What goes up must come down

Very few relationships in this world are monotonic. Not the price of stocks, not the traffic on this blog, and not global climate trends. Maybe if more people understood this, we'd have less nonsense about climate change clogging the media.

By monotonic, I mean, if you plot a trend on a standard x-y graph, monotonic lines will always go in the same direction, whether it's up, down or flat, no matter what the scale. The fact that most functions aren't monotonic should be obvious to anyone who thinks about it for a while. The real world has a habit of being a little wonky. Even those things that display long-term trends, like the cost of gasoline, sometimes move in the opposite direction.

This is why climatologists don't particularly care too much what the global average temperature is today. They don't even care too much what the global average temperature change was for the last 10 years. What they do care about is decades or centuries-long trends, because the scale at which long-term forces overwhelm the irregular and short-term forces that don't really matter.

There's a new paper coming out in Geophysical Research Letters that includes some very useful graphs that make this point crystal clear. Here's one.

There's nothing much new in the paper, which carries the rhetorical title "Is the climate warming or cooling?" David R. Easterling of the National Climatic Data Center in Asheville, NC (just up the interstate from my neck of the woods) and Michael F. Wehner of the Lawrence Berkeley National Laboratory out of the west coast appear to be driven only by the need to better inform the public -- using easy-to-grasp illustrations -- about what's going on with the Earth's climate. Their paper is a touchstone of clarity and I reproduce the abstract (of what I hope the publishers at GRL will decide to make freely available) here with glee:

Numerous websites, blogs and articles in the media have claimed that the climate is no longer warming, and is now cooling. Here we show that periods of no trend or even cooling of the globally averaged surface air temperature are found in the last 34 years of the observed record, and in climate model simulations of the 20th and 21st century forced
with increasing greenhouse gases. We show that the climate over the 21st century can
and likely will produce periods of a decade or two where the globally averaged surface
air temperature shows no trend or even slight cooling in the presence of longer-term
warming.

Their timing is almost perfect. It could only have been better if the paper was released the same day as Washington Post columnist George F. Will's most recent effort to misrepresent the science of global warming. Even after being raked over the coals by Chris Mooney for a previous obfuscatory column and the head of the World Meteorological Organization for the same, Will wrote last week that

... according to statistics published by the World Meteorological Organization, there has not been a warmer year on record than 1998.

This may or may not be technically true. NASA reports that 2005 was slightly warmer than 1998, but the difference is tiny and all comes down to how close the thermometers are to the Arctic Ocean and other methodological preferences. But what's not true is that it means anything.

Easterling and Wehner show that we can expect there to be periods of time as long as a decade or so during which the global average temperature will decline. This is the way the world works. It makes no difference to the long-term trend of gradually warming temperatures. To arbitrarily look at the last decade is anything but scientific.

Why isn't the world's climate monotonic? Because it's complicated. To take one example, there's this phenomenon involving the heat transferred between the ocean and the atmosphere called the El Niño Southern Oscillation (ENSO). It swings back and forth, somewhat irregularly, between two states. In one state (El Niño) more heat ends up in the air, so the thermometers that record the temperatures we experience tend to say it's getting warmer. In the other, La Niña), the process is reversed and more heat is transferred to the deep oceans, leaving our air a little cooler.

Even though our surface temperatures change, the total amount of heat on the Earth doesn't. We're not sure exactly what drives this oscillation, which is why it's called a phenomenon. Still, it is reasonable to adjust our records to compensate for its effect. If you do, and plot the results on another standard two-dimensional graph, the same long-term trend is there, only it's a little easier to see.

Here's one such graph, courtesy of RealClimate.org

If you look closely at the right end of the graph, you can see how recorded temperatures (dotted lines) in 1998 make any trend that starts there and runs into the future negative. That's because 1998 saw a major El Niño. Corrected for ENSO, the positive trend is obvious. Actually, it was obvious before, but now it's even more obvious. Incidentally, 2008 was an La Niñ year, and that explains a good deal of the cool weather we've been having of late. So choosing 1998-2008 turns out to be remarkably poor choice to examine global warming patterns.

This isn't a particularly difficult concept to grasp. We all know that it gets warmer in the summer and cooler in the winter in the Northern Hemisphere, so no one would dare use a temperature trend of half a year to reach a conclusion about global warming. Looking at five or even 10 years can lead to similar errors. Yet otherwise intelligent critics continue to do just that, even after the problem has been pointed out to them.

I'll leave the implications of such stubborn behavior to others.