Eos, Vol. 95, No. 21, 27 May 2014
VOLUME 95
NUMBER 21
27 May 2014
EOS, TRANSACTIONS, AMERICAN GEOPHYSICAL UNION
PAGES 173–180
Seafloor in the Malaysia Airlines
Flight MH370 Search Area
mercial fishing gear or on whales to track
their migration.
The dashed lines in Figure 1 roughly parallel the estimated flight paths suggested by
the Doppler handshake analysis and encompass the areas of the various reported acoustic contacts. Here the swath defined by these
dashed lines is referred to as the acoustic
search zone.
PAGES 173–174
On the morning of 8 March 2014, Malaysia
Airlines flight MH370, from Kuala Lumpur to
Beijing, lost contact with air traffic control
shortly after takeoff and vanished. While the
world waited for any sign of the missing
aircraft and the 239 people on board, authorities and scientists began to investigate what
little information was known about the plane’s
actual movements.
As days and weeks passed, the search
began to focus on the Indian Ocean to the
west of Australia—far from the flight’s intended path. Clues to how the plane got so
far off course may be in the plane’s “black
boxes”—its flight data and cockpit voice
recorders. Finding the recorders is therefore a
top priority.
Little is known about the seafloor from
ship-borne echo sounder measurements in
the region where flight MH370 is believed to
have crashed. Available depth measurements
cover only 5% of the 2000 by 1400 kilometer
area in Figure 1 (a high-resolution copy of this
figure may be found in the additional supporting information in the online version of
this article), and only a very few of them were
acquired with modern acoustic and navigational systems. This lack of data makes the
search for MH370 all the more difficult. It also
highlights how most seafloor features are very
poorly resolved. However, satellite altimeter
measurements provide global bathymetry
estimates at a resolution of about 20 kilometers
[Smith and Sandwell, 1997], making it at least
possible to map the major seafloor features
in the search area.
Constraining the Search Area
When MH370 went missing, the search
initially focused in the Gulf of Thailand and
the South China Sea, along the intended flight
path from Kuala Lumpur to Beijing. After a
few days, reports of possible radar contacts
caused the search to expand westward,
across the Malay Peninsula to the Strait of
Malacca and the Andaman Sea.
BY W. H. F. SMITH AND K. M. MARKS
Later, the search was extended even farther
west into the Indian Ocean, when it was
realized that the aircraft’s satellite communications system had remained operating for
several hours. Although the aircraft was not
actively transmitting information about its
health and position, it was automatically
replying to brief queries from a ground station
in a process known as a handshake. These
handshakes, which took place approximately
hourly, were relayed through an Inmarsat
telecom satellite in geostationary orbit over
the central Indian Ocean. The round-trip
travel time of each handshake gave a crude
estimate of the distance between the aircraft
and the satellite while Doppler shifts in the
handshake allowed a rough estimate of the
aircraft’s velocity away from the satellite.
This analysis, completed about 10 days
after the disappearance, was combined with
estimates of when the plane might have run
out of fuel. Together they suggested that the
aircraft might be anywhere in a large area
of the Indian Ocean west of Australia.
MH370’s black boxes were equipped with
“pingers” programmed to emit acoustic signals
if the boxes fell into the sea. The expected
battery life of these pingers was approximately 1 month, so there were only a few days
of expected pings left when it was reported
that the Chinese vessel Haixun 01 had
detected pings on 4 and 5 April in the water
above the east flank of the Batavia Plateau
(see black circle in Figure 1). Over the next
3 days the Australian vessel Ocean Shield
reported three other contacts, one contact
apparently hearing pings emitted by two
distinct devices, in an area above the north
flank of the Zenith Plateau (see red circle in
Figure 1).
The Batavia and Zenith contact locations
are approximately 600 kilometers apart, and
it seems unlikely that pingers at the end of
their battery life could be heard over such
distances, yet sound propagation in the
ocean is quite complex. Nonetheless, Chinese
and Australian authorities seemed confident
that the carrier frequency, duration, and
pulse repetition interval of the pings heard
were likely those of an aircraft black box
pinger rather than those mounted on com-
This paper is not subject to U.S. copyright. Published in 2014 by the American Geophysical Union.
Seafloor Data From Ship Surveys
Data collected by research vessels Vema
and Robert D. Conrad provide the only available depth surveys [International Hydrographic
Organization Data Center for Digital Bathymetry (IHO DCDB), 2014] over the Batavia and
Zenith plateaus (see Figure 1, B and Z). The
Batavia Plateau was crossed by Vema in 1962
and by Conrad in 1965; both dates are prior
to when the U.S. Navy–run Transit satellite
navigation system became standard for
remote ocean navigation. Vema also crossed
the Zenith Plateau in 1977, when Transit was
available.
