Weather forecasting has always been a primary concern for boaters, who nowadays rely on all kinds of sensors and satellites to try and stay ahead of storms. But is it possible that in addition to looking at new technology for better forecasts, we should also be looking backward in time? As it turns out, some of the information we need to improve today’s forecasts could actually be buried in the pages of handwritten logbooks from whaling ships that date to 1780.

That’s the theory behind a research project led by Timothy Walker, a professor of history at the University of Massachusetts Dartmouth, and Caroline Ummenhofer, a climate scientist at the Woods Hole Oceanographic Institution in Falmouth, Massachusetts. Together with their team of researchers, they are combing through thousands of logbooks that remain from the heyday of New England’s whaling industry, pulling out information from the handwritten entries and converting it to computer data for modern-day climate modeling research.

“The whalers went to pretty remote ocean areas 250 years ago, where we have absolutely no clue what the weather and conditions were like,” Ummenhofer says. “Some of those areas, today, are experiencing pretty big changes in wind patterns.”

Walker got the idea for the research project while serving as crew on the schooner Ernestina-Morrissey, which was built in 1894 and is based in New Bedford, Massachusetts. He was on board with a colleague who is an oceanography expert, and they got to talking. They had heard about the Old Weather project, in which volunteers transcribe ships’ logbooks from the 19th and early 20th centuries, gathering information about everything from sea ice to weather conditions. They brainstormed ways to expand that concept.

Walker lives near the New Bedford Whaling Museum and the New Bedford Free Public Library, which, combined, have about 3,000 whaling-ship logbooks—the museum’s holdings are reportedly the world’s largest collection. So, Walker asked Ummenhofer about the concept of mining those logbooks for information that could be turned into data for climate research. “Within about 10 minutes of describing to her what we had access to in terms of archival information,” he says, “she got very excited.”

Ummenhofer’s research focuses on the ocean’s role in climate, including how it impacts rainfall. Why is one year dry, but another wet? What role is the ocean playing with extreme weather events? To answer those kinds of questions, Ummenhofer needs data that goes back much farther than, say 30 or 50 years—and she needs data from parts of the ocean where, most of the time, nobody has been. Except, in many cases, for the whaling ships. They went looking for their prey far from the traditional sea lanes that commercial and military ships have plied for hundreds of years.

“They’re searching for wild animals in the ocean, so they go to some of the most remote places and therefore gather data that we don’t have from anywhere else, going back 250 years,” Walker says. “They were certainly paying very close attention to wind direction, wind speed, wind force, ocean sea state—a number of things that are very valuable for modern climate modelers.”

Mining these logbooks for information presents numerous challenges. The good news is that most of the historical records are in surprisingly decent shape, kept in archival collections and available for viewing only in special reading rooms. Most of the logbooks were written on cotton rag paper, which predates wood pulp paper and holds up better over the years.

But the key word there is “written.” The logbooks were handwritten aboard whaling ships, which means that no two logbooks are the same. A single page of a logbook can have four days’ worth of information, or 25 days’ worth of information, or anything in between. And the people who wrote that information had different styles of handwriting, often in cursive, which today’s college-age researchers have a hard time reading because penmanship is no longer taught in a lot of schools.

“The real challenge is reading the handwriting—and once you work your way through one logbook, you have to start over and relearn again with a new log keeper’s style,” Walker says. “It’s one of the reasons these things can’t be scanned and transcribed by computers. We need a human who learns the handwriting, deciphers the nautical language they’re using, and records the data day by day.”

Deciphering that nautical language is another challenge. Whaling ships from that era didn’t have the kinds of wind and temperature instruments that even some of the smallest recreational boats have on board today, so there’s not going to be an entry for “wind gusts up to a precise 35 knots.” Instead, crewmen would write terms such as “a strong day” or a “fresh breeze,” language that came from the common Yankee vocabulary, and that was written in the logbooks to create the official legal record of whatever happened on board. To the sailors on the whaling ships, those terms had more than just generalized meanings; today’s researchers need to translate those meanings into modern weather information.

Walker and Ummenhofer’s team relied on previous research that matched the historical terms with what boaters know today as the Beaufort Wind Scale. A “gentle breeze” in a whaling-ship logbook equates to Beaufort Force 3. A “strong breeze” means Beaufort Force 6. A “storm” means Beaufort Force 10.

“It’s interesting: Not all of the terms were used quite as much, or there are 30 or 40 terms that all correspond to a 2 on the Beaufort scale,” Ummenhofer says. “What we find very little about is Beaufort scale of 9. That doesn’t necessarily mean they never encountered 9, 10, 11 or 12, but it might potentially be a survivor-ship bias. The ships that encountered the very strongest storms maybe didn’t come back.”

Once the researchers collect all the information from the logbooks, they enter it into a custom database and Ummenhofer extracts it. Quality control of the data is handled at the Woods Hole Oceanographic Institution—things like making sure the information’s latitude and longitude are identifiable. So far, from about 110 logbooks they’ve looked at since 2017, they’ve gleaned some 60,000 total entries and 32,000 usable entries.

“I always have to tell my researchers—and we have upwards of a dozen now who worked on the project—you’re going to find all these real human-interest stories. You get people who jump ship, there are fights between crewmembers, all kinds of things happen,” Walker says. “But we have to keep our eyes on the prize. We need the weather data.”

Now that they have proved the concept, their goal is to grow the project. Just five New England locations—the two New Bedford institutions, along with the Providence Public Library in Rhode Island, the Nantucket Historical Association in Massachusetts, and Mystic Seaport Museum in Connecticut—possess 85 or 90 percent of all the known whaling logbooks in the United States, Walker says. There are some 5,500 logbooks ready to be researched, depending on how much funding they can get and how many researchers they can add.

“I think the potential is huge,” says Ummenhofer. “The more data we have, the more we can do with that data. There’s a lot more cross-referencing and quality control. Getting started is the hard part, and we’ve done that. Now, it’s just a matter of support and funding.” 

To learn more, go to https://ummenhofer.whoi.edu/historical-records — Kim Kavin

This article was originally published in the February 2023 issue.