Wednesday, January 26, 2011

Using weather forecasting to predict pirate operations

That's the topic of a paper presented at the American Meteorological Society Annual Meeting in Seattle on Monday by James Hansen, L. Esher, E. Regnier and W. Lingsch Abstract: Pirate probabilities: On the importance of the dynamic coupling of meteorological and intelligence information for piracy interdiction (91st American Meteorological Society Annual Meeting).

Or, as covered by a local news source, As 3,500 meteorologists meet, one man's forecast: Chance of pirates:
While his colleagues swapped probabilistic equations, James Hansen came to talk about pirates.

Hansen rolled out some math of his own Monday as he explained his work to estimate the likelihood of attacks off the Horn of Africa and other hot spots, using weather data and an understanding of pirate behavior.

"Usually, I'm doing theoretical stuff down in the weeds," said Hansen, a Seattle-area native and applied mathematician at the Naval Research Laboratory in Monterey, Calif. "This is the only project where I can actually show pictures of the impact," he said, projecting images of Somali boatmen armed with missile-launchers and automatic weapons.

Though news coverage of pirate attacks has waned, the problem has worsened, Hansen said. Last year, there were nearly 450 attacks, with 53 ships captured and 1,181 crew members taken hostage. The estimated economic impact of the raids is $10 billion a year.

Even knowing the general location of pirate bases and their favored target areas, it's challenging for Navy and other patrol vessels to be in the right place at the right time, Hansen said.

"The Indian Ocean is really big."

The Navy will begin testing Hansen's model next month to see if it proves useful in helping direct patrol vessels and warn commercial ships when the risk of attack is high.
Thanks to Chris who alerted me to the Seattle Times article and to Sid, who sent on the AMS bit.

As regular readers will have noticed, wind and wave reports have been a part of the analysis done here, by NATO, by the U.S. Office of Naval Intelligence and many others over the past few years of predicting (in rough terms) the potential for piracy activity off Somalia. If Mr. Hansen, et al have devised a better tool, perhaps a more directed prevention of piracy effort will follow.

UPDATE: An example of a weather driven piracy prediction from ONI's Piracy Analysis and Warning Weekly (PAWW) Report dated 20 Jan 2011 (click to enlarge):


  1. I admire peoples predictions, but having been involved in this industry 24/7 for the last 2 1/2 years its pretty easy to predict where the pirates will be....

    There are 5 main shipping routes that extend across the Indian Ocean - they will move along those well trodden trade routes waiting for prey... they know where most of the warships are and they will happily use the captured ships as piggy back vessels to launch further attacks - they also have much better operating conditions due to their ability to launch and attack in heavier weather as long as they attack down wind, down swell.

    The Samho Jewelry incident will further change their tactics, to prevent military storming... Watch this space and remember........ not long now till the winds ease....! Then it will be chaos.

  2. You know, Nick, that the models will reveal that pirates are most likely to attack along major shipping lanes (because that's where the money is - um - merchant ships are) during periods of weather suitable enough for their boat ops.

    Sort of what you just said, but with mathematical precision. You know, "There's a 55% chance of a pirate attack east of the Seychelles today, so you merchant captains may want to double check those barbed wire barriers and add another lookout or two . . ."