Navigation on all these expeditions was by
dead reckoning between sparse navigation
points. In the Transit era these points could
be obtained once every few hours and had
accuracies of a few kilometers. Prior to Transit,
these points would be celestially determined,
perhaps days apart in bad weather. Both
expeditions occurred prior to the use of
modern multibeam echo sounders (MBES),
so depth measurements were collected by
single, wide-beam echo sounders that recorded on analog paper scrolls, the digitizing
of which is often in error by hundreds of
meters [Smith, 1993].
Of the publicly available data, ships carrying MBES and navigating by GPS have passed
through the area only twice; the R/V Melville
and the R/V Knorr each made a passage in
1996, but neither surveyed any of the major
topographic features. Apart from those two
transits, all other data publicly available in the
area are analog, low-tech soundings, and
most predate the era of GPS navigation.
The estimate that 5% of the seafloor in the
area has been mapped comes from inspection of the data in IHO DCDB [2014], the
General Bathymetric Chart of the Oceans
(GEBCO) [2010], and versions 15 and 16 of the
bathymetric model by Smith and Sandwell
[1994, 1997]. Some additional data may exist
but not reside in openly available databases.
For example, the Zenith Plateau was named
Eos, Vol. 95, No. 21, 27 May 2014
In addition, depth estimates from satellite
altimetry are most accurate where the seafloor topography is moderate and composed
of oceanic crust overlain by less than 200
meters of sediment [Smith and Sandwell,
1994]. Whittaker et al. [2013] estimate sediment thicknesses in the area varying from
12 meters to 1.5 kilometers, and Deep Sea
Drilling Project site 256 (gray dot in Figure 1)
found 251 meters of sediment [Davies et al.,
1974].
The seabed in the MH370 search area
records a complex geologic history of the
breakup of Australia, India, and Antarctica
approximately 130 million years ago [Williams
et al., 2013a]. The shallowest depth in the area
shown in Figure 1 is about 237 meters on
Broken Ridge, a structure related to the separation of Australia and Antarctica whose
conjugate, the Kerguelen Plateau, lies on the
Antarctic plate. Within the acoustic search
zone of Figure 1, the shallowest depth is about
1637 meters at the summit of Batavia Plateau.
The deepest point in the area shown also lies
within the acoustic search zone, where the
trough of the Wallaby-Zenith Fracture Zone
plunges to an estimated 7883 meters, just
south of the Zenith Plateau. These plateaus
are fragments of continental crust, leftovers
of Indo-Australian continental breakup
[Williams et al., 2013b], and are embedded in
old, deep seafloor.
Unknown Seafloor Depths
Complicate the Search
Fig. 1. Seafloor topography in the Malaysia Airlines flight MH370 search area. Dashed lines
approximate the search zone for sonar pings emitted by the flight data recorder and cockpit
voice recorder popularly called black boxes. The first sonar contact (black circle) was reportedly
made by a Chinese vessel on the east flank of Batavia Plateau (B), where the shallowest point in
the area (S) is at an estimated depth of 1637 meters. The next reported sonar contact (red circle)
was made by an Australian vessel on the north flank of Zenith Plateau (Z). The deepest point in
the area (D) lies in the Wallaby-Zenith Fracture Zone at an estimated depth of 7883 meters. The
Wallaby Plateau (W) lies to the east of the Zenith Plateau. The shallowest point in the entire
area shown here is on Broken Ridge (BR). Deep Sea Drilling Project (DSDP) site 256 is marked
by a gray dot. The inset in the top left shows the area’s location to the west of Australia. Seafloor
depths are from the General Bathymetric Chart of the Oceans [2010].
for R/V Zenith (see GEBCO Undersea Feature
Names Gazetteer, http://www.ngdc.noaa.gov/
gazetteer), which apparently discovered it in
the 1930s; Veevers et al. [1985] allude to a
crossing of the Zenith Plateau by Royal Australian Navy vessel Cook in 1983. Williams et al.
[2013b] report that R/V Southern Surveyor
obtained some data over the Batavia Plateau
in 2011, although the abyssal depths in the
region adjacent to the plateau exceeded the
designed operating range of its MBES system.
Data from the Zenith, Cook, and Southern
Surveyor expeditions could not be found in
digital form in open source databases, and
they are not used in the Smith and Sandwell
or GEBCO compilations.
Another confounding factor is that the feature names in this area are not universally
agreed upon. Google Earth™ uses the name
Wallaby Plateau for the Zenith Plateau because Google Earth™ displays names used by
the U.S. Board on Geographic Names. However, the international community uses Zenith
Plateau to avoid confusion with another
Wallaby Plateau [Sayers et al., 2002] that lies
closer to the coast of Australia (Figure 1).
Greater standardization of nomenclature and
sharing of data could improve ocean mapping.
