Our award certificate from the RMetS.

At the latest count I think we need 17906 copies of this document. Here’s yours – display it with pride.

Zooniverse HQ, Adler Planetarium, Chicago USA

Those of us who use the oldWeather website every day have got thoroughly used to it, and take it for granted. But there is a lot of magic going on behind the scenes to make a complex interaction with the logbook images simple to use on each of many different computers and browsers, and keep working reliably however many people are using it simultaneously.

This magic is actually skill and effort, of course, and it’s made possible because oldWeather is one of the Zooniverse suite of projects. Zooniverse is an infrastructure for online public science: a core team of experts in building and using websites that let everyone do science, a group of 15 projects (oldWeather is one) each doing a different sort of science using Zooniverse tools, and an ever-growing community of volunteers working on the projects.

Most of the time I only concern myself with oldWeather, but last week I went to a Zooniverse-wide workshop in Chicago; where developers, scientists and volunteers from many of the projects got together to talk about technical innovation, good practice and new ideas. It was an excellent meeting, ably described here by our own Janet Jaguar, and in more detail here by Jules Wilkinson. I came away with some new friends, some new ideas, but perhaps most of all, with a renewed appreciation of how much we owe to the development team.

When reading and annotating the logbooks, and even when analysing the results, we don’t think much about user-interface design, database schemas, servers, JavaScript, website responsiveness and the like. But all that has to be done, and to make a website like oldWeather that ‘just works’ it has to be done WELL. So it was a privilege for me to invade the high-tech (if rather dark) office on the shores of Lake Michigan where the developers work their magic – thanks to them all.

Dates are in grey, positions in brown, weather in blue and events in red. Pick your ship.

We’ve known for a long time that oldWeather is awesome, and now it’s official: We’ve won the Royal Meteorological Society’s IBM award for ‘Innovation that matters’.

Well done to everybody – there is innovation in just about all components of oldWeather: our weather and climate contributions, the design and user interface of our website, our links to archives and the humanities, the ship histories we are constructing, the surrounding and supporting work of the Zooniverse, and especially the extraordinary contributions of many volunteer participants, who have gone above and beyond what we asked in many productive directions.

When we started oldWeather it was just a climate data rescue project. It still does that and does it well, but we’ve grown into something much richer, more powerful, and more fun; and that’s the doing of the volunteer participants more than the science team – ‘innovation that matters’ indeed.

Grand opening: December 24th, 1880

In the evening we had a minstrel entertainment in the deck-house, somewhat improved over last year… The success of the evening, however, was Sharvell as a young lady, in an after-piece… A feature of the evening was presenting each guest, on entering, with a little button-hole bouquet of colored paper leaves… The jokes were of the usual order, some broad ones being inevitable… The most acceptable occurrence was the issue of a double ration of whiskey, with which, hot water, and sugar, we tried to be cheerful, and make Christmas Eve rather less dreary than many of our days now seem.

By December 1880 Jeannette had been stuck fast in the pack ice for well over a year, but Christmas Day was made as acceptable as possible by a good dinner.

And I think we may refer to our bills of fare with pardonable pride. Our mince pies were a work of art; though they were made from pemmican and flavored by a bottle of brandy, they were as delicate to the taste as if compounded from fresh beef from market. CHRISTMAS DINNER, 1880: the usual Saturday Soup, Roast Seal – Apple Jelly, Tongue, Macaroni, Tomatoes, Mince Pies, Plum Pudding, Figs, Raisins, Dates, Nuts, Candy, Chocolate and Coffee.

Reading these snippets from De Long’s journal from the Jeannette would strike a chord with any sailor. Christmas at sea seems much the same today – people still celebrate as best as they can even though they are far from home and hearth. The banquet remains the highlight of the holiday. But the work must go on, as the ship’s log shows: along with the note about the entertainment there are all the usual weather observations recorded. Considering the perilous journey the Jeannette logbooks endured from the Arctic ice to us, it is a miracle that these data still exist and can be used for science again, in new and powerful ways that De Long and his crew could not have imagined when they sledged these heavy volumes across the sea ice after the wreck of their ship, and then carried them on their backs to the cold Siberian shore.

