This work was completed as part of the SIPEX - Sea Ice Physics and Ecosystem eXperiment - voyage. September/October 2007. The work formed part of AAS (ASAC) projects 2337 and 2767. Aspects of krill (Euphausia superba), growth and condition during late winter-early spring off East Antarctica (110 - 130 degrees E) were investigated. We assessed diet and condition of larval and postlarval krill collected from open water and below the ice. Condition was assessed using lipid content, growth rates and digestive gland size; feeding history was assessed using fatty acid profiles and stomach content analysis; and a starvation study investigated the response of krill to long-term food deprivation. Potential food items were analysed for lipid and fatty acid composition. Fatty acid profiles and stomach content analysis revealed winter/early spring feeding strategies of both larval and adult krill. This work was completed as part of AAS (ASAC) project #2337

Instantaneous growth rates (IGR) of Antarctic krill kept under experimental conditions were measured. The measured appendages included the uropods, telson (both standard length measurements with the IGR technique) and the pleopod endopodite and pleopod exopodite were investigated as an alternate length measurement. IGR measurements were recorded on 90 experimental animals. The total carbon content of 45 krill of various size ranges (collected directly from the field) was determined. The relationship between the change in length in carbon as a function of growth was investigated. The parameters measured were total length, mean uropod length, telson length, wet weight, dry weight and total carbon content. This dataset was collected as part of ASAC project 141. See metadata record ASAC_141 - Collection of live Antarctic krill 'Euphausia superba'. The fields in this dataset are: Krill Total length (mm) Telson length (mm) Mean uropod length (mm) Wet weight (g) Dry weight (g) Dry Weight (mg) Carbon content as a % of dry weight Total carbon content (g) Moult Sex

This data may be used to reproduce the analyses (including figures and tables), of 'Two scales of distribution and biomass of Antarctic krill (Euphausia superba) in the eastern sector of the CCAMLR Division 58.4.2 (55°E to 80°E)'. The data describe krill biomass density distribution and krill net samples (krill total length and krill wetmass) collected during the 2021 TEMPO voyage on R/V Investigator. During the TEMPO voyage krill biomass was estimated using observations from two sampling instruments: a calibrated EK80 scientific echosounder operating at 120 kHz and an rectangular midwater trawl (RMT 1+8). The supporting data sets, all in CSV format, are split by instrument type. The EK80 has two datafiles: krill_density.csv – krill areal density from the TEMPO transects, and krill_swarms.csv -krill swarms detected during the TEMPO transects. The RMT1+8 has four datafiles net_locations.csv krill_lengths.csv krill_wet_mass_to_length.csv krill_wet_mass_to_length_model_predictions.csv The fields (columns) in each data file are: krill_density.csv "lat_M" – centre latitude of an echo integration interval [degrees] (dd.ddddd) WGS84 spheroid (GPS latitude) "lon_M" - centre longitude of an echo integration interval [degrees] (dd.ddddd) WGS84 spheroid (GPS longitude), "areal_biomass_density_g_per_m2" – Echo integration interval krill areal biomass density [g wet-mass / m^2] "daynight" – flag for when the sampling took place [day/night] "survey" – Either the main survey for the TEMPO biomass survey or the smaller-scale ‘Mawson box’ survey krill_swarms.csv "transect" – transect number "lat" – latitude [degrees] (dd.ddddd) WGS84 spheroid (GPS latitude) "swarm_depth_m" – mean depth of a krill swarm [m] "daynight" – flag for when the sampling took place [day/night] ”volumetric_density_g_per_m3" – krill swarm internal volumetric biomass density [g wet-mass / m^3] net_locations.csv "station" – Station name for net trawl R for routine haul, T for target trawl "lat" – mean latitude of a net trawl [degrees] (dd.ddddd) WGS84 spheroid (GPS latitude) "lon" – mean longitude of a net trawl [degrees] (dd.ddddd) WGS84 spheroid (GPS longitude) "daynight" – flag for when the sampling took place [day/night] krill_lengths.csv "station" – Station name for net trawl R for routine haul, T for target trawl "total_length_mm” – total length of an individual krill [mm] krill_wet_mass_to_length.csv "total_length_mm" – total length of an individual krill [mm] "wet_mass_g" - wet-mass an individual krill [g] krill_wet_mass_to_length_model_predictions.csv "total_length_mm" - total length of an individual krill [mm] "predicted_wet_mass_g" – predicted mean wet-mass an individual krill of length ("total_length_mm" ) [g] "LB_wet_mass_g" – Lower bound (lower 95% confidence interval) for the predicted_wet_mass_g [g] "UB_wet_mass_g"– Upper bound (upper 95% confidence interval) for the predicted_wet_mass_g [g]