Satellite Interpolation of Seafloor Data Gaps
The depths in Figure 1 are from GEBCO
[2010], which uses satellite altimetry to interpolate gaps between ship survey data publicly available in open sources [Smith and
Sandwell, 1994, 1997]. The accuracy of the
positions and depths in these survey data
limits the accuracy of the satellite estimates.
This paper is not subject to U.S. copyright. Published in 2014 by the American Geophysical Union.
The search for airplane debris in the open
ocean is not without precedent: On 1 June
2009, Air France flight AF447 disappeared
over the Atlantic Ocean. However, the present
search for missing Malaysia Airlines flight
MH370 takes place under very different
circumstances.
When AF447 disappeared, there was little
doubt about where it would be found. The
flight had not deviated significantly from its
intended path from Rio de Janeiro to Paris,
and the aircraft was routinely sending
messages monitoring the health of its flight
systems along with its position. Warning and
failure messages generated by these systems
in the last few minutes of flight helped to
locate the crash site, and a surface search
there found floating debris and fuel slicks the
very next day.
In addition, the AF447 crash site was in an
area already 100% covered by a previous
state-of-the-art bathymetric survey (MBES and
GPS), and this knowledge of the undersea
terrain helped searchers select and program
autonomous underwater vehicles (AUVs) to
search for the black boxes. Even so, they were
not recovered until nearly 2 years after the
crash.
In comparison, the MH370 crash site is very
poorly known. There are no measured depths
in public databases at the locations where
ping contacts were reported. Satellite altimetry estimates that depths at the Chinese and
Australian contact locations are about 4300
Eos, Vol. 95, No. 21, 27 May 2014
and 5160 meters, respectively, but these estimates are quite uncertain and might be in
error by approximately 250 meters or more.
Selecting an appropriate AUV and programming its search path require knowledge of the
terrain. A Bluefin 21 AUV initially deployed
over Zenith Plateau to search for debris, for
example, was not designed to operate at
depths below 4500 meters.
Lack of knowledge of seafloor topography
has other consequences. Bottom topography
steers surface currents [Gille et al., 2004]
while bottom roughness controls ocean mixing rates [Kunze and Llewellyn Smith, 2004],
and poor knowledge of these characteristics
limits the accuracy of forecasts of everything
from the path of floating debris to the path of
tsunamis [Mofjeld et al., 2004] and the future
of climate [Jayne et al., 2004].
Seafloor Features: Not as Well Resolved
as Martian Topography
The state of knowledge of the seafloor in
the MH370 search area, although poor, is
typical of that in most of Earth’s oceans, particularly in the Southern Hemisphere. In many
remote ocean basins the majority of available data are celestially navigated analog
measurements [Smith, 1993] because systematic exploration of the oceans seems to have
ceased in the early 1970s [Smith, 1993, 1998;
Wessel and Chandler, 2011], leaving the ocean
floors about as sparsely covered as the interstate highway system covers the United States
[see Smith and Sandwell, 2004, Figure 2].
When these sparse soundings are interpolated by satellite altimetry, as in Figure 1, the
resulting knowledge of seafloor topography is
15 times worse in the horizontal and 250 times
worse in the vertical than our knowledge of
Martian topography [Smith, 2004].
Although a new bathymetric satellite
altimeter mission could improve this situation
significantly [Smith and Sandwell, 2004], ships
with echo sounders remain the best technology for ocean mapping. The global ocean
deeper than 500 meters (that is, deeper than
the continental shelves) could be fully surveyed with state-of-the art navigation and
acoustic multibeam systems with a total effort
of about 200 ship-years of vessel activity at a
total cost less than that of a typical planetary
exploration mission [Carron et al., 2001].
Until there is such an effort, knowledge of
Earth’s ocean floors will remain limited to the
resolution available from satellite altimetry,
which is vastly poorer than our knowledge of
the topographies of Earth’s Moon, Mars, and
Venus. Perhaps the data collected during the
search for MH370 will be contributed to
public databanks and will be a start of greater
efforts to map Earth’s ocean floor.
Acknowledgments
Reviews by S. Stein and additional suggestions from E. W. Leuliette improved the manuscript. W.H.F.S. was formerly and K.M.M. is
currently chair of the Technical Sub- Committee
on Ocean Mapping of the United Nations Intergovernmental Oceanographic Commission
(IOC) and IHO joint committee for GEBCO.
The paper’s contents are solely the opinions
of the authors and should not be construed
as an official statement of policy, decision,
or position on behalf of the National Oceanic
and Atmospheric Administration (NOAA),
the U.S. government, GEBCO, the IOC, or
the IHO.
References
Carron, M. J., P. R. Vogt, and W.-Y. Jung (2001), A
proposed international long-term project to systematically map the world’s ocean floors from
beach to trench: GOMaP (Global Ocean Mapping
Program), Int. Hydrogr. Rev., 2(3), 49–55.
