Archive for July 10th, 2008

Stuck in a corner – Altimetry to the rescue

Thursday, July 10th, 2008

So where have the bloggers gone? - no entries since monday.  Yep, something was awry. You can usually tell how the day is going based on the number of blog entries.   Here we had a string of 4 days with no entries - our longest string yet. We were all pretty  busy trying to unravel this puzzle while simultaneously building the new tools we needed to unravel it.  It took us till yesterday, when a combination of new tools and new altimetric data from Colorado showed us that we had flown into a corner, and the only way out was to circle around and try again farther north. 

The story begins Sunday, where RU17 was heading northeast through a gap between two westward flows based on the Navy model.  The uncertainty issue was the eastern side of Romora Ring, the strong cold eddy we had just left.  None of the spatial datasets we had (satellite SST, SSH or forecasts) agreed on the shape or location of the eastern side of that ring.  We had planned on being advected north by the eddy on the eastern side, but the northward currents never appeared.  Instead the currents stayed to the northeast, continuing to decrease as we left the eddy early on monday. We continued in this direction, but by monday evening, the currents flipped, and were slowly increasing against us.  By tuesday, the currents to the southwest had inceased, turning into a headwind that stopped us dead. By evening we where being pushed backwards. The Navy model said the gap we were shooting for was closing, but it didn't really matter how big the gap was, we couldn't fight this current. 

So we needed a way out. Glider exit strategies from a strong current are well known.  Just like a swimmer caught in a rip current, you swim perpendicular to the current until time or space results in a change.  So which way do we turn?  RIght or left?  Usually we would choose the direction perpendicular to the current that had at least some component towards the east, towards the Azores or even Spain, our ultimate destination. But in this case we had the Navy model saying we would encounter even stronger currents in the wrong direction if we chose the direction to the east.  So on tuesday we decided we had to trust our models, and we turned RU17 from northeast, heading it first to the north, then the northwest, and finally to the west as the currents were pushing us due south.  It really hurts to give up hard fought ground like that, but we had no choice, we had to find a favorable current ride or spend valuable energy just station keeping at best.

Then on wednesday, as we were trying to fly out of the southward current, some new altimetry came in.  We did not check were the satellite overpasses where that day, we didn't have to.  The large oval cold ring we had just left was now a perfect circle in the altimetric data, and was now in agreement with the model.  The discrepancies of the last few days went away with the new data, features were in the same locations, and the currents being experienced by the glider matched what was being seen from space.  Finally, we had a roadmap with a way out (see below).  And not only did we have a roadmap, we now had the roadmap displayed in our google earth interface.  For the last few days we had been staring at tiny images on computer screens, or printing out hardcopies, trying to plot glider positions and velocity vectors by eye. Now we had an interface that did all this automatically, allowing us to zoom in and out as we needed.


With the new roadmap, we charted a course to the west that would take us into a strong region of northward velocities that we would use to slingshot us around the high in the sea surface height that is presently blocking our path east.

Below is the zoom in of the path of RU17.   It looks like strong currents to the north are about a days run to our west, and nothing but trouble is found to our east. For the first time on this mission, we have had to turn west to get to water flowing east.