Data were obtained from a laboratory growth experiment conducted from April to May 2021 at the Institute for Chemistry and Biology of the Marine Environment (University of Oldenburg). Three Antarctic phytoplankton species (Chaetoceros simplex, Phaeocystis antarctica (single cell culture) and Geminigera cryophila) were grown in monocultures and a mixture of all three species. The phytoplankton strains were previously isolated from the Southern Ocean and maintained at the Alfred-Wegener-Institute in Bremerhaven. Prior to the experiment, semicontinuous stock cultures of the diatom Chaetoceros simplex, the prymnesiophyte Phaeocystis antarctica (single cell culture) and the cryptophyte Geminigera cryophila were grown in exponential phase in 0.2µm filtered natural Antarctic seawater. The natural seawater was enriched with nutrients, trace metals and vitamins according to the F/2 medium (Guillard and Ryther, 1962) but with only 1/6 of the original F/2 iron(Fe)-concentration. The stock cultures were kept at 2°C and an irradiance of 100 µmol m₂ sek⁻¹ with a 16:8 light/dark cycle. We used a factorial design including four different nitrogen/phosphorus (N/P) treatments (control plus three different N/P levels) and 10 different temperatures (-2, -1, 0, 1, 2, 3, 4, 6, 8 and 10°C). Two of the four N/P treatments were applied with an N/P of 12 (control and ambient N/P with higher N and P concentrations), one with a molar N/P ratio of 6 (30 µmol N, 5 µmol P) and one with a molar N/P of 24 (60 µmol N, 2.5 µmol P). The nutrient treatments were generated by adding only 30 µmol L⁻¹ N (N/P 24, +N), 2.5 µmol L⁻¹ P (N/P 6, +P) and both (N/P 12, +N,P) to 0.2 µm filtered Antarctic seawater with an ambient N/P of 12 and an concentration of 30 µmol N and 2.5 µmol P. The ambient control treatment (here N/P 12 control) received no additional N and P. All other nutrients were added according to F/2 with 1/6 of iron to avoid limitation by other elements (Guillard and Ryther, 1962). The experimental temperatures were kept constant. Light was applied with LED panels (Solar Stinger LED SunStrip, Econlux) at 60 µmol m₂ sek⁻¹ with a 16:8 light/dark cycle. Culture flasks (250 ml) were used as experimental units filled with 200 ml medium. Each experimental unit was inoculated with the same biovolume. The experiment was run in triplicates, resulting in a total of 480 experimental units. Optical density (440 nm) measurements (every second day) from 250µl subsamples were used to calculate potential maximum biomass (carrying capacity, K) and the maximum growth rate (µmax / day) of the monocultures and the mixed assemblage. Subsamples were preserved with 1% Lugol's iodine for later cell counts that were transformed biovolume and OD based on biovolume-OD calibrations (see supporting material in the original publication). The final sampling of each culture was conducted during stationary phase, when increase in optical density was stable during three sampling days. The stationary phase was reached between 30 and 48 days, depending on the temperature and nutrient treatment. At the end of the experiment, 2 x 50 ml from each culture were taken and filtered onto acid washed and pre-combusted glass-fibre filters (GF/C, Whatman, UK). One filter was used for particulate organic carbon (C) and nitrogen (N), the second filter for particulate organic phosphorus (P).
