Doug Levin is the Associate Director for the Center for
Environment and Society at Washington College in Chestertown, Maryland,
and is an expert in underwater exploration technology, as well as
designing fun programs that teach complex engineering concepts.
The following is a dissection of the March 12th email from James Cameron
to Don Walsh–co-pilot of the bathyscaphe Trieste that went down to
Challenger Deep in 1960–following Cameron’s successful 8,000-meter dive
to the bottom of the New Britain Trench.
This email speaks
volumes to what goes on in and outside of the submarine. I’ve dissected
each paragraph and inserted my interpretation of what might be going on
behind the words. Of course, including my explanation has lengthened the
email by an order of magnitude. However, I was blown away by how much
was revealed within the short communication, and I wanted to share my
reactions with you. In the copy below I’ve kept Mr. Cameron’s type
normal. My responses are preceded by several dashes (—-) and are written in bold italics.
remain on the sea floor for up to six hours to conduct science
Illustration Courtesy Acheron Project Pty Ltd.
8000m dive went very well. Not an unqualified success, since the
manip(ulator) was balky and my push core sediment sample washed out on
ascent because the sample door wouldn’t stow all the way, and because of
the speed of the flow over the vehicle on ascent (5 knots average).
—- At 8,000m, Mr. Cameron’s mechanical appendages would have been working
at pressures 800 times that experienced at the sea surface. The
manipulator may have been balky because the pressures caused the
fittings that normally give the arm its agility to tighten.
—- A push core does exactly what its name suggests: A tube is
shoved into the sediment at the trench bottom and pulled back out to
collect a sample. First, a “core catcher” is placed in the end of the
core tube that enters the sediment; inverted metal or plastic fingers
keep the sample from falling out. Once the core is pushed to its
refusal (i.e. the point at which it won’t go any deeper), the top is
capped to create air pressure. (If you’ve ever put your thumb over a
drinking straw, you know that you can remove the straw from a beverage
container without losing the liquid within. The liquid will then be
released when you remove your finger from the straw top. Putting the
cap on top of the core should keep the contents from spilling out in
much the same manner.)
—– That the sediment
washed out of the push core suggests two things. One
possibility is that there may not have been a “piston” in the core
tube, or it may not have been functioning properly. [Here's more on how it works: A piston slightly smaller than the
diameter of the core tube is placed inside with a rubber "O" ring around
it, which creates a waterproof seal between the piston and the core
barrel wall. Then, the piston is positioned at the point in the core
barrel where it would touch the bottom at refusal. Then, the core barrel
is pushed into the sediments. While the core barrel advances into the
seafloor, the piston slides upward, creating a "suction." When the core
barrel reaches refusal, the piston remains in place over the core
sample, sealing the top of the core and preventing the sample from
dropping out.] The more likely problem here is that the sample is being
taken at “wicked” high pressures of depth and great speeds, so I don’t know how this
process would work under such extreme conditions. Presumably, all of
these systems were checked in the same pressure tanks used to determine
the safety of the pilot’s chamber.
—- Two, the
fact that sediment washed out tells me something about the nature of the
sediment itself: that the sediment particles (the extremely fine
materials described in latter paragraphs) are not cohesive; they do not
stick together. Clays stick together like you’d experience Playdo ®.
But this fine material collecting in the trench bottom is not sticky.
overall, the vehicle performed like a champ. Plenty of power, and even
though I lost one thruster, I still had 11 left, so the
massive-redundancy approach worked. I never lost functionality.
*OK, readers, tell me why three thrusters would be the minimum number
required for a submarine to move in all directions within the water
All the lights and cameras worked.
That all of the lights and cameras worked does not surprise me–that is
one of Mr. Cameron’s main areas of expertise, afterall! A limitation of the
light system is that, no matter how bright, light does not transmit very
well in the dankest darks of the deep ocean. I wonder (out loud) about
the physics of light transmittance underwater…I know that this
property has a lot to do with why water is blue. Even sunlight has a
difficult time penetrating the ocean depths. I wonder what the
calculated penetration of the sub’s light system was expected to be?
I’m pretty certain that light travel in water is not dependent on depth
or water pressure…I’ll have to look into that question more later on.
Sonar was balky…that’s gonna need some work.
