Ship logs document climate of the past: Arctic Rediscovery

3 November 2014

Using naval logbooks to reconstruct past weather—and predict future climate

Dan Drollette Jr

Dan Drollette Jr

Dan Drollette, Jr. is a science writer/editor and foreign correspondent who has filed stories from every continent except Antarctica. His stories have appeared in Scientific American,...


Pity the poor navigator who fell asleep on watch and failed to update his ship’s logbook every four hours with details about its geographic position, time, date, wind direction, barometric readings, temperatures, ocean currents, and weather conditions. If he was with the British Royal Navy, the penalties could be harsh; it was—and still is—a serious offense to falsify the data in a logbook. (For example, sleeping on watch—a practice that could lead to logbook-faking—was historically considered a crime punishable by death, according to item 26 of the statutes of the British Admiralty’s Articles of War.) “Anything you read in a logbook, you can be sure that it is a true and faithful account,” said Clive Wilkinson of the UK’s National Maritime Museum.

Centuries later, that rigorously acquired information is being put to good use, providing a trove of archival data to scientists who are trying to fill in the details of our knowledge about the atmosphere and the changing climate. This material is being steadily scanned and put online by the Old Weather Project, an effort in which everyday people armed with laptops comb through old logbooks for climate data and enter it into a centralized, uniform database. These volunteers are making raw information that would otherwise be overlooked into material of use to today’s meteorologists and climate researchers.

The transcription and digitalization of these old observations are providing modern researchers with a stream of reliable raw data that had been recorded on a clockwork-like basis. The material from the Old Weather Project is one of several independent lines of investigation that have been helping to create a three-dimensional computer simulation that will provide a continuous, century-and-a-half-long profile of the entire planet’s climate over time; this effort is known as the 20th Century Reanalysis Project and run by the US National Oceanic and Atmospheric Administration. (Other information sources include ocean buoys, aerial photos, weather balloons, coral growth rates, ice cores, ancient tree rings, and the pollen counts in layers of sediment created over centuries, to cite just a few examples from climatologists’ lines of inquiry).

Members of the public are a central part of the Old Weather Project—by design. Volunteers are looking through scans of the old logbooks and putting the information in them into usable form for analysis. Through this process, the volunteers in effect become “citizen scientists”—joining an emerging trend that includes such organizations such as Galaxy Zoo, in which volunteers do the grunt work of classifying galaxies, and [email protected], in which the down time of home computers is used to help comb the universe for radio transmissions from deep space, one of the most likely bits of evidence for extraterrestrial intelligence. Most such groups have an affiliation with an umbrella organization, the Zooniverse project.

Human eyes over optical scanners. A key part of the Old Weather Project involves the reading of old, handwritten naval logbooks that have been posted online. The written entries were sometimes made hurriedly, under difficult environmental conditions, by human beings under stress, and recorded on paper that was exposed to the ravages of sun, sea, and salt. Making sense of this handwriting is still best done by human eyeballs; reading handwritten text is something that computers still aren’t very good at. (Just to be sure the transcriptions are correct, the data is checked and rechecked a minimum of three times by different people.)

To make this repetitive toil more interesting and keep the volunteers going, the Old Weather Project has made it into a game, with points and ranks awarded to volunteers, based on the amount of data they enter. (For example, it takes 30 weather reports to be promoted to lieutenant.) They also keep a sharp eye out for any information about icebergs, sea ice, animals, battle debris, aircraft, and any mention of volcanoes. In the case of some old naval logbooks, the action described is anything but dull. Volunteers get to follow the minute-by-minute account of a ship going into battle—perhaps the British battlecruiser HMS Invincible going against two German armored cruisers off the Falkland Islands during World War I. Both German ships sank, while only one crewman was injured aboard the Invincible, earning the vessel a place in naval lore.

But what the project seeks from Invincible is another history: all the daily weather, wind, barometric, and temperature observations at specific latitudes, longitudes, times, and dates, faithfully recorded in its logbooks six times a day during the ship’s nine-year life, resulting in 1,200 pages of data from that one ship’s various voyages. Because of its size and depth of detail, such a trove of information is invaluable to modern scientists trying to reconstruct the past climate. There are an estimated quarter-million such logbooks in the United Kingdom alone, from the Royal Navy, other government vessels, the British version of the Merchant Marine, and, of course, the East India Company. There would be millions of such records, if logbooks from the navies and merchant ships of the rest of the world were included. A good deal of the material is already scanned and available on-line; alone hosts more than 300 logbooks and ship histories of the Royal Navy during the World War One era, along with a growing number of US logbooks.

The navy logbooks are considered some of the most desirable; the recordings in them were carried out with military precision, and the description of the weather conditions followed a standardized set of readily observed wind effects known as the Beaufort scale—named after the British Rear Admiral who devised it. “Royal Navy logs are the low-hanging fruit of historical weather observations,” explained Philip Brohan, a climate scientist from the United Kingdom’s Met Office Hadley Centre. “Mariners really care about the weather, and have done so for centuries.”

The logbook measurements are particularly useful to climatologists because they were made at sea. About 70 percent of the globe is covered by water, and the oceans absorb and transport vast amounts of heat, which affects our weather. Once the information is entered into a computer-friendly database, the researchers can fill in data points in computer models of the atmosphere for that time period. And by looking at the past with these models, climate scientists today can improve climate modeling and better predict changes in climate.

Beyond the NOAA project to model the past 140 years of climate, the Old Weather Project is contributing to the dataset being compiled by Atmospheric Circulation Reconstructions over the Earth (ACRE), an international coordinating body for climate research.

The value of the citizen scientist. The Old Weather Project is now working on a collection of logbook pages from ships operating in the Arctic region and from ships sailing in the latter half of the 19th century. Its community has already transcribed 44 percent of these. The project will expand to include larger collections of logbooks of historical weather observations after these Arctic logs are digitized.

Meanwhile, scientists in the 20th Century Reanalysis Project have been combining Old Weather’s data with other global observation archives.

To process the climate data, the 20th Century Reanalysis project uses high-performance supercomputers at Oak Ridge National Laboratory in Tennessee and at Lawrence Berkeley National Laboratory in California. Such large and expensive machines—known colloquially as “big iron”—are essential for handling all the material once it’s been digitized, with file systems capable of supporting petabytes (millions of gigabytes) of statistics. (To give a sense of scale, 2 petabytes is equal to all the information in all US academic research libraries.)

There is plenty more raw material to be tapped from the navies of the world—especially the US Navy—and from a multitude of non-naval institutions as well, such as the files from the extensive collections of whaling museums and nautical organizations. The Old Weather Project, which started four years ago and originally focused more or less exclusively on naval ships from the United Kingdom, has formed a partnership with the US National Archives so it can include ships’ logs from the US Navy and Coast Guard (and their predecessors, including the Revenue Cutter Service, the Life-Saving Service, and the Coast Survey) and from non-military American voyages in the Arctic made between 1850 and World War II. Just a few weeks ago, the Old Weather Project was able to add 125,000 observations about sea-ice zones in Baffin Bay and the Bering Strait, from the records of just eight American ships in the Arctic. “There are logbooks in America, there are logbooks in South America, in Asia. There are literally billions of observations to be captured,” Wilkinson said in a UK National Maritime Museum video.

“We need to collect as much historical data as we can over the oceans, because if we wish to understand what the weather will do in the future, then we need to understand what the weather was doing in the past.”