Simulated Sea Surface Salinity Data from a 1/48° Ocean Model.

To study the validation process for sea surface salinity (SSS) we have generated one year (November 2011- October 2012) of simulated satellite and in situ "ground truth" data. This was done using the ECCO (Estimating the Circulation and Climate of the Oceans) 1/48° simulation, the highest resolution global ocean model currently available. The ground tracks of three satellites, Aquarius, SMAP (Soil Moisture Active Passive) and SMOS (Soil Moisture and Ocean Salinity) were extracted and used to sample the model with a gaussian weighting similar to that of the spaceborne sensor ground footprint. This produced simulated level 2 (L2) data. Simulated level 3 (L3) data were then produced by averaging L2 data onto a regular grid. The model was sampled to produce simulated Argo and tropical mooring SSS datasets. The Argo data were combined into a simulated gridded monthly 1° Argo product. The simulated data produced from this effort have been used to study sampling errors, matchups, subfootprint variability and the validation process for SSS at L2 and L3.

Summertime increases in upper-ocean stratification and mixed-layer depth.

The surface mixed layer of the world ocean regulates global climate by controlling heat and carbon exchange between the atmosphere and the oceanic interior1-3. The mixed layer also shapes marine ecosystems by hosting most of the ocean's primary production4 and providing the conduit for oxygenation of deep oceanic layers. Despite these important climatic and life-supporting roles, possible changes in the mixed layer during an era of global climate change remain uncertain. Here we use oceanographic observations to show that from 1970 to 2018 the density contrast across the base of the mixed layer increased and that the mixed layer itself became deeper. Using a physically based definition of upper-ocean stability that follows different dynamical regimes across the global ocean, we find that the summertime density contrast increased by 8.9 ± 2.7 per cent per decade (10-6-10-5 per second squared per decade, depending on region), more than six times greater than previous estimates. Whereas prior work has suggested that a thinner mixed layer should accompany a more stratified upper ocean5-7, we find instead that the summertime mixed layer deepened by 2.9 ± 0.5 per cent per decade, or several metres per decade (typically 5-10 metres per decade, depending on region). A detailed mechanistic interpretation is challenging, but the concurrent stratification and deepening of the mixed layer are related to an increase in stability associated with surface warming and high-latitude surface freshening8,9, accompanied by a wind-driven intensification of upper-ocean turbulence10,11. Our findings are based on a complex dataset with incomplete coverage of a vast area. Although our results are robust within a wide range of sensitivity analyses, important uncertainties remain, such as those related to sparse coverage in the early years of the 1970-2018 period. Nonetheless, our work calls for reconsideration of the drivers of ongoing shifts in marine primary production, and reveals stark changes in the world's upper ocean over the past five decades.

Locally stationary spatio-temporal interpolation of Argo profiling float data.

Argo floats measure seawater temperature and salinity in the upper 2000 m of the global ocean. Statistical analysis of the resulting spatio-temporal dataset is challenging owing to its non-stationary structure and large size. We propose mapping these data using locally stationary Gaussian process regression where covariance parameter estimation and spatio-temporal prediction are carried out in a moving-window fashion. This yields computationally tractable non-stationary anomaly fields without the need to explicitly model the non-stationary covariance structure. We also investigate Student t-distributed fine-scale variation as a means to account for non-Gaussian heavy tails in ocean temperature data. Cross-validation studies comparing the proposed approach with the existing state of the art demonstrate clear improvements in point predictions and show that accounting for the non-stationarity and non-Gaussianity is crucial for obtaining well-calibrated uncertainties. This approach also provides data-driven local estimates of the spatial and temporal dependence scales for the global ocean, which are of scientific interest in their own right.