The Crowd and the Cosmos: Adventures in the Zooniverse
Page 16
added twist: the Apollo missions brought back a treasure trove
of lunar rocks.
A small proportion of this bounty was used for symbolic pur-
poses, with fragments of rock distributed around the world as
124 Into the ZoonIverse
symbols of American largess and technological superiority.
(Though Britain’s allocated lump sits proudly in the Natural
History Museum in South Kensington, many countries have lost
theirs; Ireland’s, for example, was thrown away after an observa-
tory fire damaged the building it was kept in.) Most, though, has
been kept in pristine conditions for scientific purposes, among
the most important of which is establishing an absolute, not a
relative date.
With the lunar rock returned to Earth, geologists and planet-
ary scientists have been able to use the full gamut of laboratory
techniques to establish the age of the samples, mostly via radio-
metric dating. One technique makes use of the fact that a par-
ticular isotope of potassium, 40K, decays to a particular isotope
of argon, 40Ar, when one of the protons in its unstable nucleus
turns into a neutron.
Argon is a gas, and if the rocks are melted through heating, it
escapes. Measuring the amount of 40Ar present and comparing it
to the amount of potassium reveals how long it’s been since the
rock was melted. Since for most of the lunar surface the last melt
corresponds to the volcanic formation of the surface, for the few
precious places visited by the twelve Apollo astronauts we know
absolutely how old the surface is. Count the craters there, and we
can calibrate our entire scale for the Solar System, and get proper
dates for major, surface-marking upheavals on any world.*
It still boggles my mind that our understanding, say, of how
old the moons of Mars are, needs to be calibrated by work with a
lump of rock picked up from the Moon by human beings, but
* It is, of course, much more complicated than that. To do the job properly one needs to consider the different rate of bombardment on Mars, neighbour to the asteroid belt, for example, and on the Moon, and of course the different rates at which meteorites burn up in Mars’ atmosphere compared to, say, in the thicker air of Earth, will matter too. But you get the idea.
Into the ZoonIverse 125
until we explore further that’s precisely the case. The trouble is
the crater counting. Most of the Moon’s surface is billions of
years old, and that is a lot of time for craters to build up. There are old surfaces on Earth, too, but here erosion due to weather
removes the scars of all but the largest or more recent impacts.
On the Moon, more or less, once a crater is in place it stays put
until aeons’ worth of subsequent impacts eventually obscure it.
Looking at parts of the Moon is a matter of picking craters out
from among the debris of previous generations of craters, which
is not an easy task. As higher-resolution images became available,
the task of crater counting became more and more difficult and
time-consuming; the arrival at the Moon of NASA’s Lunar
Reconnaissance Orbiter, which mapped almost the whole surface at Figure 18 The Apollo 17 landing site as seen by the Lunar Reconnaissance Orbiter. The remains of the lunar module and blast marks from its take-off are clearly seen, as are tracks from the lunar rover the astronauts used to explore.
126 Into the ZoonIverse
a resolution of a hundred metres per pixel, threatened to over-
whelm scientists (Figure 18).
The solution was to turn to volunteers, and using the same
software that powered Galaxy Zoo we set up and ran a project
called Moon Zoo, producing catalogues of craters at important
sites. The project wasn’t as successful as its predecessors—partly
because there turned out to be pretty big differences in what
experts were prepared to count as a crater—but it was another
early reminder that we galaxy experts were not the only ones fac-
ing a flood of images and data that people were struggling to deal
with. Nor was it just astronomers. With the Sun, Moon, and the
distant Universe covered, the fourth Zooniverse project brought
us back down to Earth.
Back down to Earth, though still dealing with planetary-scale
phenomena. Understanding the Earth’s atmosphere and climate
is of inherent interest even to an astronomer like me—how else
are we to get a sensible understanding of how all of these planets
around other stars must behave—but being able to predict how
the climate might change in the next few decades is an urgent,
vital question. As it’s not possible (or at least, not sensible) for
them to conduct experiments on a planet-wide scale, climate sci-
entists have reached the same solution as cosmologists who face
similar struggles experimenting on the Universe, turning to large
computer simulations.
In the powerful computers owned and operated by places like
the UK’s Met Office, it’s possible to run simulation after simula-
tion of how the weather and climate might play out over the next
few decades given different scenarios. It’s a complex, difficult
problem, replete with the kind of feedback loops that mean it’s
hard to make even simple predictions. For example, warmer air
may lead to more clouds. But think about an image of the Earth
from space, with bright white clouds flecked across the darker
Into the ZoonIverse 127
blue of the oceans and the green and brown of the land. It’s not
hard to realize just by looking that the clouds reflect more light
than the surface they cover, so they have a cooling effect. But
more clouds means more rain, and so the composition of the
atmosphere changes, and so on and so on.
