Inverse estimation of diffusivity coefficients from salinity distributions on isopycnal surfaces using Argo float array data

• Published in: Journal of oceanography Vol. 77; no. 4; pp. 615 - 630
• Main Author: Kouketsu, Shinya
• Format: Journal Article
• Language: English
• Published: Singapore Springer Singapore 01.08.2021; Springer Nature B.V
• Subjects: Coastal zone; Coefficients; Diffusion coefficients; Diffusivity; Earth and Environmental Science; Earth Sciences; Equatorial regions; Freshwater & Marine Ecology; General circulation models; Gyres; Inverse estimation; Mass; Mixed layer; Ocean basins; Ocean circulation; Oceanography; Original Article; Salinity; Salinity effects; Sea surface; Spatial variations; Subduction; Thickness; Tracers; Water; Water masses; Horizontal mixing; Vertical mixing; Overturning circulation
• ISSN: 0916-8370; 1573-868X
• DOI: 10.1007/s10872-021-00595-5

Distributions of diapycnal, isopycnal, and thickness diffusivity coefficients were estimated using long-term (2001–2015) mean salinity distributions on neutral density ( γ ) surfaces in seven oceanic basins (except the equatorial region). Diffusivity amplitudes and spatial differences were consistent with distributions previously estimated by various parameterizations of fine-scale structures. This result suggests that these parameterizations are consistent with water mass modifications shown by salinity distributions. Although the estimated diapycnal diffusivity coefficients were the lowest among those used by general circulation models to reproduce tracer distributions, they were closer to directly observed values. Because diapycnal diffusivity was generally small, meridional transports in the upper layers (above about 1600 dbar) associated with diapycnal mixing were relatively small. This result may indicate that diapycnal mixing in coastal regions has relatively large effects on meridional transport, along with water mass formation and mixing and wind forcing at the sea surface. The major water mass transports in subtropical gyres estimated in this study were consistent with previous estimates and corresponded to subduction rates determined using mixed layer thickness budgets and transient tracer inventories. These results, based on recent observation network data, show that understanding of water mass formation and modifications can be improved by taking into account diffusive effects, but more detailed assessments of estimation errors are needed to clarify the role of mixing in water mass modification.