Friday, June 10, 2016

If the ocean was transparent : the see-through sea

Manuscript painting of Bruce C. Heezen & Marie Tharp "World ocean floor" map
by Heinrich C. Berann, (1977)
(Library of Congress Geography and Map Division Washington) 


The ability to peer unhindered into the deep would reveal a host of wonders—and have huge practical consequences

THE surface of Mars is better mapped than that of the Earth.
Every dry, dusty square metre of it has been peered at by cameras and illuminated by altimeters.
The lion’s share of the Earth’s surface has never been shown any such attention.
This is not because Mars is more interesting.
It is because it suffers from an insufficiency of ocean.
In most respects, this is to its detriment; seas are fascinating things that make planets far more habitable.
They also allow paddling, whalesong and other delights.
But they do rather cover things up.

Water absorbs light.
Despite this, seeing through a few metres of it is not too hard, sediment permitting.
And some wavelengths can penetrate a lot more.
A ray that is just the right shade of blue will still be half as bright after passing through 100 metres as it was when it started.
If you were to sink into the ocean looking up, that shade of blue would be the last thing you would see.
But even it would eventually fade to black.
Almost the whole ocean floor is dark to those that inhabit it, and invisible to all.

What if it were not—if light could pass through the ocean as easily as it does through the atmosphere?
What if, when you looked down from a trans-Atlantic flight, the contents of the ocean, and its floor, were as clearly visible as if seen through air: what would you see?

The most persistent feature would be a thin green mist extending a few tens of metres down from the surface.
It would be too sparse to be seen over much of the planet; but in some patches, and close to some shores, it would be a visible layer of light and life.
This is the world’s stock of phytoplankton, tiny photosynthetic algae and bacteria.
Its total mass is far less than that of the plants that provide photosynthesis on land, but every year it takes 50 billion tonnes of carbon out of the atmosphere, turning it into organic matter for the ocean’s inhabitants to eat.
Scant though the planktonic biomass is, it does roughly as much biogeochemical work as all the continents’ forests, savannahs and farms.


 Map of the world without water, 1694

Water, water, everywhere

From the smallest of the surface features to the largest, you would also see more than 111,000 ships hanging as if suspended in empty space, according to estimates of the size of the world’s merchant fleet from IHS.
They are the workplaces, and sometimes homes, of at least 1.5m seafarers, and more than 500 liners provide temporary accommodation to hundreds of thousands of passengers, too.
This disassembled city of steel carries some 90% of all international trade by weight.
Its wandering buildings can carry, between them, over 1 billion tonnes of cargo: a mass equivalent to one cubic kilometre of water, a little less than a billionth of the total volume of the ocean.

That brings home the most striking feature of the see-through ocean: its emptiness.
People tend to focus on the bits of the ocean that are full of life (such as reefs) or of trade (such as shipping lanes).
But these are only a tiny fraction of everything there is.
And in much of that everything, there is close to nothing.
Spread those ships out evenly and each one of them would have 3,000 square kilometres of ocean to herself—the size of the state of Rhode Island.

Ships are not the only man-made artefacts that float across the seas.
There is an alarming amount of rubbish—in some places it outweighs the phytoplankton.
As ecologically delinquent as this is, in terms of its bulk the problem would still be easy to overlook in a transparent sea.
The “great Pacific garbage patch” consists of millions of tonnes of rubbish floating in the slowly circulating North Pacific Gyre.
But the size of the gyre is such that the rubbish adds up to just five kilograms per square kilometre.

Indeed, rather than filling the ocean, humankind has been working hard at emptying it.
Tuna stocks are thought to be half of what they were before modern commercial fisheries.
Estimates of Atlantic whale populations based on DNA suggest they used to be between six and 20 times greater than they are today.

