PPRuNe Forums - View Single Post - AF447
Thread: AF447
View Single Post
Old 20th Dec 2009, 02:07
  #4569 (permalink)  
mm43
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
A translated portion of the Annex

1.3 Reverse drift calculations

The knowledge of currents and winds in the area of the accident can provide a theoretical estimate of the previous positions of each body and debris identified by calculating a trajectory in reverse. An estimate of a possible impact zone at the time of the accident on 1 June 2009 at about 2 h 15 UTC, can be had by taking this route. This calculation is called retro-drift or reverse drift.

A team of experts from Météo-France, SHOM, IFREMER, Mercator Ocean and CROSS Gris-Nez worked on the calculations of drift. The U.S. Navy, Brazilian Navy and U.S. Coast Guard (USCG) also provided the results of their calculations.

The different points of retro-drift were calculated from the positions of Debris and bodies found on 6 and 7 June, and especially from the position of the vertical stabilizer found on 7 June 2009.

Calculations by the USCG for an estimated position of the vertical stabilizer on 1 June at 2 h 15 placed it at 30 NM south-east of its position on 7 June. Furthermore, an estimated position for the bodies on 1 June was also calculated, assuming that they drifted to the surface from the scene of the accident.

Calculations by the Brazilian Meteorological Service and the U.S. NAVY results approximate the USCG position. These simulations used the same NCOM model (1), and the results are in the Alpha area that was explored by TPLs (US Navy's Towed Pinger Locators).

Calculated retro-drift of the vertical stabilizer made by Météo - France has in turn provided an estimated impact position on 1 June at 2 h 15 approximately 50NM to the south-west of its position on 7 June. This zone extends about 25NM depending on an assumed vertical immersion of the tail of between 80 and 100%. These results differ from those provided by other simulations of backward drift.

Calculations by Météo France were based in particular on the current model developed by Mercator (2). As the results from Météo France were in a relatively remote part of the Alpha area, this area to the west was explored by the SNA (Nuclear Attack Submarine "Emerald") with the main objective to remove any doubts.
(1) NCOM model (Navy Coastal Ocean Model) is based on a 1/8° resolution grid (ie approximately 15 km). It is fed by the Air NOGAPS model (resolution 50 km) and receives daily U.S. Navy Data observations and data forecasts at hourly intervals. This data is regularly assimilated and compared internally, along with data returned by drift buoys.
(2) The Mercator model comprises two sub models, PSY2 and PSY3. - Mercator PSY2 is an Atlantic and Mediterranean High resolution model on a 1/12° (or 9 km) grid and with fifty vertical levels. The surface wind model is the ECMWF (European Center for Medium - Range Weather Forecasts) at 25 km resolution. It produces current fields daily. - Mercator PSY3 gives an over all Oceans average resolution of 1/4°, i.e. approximately a 25 km grid.
I'll try and translate other parts of the Annex when I get time.



I created the graphic above last August to give a general idea of how the currents were behaving at the time of the crash. The reverse drift positions for the vertical stabilizer provided in the BEA Interim Report No.2 have been added in magenta.

mm43

Last edited by mm43; 21st Dec 2009 at 16:12. Reason: expanded some acronyms, i.e. SNA and TPL
mm43 is offline