Table Of Contents

This Page

12.20.10: Evolution of potential density at WHOTS location

Main conclusion

To the light of the HYCOM and 1/10th deg. OFES simulations, the short events of intrusion of deep water into the surface layer are most likely due to cold water inside mesoscale structures. These structures are generated in Winter by the instabilities of the Subtropical Front (STF) to the north and migrate southward, arriving at the WHOTS mooring further south in Spring.


I have here a first look at Yanli’s HYCOM simulation of the region around Hawaii with a 1/25th deg. resolution in the vertical and 33 levels in the vertical. The time series of potential density σ resembles that observed by WHOTS (compare Fig. 1 with Fig. 1a in this note), especially the short events of deep water intrusion during Winter-Spring.


Figure 1: σ at the WHOTS location.

In the HYCOM simulation, such deep intrusion is associated with a mesoscale (about 100 km in diameter) perturbation that has migrated from the North (Fig. 2 and animation of Fig. 3). This feature is small enough that it would not be captured well by AVISO SSH.


Figure 2: σ at 30 m depth on March 3, 2010, at the time of the large intrusion of deep water into the surface layer (Fig. 1).


Figure 3: σ at 30 m depth.

If these anomalies are due to mesoscale features then a time series of σ in the 1/10th deg. OFES simulation should also include these short events. Indeed, these events seem to appear (Fig. 4) even if only 3-day snapshots are available (which would tend to underestimate the number of short –typically 1-day- events). This reinforces the conclusion that the events seen in Winter-Spring in the observations and in the HYCOM run are most likely due to the southward migration of colder water inside mesoscale structure. This conclusion is also reinforced by the observations that all events seem to occur more in Spring than in Winter (see Fig. 1 in this note); this time lag is the time the mesoscale structure takes to close the distance between the Subtropical Front (STF) where the mesoscale structures are generated and the latitude of the WHOTS mooring.


Figure 4: σ at the WHOTS location but in the 1/10th deg. OFES simulation.

Finally, Fig. 5 shows that the short events are well correlated with similar events in nitrate (with the exception of the strong nitrate event after day 200) so that it would also explain the nitrate events observed in ARGO data by Johnson et al. (2010). It would be interesting to have a look at a time series in the 1/10th deg. OFES simulation similar in characteristics to the ARGO time series and see if the events in density and nitrate are also qualitatively similar to the observed events.


Figure 5: nitrate at the WHOTS location in the 1/10th deg. OFES simulation.