The opacity of the ocean makes a straightforward numerical census of what remains impossible.
But Simon Jennings of CEFAS, a research centre in Lowestoft, in England, and Kate Collingridge have made a brave stab at estimating how many fish there are in the sea by applying ecological modelling.
Their result is strikingly small: 5 billion tonnes of fish weighing between a gram and a tonne.
If piled together, those fish would not even fill Loch Ness, which though an impressive body of water is nugatory compared with the whole ocean.
Even if Dr Jennings is off by a factor of ten, the volume of fish would still be less than that of Lake Geneva.
Broadly, the world boasts less than a minnow for every Olympic swimming pool of its seawater.

Yet life in the ocean can still mount sublime spectacles.
Nicholas Makris of MIT and his colleagues have observed fish in the Gulf of Maine using a sonar system that comes as close as almost any technology to making this article’s premise real, and rendering the ocean transparent.
Employing longer wavelengths of sound than most sonars, and taking advantage of lightning-fast processing power, it is possible to create time-lapse movies of sea life over tens of thousands of square kilometres.

Dr Makris’s team have been able to quantify the processes by which herring can gather themselves into shoals many kilometres long comprised of hundreds of millions of fish, watching their depth and behaviour change with the time of day.
And in the Gulf of Maine they were able to distinguish the calls and songs of various species of whale attracted by the herring shoals, to track them as they communicated with each other and to distinguish their different herring-snaffling strategies.

 Marie Tharp working on a map of the ocean floor at Columbia in the 1960s.
courtesy of Lamont-Doherty Earth Observatory

And a thousand thousand slimy things

Other acoustic research has revealed a fundamental feature of ocean life invisible from the surface—a layer of small fish and other creatures that spend their days at depths of a few hundred metres before rising to the surface at night.
In the early days of sonar this was regularly confused with the sea floor, because of the way the fish’s bladders resonated with the sonar’s sound waves.
The daily rise and fall of this “deep scattering layer” would, in a transparent ocean, be revealed as one of the largest mass movements of the animal kingdom.

Acoustic techniques produce pictures of the ocean’s floor, as well as its contents.
For most of the 20th century, though, the relevant measurements were sparse.
Thus the pioneering maps put together by Marie Tharp and Bruce Heezen of Columbia University in the 1950s and 1960s—which first identified the structure of the mid-Atlantic ridge, and of the faulted “fracture zones” perpendicular to it—often relied on depth data from just a few ships making single crossings of the ocean to get a sense of vast swathes of the terrain below.
The maps were works of extrapolation, interpolation and inspiration, not mere measurement.

Nevertheless, they had a huge impact.
They let geologists visualise the processes at work in the nascent theory of plate tectonics; those mid-ocean ridges and fracture-zone faults turned out to be the boundaries of the “plates” into which plate tectonics cut the surface of the Earth.
They were mind-expandingly right in their synoptic vision, if frequently inexact and sometimes mistaken in their specifics.

The side-scanning and “multibeam” sonar introduced for civilian use in the 1980s allowed a ship to map not just a thin strip of sea floor directly beneath it but a rich swathe to either side, and to provide detail on its texture, not just its depth.
At first this acuity was used mostly for sites scientists wanted to focus on, or artefacts of particular interest.
UNESCO estimates that there are 3m wrecks on the sea and ocean floors: 30 for every ship that now sails the surface.
Sophisticated sonar has found some of the spectacular ones, such as Bismarck, and others whose cargoes are of commercial interest for salvage.
It has also helped in the laying of ever more cable ever more precisely across the abyss; according to TeleGeography, there are now a million kilometres of submarine cable.
Every second they can carry 31 terabits across the Pacific, 55 across the Atlantic.

Because GPS satellites allow ships to know exactly where they are, and thus exactly which bit of the sea floor they sit above, new sonar technology has also revolutionised mapping.
The 2014 edition of the General Bathymetric Chart of the Oceans (GEBCO), an enterprise begun by Albert I of Monaco in 1903, includes sonar depth data from thousands of voyages, covering more than 60m square kilometres of the ocean floor.
But even that represents only 18% of the ocean floor.
The rest is mapped indirectly, by satellites.

