New research points to the crash site of missing plane MH370
This article by Tom Rippeth of the School of Ocean Sciences appeared on The Conversation. Read the original article.
Two years on and Malaysia Airlines flight MH370 is still missing. The plane disappeared on March 14 2014, probably over the southern Indian Ocean to the west of Australia. Despite an estimated $130m search by Australian, Chinese and Malaysian authorities, covering 120,000 square km of ocean (an area around half the size of the UK), the crash site and the bulk of the aircraft have not been found.
However, in July 2015 the first evidence of the crash was discovered. This debris washed up on the Indian Ocean island of Réunion, some 4,000km to the west of the area where we think the plane disappeared. Now researchers at the Euro-Mediterranean Center on Climate Change in Italy have used information about that debris and more found since to make a more accurate estimate of the location of the crash site.
In the new prediction, published in the journal Natural Hazards and Earth System Sciences, the researchers found that the crash site was likely to be between 28 and 35 degrees south in an area roughly west of Perth (on the Australian west coast). This partially overlaps the present underwater search area, but also extends the likely area for the plane wreckage further to the north.
These new results suggest that if the present search proves unsuccessful the search area should be extended in that direction. The research also indicates that search teams looking for more debris should focus their efforts on the coasts of Tanzania and Mozambique, as well as the islands of Madagascar, Réunion, Mauritius and the Comoros.
To make their estimate, the researchers built a computer model combining recorded information about Indian Ocean surface currents and winds, from March 14 2014 to the present date. This allowed them to simulate the forces that would have affected the plane’s debris since the crash and so effectively follow the route of the floating debris back to the likely crash site.
They represented the aircraft debris with floating particles (assuming it was light enough to float) and allowed them to be dispersed by the surface ocean currents and winds which are reproduced in the model. Because they didn’t know the exact position of the crash, they ran the model many times, each time starting with the particles in a slightly different location within the region where the crash was thought to have occurred.
The simulations that ended with particles in a similar location to where the real debris was found were then used to narrow down the likely area of the crash site. As further debris is discovered, the new information can be incorporated into the model to help pinpoint the crash site still further.
Oceanographers have been applying such “particle tracking” techniques to drifting objects (for example, 80,000 trainers that fell into the sea in 1990) to help identify the pathways of ocean currents for many years. In combination with numerical models, this technique has been applied to numerous problems, such as mapping the buildup of litter in the major ocean currents and predicting the drift of missing persons from the point they were last seen.
What makes this latest study unique is that the researchers were able to take advantage of the greatly increased computing power of modern supercomputers. As such, they were able to make a series of many simulations and estimate the likelihood of specific outcomes. This has enabled the authors to back track over a relatively large distance and time scales with relatively high precision.
One limitation is that the impact of the wind in moving the debris is slightly uncertain. But to account for this, the researchers have run the model with a range of different “wind drag coefficients” and incorporated this uncertainty in the stats.
The results of this study will therefore help guide the ongoing seabed search efforts for MH370 and also demonstrate the potential of such “hindcasting” in aiding future search efforts.
Publication date: 27 July 2016