By Stuart Pimm
“Now is the winter of our discontent,” the soon-to-be Richard III declares in opening Shakespeare’s play. He then quips to his brother, the current House of York king, that the future is surely sunny: the king has two sons and two brothers, so the York succession is certain.
Life is full of nasty surprises.
Large snowfalls disrupted travel across North America and Europe on an off for the last few weeks, for the second winter in a row. Extreme flooding has killed hundreds in Australia, Brazil, and Sri Lanka.
Yet, NOAA has just announced that 2010 was the hottest year on record (tied with 2005). And the latest decade was the hottest decade. How can this be so?
First the facts.
- Combined global land and ocean annual surface temperatures for 2010 tied with 2005 as the warmest such period on record that began in 1880 at 1.12 F (0.62 C) above the 20th century average.
- According to the Global Historical Climatology Network, 2010 was the wettest year on record, in terms of global average precipitation.
How can “warm and wet” translate into “cold and a lot of snow”? As John Holdren — Director of the White House Office of Science and Technology Policy — has pointed out, “global warming” is too simple and sounds far too gentle. “Global climate disruption” is a better term to describe the changes that are taking place to our world as atmospheric greenhouse gasses increase.
The world does not simply warm — there are changes to rainfall, too. Nor does it do so uniformly. NOAA provided a map for 2010. Some areas, including much of northwest Europe were actually cooler than normal, in 2010. Yet, parts of eastern Canada and Greenland were strikingly warmer in 2010 — up to 5 degrees C (9 F). Western Russia experienced devastating heat waves that caused massive fires and severe disruptions to the food supply.
Moreover, annual averages hide considerable seasonal variation. January and February were colder than usual in the U.S., but averaged over the year, the country was warmer.
Temperatures in 2010 compared to the average for the 20th century.
Map courtesy of NOAA
On top of regional and seasonal variations, there are longer-term oscillations and 2010 had those too! The El-Niño conditions early in the year that come with warm temperatures in the equatorial Pacific Ocean gave way to the opposite La Niña conditions — and much flooding — later on. Then there was the Arctic Oscillation, another large-scale event responsible for sending the U.S. and Europe some nasty winter weather while the Arctic was, well, balmy by comparison to normal.
John Holdren, the late Steve Schneider, and I are sitting in a Washington DC restaurant three years ago. At the next table, a couple are dismissing climate change with the usual cherry-picked examples and doing so noisily. Steve exclaims “Bayesian priors!” and we go back to talking about our wives, families, and wine, which is what male scientists do spend time talking about.
It was the kind of remark that we all loved Steve for making. It requires explanation.
Bayes was an 18th century mathematician and his famous theorem has to do with how probable we think events are in terms of what we expect. Now, scientists know with remarkable precision that the carbon dioxide is increasing in the atmosphere. The resulting physics has been well-known for over a century — the atmosphere traps more energy and, indeed, scientists know how much quite accurately too.
So, let’s suppose that only for 1 percent of all the places with temperature records are there real cooling trends. (That’s our “prior” in Steve’s words.) Obviously, the remaining 99 percent have warming trends.
Given all these messy oscillations, seasonal patterns, and geographic variation, and the like, it’s not easy to detect the true underlying temperature changes.
Suppose that 80 percent of all temperature records detect cooling when it is really happening and so 20 percent miss it. Further suppose that 10 percent of temperature records suggest cooling when it’s not really happening and, so, 90 percent are of the records are correct.
I’ve made these numbers up, of course — well almost. I cribbed them from a website about Bayes. They illustrate an important point — and Steve’s cryptic remark.
What we want to know is what the odds are that when we observe a cooling trend, that it’s really happening? You might think, well, the tests are 80 percent to 90 percent accurate so if we see cooling it has a good chance of being real. Wrong. If you see cooling, the odds of it being real are only 7.5 percent.
If this boggles your mind, remember that by far the greatest number of apparent cooling trends come from the sites that aren’t cooling — but are misclassified as if they were.
Or, one discontented winter does not reject the trend towards ever more warm years.
Which bring us back to history. The king dies, Richard kills the king’s heirs in the Tower, then his other brother, becomes king, but ends up impaled by a pike by a rival claim to the throne. The House of York ends. It gets worse. Shakespeare was a great spin doctor and vilifies Richard for all time. The rival is glorified — good politics, since he was the grandfather of Shakespeare’s queen (Elizabeth I).
From which I draw two conclusions. First, life is unexpected. The House of York seemed secure; it wasn’t. The planet’s species that live in the places where it gets unexpectedly hot or cold, wet or dry, get too much or too little snow, get clobbered. Whatever the long-term outlook, it’s the unexpected that matters.
Second, you only die once. Richard only needed a horse, and he didn’t get one, and so did not fight another day. Once species are gone, that’s it. Extreme conditions may only be transient and local, but that may be enough to eliminate a species forever.
Professor Stuart L. Pimm is a conservation biologist at Duke University, North Carolina. A former member of the National Geographic Committee for Research and Exploration, Pimm is the author of dozens of books and research papers, including the book “The World According to Pimm: A Scientist Audits the Earth.”