57

Slate published an article recently, titled The Global Internet Is Being Attacked by Sharks, Google Confirms. Despite being catchy and all, most of the references seem to be other news outlets or actually contradict the claim.

Are there any sources supporting or contradicting that sharks are a threat to undersea cables?

Andrew Grimm
  • 38,859
  • 36
  • 141
  • 342
Édouard Lopez
  • 983
  • 7
  • 16

2 Answers2

70

Below are points of evidence supporting the fact that sharks are a threat to undersea cables.

EVIDENCE 1: The first report of sharks attacking cables came from the Canary Islands in 1985, when sharks' teeth were found embedded in an experimental cable. I found the report in an old news paper.

enter image description here

According to a report by the United Nations Environment Programme and the International Cable Protection Committee Ltd.,

Fish, including sharks, have a long history of biting cables as identified from teeth embedded in cable sheathings. Barracuda, shallow- and deep-water sharks and others have been identified as causes of cable failure. Bites tend to penetrate the cable insulation, allowing the power conductor to ground with seawater. Attacks on telegraph cables took place mainly on the continental shelf and continued into the coaxial era until 1964. Thereafter, attacks occurred at greater depths, presumably in response to the burial of coaxial and fibre- optic cables on the shelf and slope. Coaxial and fibre-optic cables have attracted the attention of sharks and other fish. The best-documented case comes from the Canary Islands, where the first deep-ocean fibre-optic cable failed on four occasions as a result of shark attacks in water depths of 1,060–1,900 m [3,478 to 6,234 feet].

See also:

Marra, L.J., 1989. Shark bite on the SL submarine light wave cable system: History, causes and resolution. IEEE Journal Oceanic Engineering 14: 230–237

EVIDENCE 2: A famous report by New York Times back in 1987 reported that the fibre optic cables linking the US, Europe and Japan were being nibbled persistently by sharks, causing phone and computer failures around the world.

The article also reported what was presented in Evidence 1, but adding as extra Dr. Nelson's claims:

In the report The finding that sharks are supersensitive to electrical signals, able to detect electric fields as faint as a few millionths of a volt per centimeter in water, is a recent significant discovery in marine science, Dr. Nelson said.

The sharks may detect a faint field near the cable and attack. "Not knowing any better, they try to eat it," Dr. Nelson said. "It's programmed in their genes. Whether the field comes from a cable or from a tin can, sharks are prone to behave as if they were encountering a food item, and try to eat it up."

EVIDENCE 3:

It seems that the funniest battle now is not Google vs. Amazon or Google vs. Microsoft, but Google vs. Sharks.

Based on the comments made by Dan Belcher, a product manager on Google's cloud team, during the opening keynote of the company's Cloud Roadshow in Boston last week; Google invests heavily in protecting its trans-continental infrastructure, including wrapping cables in Kevlar to thwart attacks by hungry sharks.

BONUS: A YouTube video showing a shark biting a submarine cable during a survey operation. It was spotted by a remotely operated underwater vehicle. .

enter image description here:

UPDATE: The cause isn't clear why Sharks bite cables.

Reasons for the attacks are uncertain, but sharks may be encouraged by electro magnetic fields from a suspended cable strumming in currents. However, when tested at sea and in the laboratory, no clear link between attacks, electromagnetic fields and strumming could be established. This lack of correlation may reflect differences between the behaviour of the deep-water sharks responsible for the bites and that of the shallow-water species used in the experiments. Whatever the cause, cables have been redesigned to improve their protection against fish biting.

UPDATE 2:

According to Submarine cables and the oceans: connecting the world report, external human aggression causes more faults for cables more than any other category, with fishing accounting for nearly half of all reported faults. Anchoring is the second major cause of faults, with dredging, drilling, seabed abrasion and earthquakes also causing significant numbers. However, natural hazards including seabed abrasion, shark bites account for less than 10 per cent of all faults.

Shark bites account for only 0.5% for all faults. They're consider a threat, but it's a minor threat.

