Recovery of a Lost Subsea Asset at Full Ocean Depth in the Mariana Trench (10,925 m ± 4) Using a Crewed Submersible

Abstract

In 2019, the Deep Submergence Support Vessel Pressure Drop embarked on the third leg of the Five Deeps Expedition. Over the course of 10 days, the full ocean depth submersible Deep Submergence Vehicle Limiting Factor made a record five dives to the bottom of the Mariana Trench: four to the Challenger Deep, one to the Sirena Deep. The submersible was supported by three scientific landers, one of which became stuck on the seafloor at 10,925 m depth. Here we describe how, during the third dive of the campaign, the expedition utilized the submersible to rescue this lost asset from full ocean depth. The expedition was not only significant for its operational jump from single full ocean depth dives to multiple dives in a short space of time, but demonstrated that assets lost in the deepest 45% of the oceans are no longer irretrievable.

Full text

TECHNICAL NOTE
Recovery of a Lost Subsea Asset at Full Ocean
Depth in the Mariana Trench (10,925 m ± 4)
Using a Crewed Submersible
AUTHORS
Alan J. Jamieson
Minderoo-UWA Deep-Sea Research
Centre, The University of Western
Australia
Patrick Lahey
Triton Submarines LLC,
Sebastian, Florida
Rob MacCallum
EYOS Expeditions, Isle of Man,
United Kingdom
Stuart Buckle
Tim Macdonald
Victor Vescovo
Caladan Oceanic LLC,
Southlake, Texas
ABSTRACT
In 2019, the Deep Submergence Support Vessel Pressure Drop embarked on
the third leg of the Five Deeps Expedition. Over the course of 10 days, the full
ocean depth submersible Deep Submergence Vehicle Limiting Factor made a re-
cord five dives to the bottom of the Mariana Trench: four to the Challenger Deep,
one to the Sirena Deep. The submersible was supported by three scientific
landers, one of which became stuck on the seafloor at 10,925 m depth. Here
we describe how, during the third dive of the campaign, the expedition utilized
the submersible to rescue this lost asset from full ocean depth. The expedition
was not only significant for its operational jump from single full ocean depth
dives to multiple dives in a short space of time, but demonstrated that assets
lost in the deepest 45% of the oceans are no longer irretrievable.
Keywords: submersible, lander, Mariana Trench, exploration, asset recovery
Introduction
T
he exploration of Earth’s marine
frontiers is fraught with challenges,
and perhaps none more so than reach-
ing the deepest parts of the oceans.
Challenges such as high hydrostatic
pressure and distance from the surface
make such endeavors infrequent and
very expensive. Decades ago, the
bathyscaphes Trieste and Archimede
proved that it could be done, al-
though the former only made one
trip to full ocean depth and the latter
did many very-deep dives but none to
the deepest points (Piccard & Dietz,
1961, and Jarry, 2003, respectively).
More recent human occupied vehicles
(HOVs) such as DeepSea Challenger
were successful in reaching full ocean
depth but did so only once (Gallo
et al., 2015). Remotely operated vehi-
cles such as Kaiko and Nereus also
reached full ocean depth but were sub-
sequently lost (Tashiro et al., 2004,
and Cressay, 2014, respectively). Cur-
rently, there are several HOVs capable
of diving to 6,500–7,000 m such as
Alvin (United States; Piecuch 2021),
Shinkai 6500 (Japan; Komuku et al.,
2007), and Jiaolong (China; Liu
et al., 2010) and, more recently, the
deeper diving submersible Fendouzhe
has reached full ocean depth multiple
times (China; Liu et al., 2021).
In 2017, construction of a new
submersible began as a privately
funded joint project between Caladan
Oceanic LLC and Triton Submarines
LLC (Jamieson et al., 2019). The ob-
jective was to produce a fully certified
submersible capable of taking two
people repeatedly to full ocean
depth on an operational basis. The
HOV would be used in an expedi-
tion, known as Five Deeps,tovisit
the deepest point in every ocean:
The Puerto Rico Trench (Atlantic),
SouthSandwichTrench(Southern),
Java Trench (Indian), Mariana
Trench (Pacific), and the Molloy
Hole (Arctic), in that order (Stewart
& Jamieson, 2019; Jamieson, 2020).
During the expedition, the HOV
successfully completed this explora-
tion challenge as well as other ven-
tures such as tourism, marine
archaeology, and deep-sea science.
The expedition began in Decem-
ber 2018. Following 14 test dives
off the Bahamas, the submersible
achieved the first of the five deeps at
8,376 ± 5 m in the Puerto Rico
Trench on 19th December 2018
(Jamieson et al., 2020); followed by
the second at 7,434 ± 3 m in the
South Sandwich Trench, 3rd February
2019 (Jamieson et al., 2021); and the
third at 7,192 ± 5 m the Java Trench
on 5th April 2019 (Jamieson et al.,
2022). The expedition arrived at the
Winter 2023 Volume 57 Number 1 151