There are several types of sonar available. The most basic of them
would be a single-beam echo sounder that sends a sound pulse to the sea
floor. Knowing the speed of sound in water and the time it takes for
the pulse to travel to the seafloor and back, and dividing that number
by 2, one can calculate the depth of water below the submarine. This in
“boat-speak” would be called a depth sounder. I would be curious to
hear back from James Cameron regarding the operations definition of the
word “balky.” As a sub pilot this, to me, translates to a system that
is difficult to operate, or data return that is questionable or
—- That said, there are several devices that I’m
hoping Mr. Cameron has installed on his submarine. Second to the
single-beam echo sounder, I’m hoping he has a device that will map the
three-dimensional sea scape of the trench floor. In the business, we
call this multi-beam. The multi-beam system can do more than measure
water depths; it can be “tuned” to take high-resolution pictures of the
seafloor using soundwaves. This capability is adapted from Side Scan
Using the variations in strength of the echoes from the transmitted
soundwaves, a picture can be painted of the seafloor. Northrop Grumman
has developed another type of underwater imaging tool that uses lasers
to create photograph-like images.
The Side Scan Image seen at left was taken in the harbor of Valparaiso,
Chile. It shows one ship hull sitting upright over a piece of another,
and a tugboat lying on its side. The shadows on either side of the sonar
targets result from weak acoustic returns (shown in dark colors).
time close to 5 hours, range of exploration about 1.5 km horizontal,
and about 300m vertical along the trench wall, which was like the Grand
Canyon, vertical faces interspersed with angled scree slopes. Dramatic
scree slopes” tells me that that the cliff walls in the trench are
weathering. The “scree” must refer to pieces of the rock wall that have
fallen to the seafloor, creating a talus at the toe of the cliff.
Because trenches are commonly affected by earthquakes, we would expect
rock pieces to frequently break apart from the main wall during these
The ponded sediment in the center of the
trench was the finest I’ve ever seen. When the thrust-wash just barely
kissed it, it formed silken veils undulating across the bottom, and then
it would rise and hang in tendrils like ectoplasm. Not at all like the
typical turbidite plains of abyssal depths.
sediments at these ocean depths consist of all of the materials that
have rained from the skies, washed off of the continents, and dropped
through nearly 7 miles of water. Note that this “pelagic” rain includes
animals that have lived and died in the sea. Pardon the visual, but the
deep sea would also be the final resting place for any fish poop. The
fine material further includes cosmic dust that rains from space, is
captured by Earth’s gravity, and enters the ocean. It might take years
for the space material to settle into the crevice called the Challenger
—- Submarine canyons cut across the continental shelf and
open onto the continental slope. These canyons are a conduit for
sediments that enter from the continental margins and eventually pass to
the abyssal plain. The turbidite material Mr. Cameron refers to are
the sediments that pass through these submarine canyons, which are cut
into the continental slope. On the East Coast of North America, these
canyons are thought to be extensions of onshore rivers, e.g. the Hudson
Canyon. On the West Coast of North America, the submarine canyons are
not necessarily correlated with West Coast rivers; rather they are
thought to follow fault lines perpendicular to the coast.
I dove the basin of ponded sediment was 1.5 km across, flat as a
billiard table, and virtually featureless. It actually ended at a
well-defined “beach” where the normal rocks and sediment commenced,
terracing upward to the fault scarps. I explored up the scarps onto a
plateau. The small exposed rock faces had large communities of white
anemones about 1 foot long. Hanging gardens. It was a completely
distinct micro-habitat from the flat basin. Out on the plain the
dominant fauna were 1′ diameter jellies that would lie on the bottom or
swim about 2 meters up. When disturbed they would fly off the bottom.
There also were large numbers of amphipods in all sizes. The baited
lander captured images of incredible aggregations, including individuals
close to a foot long.
—- The “flat ponding” of the
fine, noncohesive sediments (noncohesive meaning that if you picked the
sediment up in your hand, it would not stick to you) suggests that there
are no currents moving across the trench floor. What blows my mind is
the image of the ecosystems at this depth. Remember that the pressures
at this depth would feel like having 50 airline jets sitting on top of
you. These critters have adapted to the pressure and the darkness, and
can apparently move with agility. If we could learn their secrets, Mr.
Cameron might be able to design his manipulator, sampling, and testing
equipment to emulate them.
