Field sampling:
Ice samples for primary production measurements were collected mid-morning from 6 stations along the western Antarctic Peninsula in November and December of 2019 on board the R/V Nathaniel B. Palmer along a north-south transect from 64.8°S to 67.8°S. For Stations (Stns) 2 and 3, the ice was "rotten" (sufficiently melted to be disintegrating structurally, present only in small pieces) and collected as an ice-seawater slurry. Small ice chunks were collected directly from the sea surface via a crane-suspended "personnel basket". At the additional 4 stations, sea ice was collected by coring the ice. At Stns 4 and 7, algal samples were collected from internal ice-core layers. Stns 4 and 7 were rafted floes with a flooded internal layer, with Stn 7 > 2500 square meters (m²) in size. Stns 5 and 6 were on landfast sea ice, where the algae were collected from the bottom 10 centimeters (cm) of the ice. At these 4 stations (Stns 4 through 7), ice cores were taken with a 7.5 cm Kovacs corer separated by at least 1 m horizontally. Cores were shaded from direct sunlight, while 5 to 10 cm of the visible algal band at the bottom or middle was sectioned with an ethanol-cleaned saw and placed into acid-washed (10% HCl) containers. Each replicate consisted of either a single core (for 10 cm sections) or pools of two cores (for 5 cm sections).

All ice samples were stored in dark, insulated containers for a maximum of 4 hours. On the ship, all ice samples used for primary production measurements were melted in a 3:1 volumetric ratio of melting solution to ice to minimize the effect of the melt on sea-ice communities. The melting solution was a 0.2 micrometer (μm) filtered artificial salt mixture containing three main sea salts plus bicarbonate according to ESAW artificial seawater (3.63 x 10⁻¹ molar (M) NaCl, 4.71 x 10⁻² M MgCl₂.6H₂O, 2.5 x 10⁻² M Na₂SO₄, and 2 x 10⁻³ M NaHCO₃⁻, salinity 35). Additional ice cores collected for ancillary biological measurements were melted in a 1:1 volumetric ratio of melting solution to ice. Melts were conducted in the dark at approximately 20° Celsius (C). To speed the melting process, ice samples were further broken into pieces with acid-washed pickaxes with most ice completely melted within 5 hours. Volume and salinity were measured as soon as the ice was completely melted, with sample temperatures remaining below 0°C. All reported volumes were corrected to the original ice volume.

Pigments:
A subsample of 1:1 melt volume was filtered on 25-millimeter (mm) GF/Fs for pigment analysis by high-performance liquid chromatography (HPLC). Filters were promptly flash-frozen in liquid nitrogen and stored at -80°C until analysis. Reverse-phase high-pressure liquid chromatography was conducted at the University of South Carolina after the method detailed by Pinckney et al. (1998). To estimate the relative abundance of diatoms, Phaeocystis, and cryptophytes (as contribution to Chl a in milligrams per liter (mg L⁻¹)), the respective diagnostic pigments of fucoxanthin, 19' hexanoyloxyfucoxanthin, and alloxanthin were used as in Everitt et al. (1990) and Arrigo et al. (2000). Pigment data was qualitatively confirmed via light microscopy.