Arctic ocean salinity and freshwater content reconstructed from satellite sea surface salinity
Year: 2026
Type: dataset
Abstract: This dataset provides daily gridded reconstructions of ocean salinity and freshwater content (FWC) for the Arctic Ocean, based on satellite sea surface salinity (SSS). The product covers 2010–2023 and is distributed on a 25 km Equal-Area Scalable Earth (EASE2) grid. Each daily NetCDF file contains reconstructed salinity profiles at 42 vertical levels, together with vertically integrated FWC and associated error estimates. The reconstructions are generated using an empirical orthogonal function–generalized additive model (EOF-GAM) framework. Empirical orthogonal functions derived from in situ hydrographic profiles represent the dominant vertical structures of salinity variability. Generalized additive models predict the corresponding mode amplitudes from satellite SSS and environmental predictors, which are then used to reconstruct subsurface salinity profiles. This approach estimates the components of observed subsurface salinity variability that are statistically associated with surface observations and environmental conditions, without prescribing a model- or reanalysis-derived subsurface background field. FWC is calculated by vertically integrating the salinity deficit relative to a reference salinity (34.8) from the ocean surface to each specified depth. The product includes reconstructed salinity profiles (sal), salinity reconstruction uncertainty estimates (sal_error), cumulative freshwater content (fwc), and FWC uncertainty estimates (fwc_error). The uncertainty estimates are derived using a perturbation-based approach to quantify the sensitivity of the reconstructed fields to uncertainty in the reconstruction inputs. Salinity and salinity uncertainty are reported in practical salinity units (numerically equivalent to 0.001 in the NetCDF metadata), while FWC and its uncertainty are expressed in metres. Latitude, longitude, depth, and time coordinates are provided in each file. The dataset supports investigations of Arctic and northern high-latitude freshwater storage, seasonal and interannual variability, and the information content of satellite SSS for subsurface ocean monitoring. It can also be used for comparison with existing gridded hydrographic and ocean reanalysis products. Reconstruction uncertainty and regional, seasonal, and depth-dependent performance should be considered when interpreting the estimates.
Source: SEANOE
DOI: 10.17882/120534
Language: English
Subfield: Oceanography
Field: Earth and Planetary Sciences
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7223027096