*At this point… I’d like to throw
out a challenge to anyone who has read this far to grab a piece of paper
and your crayons. Using the imagery painted by Mr. Cameron’s words,
why don’t you try and draw what he’s seen? Take a picture of your
artwork and send it to NatGeoEd@ngs.org for us to post on the blog!*
tried but was unable to rendezvous with the lander because the sonar
was not cooperating. Normally the lander is a very bright target, and it
should have been easy to find on that flat plain. But without sonar,
nor accurate coordinates from the surface, it was a visual search, which
is very limited. It might have been 50 meters to my left and I went
right by it. I could have done an expanding-square search pattern, but I
decided it wasn’t the best use of my power, when there was real
exploring to do.
—- OK, let’s dissect this part of the
email. The sub’s “toolbox” includes a couple of landers. The details
of the landers are not readily available, so I’m doing a little arm
waving here (Read: this next analysis will be an educated guess based on
my knowledge of similar submarine systems.). The landers are not
attached to the sub, and so they use soundwaves to let the sub know
—- Here’s a
bit more about how this would work: When the landers initially leave
the ship it “knows” where they are; the ship’s GPS has their position
pinpointed to a “gnat’s tush” (i.e. very precisely). The sub and the
landers then continually talk to the ship and use an internal motion
detector (Doppler Velocity Log (DVL)) to calculate their position
relative to the ship’s location. According to my buddy, Dr. Art
Trembanis at the University of Delaware, this is the underwater
equivalent of a game of “Marco Polo.” The sub and lander need to do this
continually to maintain accurate position.
—- I’m guessing
that Mr. Cameron’s landers lost communication with the ship at some
point, and thus were calculating their position based on the last known
location. This can be a problem for obvious reasons: The more time
elapses from a known position, the less confident we can be with the
actual position. So Mr. Cameron had released the landers, but was unable
to communicate with them, and his sonar could not provide a visual
location for them. Essentially, it appears that Mr. Cameron was flying
his sub without knowing his exact position. Without knowing an exact
position, it would be difficult to make a map using only the information he
Actual deepest depth for the dive was
26,791′ (8221m). Initial descent speed was 4.5 knots, attenuating near
the bottom to about 1.5kt, before I trimmed neutral with a few small
shot dumps totaling about 50lbs.
—- What are shot
dumps? Essentially, the sub is built to be positively buoyant (you know
what that means if you’ve read my earlier posts Kitchen Drogues and
Real-World Buoyancy). In the event of a malfunction, the sub can release
weight and it will float to the surface. Historically, the “shot”
consisted of lead pellets. I would be curious to hear if there was an
alternate, biodegradable material used today for this task. If it were
me, I would want the shot release mechanism to be manual and
reliable–without the use of electricity. If not, and there were a power
failure, an electronic release would prevent Mr. Cameron from releasing
the weight that would allow the sub to return to the surface safely.
drove the final 100m down on thrust, very slowly (because I didn’t
trust my altimeter yet… we’d just met and were only dating)… and
parked on the bottom using about 10% downthrust.
first thought…Why would you want to minimize the use of thrusters at
the top of the sub when landing? Because this would eliminate, or at least reduce,
disturbance to the seafloor. And, also note…this statement confirms
that the sub is positively buoyant by the need for 10% downthrust to
keep the vehicle on the seafloor.
Ascent speed was 5.7
knots slowing in the upper water column to 4.8. The soft ballast system
functioned perfectly, giving the sub an additional 400kg of lift. The
bag pops out automatically at about 200m depth and inflates slowly after
the sub reaches the surface. It is an oil-over-gas system of our own
design, which uses a spring-loaded poppet valve to open a bottle of
nitrogen at 3500psi when the external pressure balances on descent. The
valve locks open, charging the bag, and a reservoir of silicon oil
fills the tank so it doesn’t implode at depth. On ascent, the gas boils
out of the silicon, filling the float bag in about 3 minutes after
Surfacing at 4.8 knots is dramatic. I point the boom
camera and the 1000w spotlight straight up. I can see the surface
shimmering from about 100′ down. There’s a real sense of “ground rush”
as the shimmering patch grows rapidly bigger, filling the “sky” above
the sub. Then BWOOSH! an explosion of foam and bubbles, and the sub
pogos back down about 5m, then rises again and comes to rest. I call it
“Splash-up “… bastardizing a term from the 60′s space missions.