Whereas light is absorbed by water, some forms of electromagnetic radiation bounce right off it.
Satellites can thus use radio waves to get a very accurate picture of the height of the ocean’s surface.
This varies from place to place, reflecting the unevenness in the solid Earth’s gravitational field that comes from the planet not being a perfect sphere.
The sea level is, for example, slightly higher above a seamount—an ocean-floor protuberance that does not make it to the surface—because the water feels the gravitational attraction of its mass.
This difference is only a couple of centimetres; but satellites can measure it.

 Tharp invents augmented reality
Altimetry has discovered at least 10,000 such seamounts.
Statistics suggest that hundreds of thousands of smaller ones remain to be found.
Added together that’s an ecologically interesting habitat about the size of Europe that was previously almost completely uncharted.

Since the 1990s radar-altimetry has allowed oceanographers to fill in the 80% or so of the ocean floor that sonar bathymetry does not cover.
The latest GEBCO map still required some interpolation.
But in both resolution and consistency such hybrid maps are far better than what went before.
In some ways looking at these maps comes as close as one can get to seeing right through the ocean.

The charm├Ęd water burnt alway

There is a subtle distortion, though.
Maps of the ocean floor are typically rendered in a “shaded relief” style (and computers now add a spectrum of “false colour”, with red for high and blue for low).
For this to make sense to the untutored eye, the relief in question has to be exaggerated, typically by a factor of ten or 20.

So people have become used to seeing the ocean-floor world as interestingly craggy.
It really isn’t.
In maps the drops that separate continental shelves from the abyssal plains far below them fall away like the edge of a flat Earth; in fact they have typical gradients of about 7%.
Were it not for the water, few features in the ocean would present an off-road car with much difficulty.

Marie Tharp drew her maps in this way in part to emphasise the new features she, Heezen and their colleagues had discovered.
But it was also because the obvious alternative was no longer legal.
Earlier 20th-century maps of the ocean floor had, like maps of the land, used contours.
In the 1950s the precise depths necessary for making contour maps were classified by the American government.
The deep seas were becoming a cold-war battlefield.

Being unseen had given submarines a tactical advantage since they entered widespread use in the first world war.
In 1960 the obscurity of the depths took on a strategic importance, too.
The nuclear-powered George Washington, launched that year, carried 16 Polaris missiles with nuclear warheads.
That her location when submerged could not be known meant there was no way for all of America’s nuclear weapons to be destroyed in a pre-emptive attack.
The appeal of this “assured second strike” capability saw missile submarines adopted by Russia, Britain, France, China, Israel and India.
These days about a dozen nuclear-missile-carrying submarines (known as SSBNs) patrol the ocean at any given time.
If water were perfectly transparent you would see them, plump tubes of menace hanging in the void.
And if you could see them, you could target them.

There is a certain irony, then, that the technologies which have done most to make the ocean transparent have come from the armed forces.
The American navy developed multibeam sonar to understand the submarine battlefield.
The gravitational-field mapping that lies behind satellite altimetry was needed so that submarines and their missiles would better know where they were and what they would hit.
The cold war produced the experts as well as the technology: Dr Makris listened for submarines at the Office for Naval Research before he listened for herring off Maine.
If you were interested in ocean remote sensing, he says, you more or less had to: “They had all the great toys.”

The end of the cold war saw a big drop in undersea sensing as a military priority, but its strategic importance is hardly diminished.
Britain, for example, is deciding whether to renew its SSBN fleet.
It matters whether the submarines will, in the 2050s, be as impossible to trace as they are today.

 Geological map of the seafloor by Nicolas Parubets

Under the keel nine fathom deep

What new technological approaches might be able to make the ocean transparent to submarine-hunters?
Two are widely discussed: drones and big data.
Uncrewed surface vessels and submersibles might be able to field far more instruments more cheaply than navies have in the past.
And new data-processing capabilities might be able to make sense of signals that would previously have been swamped by noise.

