Hello all,

	   We just finished the latest FCAST with the following to report:

	1) We compared model trajectories and drifter trajectories
	over the 3 day period 12 May (4am EDT) to 15 May (4am EDT).
	These are drifters released by Jim as follows:

	   - two of them over the 50m isobath, one drogued at 13m and 
	         the other at 33m
	   - the other two over the 65m isobath, one drogued at 8m and 
	         the other at 33m

	2) The observed trajectories of the 50m isobath releases were WSW of
	the release points, while the observed trajectories of the 65m
isobath releases 
	were almost due south-southeast of the release points.

	3) These observations held for each set of drifters, i.e., those on
the 50m isobath
	went WSW and those released on the 65m isobath went SSE
"barotropically".
	 
	4) This pattern suggest a "divergence(?!)" away from the front in
the upper 35m 
	of the water column.  Certainly we can't say this for sure, but this
is one
	interpretation.

	5) The model results do not fully capture this structure.  The model
results
	are best for the 50m isobath releases - the model drifter
trajectories are in the
	correct direction, but appear to be a little fast.  Recall that on
the 13th of May
	there was a forecast 14-hour period of northerly winds (from the
north) at 20 knots
	that we never observed at sea and this may be a factor (in the
simple Ekman
	sense) in pushing the drifters further west than the observed drift.


	6) The model trajectories for the drifters released along the 65m
isobath did not 
	show the SSE direction observed in the real drifters.  The model
trajectories are 
	in the SW direction, resulting in clear visual discrepancies when
plotted.  (I do 
	not have numbers, but the observed drifters are more than one tidal
ellipse south of 
	the modeled location, i.e., the real drifters are about 15 kms south
the model 
	drifters).

	6) The good news remains that the assimilation of the ADCP data
strongly arrests
	the SW flow that otherwise would result.  If the assimilation does
not take place,
	the model drifters would overshoot the observed positions
substantially. The 
	additional SSE drift in the 65m isobath released drifters could be
related to some
	of the structures seen in the satellite imagery of several days ago.

	7) The ADCP data assimilated for the above runs are those of the
Endeavor plus those
	of the Edwin Link through yesterday (14 May noon EDT).  We still
have not included 
	drifter m3d data.

	8) We will also run a case tomorrow with the new T and S (.s3r)
fields that Jim computed
	from the bongo surveys (and OAXed into the initial T and S fields).
Perhaps these will 
	provide additional help.

	We will discuss these results further with the Endeavor team by
radio at 1000h on 
	15 May (EDT).  We understand there will be a zodiac-transfer of
supplies for some
	of the bio-labs.  Let us know of any additional data transfers you
wish, floppies, hardcopy 
	plots, etc.  

	Cheers,
	Cisco and Jim.