George Chalhoub
  • 30,246
  • 14
  • 129
  • 136
  • 1
    It makes sense to me too that sharks would be attracted to the cables because of their sensitivity to electrical fields. Is there no marine biologist that confirms this somewhere? – Spork Aug 20 '14 at 13:39
  • @JörgWMittag Ah, that would explain it. Doesn't necessarily mean there's a *stronger* field, but it does explain the existence of one, and that's enough. georgechalhoub, I think it's not entirely useless, because without it there's a missing link in the explanation for attacks on fibre-optic cables. – Bob Aug 20 '14 at 15:25
  • I up-voted, but your answer only support pro-threat POV missing on insignificant the threat is – Édouard Lopez Aug 20 '14 at 15:46
  • According to [one source](http://www.shark.ch/Information/Senses/) (and I've heard this before, source just supports it), sharks will explore anything unknown in their environment by biting/tasting it. They also shed teeth easily. Given that people are alarmist about sharks, it occurs to me that finding teeth in cables might cause a leap to "attack" (with all the viciousness that people imagine), when it might actually be much more sedate (but still damaging)? – SevenSidedDie Aug 21 '14 at 18:12
  • "The best-documented case comes from the Canary Islands, where the first deep-ocean fibre-optic cable failed on four occasions as a result of shark attacks…" <- This would suggest that EM fields are not attracting the sharks, as fiber optic cables do not generate measurable EM fields (which also makes them better for security purposes). – Ben Hocking Jul 05 '15 at 20:34
  • Its easy to see how sharks would have evolved to bite at electrical fields. Animal muscles are controlled by electro-chemical impulses in nerves. Salt water conducts electricity, so these would be detectable. Before humans started laying cables just about the only source of weak electrical signals in the water would be shark food, so of course sharks would evolve to bite it. – Paul Johnson Mar 22 '16 at 17:18
  • @BenHocking: fibre-optic lines themselves may not generate EM fields, but the power systems for the optical repeaters certainly do. – Paul Johnson Mar 22 '16 at 17:21
38

Sharks are an extremely small threat to undersea cable in regards to others causes as shown in the graphic below (from a report cited in @georgechalhoub's answer).

 Proportion of cable faults by cause, from a database of 2,162 records spanning 1959–2006

Data

Proportion of cable faults by cause, from a database of 2,162 records spanning 1959–2006

  1. Fishing 44.4%;
  2. Anchor 14.6%;
  3. Component 7.2%;
  4. Abrasion 3.7% ;
  5. Geological 2.6% ;
  6. Dredge/drill/pipeline 0.9% ;
  7. Fish bite 0.5% (include sharks) ;
  8. Iceberg 0.1% ;
  9. Other 4.8% ;
  10. Unknown 21.3%.

Take-away

  • Geological causes damage cables more frequently than sharks' bites.
  • Sharks' bites are only 5 times more frequent than iceberg damage.
  • 14 fish bites (all species) in the last 50-60 years.

Reference

Édouard Lopez
  • 983
  • 7
  • 16
  • Yep, as I said earlier "I should have emphasis on the seriousness of the attack in my question" – Édouard Lopez Aug 20 '14 at 15:50
  • @georgechalhoub in one comment to your question. – Édouard Lopez Aug 20 '14 at 15:51
  • 5
    More accurately, sharks are only *known* to be an extremely small threat to undersea cables. That data set does not rule out the possibility that sharks are the second-largest threat. – Kevin Krumwiede Aug 20 '14 at 15:52
  • 2
    I dislike this answer a lot. It does not answer the original answer fully, and there is nothing in this pie chart to suggest it as being 'non serious'. All it says is that it doesn't happen often. An undersea cable failure is extremely serious. With a 0.1% chance of nuclear disaster, would you say that's not a serious chance based on a pie chart? Solid -1. – Spork Aug 21 '14 at 00:43
  • 1
    I think nobody in their right mind thought that sharks were the *main* source of undersea cable failures. – Spork Aug 21 '14 at 00:44
  • What is "component"? – Robert Aug 24 '14 at 00:52
  • 1
    @Robert it refer to cables' components – Édouard Lopez Aug 25 '14 at 19:58