8 September 2009 - for 8-10 September 2009

Hi Glen and all,

7 September 2009 - for 7-9 September 2009

Hellow Glen,

Thanks. Thanks a lot for the good data! We had high uncertainty in that region, so this is great. Thanks to the whoi support and OR1 crew too.

I am as puzzled as you are since conditions seem different. This is good, we should learn from it. This is the Q of QPE!

A few very preliminary ideas:

1. we may have hit a internal tide-wave (we have sections on the web, internal tidal effects are large now).
2. I have seen bands of low salinity in the region, most likely advected from the Kuroshio as it decelerates locally. I assume this could lead to downwelling to the left of the Kuroshio by geostrophic adjusment to lower speeds, hence the lower salinities, higher T. The Kuroshio accelerates as it goes over the I-lan ridge (hence the upwelling to its left), then decelerates as it hits the shelf, hence the lower salinities I guess you see further to the northeast. See for example:
   • T and S forecast at 100m on 03Sep06z (last seasoar data used for that run was 31 Aug). Sal at the location of the survey 8 is around 34.4 to 34.55. Even though this forecast has not assimilated any seasoar data since leg 1, it would match your data of the survey 8.
   • and S forecast at 100m on 07Sep06z (this is the forecast we issued yesterday, which assimilated data up to 06 Sep 10z): S at the location of the survey 8 in that forecast is 34.5 to 34.6. Your seasoar data shows 34.35 to 34.5 at that depth.
3. we hit some waters coming from the shelf along the tip of Taiwan in a mean mesoscale cyclonic circulation around the dome (dome that is moving to the northeast). I have seen this in the runs at times but with strong tides, it is hard to detect now.

Assuming you stayed long enough on the western edge of your survey (3 to 6 hours), the internal tides are unlikely. This means that the Kuroshio acceleration and deceleration is perhaps a better guess. It will take time to figure this one out.

We have launched new forecasts (as you know we made a typo last night so a file that did not exist killed the runs). We also have uncertainty fields. Once they are done, we will let you know.

The Canyon plan is good.

Thanks a lot again for the data!!

Pierre

Hi Tim,

Thanks for the news.

I just quickly looked at the Seasoar survey, they should be very helpful. Thanks a lot.

In the predictions, the "cold dome" has been eroding to the northeast and it now looks more or more looks like a patch of eddies and filaments, modulated by strong tides and waves. I have also seen some low salinity bands coming in from I-Lan (this may happen when the Kuroshio decelerates, by ageostrophic vertical velocity cells) . We have a lot of uncertainty at the cold dome inflow area in our estimates, so the data will help a lot. In any case, I hope we went northwest enough to be sure on the dome.

Thanks,

Pierre

6 September 2009 - for 6-8 September 2009

Adaptive Sampling Recommendations

Hi Glen,

For the Cold dome survey, I made some suggestions in my "Ocean Physics Forecast, update up to 08 Sep 00Z, 2009" email.

In your plan, you have:

September 7- Day 0600-1800 SeaSoar Operations Repeat of Survey 3 of Leg 1 in center of Cold Dome. Adjust as needed for Cold Dome structure and/or Kuroshio Intrusion. Along-shelf lines at 130 m isobath and further onshore to determine Cold Dome structure, adaptive.

5 September 2009 - for 6 September 2009

Hi Glen and Chris,

We have posted the acoustic predictions for day 3, a cross-slope event in the cruise plan. These acoustic predictions are for Sep 06 15Z, 18Z and 21Z.

Attached is the forecast expected Range (km) in the cross-slope direction on Sep 06 15Z for a source at (xs,ys) , i.e. this is the Mean Range on 06Sep 15Z at which TL in the cross-slope direction drops to 60±5 db averaged over 20-100 m depths, for source at 900Hz, 60m. The variability in space of these results seem to match those of Kevin. The mean ranges maybe a bit off, they depend on the bottom we use which Jinshan and I estimated (using the QPE Pilot data) to be, "for a depth < 140m: Sand sediment (sound speed ratio: 1.038, ρ: 1.90 g/cm3, α: 0.9 dB/λ)". Let us know what you find.

I also included the standard deviation of the TL in depth, with respect to a mean defined as "TL for source at 900Hz, 60m averaged over all bearings, 6-8km ranges and 20-100m depths". Here, the variability in time is interesting.

