Publication Date:
2017-01-12
Description:
A new 11-year (2004-2014) monthly 1-degree gridded Argo temperature and salinity dataset with 49 vertical levels from the surface to 1950 m depth (named BOA-Argo) is generated for use in ocean research and modeling studies. The dataset is produced based on refined Barnes successive corrections by adopting flexible response functions based on a series of error analyses to minimize errors induced by non-uniform spatial distribution of Argo observations. These response functions allow BOA-Argo to capture a greater portion of mesoscale and large-scale signals while compressing small-sale and high frequency noise relative to the most recent version of the World Ocean Atlas (WOA). BOA-Argo dataset is evaluated against other gridded datasets, such as WOA13, Roemmich-Argo, Jamestec-Argo, EN4-Argo, and IPRC-Argo in terms of climatology, independent observations, mixed layer depth and so on. Generally, BOA-Argo compare well with other Argo gridded datasets. The RMSEs and correlation coefficients of compared variables from BOA-Argo agree most with those from the Roemmich-Argo. In particular, more mesoscale features are retained in BOA-Argo than others as compared to satellite sea surface heights. These results indicate that the BOA-Argo dataset is a useful and promising adding to the current Argo datasets. The proposed refined Barnes method is computationally simple and efficient, so that the BOA-Argo dataset can be easily updated to keep pace with tremendous daily increases in the volume of Argo temperature and salinity data.
Print ISSN:
0148-0227
Topics:
Geosciences
,
Physics
Permalink