This is the first of two stories on the Hurricane Reanalysis Project.
Well-practiced in the discipline of “hunkering down” for approaching and passing storms, Virgin Islanders might be surprised to learn just how many hurricanes are officially listed for the territory in the last 70 years.
From 1946 through 2019, the official record at NOAA (the National Oceanic and Atmospheric Administration’s) National Hurricane Center lists just three V.I. hurricanes: Hugo in 1989; Lenny in 1999; and Dorian in 2019.
The catastrophic dual destroyers of 2017 are listed only for Barbuda, St. Martin, the British Virgin Islands and the Bahamas (Irma) and for Dominica and Puerto Rico (Maria.) Not the USVI, which is still recovering from millions of dollars’ worth of damage and total disruption of its health, education and other social systems.
Hurricane Marilyn, which destroyed or badly damaged an estimated 80 percent of structures on St. Thomas in 1995, is listed solely at Dominica, and only as a minor, Category One hurricane.
The most obvious explanation for what looks like blatant omissions involves terminology.
The listing, which appears on the NOAA website, is headed “International Atlantic Hurricane Landfalls.” The “international” designation is geographic, not political, and serves to distinguish between hurricanes affecting the Caribbean region and those that made landfall on the U.S. mainland. There is a separate listing for the mainland.
“Landfalls” is the key word for both. It denotes that the eye of a listed storm actually passed over the coastline of a given land area.
Considering the small size of each of the Virgin Islands, that makes “landfalls” rare, according to expert Christopher Landsea who is chief of the National Hurricane Center’s Tropical Analysis and Forecast Branch.
“With islands, though, you can have a huge impact without landfall,” he said. A storm may “strike” an island, without making landfall.
Landsea is also open to the possibility that there could be errors in the database.
In fact, he has been a leader in making corrections to the database for more than 20 years and is in charge of the effort that is designated the Hurricane Reanalysis Project.
When the project started in the 1990s, the formal U.S. government record of tropical cyclones (hurricanes and tropical storms) in the Atlantic began with the year 1886. Using historical records, project researchers were able to extend the database back in time to 1851. They also began moving forward chronologically, reviewing and reassessing whatever information they could find, and periodically issuing reports, which are available on the NOAA website.
To date, the researchers have made adjustments to the record of U.S. storms up through 1965. There are also a handful of analyses on individual hurricanes deemed important enough to warrant their own report.
In a recent interview with the Source, Landsea said he hopes to complete and publish the analysis of 1966-1970 in a few months.
That report will contain information about a hurricane that lasted for two days north of the Azores. “It wasn’t even in there (in the database) – not even as a storm,” he said, adding it is always exciting to discover a hurricane that was previously unknown. In the 1960s, he said, there were an average of two to four storms each year that didn’t make it into the old record.
The project is not just about adding storms to the record; it is also about changes concerning the path of a storm, its intensity, its duration, and its lifespan.
The report covering the period 1851 to 1910, for instance, says there were “several thousand alterations and additions” to that period made as a result of the reanalysis project, and there could be more in the future “once new information is made available.”
“Our understanding (of tropical cyclones) has evolved,” Landsea said. The technology available to find, track, and assess storms has changed dramatically over time. Now, “If there’s a storm out there, we know about it.”
Landsea said he got involved in the reanalysis project “kind of by accident.” He was in graduate school, researching climate change at Colorado State University, where he would earn his master’s degree and a doctorate in atmospheric science in the 1990s. He was using the hurricane database and it was evident that it was lacking.
He credited the late Charles Newmann, a meteorologist specializing in tropical cyclone forecasting and research, with spearheading the revision of the record.
Most of the work has been done with a very small staff. The team size fluctuates. Currently it consists of two forecasters and himself, Landsea said. And they can’t devote full time to the project, so work is moving slowly.
“Our day job is forecasting,” he said. Tracking storms in real-time and issuing warnings about them takes precedence over historical data dives.
Next: Some of the hurricane reanalysis findings.