Measurements for a 92 day climate manipulation experiment on replicate macrofaunal mesocosms collected on the cruise JR18006 in the Western Barents Sea and at Rothera research station in 2019

Individuals of Astarte crenata, Ctenodiscus crispatus and Cistenides hyperborea were obtained from replicate SMBA (Scottish Marine Biological Association) box cores and 15 minute Agassiz trawls in the Barents Sea (stations B13, 74.3 degrees N, 30.0 degrees E; B16, 80.3 degrees N, 30.0 degrees E; JCR18006, RSS James Clark Ross) in July 2019. Individuals of the Aequiyoldia eightsi and Laternula elliptica were collected by SCUBA-assisted divers at Rothera Point, Adelaide Island, West Antarctic Peninsula (67.3 degrees S, 68.1 degrees W) in March-April 2019. Surficial sediment (less than 5 cm depth) for the Arctic species was collected using SMBA box cores at stations B13, B14 (76.3 degrees N, 30.3 degrees E) and B16 in July 2019 and, for the Antarctic species, by hand from the Hamble, UK (50.9 degrees N, 1.3 degrees W). Sediment was sieved (500 micrometer mesh) within a seawater bath to retain the fine fraction and remove macrofauna and debris. Species were distributed across 42 clear acrylic aquaria (internal LWH: 12 x 12 x 33 cm, 3 replicates treatment-1: species × location × climate scenario), designed to accommodate the appropriate density (Arctic, 2 ind. mesocosm-1; Antarctic, 1 ind. Mesocosm-1; size and burrowing requirements of each species (sediment depth: A. crenata, C. crispatus & C. hyperborea, 16 cm; A. eightsi, 12cm; L. elliptica, 19cm), and exposed to ambient (1 ± 0.5 degrees Celsius, around 400 ppm [CO2]) and indicative near-future (3 ± 0.5 degrees Celsius, around 550 ppm [CO2]) environmental conditions. Aquaria were randomly placed within one of two insulated seawater reservoirs (LWH: 1.2 x 1.2 x 0.8m, Tanks Direct, UK). All aquaria were overlain with seawater (salinity 33, 10 micrometer sand filtered, UV sterilized) to a combined sediment and water depth of 31 cm and maintained in the dark. After acclimation to ambient mesocosm conditions (21 days, 09-29/09/2019), the water temperature and atmospheric CO2 of the aquaria in the future treatment was gradually increased at 0.5 degrees Celsius and 50 ppm increments every 7 days to reach the desired near-future climate conditions. During both the acclimation and experimental period (92 days, 21/10/2019 - 21/01/2020), C. crispatus and C. hyperborea were fed ad libitum once a week with commercially available fish food (Aquarian Tropical Flake; 0.03 g), and A. crenata, A. eightsi and L. elliptica were fed ad libitum three times per week with 100 ml of precultured phytoplankton (33:33:33 mix, Isochrysis sp., Tetraselmis sp., and Phaeodactylum sp.; Algal Culture Laboratory, National Oceanography Centre Southampton). To avoid accumulation of excess food and associated nutrients, partial seawater changes (weekly 50 percentage seawater exchange) were undertaken. Temperature, pH, salinity and total alkalinity were periodically measured across all experimental mesocosms (T,pH,S: weekly, AT: week 2, week 6, week 11). Total alkalinity analysis followed the standard HCl titration protocol of the Carbonate Facility at the University of Southampton. From which, DIC, [pCO2], [calcite], [aragonite], [HCO3] and [CO3] were calculated using the CO2calc software. CO2(ppm), H2O(ppt), H2O(C), Cell_Temperature(C), Cell_Pressure(kPa), CO2_Absorption and H2O_Absorption were measured continuously in the ambient air and future climate regime gas mixture from the start (21/10/2019) to end (21/01/2020) of the experimental period. All measurements were calculated using an Infra-red gas analyzer LiCOR LI-820 model and recorded using the interface software on a standard laptop. Behavioural activity was quantified using measurements of movement and burial behaviour at the sediment surface. Individuals (morphology, +/- 0.01 mm; blotted wet weight, +/- 0.001 g) were placed in temperature-acclimatised viewing trays containing sediment (depth 5 cm) overlain with sea water (depth 3 cm) and viewed (less than or equal to 60 minutes) with a benchtop video camera (Logitech C920 HD Pro Webcam, 1080p). The time taken for each individual to initiate movement (response time, s) and to complete burial (burial time, s) was recorded (SkyStudioPro time-lapse software; 3 frame s-1, autosave every 300 frames) and analysed frame by frame (VLC Media Player). Faunal mediated sediment particle reworking metrics were estimated at the end of the experimental period by establishing the redistribution of optically distinct particulate tracers (luminophores: 30g aquaria-1, fluorescent green, 125 - 250µm silica sand,; Glass Pebbles Ltd., UK). Luminophores were evenly distributed across the sediment surface immediately after the partial seawater change. After 12 days, images (RGB colour, JPEG compression) of the vertical redistribution of luminophores were taken from each side of each aquarium using a digital SLR camera. Extracted profile data was obtained using a custom-made semi-automated macro that runs in ImageJ (version 1.47s, released 3rd June, 2013), a java-based public domain program developed at the US National Institutes of Health (http://rsb.info.nih.gov/ij/index.html). The ventilatory behaviour of the infauna (bioirrigation) was estimated from absolute changes in the concentration (10mM, 1.0289g NaBr dissolved per 1l seawater mesocosm-1) of the inert tracer sodium bromide (Delta [Br-], mg L-1; negative values indicate increased infaunal ventilatory activity) over an 8 h period on day 12, determined using a Tecator flow injection auto-analyser (FIA Star 5010 series). Accumulated water column concentrations (delta micromol L-1) of NH4-N, NOx-N (i.e. NO3-N + NO-N) and PO4-P were determined at the start and end of the experimental period from standardized samples (taken from the centre of each mesocosm at around 5cm depth, 0.45 micrometer NALGENE filtered) following standard protocols using a Lachat Quikchem 8500 flow-injection auto-analyser. Sediment was analysed using a Malvern Mastersizer 2000 He-Ne LASER diffraction sizer for sediment particle size at the University of Cambridge following a standard protocol (provided). Sediment organic material content (percentage) was analysed at the University of Cambridge following a standard protocol (provided). Full descriptions and additional information provided in Williams et al. [submitted for publication] in main manuscript and supplementary information.