Mariana Trench on 26th April 2019,
and given the prestige of this location
and its extreme depth, a greater
amount of time (10 days) was allocated
to this site. The objectives were to
complete four dives, comprising two
single occupant “solo”dives, a dive
for DNV-GL (Det Norske Veritas-
Germanischer Lloyd) classification,
and a dive for the submersible manu-
facturers. This many dives to full
ocean depth had never been attempted
before making it a highly ambitious
undertaking. Here, we describe the
achievement of more dives to Challeng-
er Deep in 10 days than were planned,
and how the HOV was used to suc-
cessfully locate and recover a lost lander
from the deepest point on Earth.
Material and Methods
The Deep Submergence Vehicle
(DSV) Limiting Factor or Triton
36,000/2 (Triton Submarines, United
States) is a full ocean depth rated two-
crewed submersible (Jamieson et al.,
2019) (Figure 1A). The submersible
is operated from the 68 m-long Deep
Submergence Support Vessel (DSSV)
Pressure Drop and supported by a
number of small surface craft and
three full ocean depth rated lander
systems (Figure 2B). The landers had
a footprint of 1 × 1.5 m and were
1.5 m high with a weight in water
of ~50 kg -ve. All these assets com-
bined are collectively known as the
Hadal Exploration System, operated
at the time by Caladan Oceanic Ltd
(United States).
DSV Limiting Factor communi-
cated with the ship, and the support-
ing landers via underwater modems
(GPM300 Acoustic Modem, L3 Oce-
ania,Fremantle,Australia).Video
data were acquired using three exter-
nally mounted high-definition video
cameras (IP Multi SeaCam 3105;
Deep Sea Power and Light, San
Diego, CA), illuminated by ten
15,500 lumen LED lights (LED-
1153-A3-SUS; Teledyne Bowtech,
Aberdeen, UK). Depth and tempera-
ture were recorded by twin Conduc-
tivity, Temperature, and Depth
(CTD) probes (SBE 49 FastCAT,
SeaBird Electronics, Bellevue, WA),
where depth was recorded with an ac-
curacy of 0.1%. It was also equipped
with a “Raptor”hydraulic manipu-
lator arm (Kraft TeleRobotics,
Overland Park, KS). The three iden-
tical landers were free-fall baited
camera landers primarily used for
scientific studies, equipped with the
same camera, lights, CTDs, and
underwater modems as Limiting Fac-
tor.BothLimiting Factor and the
landers descend to depth by virtue of
FIGURE 1
(A) The full ocean depth rated two-crewed submersible DSV Limiting Factor, and (B) one of the
supporting landers (Skaff). The Limiting Factor image was taken by Skaff at 10,920 m in Chal-
lenger Deep, and the Skaff image was taken by the Limiting Factor at 3,000 m in the Puerto Rico
Trench.
FIGURE 2
Images from the lost lander Skaff at 10,925 m, showing (A) Closp landing in the distance, (B)
the submersible arriving at Closp, (C) the submersible approaching the lost lander, and (D) the
submersible–lander rendezvous. The bait arm bars in (D) are 50 cm apart.
152 Marine Technology Society Journal

