We study here the relationship between the unstable manifold –computed via the backward-in-time finite-size Lyapunoc exponents (FSLE)– and the location of surface mixed layer instabilities (SMLI) in the regional HYCOM simulation.
In Fig. 1 is show the FSLE for Jan. 15, 2010. The FSLE on that day is co-located with the sharp gradients of sea surface temperature (SST) that indicate where the SMLI occur. In order to test if this is also the case in the ocean, we try next to relate the FSLE (that we can compute from AVISO SSH) to the anomalies of buoyancy frequency (N) near the surface (Fig. 3). We also find that in most cases, the anomalies in N are close to ridges of FSLE.
We constructed a first test that could be used for the observations. For each location, I relate the value of N with the maximum value of FSLE within 10 km around. The result is shown in Fig. 4. We see that the values are distributed over the upper right corner; all large values of N are associated with large values of FSLE (but low values of N can be associated with low or large values of FSLE).
I perform the test for different days in the year but I think it is going to be difficult to use that test for the observations and I need to find another one.
Figure 1: FSLE on Jan. 15, 2010 in the HYCOM simulation. The FSLE were computed using 90-day of surface geostrophic flow prior to that date.
Figure 2: Magnitude of gradient of SST on Jan. 15, 2010.
Figure 3: As in Fig. 1 with contours of N at 10 m.
Figure 4: N at 10 m versus the local maximum value of FSLE. For each location, we keep the maximum value of FSLE found within 10 km.
See fsle_HYCOM.m in RESEARCH/PROJECTS/MARINE_BIOLOGY/SUBMESOSCALE_PROCESSES/FSLE/analysis/HYCOM on ipu1.