Characterization of intermittency in the upper ocean

• Main Authors: Isern-Fontanet, Jordi, Turiel, Antonio
• Format: Conference Object
• Language: unknown
• Published: 2019
• Subjects: North Atlantic
• Online Access: http://hdl.handle.net/10261/207663

Waves, Instabilities and Turbulence in Geophysical and Astrophysical Flows (WITGAF 2019), 8-20 July 2019, Cargèse, France.-- 1 page, 1 figure Intermittency is a key property of turbulent flows such as the ocean. In this study, we investigate it in the upper ocean (z≤−500 m) from in situ measurements of velocities provided by the Oleander project. The dataset consists on weekly profiles of horizontal velocities between New Jersey and Bermuda in the North Atlantic obtained by an ADCP mounted on a ship-of-opportunity. Rather than using the canonical approach of computing the velocity structure functions and, then, characterize the anomalous scaling, we computed the singularity exponents h(~x) of the gradient of longitudinal and traverse velocity components and the associated singularity spectra D(h). The preliminary analysis has been focused on the symmetry about the mode of theD (h); the value of the most singular exponent h∞, here approximated by h∞≈ min(h); and the amplitude of D(h), here defined as ∆h−≡hd−h∞, where hd is the mode. The histograms of these three quantities can be found in the first row of the figure. Results unveils that the D(h) are asymmetric about the mode, which is incompatible with symmetric models such as the Log-Normal model. Besides, ∆h− provides a quantification of the anomalous scaling of the structure functions as it can be seen by applying the Legendre transform to the D(h) and h∞ can be interpreted in oceanographic terms as the intensity of the strongest front. The comparison of these two quantities (second row of the figure) reveal a linear relationship between the intensity of fronts given by min(h) and the anomalous scaling given by ∆h−. Furthermore, the slope between these two quantities, which can be related to the intermittency parameter, shows a clear dependence with depth. Finally,our preliminary analysis also show that observations are compatible with the Log-Poisson mode