These difficulties can be addressed, and indeed they have been
to such an extent that we can be confident that the changes now
visibly underway in the Earth’s climate are due to the steady
input of man-made carbon dioxide into the air. Unless politi-
cians the world over (and, yes, the rest of us) get our act together, these sobering simulations show our future of more extreme
weather events, unbearable summer heat, and dramatic change
across the face of the planet. Planning for these changes, much
less averting them, depends on being able to rely on the models,
and that means doing the best possible job of testing them.
One option is to wait. I did attend one meeting of climate sci-
entists studying the Arctic, memorably held in a dark and stormy
Reykjavik one October. With storm clouds gathering over the
ocean and cutting the days short, it appeared to me to be a suit-
able location to discuss the fate of the world. I’d expected the
speakers to have ideas on how to engage the public in averting
climate catastrophe, but the mood of the meeting was different.
Everyone who was involved in trying to create explanations, the-
ories, and models of the Arctic had results that made sense of
current conditions. That, after all, was the price of entry—any
new theory that couldn’t explain what was happening in the
Arctic today would get sho
rt shrift. However, different models
predict wildly different futures, and amid a slightly manic sense
of end-of-the-world excitement, the scientists in Iceland dis-
cussed how the changing climate might provide experimental
proof of one idea or another.
128 Into the ZoonIverse
Given that all of the models predicted significant rises in sea
level, I prefer to find a means of testing climate models which
doesn’t involve permanently ruining the delicate balance of the
Earth’s atmosphere and ecosystems. It seems a bit much to pre-
cipitate the greatest environmental catastrophe in the history of
our species to prove one Nature paper right over another. Luckily, we can test climate models by seeing how well they predict not
only the future, but the past.
By gathering data on the weather and climate going back
centuries we can make new demands on the supercomputers
and the theories which are programmed into them. As well as
explaining what we see today, we can insist they explain the
past too, and thus gain confidence that they will be able to guide
us towards the future. The only problem is that we don’t have
decent records of the weather across much of the world from
before the middle of the twentieth century. Western Europe
and the Atlantic coast of North America are OK, but the rest is
pretty hazy. My main contact at the Met Office, a physicist
called Phil Brohan, describes the picture as being clouded by a
fog of uncertainty.
Climate is global, and the effort to understand how the atmos-
phere is behaving greatly benefits from a worldwide picture. One
part of the world might be having an unseasonably rainy sum-
mer while others bake in conditions of severe drought. If Europe
is cold one winter, it doesn’t mean that China isn’t having a mild
time of it. Records do exist, but they are locked away, hidden in
handwritten notebooks and stuffed in drawers in half-forgotten
archives the world over.
Phil reckons there are a billion or so observations out there to
find, in need of rescue and conversion into digital data that can
be fed into the computers. Some have yet to be unearthed from
those dusty drawers. Most still sits stubbornly on paper, awaiting
Into the ZoonIverse 129
the digital photography that is the first step in the long process
of digitization. Some, though, just need typing up, a data entry
task that seemed both urgent and so large as to be intractable.
We thought that volunteers might be able to help, but their work
needed focus.
Phil and colleagues wanted to concentrate on the most useful
data, something that would give them worldwide coverage as
quickly as possible. You really want a fleet of mobile weather-
observing platforms, touring the world and making scientific
measurements as they went. The word ‘fleet’ is the right one—
land makes things complicated, what with hills and mountains
and valleys and other factors that affect the local weather, so to
keep things simple you would, for preference, send your weather
detectors out to sea.
Luckily, the British Royal Navy has been systematically record-
ing the weather every four hours on board every ship since the
late nineteenth century, sticking to the task come hell or literally
high water. The ships’ logs held in the National Archives in Kew
contain page after page of these observations in table after table,
with air pressure, temperature, and information about the wind
recorded next to the everyday business of loadings, unloadings,
and navigation. These logs aren’t brilliantly written literary jour-
nals penned by officers who would, were it not for Naval service,
be dashing off novels in London, but to the Met Office team they
were priceless treasures, and we decided to start with the logs
from the early part of the twentieth century, particularly around
the First World War.