Davies, T. A., et al. (1974), Site 256, Initial Rep. Deep
Sea Drilling Proj., 26, 295–325.
General Bathymetric Chart of the Oceans (GEBCO)
(2010), GEBCO_08 grid, version 20100927, http://
www.gebco.net/data_and_products/gridded
_bathymetry_data/, Br. Oceanogr. Data Cent.,
Liverpool, U. K.
Gille, S. T., E. J. Metzger, and R. Tokmakian (2004),
Seafloor topography and ocean circulation,
Oceanography, 17(1), 47–54, doi:10.5670/oceanog
.2004.66.
International Hydrographic Organization Data
Center for Digital Bathymetry (IHO DCDB) (2014),
Marine Trackline Geophysical Data, http://ngdc
.noaa.gov/mgg/geodas/trackline.html, Natl.
Geophys. Data Cent., Boulder, Colo.
Jayne, S. R., L. C. St. Laurent, and S. T. Gille (2004),
Connections between ocean bottom topography
and Earth’s climate, Oceanography, 17(1), 65–74,
doi:10.5670/oceanog.2004.68.
Kunze, E., and S. G. Llewellyn Smith (2004), The
role of small-scale topography in turbulent mixing
of the global ocean, Oceanography, 17(1), 55–64,
doi:10.5670/oceanog.2004.67.
Mofjeld, H. O., C. Massell Symons, P. Lonsdale,
F. I. González, and V. V. Titov (2004), Tsunami
scattering and earthquake faults in the deep
Pacific Ocean, Oceanography, 17(1), 38–46,
doi:10.5670/oceanog.2004.65.
This paper is not subject to U.S. copyright. Published in 2014 by the American Geophysical Union.
Sayers, J., I. Borissova, D. Ramsay, and P. A. Symonds
(2002), Geological Framework of the Wallaby
Plateau and Adjacent Areas, Geosci. Aust. Record,
vol. 2002/21, 133 pp., Geosci. Aust., Canberra.
Smith, W. H. F. (1993), On the accuracy of digital
bathymetric data, J. Geophys. Res., 98(B6),
9591–9603, doi:10.1029/93jb00716.
Smith, W. H. F. (1998), Seafloor tectonic fabric from
satellite altimetry, Annu. Rev. Earth Planet. Sci., 26,
697–747, doi:10.1146/annurev.earth.26.1.697.
Smith, W. H. F. (2004), Introduction to this special
issue on bathymetry from space, Oceanography,
17(1), 6–7, doi:10.5670/oceanog.2004.62.
Smith, W. H. F., and D. T. Sandwell (1994), Bathymetric prediction from dense satellite altimetry
and sparse shipboard bathymetry, J. Geophys.
Res., 99(B11), 21,803–21,824, doi:10.1029/
94jb00988.
Smith, W. H. F., and D. T. Sandwell (1997), Global
sea floor topography from satellite altimetry
and ship depth soundings, Science, 277(5334),
1956–1962, doi:10.1126/science.277.5334.1956.
Smith, W. H. F., and D. T. Sandwell (2004), Conventional bathymetry, bathymetry from space, and
geodetic altimetry, Oceanography, 17(1), 8–23,
doi:10.5670/oceanog.2004.63.
Veevers, J. J., J. W. Tayton, B. D. Johnson, and
L. Hansen (1985), Magnetic expression of the
continent- ocean boundary between the western
margin of Australia and the eastern Indian Ocean,
Z. Geophys., 56, 106–120.
Wessel, P., and M. T. Chandler (2011), The spatial
and temporal distribution of marine geophysical
surveys, Acta Geophys., 59(1), 55–71, doi:10.2478/
s11600-010-0038-1.
Whittaker, J., A. Goncharov, S. Williams, R. D.
Müller, and G. Leitchenkov (2013), Global
sediment thickness data set updated for the
Australian-Antarctic Southern Ocean, Geochem.
Geophys. Geosyst., 14(8), 3297–3305, doi:10.1002/
ggge.20181.
Williams, S. E., J. M. Whittaker, R. Granot, and D. R.
Muller (2013a), Early India-Australia spreading
history revealed by newly detected Mesozoic
magnetic anomalies in the Perth Abyssal Plain,
J. Geophys. Res. Solid Earth, 118(7), 3275–3284,
doi:10.1002/jgrb.50239.
Williams, S. E., J. M. Whittaker, and R. D. Müller
(2013b), Newly-recognised continental fragments
rifted from the West Australian margin, paper
presented at the West Australian Basins Symposium 2013, Pet. Explor. Soc. of Aust., Perth,
Western Australia, Australia, 18–21 Aug.
Author Information
WALTER H. F. SMITH and KAREN M. MARKS, Laboratory for Satellite Altimetry, National Oceanic and
Atmospheric Administration, College Park, Md.;
email: [email protected]