Searchers listening for the pinger aboard Malaysia Airlines Flight 370 are entering their final days of hope. The pinger's batteries are expected to die very soon, even though there are better indications that searchers are getting closer. This is day 31 and batteries are supposed to last 30-45 days. Associate Professor Annalisa Braccois is an oceanographer and is keeping tabs on the search.
The body of the plane likely lies on the ocean floor somewhere in the Southern Indian Ocean under more than 10,000 feet of water. At these latitudes, currents are highly variable, may reach two cm/s in strength and are often accompanied by high waves and bad weather. The floating debris identified in numerous satellite images as likely belonging to the airplane can move as much as 100 km per day. Additionally, due to the difficult weather conditions, their recovery and certain attribution has not been possible yet.
I think it is worth remembering that a similar (but, in many respects, far easier) search took place off the coast of Brazil in 2009. The search for Air France flight 447 took two years to complete. In that case, the radars had tracked the airplane for most of the flight, debris were found and attributed within two weeks and the ocean conditions are more accessible. Nonetheless, the body of the missing airplane was located and recovered in the summer of 2011.
I believe that only a strategy similar to the one adopted for that flight has some potential to succeed for MH370. But recovering the debris (and being able to track them back in time on the satellite images) is a required ingredient.
If this can be done, it will be possible to use numerical models of the ocean circulation run in the so-called ‘assimilative mode’ (i.e. assimilating the mean current velocities derived from satellite measurements of surface ocean elevation) to track back in time the origin of the debris to the day when the plane likely crashed. Practically, a model will be able to reconstruct the pathway the debris traveled from the day of recovery to the day of the crash. Such pathway is not exempt from errors. By running the model multiple times with slightly different initial conditions, it will be possible to create a probability distribution of the location of the impact, allowing for restricting the radius of the search and hopefully account for the model error.
Weather permitting, ships equipped with sonars and AUVs (autonomous underwater vehicles) will then be able to perform a more restricted search and possibly identify the remainder of the plane underwater. However, both planning and performing those operations requires time. And luck.
For more information, or to schedule an interview, please contact: