Climate, Weather, Severe Events

Is it getting weird yet?

This is a difficult question. Weird is not a quantifiable concept, at least not easily. And confirmation bias colors what we are seeing - those who expect weirdness see weirdness; those who don't simply experience nature's ordinary orneriness.

But let me try to explain what I mean when I say I expect weirdness. I do not mean that the atmosphere is slightly warmer and slightly wetter, so that warm extremes and wet extremes will slightly increase. That is easy to argue, and a lot of climate folk are falling back on that argument, but I think there is more to the case than that.

Let me go back to an analogy I have used to explain the difference between climate and weather. Imagine a large tropical fish tank on a table, say at a mid-range Chinese restaurant. The fish swim hither, the fish swim yon. As you watch the fish over many visits (perhaps you especially like to sit by the fish tank as you roll up your mu shu pancakes) you come to know their habits. The striped one likes to hang out near the surface but occasionally goes for a dive, the reddish one hangs out by the coral, and so on. As you learn the habits of the fish, you develop a sort of set of expectations for where they will be as you look up from your meal. Occasionally they will be a bit out of place; you'll think "that's a little weird" but not lose any sleep over it.

As an expert on the fish, you sometimes strike up a conversation with your companions. "Look, I bet that one's getting ready to dive to the bottom..." Congratulations, you are now an expert on fish prediction! This is like weather prediction, only with fish. Then someone asks you to predict where the clownfish will be in ten minutes. You chuckle and say you can only guess, based on its habits; it's current position gives you no information about that. Good. Now you have said something about fish climate.

Now imagine that your obnoxious Young Republican nephew us joining you for dinner, and leaning on the tank so hard as to make you worry that he will knock it over. You tell him so, and ask him to please stop messing with the fish tank, which after all is not his property, and which the restaurateur will surely come after you for rather than your ne'er-do-well nephew. And besides, you really do rather like the fishtank.

Of course, your nephew persists, invoking his freedoms. You see the corner of the tank lift off the table and become alarmed. You say you are pretty sure he will break the fishtank if he keeps pushing it harder and harder. Rather than complying, he asks you again where the clownfish will be in ten minutes.

"I have no idea!" you reply, exasperated. "What the heck does that have to do with it? That tank will be smashed before you know it; the fish will be flopping on the floor, I'll be out twelve hundred bucks, and I'll be persona non grata at my favorite Szechuan joint, thanks to you!"

"Aha", quoth he. "So you know nothing about tropical fish displays at all! You admit it! If you don't even know where a single fish will be in ten minutes, how can you make predictions about when it will hit the floor?"

I thought this analogy could make the difference between weather and climate clear. The weather is the unpredictable behavior of the fish, and the climate is the predictable behavior of the fish tank that influences the behavior of the fish.

---

I've come up with an extension to the analogy which I think may start to give the flavor of why I think extreme events ought to increase. This time, my nephew decides that as an exercise in his freedom, he will be pitching all his beverages into the fish tank. Clearly, the behavior of the fish will change. But the water level in the tank will also change. Some of the fish may end up spending some of the time at the top of the tank. That is above the level that they have ever been able to attain under normal conditions. The changed conditions make new fish configurations possible that had been impossible before.

---

I can make a somewhat less handwavy case using something like the Lorenz system, but there's no solid proof of this speculation. However, it is at least possible, through this analogy, to understand that we aren't just loading the dice. In another analogy we've referred to around here, we might roll a thirteen.

Are we doing so already? That's a harder question. I do not know that deferring to statistical arguments can tell us anything. An April tornado season without known precedent follows upon no detectable trend in tornadoes. So the idea that it's part of a trend in tornadoes is unsupported. But it's still a very weird event.

I think we have to expect the unexpected. And when it happens we are going to have to do the hard work of analyzing the underlying physics. Subdividing it by individual phenomenon will tell us nothing because we haven't got enough data to tell us about the distribution of extreme outliers.