The embryonic development of Antarctic krill (Euphausia superba) is sensitive to elevated seawater CO2 levels. This data set provides the experimental data and WinBUGS code used to estimate hatch rates under experimental CO2 manipulation, as described by Kawaguchi et al. (2013). Kawaguchi S, Ishida A, King R, Raymond B, Waller N, Constable A, Nicol S, Wakita M, Ishimatsu A (2013) Risk maps for Antarctic krill under projected Southern Ocean acidification. Nature Climate Change (in press) Circumpolar pCO2 projection. To estimate oceanic pCO2 under the future CO2 elevated condition, we computed oceanic pCO2 using a three-dimensional ocean carbon cycle model developed for the Ocean Carbon-Cycle Model Intercomparison Project (2,3) and the projected atmospheric CO2 concentrations. The model used, referred to as the Institute for Global Change Research model in the Ocean Carbon-Cycle Model Intercomparison Project, was developed on the basis of that used in ref. 4 for the study of vertical fluxes of particulate organic matter and calcite. It is an offline carbon cycle model using physical variables such as advection and diffusion that are given by the general circulation model. The model was forced by the following four atmospheric CO2 emission scenarios and their extensions to year 2300. RCP8.5: high emission without any specific climate mitigation target; RCP6.0: medium-high emission; RCP 4.5: medium-low emission; and RCP 3.0-PD: low emission (1). Simulated perturbations in dissolved inorganic carbon relative to 1994 (the Global Ocean Data Analysis Project (GLODAP) reference year) were added to the modern dissolved inorganic carbon data in the GLODAP dataset (5). To estimate oceanic pCO2, temperature and salinity from the World Ocean Atlas data set (6) and alkalinity from the GLODAP data set were assumed to be constant. Marine ecosystems of the Southern Ocean are particularly vulnerable to ocean acidification. Antarctic krill (Euphausia superba; hereafter krill) is the key pelagic species of the region and its largest fishery resource. There is therefore concern about the combined effects of climate change, ocean acidification and an expanding fishery on krill and ultimately, their dependent predators—whales, seals and penguins. However, little is known about the sensitivity of krill to ocean acidification. Juvenile and adult krill are already exposed to variable seawater carbonate chemistry because they occupy a range of habitats and migrate both vertically and horizontally on a daily and seasonal basis. Moreover, krill eggs sink from the surface to hatch at 700–1,000m, where the carbon dioxide partial pressure (pCO2 ) in sea water is already greater than it is in the atmosphere. Krill eggs sink passively and so cannot avoid these conditions. Here we describe the sensitivity of krill egg hatch rates to increased CO2, and present a circumpolar risk map of krill hatching success under projected pCO2 levels. We find that important krill habitats of the Weddell Sea and the Haakon VII Sea to the east are likely to become high-risk areas for krill recruitment within a century. Furthermore, unless CO2 emissions are mitigated, the Southern Ocean krill population could collapse by 2300 with dire consequences for the entire ecosystem. The risk_maps folder contains the modelled risk maps for each of the climate change scenarios (i.e. Figure 4 in the main paper, and Figure S2 in the supplementary information). These are in ESRI gridded ASCII format, on a longitude-latitude grid with 1-degree resolution. Refs: 1. Meinshausen, M. et al. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109, 213-241 (2011). Orr, J. C. et al. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437, 681-686 (2005). Cao, L. et al. The role of ocean transport in the uptake of anthropogenic CO2. Biogeosciences 6, 375-390 (2009). Yamanaka, Y. and Tajika, E. The role

