This dataset contains processed data from the physical oceanography instrumentation (temperature, salinity, pressure, oxygen, ocean currents) on the Y5-1 mooring. This mooring was part of a mooring array (Y1-1, Y2-1, Y3-1, Y4-1, and Y5-1). Y1-1 and Y2-1 were within 1.7 km distance, Y3-1 and Y4-1 were within 1.9 km distance from each other, and the whole array formed a line (ca. 132 km length) approximately perpendicular to the sea ice edge. The moorings were deployed during RV POLARSTERN expedition PS131 (ATWAICE) in July 2022, and the physical-oceanographic instruments were recovered during PS137 (ALOIS) in June 2023. The scientific objective of the mooring array was related to key mechanisms of the rapid Arctic sea ice decline and Arctic Amplification. These include processes affecting heat fluxes in the air-ice-ocean system, ocean mixed-layer halocline coupling, ice melt, and ice edge dynamics in the marginal ice zone. Therefore, the mooring instrumentation targets the expected strong seasonal variability of the upper ocean. Here, we provide the processed data; the raw data are also published in PANGAEA and are linked below. Note that there can be differences in the listed depths for each instrument compared to the raw data submission. For the raw data, the nominal depths were listed; here, we provide the minimum observed depth after processing, assuming it represents a situation without mooring blow-down. In general, the processing comprises the following steps: for each instrument, the measured time series is cut to the time period the sensor was at target depth, i.e., from after the deployment was completed to before the recovery started. Then, each instrument without its own pressure record is assigned a depth by a constant offset to neighboring pressure sensors, based on the mooring drawing. For ADCPs, the (variable) depth record and cell range information are used to create a time-depth matrix indicating the depth of each ADCP cell at each time step. Data above the 99.98% speed percentile are removed. ADCP bins that showed interference with other gear in the mooring (e.g., large buoyancy floats) were excluded manually. Also, data from ensembles with fewer than 30% good samples were removed, and a threshold of 0.1 m/s for the error velocity is applied. The derived ocean current directions are corrected for magnetic declination. Where applicable, practical salinity is calculated from conductivity. Outliers (upper and lower 0.02% percentile) are removed. Temperature values lower than -3 °C and greater than 30 °C are removed directly. For all SBE37 with dissolved oxygen measurements, the first two days of data are discarded to exclude unreliable data from the sensor membrane adjusting to the water pressure. The attached data contain processed measurements from five Seabird SBE37 MicroCATs (at depths: 18 m, 38 m, 71 m, 98 m, 244 m; sampling interval 1 h for the shallowest two, 10 min for the remaining ones), nine Seabird SBE56 temperature loggers (at depths: 23 m, 28 m, 48 m, 61 m, 81 m, 148 m, 194 m, 402 m, 470 m; sampling interval 30 s), one upward-looking Teledyne RDI 300 kHz Workhorse ADCP (at depth: 48 m; sampling interval 15 min), one upward-looking Teledyne RDI 75 kHz Longranger ADCP (at depth: 290 m; sampling interval 30 min), and one RBRDuo wave16 recorder in 4 Hz burst mode (nominal depth: 17 m; sampling interval 30 min). From the latter, we only provide the averaged pressure records; the burst data is available from the raw data submission (see below). The 300 kHz ADCP stopped recording on 5 Jun 2023 due to a low battery. The 75 kHz-ADCP exhibited some unexpected behavior. From approximately 130 m depth toward the surface, the velocity amplitude appears slightly reduced compared to deeper measurements. Relative to the magnitude of the tidally dominated velocities, these deviations in each velocity component are small ( 0.03 m/s. This behavior could be caused by a fouling plate or by buoyancy floats in the mooring line affecting the backscattered ADCP signal. Alternatively, given the mooring position near steep topography with strong tidal currents, the behavior may also be related to flow-topography interactions affecting the mean and variance of the velocity signal at a certain depth. No modifications were made to account for this behavior, so the data should be interpreted with caution. The SBE56 temperature measurements (native 30-s sampling) are smoothed using a running mean with a 20-min window size, and a data subset with 20-min resolution is stored in the present dataset.

The authors are grateful to the captains, crews, and technical/scientific staff of the expeditions PS131 and PS137 onboard RV Polarstern. Many individuals have contributed to the conception of the research, the preparation of the instruments, the deployments and recoveries, as well as to the retrieval of the data, which we greatly appreciate. We specifically thank Torsten Kanzow, Vera Schlindwein, Jacob Allerholt, Nicolas Dettling, Carina Engicht, Rainer Graupner, Normen Lochthofen, Matthias Monsees, and Jutta Vernaleken. We acknowledge support from the Helmholtz infrastructure program Frontiers in Arctic Marine Monitoring, and the Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung. This mooring contributed to the Arctic PASSION project.