Ten sea ice cores were drilled close to each other with a Kovacs 9 cm diameter corer (Kovacs Enterprise, Roseburg, USA) on April 8, 2020, near the main sampling site (MCS) for FYI and SYI during leg 3 of MOSAiC (station PS122/3_35-80). The ice cores were immediately transferred into plastic bags and stored at −20 °C together with a snow sample, which was collected before core extraction. In the home laboratory, the ice cores were rinsed with deionized water and the wet top surface was scraped off before the cores were sectioned into 10 cm pieces and the resulting intervals of the corresponding ice core depths were combined. Samples from one sea-ice core were not merged with the samples from the other nine ice cores to investigate the effects of sample pooling and possible related sources of bias. After melting, the meltwater was filtered through 0.45 µm Merck Millipore® cellulose acetate filters. Following homogenization and sub-sampling for salinity and δ18O analysis, the filtered samples were acidified to pH ≈ 2.2 with ultra-pure concentrated hydrochloric acid. Another aliquot was separated for REE and Nd concentration analyses. The snow sample was treated similarly, excluding the deionized water rinse and sectioning steps. The neodymium isotopes are reported as 143Nd/144Nd and as εNd and were determined at GEOMAR using an multicollector inductively coupled plasma mass spectrometer (MC-ICP-MS). Their internal and external reproducibility is given in the data table. After pooling, the sea ice samples still had concentrations too low for εNd analysis, so two or three ice core intervals had to be combined to allow determination of εNd with relatively small uncertainties. The repeatedly reported εNd values for individual ice core samples are therefore the same measurement, not the same εNd values measured twice. Rare earth element concentrations are given in pmol/kg and were pre-concentrated offline using a SeaFAST system and determined using an inductively coupled plasma mass spectrometer (ICP-MS) at GEOMAR. Their external reproducibility is better than ~11% for La, Ce, Pr, Nd, Sm and Eu, better than ~13% for Gd, Tb and Dy and better than ~30% for Ho, Er, Tm, Yb and Lu. Stable oxygen isotopes were determined at the Stable Isotope Laboratory of the College of Earth, Ocean, and Atmospheric Sciences at Oregon State University using the CO2-water isotope equilibration technique and are reported as the measured 18O/16O ratio normalized to Vienna Standard Mean Ocean Water (V-SMOV) in δ-notation, with an external reproducibility of ± 0.05 ‰ or better.