All of these results seem to match well with Kevin GA's plans and suggestions, so I also suggest to do Kevin's plan (western side). This assumes the ocean forecasts and bottom estimates are not too bad, but the Seasoar data should help us find out what is right/wrong, so we all win with this choice.

Best wishes,

Pierre

Hi Glen and all,

4 September 2009 - for 5 September 2009

Hi Glen and Chris,

Please find attached the acoustic forecast and adaptive sampling suggestion for day 2, a 130m isobath run (as in the modified cruise plan). The predictions attached are for 15Z, Aug 05, which is 2300 CST. To reduce email load, next I will email only both of you and Phil the forecasts for 18Z and 21 Z on Aug05 (later in the night in CST). I assume you can distribute these forecasts on the ship (all plots will be on the web). As yesterday, all of the information is along the 130m isobath (see plot with location attached).

In these plots, we still use reciprocity. On two images, you have two vertical sections, which contain a 2D color plot and a 1D std over depth. One image corresponds to an eastward propagation, the other to a westward propagation. Consider the eastward propagation, for example. The 2D color section tells you that if the receiver is at that depth and lat-lon along the line (a pixel), it will hear a source at 6-8km to the west (to the left) with XX db (as colored in the pixel). Below, you have the standard deviation with depth. Similar things can be said for the westward propagation 2D plots.

Then, we give the standard TL section plots, for a eastward propagation and for a westward propagation, selecting the location of maximum std with depth and the location with minimim std with depth. You could select either of them depending of what the goals are.

Comparing our forecasts for 15Z, 18Z and 21Z, the forecast indicate that propagation is better earlier (e.g. around 15Z than 21Z).

You could also decide to launch at a location for which the receiver will have minimum loss or maximum loss.

Of course, these are mean and variability estimates, which are both uncertain.

3 September 2009 - for 4 September 2009

Hi Glen and Chris,

Please find attached info on the acoustic forecast and adaptive sampling for day 1. The predictions are for 15Z, Aug 04, which is 2300 CST. Of course, 3h and 6h later, the situation will be different, but around 15Z, Aug 04, the forecast should be representative, with perhaps the mean off due to bottom uncertainties. All of the information is along the 130m isobath (or close to it, it is along a along-slope line in our horizontal map plots, see plot with location attached)

In these plots, we used reciprocity as I was telling Tim this morning on skype. I thought this would be less confusing for the user.

On two images, you have two vertical sections, which contain a 2D color plot and a 1D std over depth. One image corresponds to an eastward propagation, the other to a westward propagation. Consider the eastward propagation, for example. The 2D color section tells you that if the receiver is at that depth and lat-lon along the line (a pixel), it will hear a source at 6-8km to the west (to the left) with XX db (as colored in the pixel). Below, you have the standard deviation with depth. Similar things can be said for the westward propagation 2D plots.

Then, we give the standard TL section plots, for a eastward propagation and for a westward propagation, selecting the location of maximum std with depth and the location with minimim std with depth. You could select either of them depending of what the goals are.

TL Vertical Sections - 4 September - 1500Z
130m Location	Mean Eastward Propagation	Mean Westward Propagation	TL Field Eastward Propagation	TL Field Westward Propagation

You could also decide to launch at a location for which the receiver will have minimum loss or maximum loss.

Note that all of this is a valid forecast around 15Z.

Of course, these are mean and variability estimates, which are both uncertain.

29 August 2009 - for 30 August 2009

Hello Jim and Frank,

I checked the lat/lon of survey 1 (in red at right). This seems great. If it is somehow possible, to try to go a bit more to the southwest, I attached a new set of positions (in blue at right) for a modified survey 1.

Frank: to get the new blue track, here is what I did using your files

 plot(txyz(:,2)-0.1,txyz(:,3)-0.1,'bx')

The southwest corner of this new blue track should see something. Of course, I am not sure if you can do this (in light of fishing and other OPS constraints, etc). As long as you can go west enough, you should see something in the data (with currents to the northwest at 50m).

Here are some updated model fields (forecasts for 29 August): see the inflows of cold water (blue) just north of Taiwan.

09Z	12Z	15Z	18Z	21Z

Hope this helps,

Pierre