup>2</sup>). On recovery all faunal samples were fixed and preserved with 10% buffered formaldehyde solution. The faunal samples were transported to Plymouth Marine Laboratory where the fauna were extracted, identified to species level where possible using the most up to date literature available, and biomassed (blotted wet weight) to species level. A reference collection has been created containing an example of each taxon to ensure maximum quality control was maintained by the three analysts that conducted the species identification and for subsequent cruise data generated. Funded by the NERC Changing Arctic Ocean Seafloor (ChAOS).

Multiple images of the seafloor at six sites across a broad latitudinal range in the Barents Sea in the Arctic were collected in July 2017 on the month long scientific cruise JR16006. The dataset includes environmental variables for each accompanied image. Each image (406 x 341mm) has density of fauna from different functional groups. We have 13 different functional groups based on other similar studies. The aim was to look at the effect of climate change in the Arctic on the biology of the seafloor. Funding was provided by the NERC Changing Arctic Oceans ChAOS project.

High-resolution X-ray computed tomography images of two deep-sea bamboo corals (Acanella arbuscula, Johnson, 1862; Keratoisis sp., Wright, 1869) collected from Baffin Bay and Davis Strait during a research expedition on board the CCGS Amundsen in July-August 2021. Corals were imaged using Micro-Focused X-Ray Computed Tomography at the Micro-Vis X-ray Imaging Centre (Southampton, UK) to non-destructively investigate their skeletal architecture, calcification strategies and growth patterns. Supported by a National Environmental Research Council Funded (INSPIRE) PhD [grant number NE/S007210/1, 2019-2027, awarded to T.J.W] and the National Research Facility for Lab X-ray CT (NXCT) [EPSRC grant number EP/T02593X/1].

Macrozooplankton and nekton were collected with a Rectangular Midwater Trawl 25 (RMT25) over several visits to the sustained observation location P3 (52.70 S, 40.26 W) in the northern Scotia Sea during November and December 2017. The work was carried out as part of the NERC Large Grant, COMICS (Controls on Mesopelagic Interior Carbon) on board the RRS Discovery (cruise DY086). The RMT25 net hauls sampled between 10 and 500 m depth, with the water column divided into 2 depth intervals (10-250 m and 250-500 m). A total of 6 hauls were obtained during 3 separate visits to station P3, each visit comprising a pair of hauls, of which one was carried out in nominal daytime and the other in nominal nighttime. Catches were immediately sorted on board and identified to the lowest taxonomic level feasible. Subsamples of the catches were retained, principally for subsequent biochemical and physiological analyses. In total, 777 fish were caught, belonging to at least 23 species, with catches dominated by the myctophids Krefftichthys anderssoni, Gymnoscopelus braueri, Electrona antarctica and Protomyctophum tenesoni. The water column below 250m was dominated by Bathylagus spp. Temperate myctophid species, such as Protomyctophum parallelum and Protomyctophum andreyeshevi were also caught in small numbers. With regards macrozooplankton, the 250m-500m depth interval was dominated by the jellyfish, Atolla and Periphylla. The tunicate Salpa thompsoni and the euphausiids Euphausia triacantha and Thysanoessa spp. were also relatively abundant. Jellyfish still dominated catches in shallower waters (250m-10m), closely followed by euphausiids and Salpa thompsoni and chaetognaths. Themisto gaudichaudii and Parandania boecki were the most numerous amphipod species caught. Decapods were only caught in the deeper depth interval, both day and night.

Macrozooplankton and nekton were collected with a Rectangular Midwater Trawl 25 (RMT25) at locations within the Benguela Current region in May and June 2018. The work was carried out as part of the NERC Large Grant, COMICS (Controls on Mesopelagic Interior Carbon) on board the RRS Discovery (cruise DY090). Depth-discrete samples were collected across four time stations (BS1, BN1-3) between 0-750 m at intervals of 750-500m, 500-250m, 250-125m and 125-10 m. At each time station, two RMT25 hauls were deployed in the hours of darkness and two in daylight, with 16 deployments being undertaken overall. The RMT25 was operated via a downwire net monitor and was equipped with a flow meter, and temperature and salinity sensors. Nets in the deep strata (750-500m and 500-250m) were sampled for approximately 40 mins. and nets in the shallow strata (250-125m, 125-10m) for approximately 20mins. Catches were immediately sorted on board and identified to the lowest taxonomic level feasible. All fishes and subsamples of the other parts of the catch were retained (frozen), principally for subsequent biochemical and physiological analyses. In total, 1917 fish were caught and preserved (not including Cyclothone spp.). Catches were dominated by the myctophids and various other mesopelagic fish species. The water column below 250m was dominated by Bathylagus spp. and genus Melamphidae spp. The most numerous fish overall were the Cyclothone spp. which occurred in large numbers below 500m. In deeper depth intervals (250m-750m), the macrozooplankton component of the RMT25 net catches was mostly dominated by Decapoda and hydromedusae of the genus Atolla spp.. Salps, smaller hydromedusa species and small euphausiids Euphausia hanseni and Nematocelis megalops dominated the shallower depths (10-250m).