takata, what if you flipped your analysis, so that the hypothesized impact points were west of 30W, and not east of it?
The Brazilian Air Force search grid for June 1 is limited to the projected track. On June 2, the Air Force does two rectangular grid searches that run eastward from the search grid of June 1. Searches on June 3 and 4 are in areas prediominately east of the track. To me that suggests that the June1-4 surface current was similar in velocity and direction (SW to NE) to what is shown in the chart for June 5, and the Air Force search pattern was based on the expected drift from the current.
If AF447 had turned west to divert around the CB complex, or even turned back toward Brazil when systems began to fail, and began breaking up around 31 or 32W, then the prevailing surface current may have carried parts of the wreckage NE to the point near 30W and 3N where the first bodies and confirmed parts of the plane were retrieved.
If the above hypothesis is correct, then I would expect that the first group of bodies that were recovered to be associated with the plane's impact with the water, and the second larger group of bodies which was subsequently found much further north to possibly have left the plane during the initial break-up phase. (There are enough bodies that the investigators should be able to match bodies with assigned seats and test the validity of the above hypothesis.) This clustering of the bodies generally into two distinct groups would be the reverse of what one might expect if the plane had remained on a NNE heading, and broken up.