expendable ballast weights, which
when released via the modems or
back-up burn-wire timers (landers) or
manually (on board the HOV) return
the vehicles to the surface via positively
buoyant syntactic foam (TG39/
11,500, Trelleborg Applied Technolo-
gies, Rochdale, UK). Recovery on the
surface was aided by Iridium satellite
beacons (iBCN, Novatech, Canada)
and Xenon strobes (MMF-7500,
Novatech, Canada).
Prior to launching the HOV, the
DSSV Pressure Drop mapped pro-
posed dive sites and surrounding
areas in the Mariana Trench using a
hull-mounted EM 124 multibeam
echosounder (MBES; Kongsberg,
Norway; see Bongiovanni et al.,
2022). The three landers were then
deployed between 1 and 2 km apart,
and their positions triangulated using
modems operated from the ship and
one of its supporting surface boats.
Limiting Factor was then deployed,
using one of the landers as a marker
to aim for where the MBES data
identified as the deepest point.
Results
Following the first dive by Limit-
ing Factor to 10,924 m on the 28
th
April 2018, a second solo dive was
undertaken on the 1
st
of May. The
landers Skaff, Closp, and Flere were
deployed prior to this second dive to
depths of 10,925, 10,912, and
10,877 m, respectively. On arrival
on the sea floor, Limiting Factor ren-
dezvoused with Skaff that had been de-
ployed at 11.3700° N/142.5883° E. It
became apparent that the lander was
leaning forward into very soft sediment
that buried the entire 1-m-long bait
arm and had accumulated across the
base of the lander. It was not recog-
nized at the time that this partial burial
may be problematic. However, follow-
ing the surfacing of the HOV, acoustic
release commands were transmitted
from the ship to all the landers. Closp
and Flere surfaced on time, yet Skaff
did not, despite both the primary
and back-up burn wire releases having
activated. Following a period whereby
a back-up timed release should have
released the ballast weight, it was ap-
parent that Skaff was indeed stuck on
the seafloor.
The following day, it was decided
to make the third dive an attempt to
rescue the lander. The lost lander had
a value of $250,000, but it was also
the only lander that had been fitted
with a $150,000 special issue
Omega Seamaster Planet Ocean Ultra
Deep model.
On May 3
rd
,Closp was deployed
aiming for exactly the same position
as Skaff to act as a navigation beacon
for the submersible. To prevent Closp
also becoming stuck in the soft sedi-
ments, the ballast weight was tethered
1 m below the main frame so that no
part of the lander was itself in contact
with the seafloor.
In the period between the sub-
mersible leaving Skaff on the 1
st
to
the rescue attempt on the 3
rd
, the bat-
tery power on the lander had reduced.
This resulted in the camera still re-
cording, but the light had since failed.
Fortuitously, this allowed Skaff to re-
cord the rescue in greater detail than
if the light had been on, and further-
more, the camera was facing toward
Closp and actually recorded the land-
ing at an estimated distance of ~200
m (Figure 2A).
Nearly 5 hr after Closp landed, the
arrival of Limiting Factor on the
sea floor was also filmed by Skaff
(Figure 2B) landing between 100
and 200 m from Closp.TheHOV
then spent 50 min making a slow ap-
proach to Skaff using the flashing
light of the recovery beacon to visu-
ally locate it (Figure 2C-D).
Upon finding Skaff, the submers-
ible did a 360° inspection of the
lander (Figure 3A), and it was decided
to use the manipulator arm to dis-
lodge the lander from the sediment
by simply pushing the lander at the
top of its frame in the opposite direction
FIGURE 3
Images from the submersible showing (A) rendezvousing with the lost lander Skaff, (B) Skaff
ascending off the seafloor after being stuck for over 2 days, (C) returning to Closp, and (D)
witnessing the ballast release and ascent of Closp.
Winter 2023 Volume 57 Number 1 153