Taken from the "Supporting Information" for the main paper. See the referenced papers for more information. Our results are based on numerical simulation of Southern Ocean sea ice, conducted using the Los Alamos numerical sea-ice model CICE version 4.0 [CICE4; Bailey et al., 2010] configured in stand-alone mode on a 0.25 degree x 0.25 degree grid, extending to 45 degrees S, with 3-hourly output [Stevens, 2013]. The atmospheric forcing for CICE4 came from the hemispheric forecasting model Polar Limited Area Prediction Systems [Polar- LAPS; Adams, 2006] and ocean forcing from the global ocean general circulation model Australian Climate Ocean Model [AusCOM; Bi and Marsland, 2010]. The model is well-constrained in its representation of processes of sea ice formation and melt, and comparison with observed areal ice extent shows minimal deviations over the 1998-2003 period, particularly during winter [Stevens 2013]. Stevens [2013] evaluates the sensitivity of the model to the number of ice thickness categories. Sea ice thickness sensitivities in the CICE model are considered in detail in Hunke [2010, 2014]. For the warm climate scenario, changes were implemented that are consistent with the A1B scenario from the Fourth Assessment from the IPCC [Meehl et al., 2007]. This is a mid-range scenario that assumes rapid economic growth before introduction of new and more efficient technologies mid century. Specifically, the following changes were applied uniformly to the current climate forcing field for a single year: a 2 degrees C increase in air temperature, a 0.2 mm/day increase in rain, a 1.5% increase in cloud fraction, a -2.3 hPa change in surface air pressure, a 25% increase in wind, a 12 Wm-2 increase in long wave downward radiation and a 20% increase in humidity. Outputs and forcings from CICE4 that are relevant for consideration of under-ice habitats for larval krill include: snow depth, ice thickness, ice concentration, movement, ridging rate, day length (dependent on day-of-year and latitude), radiation above the ice (influenced by cloud cover), and radiation below the ice (influenced by ice and snow depth). Table 1 in the main text describes how these were used in the following two filters and one overlay for evaluating the location and suitability of potential larval krill habitat during winter. Taken from the abstract of the main paper: Over-wintering of larvae underneath Antarctic pack ice is a critical stage in the life cycle of Antarctic krill. However, there are no circumpolar assessments of available habitat for larval krill, making it difficult to evaluate how climate change may impact this life stage. We use outputs from a circumpolar sea-ice model, together with a set of simple assumptions regarding key habitat features, to identify possible regions of larval krill habitat around Antarctica during winter. In particular we assume that the location and suitability of habitat is determined by both food availability and three dimensional complexity of the sea ice. We then compare the combined area of these regions under current conditions to that under a warm climate scenario. Results indicate that, while total areal sea-ice extent decreases, there is a consistently larger area of potential larval krill habitat under warm conditions. These findings highlight that decreases in sea-ice extent may not necessarily be detrimental for krill populations and underline the complexity of predicting future trajectories for this key species in the Antarctic ecosystem.

This dataset contains estimates of krill swarm characteristics from statistical models based on underway acoustic observations along with underway and remote-sensed environmental data. Estimates of internal swarm density and depth across the study region (60-80 degrees E) are included for the time of the survey (Feb 2006). Estimates of February internal swarm density across the broader East Antarctic region (30-120 degrees E) are also included for the period 2001-2010.