Old Weather citizen-scientists are making the recovery and analysis of these hard-won data possible, while also getting to read the unedited stories from the Jeannette and many other ships in the new (and growing) Old Weather – Arctic fleet.

Kevin

They don’t mention what it sounded like, but the log of the Thetis does say a lot about sea-ice. This isn’t surprising: they were sent to rescue the Lady Franklin Bay Expedition in 1884 because two previous resupply expeditions (in 1882 and 1882) had been blocked by the sea ice, and one ship (the Proteus) had been crushed by the ice and sunk. So manoeuvring through the ice was their major challenge.

We’d like to compare the Thetis’ observations with modern observations of sea-ice in the same place – as we do with temperature, pressure wind etc. This is a challenge, because modern observations come mostly from satellites; they are precise and quantitative and tell us the fraction of the sea (in each region) that was covered with ice on each day. Thetis’ log, on the other hand, contains statements like ‘working through slack ice and broad leads’ or ‘forcing a lead through heavy pack ice’; is this more or less than ’75% coverage’? The best place to start is generally just to plot the data, so the video below shows the movements and ice observations we’ve taken from the log, and a modern satellite ice climatology (HadISST).

We can tell quite a lot just from this: it looks as if they first met serious ice around Disko island, on May 22nd, while the modern estimates suggest that we’d expect this a little further north. But in general they are seeing ice in much the same times and places that the satellites do.

It’s not quite enough to compare the ship records with a single satellite field. Sea-ice severity varies from year to year, so we’d like to compare Thetis’ experiences with a range of modern years, not just an average year (a climatology).

Thetis’ ice observations (black text) compared with max, mean and min coverages from HadISST 1979-2004 (light, mid, and dark blue curves)

I’m calling this an iceberg plot. You’ll need to look at the bigger version, but it shows that the period where the ship made observations of ice fits pretty well with modern expectations, but that those observations generally imply heavy ice cover, and the modern observations suggest heavy ice cover only in particularly severe years. So we can tentatively conclude that, in summer 1884 in Baffin bay, there was more sea-ice than usual (by modern standards) but that the coverage was within the expected current range of year-to year variability. This is consistent with their air temperature observations, which were colder than the modern average, but not dramatically so.

A remaining oddity, however, is the time of year of this voyage. Thetis set out in May, and reached her farthest north in June. Sea-ice cover in Baffin bay is still heavy in June, and doesn’t reach its minimum until September. If they had waited two more months before setting out they could have reached their destination with much less effort (look at the ice-cover at the end of the video, when they are back in New York). Thetis could not wait without leaving the men they were rescuing to starve, but the resupply missions in 1882 and 1883 could have travelled at more favourable times of year. If Thetis could get there in June 1884, but Neptune, Proteus and Yantic could not get there in August of 1882 or 1883 – does that mean that 1882 and 1883 were remarkably heavy sea-ice years?

I wonder if the National Archives have the logs from the voyages in 1882 and 1883.

Back Row, left to right: Mark Mollan (Maritime/Navy reference Archivist), David Ferriero (Archivist of the United States), Jane Lubchenco (under secretary of commerce for oceans and atmosphere, and administrator of NOAA), Kathy Wendolkowsky (oldWeather citizen scientist), Linda Joy (NOAA Communications), and Kevin Wood (oldWeather-Arctic principal investigator). Front row: logbooks of USS Jeannette, USRC Bear, USRC Corwin, and USS Vicksburg.

It’s not enough just to launch an exciting new project. We also have to tell everybody it’s started so they know that they can join in. So as part of the oldWeather-Arctic launch we held a press event, hosted by our friends at the National Archives in Washington and splendidly organised by Linda Joy, from NOAA communications.

Most of the project participants live too far from Washington to attend (I phoned in, to help answer questions, but it’s not the same). However we were ably represented by Kathy Wendolkowsky and Kevin Wood, who explained to a fascinated audience of dignitaries and members of the press, what we are doing, why it’s important, and why it’s fun.

The Archivist and the Head of NOAA both spoke at the event, here’s what they said:

• Remarks by David Ferriero (Archivist of the United States)
• Remarks by Jane Lubchenco (under secretary of commerce for oceans and atmosphere, and administrator of NOAA)

Photo credit: Amy Cox (NOAA) – it’s a little blurry because she could not use the flash (it’s bad for the logbooks).