Using powerful lights , Mr. Cameron was able to spot the “surface
shimmering” from a depth of about 30m (100ft). So, this tells me that
his visibility at depth was also probably limited to 30m, even with 1000
watts of light. I note from looking at the sub that the banks of
lights are on the upper part of the Sub fuselage. I’m wondering why they
were placed so high.
*How would you position the lights for underwater viewing?*
CHALLENGER, the submersible that explorer and filmmaker James Cameron
will pilot to the bottom of the Mariana Trench. The vessel is the
centerpiece of DEEPSEA CHALLENGE, a joint scientific project by Cameron,
the National Geographic Society and Rolex to conduct deep-ocean
research. Photo by Mark Thiessen, National Geographic.
only significant problem on the dive is that one of the six battery
buses failed without adequate warning as I was making preparations to
ascend. Some fault in the battery management system comms inside an
external PBOF multi-bus box, probably related to water ingress, but I
haven’t gotten a report yet from the electronics guys.
This is a tough one…the connectors for this sub are highly
specialized. The acronym PBOF translates to Pressure Balanced Oil
Filled. Each system external to the sub experiences full ocean depth
pressures. In this case, it appears that the circuit box that “lives”
outside of the sub may have leaked.
failure took out the A-comms system, which was on that particular bus,
so I lost comms completely. Fortunately the back-up modem, which is
powered with its own independent battery, kept transmitting, so they
knew I was coming up when the depth numbers started changing. They
cleared back about a klick from my last known position and waited. I
surfaced about 1500m from the ship, but plainly visible. That’s why I
personally like night recoveries. The sub has so many lights and strobes
it’s like a UFO mothership, visible to the horizon at night, from the
bridge wing of a ship the size of Mermaid Sapphire.
Mr. Cameron’s sub apparently has only one main communication system,
which he calls the A-comms. Without this, he was without direct two-way
communications. If it were me, I’d have two, probably three, completely
identical but separate ways to maintain “fluid” contact with the mother
Sitting down there at 27000′, alone in the dark,
with no comms, no contact whatsoever with the world so far above, and
nothing but the ingenuity of the engineering to get me back…it’s
simultaneously scary and exhilarating. It’s the precipice we put
ourselves on by choice, to test ourselves and our machines. I configured
the cameras to get a good shot of the weights coming off and hitting
the seafloor in 3D, but I can’t say I spent an undue amount of time on
the lighting. I wanted to see those babies jettisoned as quick as
possible. It’s a good feeling when 350kg comes off, with the
characteristic “SHOONK” as the weight carriages run down the
Then I pulled the breaker on the shot-hopper magnet,
and let the other 150kg of shot pour out, watching on the boom camera
as it spiraled down into darkness in the trailing vortices under the
sub. Then I just powered down everything I didn’t need and sat hunched
in the dark, waiting… watching the numbers on the depth indicator
count down toward the surface.
It was an interesting ascent.
Virtually silent except for the soft whir of the scrubber fan, and the
rustle of water vortexing down across the fairing at 5 knots. There was a
slight rhythmic rock to the sub, due to vortex-shedding, which I
normally didn’t notice because I’m usually too busy doing things…comms, photography etc. But in this low power contingency, I was just
sitting there in the dark listening and feeling the sub. It was
fascinating to imagine 8 kilometers of water speeding by vertically. I
imagined the pressure coming off slowly as the ocean loosened its
—- Unbelievable. This seemingly short
email from Mr. Cameron to Mr. Walsh told more of the story behind the
sub and the mission than the entire rest of the [deepsechallenge.com] website–and elicited
more questions. Thank you, Mr. Cameron, for this colorful insight.
You’d have loved it.
Yes, I imagine Mr. Walsh would have…and I’d volunteer to be right by your
side…if there was room… and the invitation extended…
More to come…
See you in Guam.
—- I wish…
–by Doug Levin
1. deepseachallenge.com (Main National Geographic site for the expedition)
2. NatGeoEd.org/deepsea-challenge (National Geographic Education site for the expedition)
3. Expedition Journal (The official blog from the deck of Mermaid Sapphire,
James Cameron’s mother ship. Our Education bloggers are using this
blog, as well as the main deepseachallenge.com website, to inform their
writing about the expedition)