Raw and processed acoustic data were collected in East Antarctica from the RSV Aurora Australis during two surveys: the Krill Availability, Community Trophodynamics and AMISOR Surveys (KACTAS) and the Krill Acoustics and Oceanography Survey (KAOS) in the East Antarctic. Seabed alias example file used to make Figure 3. 38_false_bottom.sv.csv (file size: 10.7 MB) This file is an example of seabed aliasing (false bottom echo) that occurred during the KAOS survey. The data in this file are acoustic (mean volume backscattering strength, Sv) sample-by-sample and in a CSV format. Ping_index - ping number Distance_GPS - along track distance from the vessel’s GPS (nautical miles). Distance_vl - along track distance from the vessel’s log (nautical miles). Not used here, so is populated by Echoview’s ‘don’t care’ value (-9.90E+37) to keep the file format consistent. Ping_date - format yyyy-mm-dd Ping_time - format hh:mm:ss Ping_milliseconds - format (integer; ms) Latitude - position from the vessel’s GPS (degrees) Longitude position from the vessel’s GPS (degrees) Depth_start - start depth of vessel echosounder logging range (m) Depth_stop - stop depth of vessel echosounder logging range (m) Range_start - start range of vessel echosounder logging range (m) Range_stop - stop range of vessel echosounder logging range (m) Sample_count number of samples in a ping. Acoustic Sv samples follow in column-wise vector (dB re 1 m-1) Transect metadata all_transects.csv (file size: 14.8 K) This is the transect metadata for both the KACTAS and KAOS surveys: Transect - transect number startDate - start date of transect dd/mm/yyyy startTime - start time of transect hh:mm endDate - end date of transect dd/mm/yyyy endTime - end time of transect dd/mm/yyyy Ping_subset - a ping subset specified between two timestamps to isolate acoustic data that occurred on transect , i.e start timestamp to stop timestamp (yyyy-mm-dd hh:mm = yyyy-mm-dd hh:mm). Direction - Direction traveled along the transect (N - north or S- south). Light - day or night when transect was observed Survey - Either KACTAS or KAOS Leg - Krill box 1 or 2 (there were two surveys only during the KAOS voyage). Pass - Sampling bout for a transect in a given direction. Example R-code FigureAndDataprocessingExample.R (12KB) This R-code provides examples of scripting acoustic data processing using EchoviewR, specifically, using ping-subsets to isolate acoustic data along a transect of interest, detecting schools and exporting echo integrations for 38, 120 and 200 kHz. The R-code for making the figures in the paper is also given. GPS vessel positions for both the KACTAS and KAOS surveys KACTASandKAOS_GPS.csv (3 MB) This CSV file gives the vessel track for both the KACTAS and KAOS surveys. GPS_date - dd/mm/yyyy GPS_time - HH:MM:SS GPS_milliseconds - integer Latitude - position from the vessel’s GPS (degrees) Longitude - position from the vessel’s GPS (degrees) x - relative grid position (x) used for plotting Figure 1 y - relative grid position (y) used for plotting Figure 1 Survey - either: KACTAS_Krillbox, KAOS_Krillbox1, or KACTAS_Krillbox2 Krill swarms for both the KACTAS and KAOS surveys KACTASandKAOSswarms.csv (2.1 MB) Krill swarms descriptors for the KACTAS and KAOS surveys in CSV format (see Table 2 for description of the data fields). Echoview file for the KACTAS survey KACTAS-survey.EV (21.8 MB) An Echoview file (version 12.0) for the KACTAS acoustic data analysis KACTAS EK60 scientific echosounder calibration values KACTAS_EK500_calibration.ecs 3,623 16/05/2022 21:21 -a-- An Echoview format calibration file for the KACTAS survey (see Table 3 for calibration values and Demer et al.20 for a description of the calibration parameters). Echoview file for the KAOS survey KAOS-survey.EV (16.4MB) An Echoview file (version 12.0) for the KAOS acoustic data analysis KAOS EK60 scientific echosounder calibration values KAOS_EK60_calibration.ecs 5,711 01/11/2021 04:36 -a– An Echoview format calibration

Antarctic krill is a key component of Southern Ocean ecosystems and there is significant interest in identifying regions acting as sources for the krill population. We develop a mechanistic model combining thermal and food requirements for krill egg production, with predation pressure post-spawning, to predict regions that could support high larval production (spawning habitat). We optimise our model on regional data using a maximum likelihood approach and then generate circumpolar predictions of spawning habitat quality. The uploaded datasets represent model predictions of seasonal circumpolar spawning habitat quality of Antarctic krill as well as composite data of the circumpolar mean annual number of weeks in which modelled spawning habitat quality is higher than the summer 80th percentile.

Crustaceans grow or shrink in size as they moult. Length of discarded moults represent length of animals before their moulting events. Therefore, by measuring length of discarded moult and length of animal after moult, growth increments at the time of moult can be obtained. IGR is defined as the growth increment expressed as a proportion of pre-moult total length (TL). IGR can be converted into daily growth rate for a given value of TL by calculating absolute growth increment and dividing by an estimate of inter-moult period (IMP). The IGR technique depends on the collection of live krill in good condition. Krill were caught with an RMT-8 net and individual freshly caught animals were randomly selected from the catch and immediately transferred to individual jars. They were then maintained onboard and checked regularly for moults for up to 5 days following capture. The experiments were run a flowthrough seawater system which used 250 ml jars with small holes to allow water exchange in a large flow-through tank of seawater maintained at ambient ocean temperature. No additional food was provided. The system allowed experiments with over 4000 krill. Each krill was checked daily after capture to ascertain whether it had moulted. If an animal had moulted, then the animal and its moult were collected and frozen in liquid nitrogen or at -85 degrees C to be measured back ashore. The growth rate will be estimated from the difference in length of the uropod of the moult and that of the whole post-moult krill. This work was completed as part of ASAC projects 2655 and 2679 (ASAC_2655, ASAC_2679).