Thousands of remote-sensing platforms are already scattered around the ocean.
The Argo array currently consists of 3,918 floats which submerge themselves to about 2,000 metres and then return to the surface, measuring temperature and salinity as they rise and fall and sending their data back by satellite.
By gauging the amount of heat stored in the ocean they are crucial to studies of climate change.
These floats go where the currents take them, but that is not mandatory.
The wings of “seagliders”, which also rise and fall by changing their buoyancy, allow them to traverse large distances as they sink.
They can operate autonomously for months at a time and traverse whole ocean basins.

There do not yet appear to be any seagliders designed for detecting or tracking submarines—but in April DARPA, the Pentagon’s developer of futuristic technology, commissioned Sea Hunter, a small non-submersible trimaran that needs no crew, but carries sensors.
It is intended to prove that once an enemy submarine is located it can be trailed indefinitely.

Sea Hunter is designed to track conventional diesel-electric submarines, not SSBNs.
The American navy got a shock in 2006 when a previously unnoticed Chinese diesel-electric boat surfaced less than 10km from one of its aircraft-carriers, Kitty Hawk, in the Philippine Sea.
If it wants to keep its carriers safe it needs to be able to keep better tabs on such craft.
But what can be used for one sort of submarine today might be adapted to track another tomorrow.
It is likely that drones above, on or below the surface will come to play a much bigger role in anti-submarine warfare; the underwater ones, though, will still have to deal with the sea’s opacity.
A swarm of airborne drones can co-ordinate itself by radio, but things are harder underwater.

New data-processing approaches could also make submarines easier to see.
America’s Ohio-class submarines displace 18,750 tonnes when submerged.
Moving such a big object, even slowly, will leave a wake of sorts on the surface.
Computers are getting better and better at picking small signals out of noisy data.
And being metal, submarines have an effect on the Earth’s magnetic field, another potential giveaway.
Flying drones equipped with new sorts of magnetometer could make submarine-hunting easier.

Turning these possibilities into operational systems could make vital parts of the ocean—for example, some of the seas off Asia—transparent.
Scaling them up to cover whole ocean basins, though, would be a huge endeavour.
Remember the first insight of the transparent ocean: very big, very empty.
That array of 3,918 Argo floats works out as one per 340,000 cubic kilometres of water.
And SSBNs are sneaky.

If the SSBNs can still find somewhere to lurk, for now, the ocean will surely become more see-through, especially at the edges.
Dr Makris would like to make sonar systems like that which he and his colleagues have pioneered available for fisheries management.
As Dr Jennings points out, the seas are already transparent for a lot of fishing fleets, thanks to short-range fish-finding sonar and spotter planes.
Letting managers see what is going on might be a boon for conservation in some fisheries.

Charting of the deep seas will continue, too.
The task is daunting: Larry Mayer of the University of New Hampshire says multibeam-sonar mapping of all the remaining deep ocean would take 200 years of a research ship’s time.
But bit by bit it will be done.
A GEBCO forum in Monaco this month will discuss the way forward.

Being able to see is only the start; then you have to learn to look, to distinguish, to understand.
What ecological patterns could be discerned from those as yet unmapped seamounts?
What secrets lie in the ecosystems of the deep sea?
What archaeological surprises may lurk in those millions of wrecks—or in the abandoned homes of those who, in the last ice age, lived in plains that today are sea floors?
Where is the heat the Argo floats are tracing ending up—and how likely is it to come back out?
What sorts of clever management could restore some of the riches that have been fished away?

There is a fear that making things visible will strip them of their mystery.
Maybe so.
But it need not strip away curiosity or wonder.
As mappers of both Mars and the ocean bear witness, there is no void, abyssal or interplanetary, that those feelings cannot fill, if given a chance.

Links :

No comments:

Post a Comment