An interview with climate scientist Michael Wehner
By Josh Chamot of Nexus Media
The Pineapple Express storms hitting the West Coast are intense, causing massive floods and landslides — and replenishing reservoirs after historic drought. But is the drought-flood pattern tied to our planet’s warming? Michael Wehner, a leading climate scientist at the Lawrence Berkeley National Laboratory, weighs in with what we now know and what we don’t. This interview has been edited for length and clarity.
Is weather really that different now compared to pre-industrial times?
In isolation, of course not — to date, any modern extreme weather event would have been possible in pre-industrial times.
That said, the likelihood of a big storm — or a severe heat wave — has changed a lot. It’s clear that since the 1950s, over much of the United States, extreme storms are becoming wetter.
We’ve only had one degree centigrade of global warming. That’s a lot, but it’s nothing compared to what is coming down the line if we stay on the same greenhouse-gas emissions scenario that we’re on. By the end of this century, you’re looking at 3 or 4 degrees of warming, which is a planet that no human has ever lived on.
We have no experience with such a planet, and it’s entirely possible that in such a warmer world, storms that were previously impossible — and certainly heat waves — will occur.
The Pineapple Express atmospheric river. Source: NOAA
So how intense is the weather in California this winter?
Yeah, the second storm has hit, and it’s raining like crazy. It’s what the forecasters call a Pineapple Express [so-called because it passes over Hawaii]. They’re part of a larger class of storms called atmospheric rivers.
These big storms that hit California are almost always atmospheric rivers, copious transports of tropical moisture to the coast. It’s one of two kinds of storms we get here, along with cold extratropical cyclones, Aleutian lows, that form near Alaska.
The Pineapple Express brought rain to California this month. Source: NASA
Just two types of storms drive all of your weather?
That’s about all we get in central California. Whereas the East Coast gets all different kinds of storms, and the Midwest gets all sorts of other things, two types account for almost all of our precipitation. The extreme precipitation typically comes from the atmospheric rivers.
California precipitation is very episodic. You either get a big storm or you don’t, and in between it tends to be clear. This week’s are a little unusual. As the forecasters will say, “the storm door is open.” You’ve got more than one coming through.
They’re separate entities, but their impact is not — it’s cumulative. There was flood and landslide risk with the first storm this week, but it’s exaggerated now because the ground hasn’t had a chance to drain… it’s totally saturated.
In 2010, severe storms and floods damaged California roads like this one in Pasadena. Source: FEMA
So how do you attribute climate change to specific weather events, or prove an event is unrelated?
There are two ways of looking at the question of whether there was human influence on a particular weather event. First, is there a change in the probability of an event, and second, is there a change in magnitude.
There are events that happen where there is no apparent human influence. But for storms like those here in California this week, or heat waves, this is less and less the case. We have not done a formal attribution study on this storm, but I’d be very surprised if somebody doesn’t.
El Niño delivered storms that flooded California’s Russian River in 1998, which as then the hottest year ever recorded. Source: FEMA
Can you expand on that a bit?
A model can do three things: it can help you understand something you think you already understand; it could shatter that belief, which it sometimes does; and occasionally — very occasionally — it tells you something you didn’t already know.
If the model makes some projection, and I have no idea why, then that’s an interesting result, but what does it mean? Do we really believe it? It’s worth publishing to get other people thinking about it, to come up with the explanation for why the model is either right or wrong.
I don’t believe anything blindly, not even observations, but I do believe in the scientific method, and trying to understand processes, and models are just one set of tools for that.
There’s no question that high-performance computational technologies are a critical part of this. But it’s not just the machine — it’s much more than that. It’s the people. Because of the landmark 2003 heat wave paper by researchers from the UK Met Office and the University of Oxford, an entire community developed. It’s a scientific curiosity driven at this particular problem at this particular time.
It will run its course, and get boring in a few years. We’ll have scoped it all out. That doesn’t mean it’s not important from a societal context, or loss and damages — but a lot of smart people are now thinking about this all at the same time, and that broadens the field.
It’s important to stress that this is 19th-century technology. We know how much moisture can fit into warm air because of steam engines.
Changes in seasonal average precipitation are projected to go both up and down — it will go up a lot in the upper Midwest in the winter, close to limits predicted by thermodynamics. In the desert southwest it will dry out, and that’s because of circulation changes.
In most places, for very rare events, extreme precipitation is expected to increase, even in places where the average precipitation is projected to decrease. So, in some places, in a 4-degree-warmer world, you will have both floods and droughts. It seems counterintuitive, but when it rains, it will rain a lot. This is kind of what it does now in the southwest.
As computers have gotten faster, scientists have been able to produce more refined climate models. Newer models divide the atmosphere into smaller segments, allowing for more accurate predications. Source: NCAR
In my view, the most important recent development in climate models’ ability to simulate past extreme storm statistics and to project future changes in those statistics is high resolution. Typical climate models discretize the planet into 100-kilometer cells, at best. Recent advances in high-performance computing have enabled that resolution to be pushed to 25 kilometers, but that takes the most powerful supercomputers in the world. Fortunately for me, we have some at Berkeley Lab.
Those high resolution models simulate storms, both extreme and run of the mill, far better than the previous generation. It is still early days, and there is much to learn, and many more simulations to do, but this is a golden age.
This interview was conducted by Josh Chamot, who writes for Nexus Media, a syndicated newswire covering climate, energy, policy, art and culture.