If we could get people to transcribe what was written down,
that is. The software was easy enough to adapt, but there was a
nagging worry that people would find the task just too boring to
be contemplated, no matter the scientific justification. My confi-
dence that we weren’t just inflicting tedious data entry on people
130 Into the ZoonIverse
wasn’t helped by the quizzical look the project name we chose
got in meetings; we called the project ‘Old Weather’, which I
thought captured nicely the everyday nature of the data being
collected and the public fascination with weather records. It
turns out that talking about the weather is, perhaps, a uniquely
British phenomenon (who knew?) and my American colleagues
in particular were deeply sceptical.
We were so worried that we did something we hadn’t done in
Galaxy Zoo, Solar Stormwatch, or Moon Zoo. We decided to try
and turn this task into something of a game. When they arrived
on the site, classifiers were expected to sign up to join a ship’s
crew. They could then follow along on its voyage, merrily typing
in weather data as they went, and would be rewarded for their
efforts by promotion within the ranks. The volunteer who’d con-
tributed most to a particular ship’s logs would be the captain,
those who had contributed a bit the officers, and so on down the
list to the new recruits who had merely dabbled.
It turned out that we were silly to worry, and Old Weather
was an enormous success. In concentrating on the weather data
our friends the climate scientists needed, we had neglected the
inherent interest of the logbooks themselves. Within their
pages, which we had seen only as tables of numbers, were
laconic records of life on-board ship. As they worked through
page after page, volunteers got to follow their ship around the
world. (This turned out to be useful, too, as volunteers who
spotted sudden jumps in recorded position could put right mis-
taken coordinates which were wrong in the original logbooks
themselves.)
They also found odd little notes providing windows into life
on the high seas. There were reports of illness, and even the
occasional death. Comings and goings were recorded; one officer
who popped up on several ships turned out to be responsible for
Into the ZoonIverse 131
distributing medals to boost morale. One, presumably long-
suffering crew seems to have enjoyed regular lectures on a variety
of improving subjects from their captain, all faithfully noted in
the log. We know, thanks to Old Weather volunteers, where a
saucepan was lost overboard just south of Iceland. Participants
were especially moved by the loss of the chocolate rations of the
HMS Mantua in a dockside loading accident (Figure 19).
These minutia have now been edited—by volunteers—and are
available online. The weather data has been fed into climate
models and the fog of ignorance Phil and his Met Office col-
leagues are fighting has receded just a little. Having completed
the Royal
Navy’s First World War logs, Old Weather volunteers
have taken on more difficult challenges, including the amazing
records of the early Atlantic whalers whose battles with the ice
preserve a record of exactly where that ice was.
And the list goes on. In 2010 I moved, temporarily, to the Adler
Planetarium in Chicago. Adler’s a marvellous place, sitting on a
peninsula that juts out into Lake Michigan; my office was under-
neath the best view of the city skyline available anywhere.
Founded in 1930 by Max Adler, a businessman who’d made his
fortune from the Sears catalogue empire, it’s nearly unique as a
place which employs academic researchers as part of the museum
staff. They’d spotted the potential for volunteers to contribute to
science long before, and were excited about Galaxy Zoo. While I
was there, a grant from the Sloan Foundation meant we could
build a proper development team. I soon moved back to Oxford,
but Arfon moved out to Chicago and the Planetarium has hosted
a large part of the Zooniverse team ever since. Before too long we
were building projects that helped researchers understand plank-
ton, study animals on the Serengeti, delve deeper into particle
physics, and much, much more.
Figure 19 Logbook from HMS Mantua for 6 July 1917. Weather observations occupy the central columns, with occasional notes on the right, including the sad loss of fifty pounds of chocolate while loading.
Into the ZoonIverse 133
From trying to understand galaxy evolution to old whaling
records and gazelle spotting was quite a journey. In each project,
I’d been worried about getting sufficient volunteers, but each
time I was elated by the sheer power of people’s desire to help.
The ability to spend a few minutes trying to understand the
Universe (or the Earth) was apparently the kind of thing people
wanted in their lives, and as we tried new and more complex
things I remained in awe of quite how each small effort could add
up to something grand.
5
TOO MANY PENGUINS
I don’t think I was ever in any serious danger, but as I slid down an icy slope, scraping against freezing rock, my thumping heart
would have disagreed. About twenty metres below was the frigid
water of the Southern Ocean, its waves turned into slush by icy
melt. Scrabbling desperately at the slope through hands hidden