to which it was leaning. This was
achieved within 12 min (Figure 3B),
and the lander then ascended toward
the surface as normal.
The Limiting Factor then navigat-
ed back to Closp and requested an
acoustic release command from the
surface (Figure 3C). The submersible
then waited to observe and confirm
Closp beginning its ascent from the
seafloor (Figure 3D) before returning
itself.
Two days later, the HOV and
supporting landers undertook a
fourth dive to Challenger Deep. The
expedition then relocated to eastward
to Sirena Deep and undertook a dive
to 10,705 m with the Limiting Factor
and the three landers, concluding the
Mariana Trench campaign of the Five
Deeps Expedition.
Discussion
The Mariana Trench campaign
was extraordinary for multiple rea-
sons. It first demonstrated that a full
ocean depth HOV could be capable
of repeated dives to full ocean
depth. The dive that saw the rescue
of Skaff was also a certification dive
as the passenger in the sub at the
time was a DNV-GL inspector who,
upon returning, issued the Limiting
Factor with the first ever classification
of “unlimited depth.”It also demon-
strated that, in the space of just 10
days, the whole system—from the
ship, multibeam, HOV, and landers
—was capable of mapping the deepest
place on earth, diving 5 times in a
submersible and achieving 20 scien-
tific lander deployments between
2,100 and 10,900 m depth. It also
demonstrated the ability to locate
and rendezvous with landers pre-
deployed on the seafloorasthiswas
done on all five dives on this cam-
paign, with two on the rescue
dive (Figure 3). This is a major oper-
ational jump from single dives to
multiple, and more complex dives
at full ocean depth. The lander res-
cue also shows that assets lost in
the deepest 45% of the oceans are
no longer irretrievable.
It is important to note that prior
to the Mariana Trench leg of the
Five Deeps Expedition, the submers-
ible had only completed four dives to
the maximum depth of trenches
greater than 6,000 m (the Puerto
Rico and South Sandwich trenches,
and two dives in the Java Trench).
The first Java Trench dive was the
only time prior to the Mariana
Trench leg that the submersible
managed to rendezvous with a land-
er, making the rescue of Skaff at
10,925 m all the more challenging
and remarkable.
The DSV Limiting Factor has
since gone on to complete a total of
22 dives to Challenger Deep and Sire-
na Deep (including another full ocean
depth lander rescue at 10,688 m in
2021), as well as several other dives
to over 10,000 m (Tonga, Kermadec,
and Philippine trenches).
Acknowledgments
We thank the entire crew, compa-
ny, and sub team of the DSSV Pres-
sure Drop, including Commander
Don Walsh USN (Ret) and Paul
Henry Nargeolet.
Corresponding Author:
Alan J. Jamieson
Minderoo-UWA Deep-Sea Research
Centre,
The University of Western Australia
35 Stirling Highway, Perth, WA 6009
Email: Alan.j.jamieson@uwa.edu.au
References
Bongiovanni, C., Stewart, H.A., &
Jamieson, A.J. 2022. High-resolution multi-
beam sonar bathymetry of the deepest place
in each ocean. Geosci Data J. 9(1):108–23.
https://doi.org/10.1002/gdj3.122.
Cressay, D. 2014. Submersible loss hits re-
search. Nature. 509:408–9. https://doi.
org/10.1038/509408a.
Gallo, N.D., Cameron, J., Hardy, K.,
Fryer, P., Bartlett, D.H., & Levin, L.A. 2015.
Submersible-and lander-observed community
patterns in the Mariana and New Britain
trenches: Influence of productivity and depth
on epibenthic and scavenging communities.
Deep-Sea Res Part I. 99:119–33. https://doi.
org/10.1016/j.dsr.2014.12.012.
Jamieson, A.J. 2020. The Five Deeps Expe-
dition and an update of full ocean depth ex-
ploration and explorers. Mar Tech Soc J.
54(1):6–12. https://doi.org/10.4031/MTSJ.
54.1.1.
Jamieson, A.J., Linley, T.D., Stewart, H.A.,
Nargeolet, P.-H., & Vescovo, V. 2020. Revi-
siting the 1964 Archimède bathyscaphe dive to
7300 m in the Puerto Rico Trench, abun-
dance of an elasipodid holothurian Peniagone
sp. and a resolution to the identity of the
‘Puerto Rican snailfish’. Deep-Sea Res Part 1.
162:103336. https://doi.org/10.1016/j.
dsr.2020.103336.
Jamieson, A.J., Ramsey, J., & Lahey, P. 2019.
Hadal manned submersible. Sea Technol.
60(9):22–4.
Jamieson, A.J., Stewart, H.A., Weston, J.N.J.,
& Bongiovanni, C. 2021. Hadal fauna of
the South Sandwich Trench, Southern
Ocean: Baited camera survey from the Five
Deeps Expedition. Deep-Sea Res Part II.
194:104987. https://doi.org/10.1016/j.dsr2.
2021.104987.
Jamieson, A.J., Stewart, H.A., Weston, J.N.J.,
Lahey, P., & Vescovo, V.L. 2022. Hadal
biodiversity and potential chemosynthesis in
the Java Trench, Eastern Indian Ocean. Front
Mar Sci. 9:856992. https://doi.org/10.3389/
fmars.2022.856992.
154 Marine Technology Society Journal

Jarry, J. 2003. L’aventure des bathyscaphes:
Marins, ingénieurs et savants au plus profond
des mers. Editions du Gerfaut.
Komuku, T., Matsumoto, K., Imai, Y.,
Sakurai, T., & Ito, K. 2007. 1000 Dives
by the Shinkai 6500 in 18 Years. In 2007
Symposium on Underwater Technology and
Workshop on Scientific Use of Submarine
Cables and Related Technologies. IEEE.
pp. 21–27. https://doi.org/10.1109/UT.
2007.370833.
Liu, F., Cui, W., & Li, X. 2010. China’sfirst
deep manned submersible. JIAOLONG Sci
China Ea Sci. 53(10):1407–10. https://doi.
org/10.1007/s11430-010-4100-2.
Piccard, J., & Dietz, R.S. 1961. Seven miles
down. Longmans.
Liu, Y., Xue, J., Yang, B., Zhu, M., Guo, W.,
Pan, F., …Zhang, L. 2021. The Acoustic
System of the Fendouzhe HOV. Sensors.
21(22):7478. https://doi.org/10.3390/
s21227478.
Piecuch, H. 2021. Overhaul to take Alvin
to greater extremes: Work on iconic sub will
put 99% of the ocean floor within reach.
Oceanus. 56(1):32–4.
Stewart, H.A., & Jamieson, A.J. 2019. The
Five Deeps: The location and depth of the
deepest place in each of the world’s oceans.
Earth Sci Rev. 197:102896. https://doi.
org/10.1016/j.earscirev.2019.102896.
Tashiro, S., Watanbe, M., & Momma, H.
2004. Loss of the full ocean depth ROV
Kaiko, Part 2: Search for the ROV Kaiko
vehicle. Proc. 14
th
Intern. Offshore and
Polar Eng Conf. 2:194–8.
Winter 2023 Volume 57 Number 1 155