That was very much a follow-up on to a paper that Kevin Trenberth, who was my co-author and head of the climate analysis section at NCAR, had written by himself about four years earlier. At that time, he had noted that the atmospheric behavior over the North Pacific ocean fundamentally changed in the late 1970s, and so our paper together in 1994 was trying to explore that change in much more detail and trying to document more widespread impacts of it.

  How did you go about doing that?

We used direct measurements of climate variables like temperature, surface pressure, and the like, that we had going back 100-150 years. We used those to try to put the climate change that occurred in the late 1970s into a longer-term perspective. We were trying to assess how unusual that change was. Had something similar occurred before or not? And by looking at the global picture, looking at what other climate changes were occurring, we hoped to learn something about cause and effect.

  And what was the result?

The result of that paper was the argument that the changes we were observing in the ocean atmosphere system over the North Pacific were directly related to changes occurring in the tropical Pacific; in particular, a warming of tropical Pacific ocean waters that is associated with El Niño. El Niño, of course, is a mode of variability that occurs every few years. If you just think about what you commonly hear, one year might be an El Niño year and the next might be a La Niña year—it goes back and forth. What we concluded was that in the tropical Pacific we were tending to have more warm-water events, more El Niños, than La Niñas, but thousands of miles away, in the middle latitudes in the North Pacific, the climate was responding to those changes in the tropics. And that response involved changes in the ocean in the North Pacific. And because the ocean responds on a much longer time scale than the atmosphere, those changes were now affecting the atmosphere over that longer time scale. Thus the term "decadal." So we documented this very slow change in climate over the North Pacific and related it to the higher frequency fluctuation of the tropical Pacific.

  And why do you think it had such influence in your field? Why was it so highly cited?

Well, it was documenting the extent of changes in the North Pacific, which has influence over North America. It influences storm activity—storms coming off the Pacific and hitting the west coast—and temperature, and those influence biological changes and so on. In effect, we live in the middle latitudes and this paper was saying, look, we've had this major change in how the atmosphere is behaving, just to the west of the United States. Obviously the climate community in general had interest in it, but it had a big impact on scientists in the U.S. The atmosphere, especially in the latitudes we live in, is chaotic. It’s unpredictable. That's why we have a hard time forecasting the weather beyond a few weeks. The atmosphere is inherently unstable. Yet this paper was saying there has been a real change in the atmosphere, that it has persisted for 10 or 15 years, and there is evidence for this much slower-term variability and this has consequences on a wide range of factors, from climate to biology to practical issues like the rise and fall in fish populations.

  Why did you choose Climate Dynamics to publish this article?

That’s not an easy question to answer. What I can say is we pulled together a lot of evidence and we wanted to publish it all in one article. We didn't want to break out or highlight one feature. And we thought we could do that in Climate Dynamics and we could get it published quickly there, as well. And it's certainly a highly respected journal.

  Now, the second paper on decadal trends—and the most-cited—was in Science in 1995, "Decadal trends in the North Atlantic Oscillation: regional temperatures and precipitation." What was the motivation of that investigation?

That paper got launched again because of our interest in El Niño and predicting the climate. The idea is if you can predict what the ocean will be like six months into the future you can start to have some skill in making longer-term forecasts of the statistical behavior of the atmosphere. I can't tell you if a cold front will be moving through New York City exactly six months from now, but I might be able to tell you it will be a hotter than average summer. That's what El Niño gives you. But there's another pattern of variability, and this one resides in the North Atlantic. It's called the North Atlantic Oscillation, and it is the dominant controller of weather and climate variability from the East Coast of the United States all the way through Europe and into Siberia, as well as from the Arctic all the way down to tropical North Africa. But researchers haven't paid that much attention to it because it resides in the middle and high latitudes in the North Atlantic where the atmosphere exhibits this stochastic, chaotic behavior. In other words, there's very little predictability. This large-scale thing just sort of flip-flops around. It's really hard to say even a month from now what it will be doing. So it was of academic interest, but not much practical interest.

I was looking back at some papers that a colleague had written in the late 1970s documenting aspects of this North Atlantic Oscillation. He had a time-series of how it had behaved, but his time series were not very up to date. They went through the mid-1970s. I just simply started, first, to update those time series and, lo and behold, I found this thing had also undergone a really interesting change in behavior over the last, say, 40 years. And then, second, just flipping through these diagnostic bulletins put out by the Climate Prediction Center in Washington—these are maps of weather and climate worldwide—I noticed that for quite a few winters, say the late 1980s through the mid-1990s, Europe kept experiencing the exact same anomalies in terms of temperature and snowfall. In particular, it was very warm, winter after winter after winter, and the Alps were very dry. I hadn't seen much written about it, but I put the two pieces of the puzzle together very quickly. And it turned out that it wasn't true, as people had previously thought, that this oscillation was undergoing purely stochastic behavior. Rather, it had been experiencing this long-term pattern of behavior and it had transitioned from one pattern, characterized by its behavior from the 1950s through the first part of the 1970s, to the exact opposite during the 1980s and early 1990s. And that's what that paper documents. This was a very real climate-change signal and there were very real consequences.

  Why did you chose Science this time, instead of a more specialized journal?

I felt that it was more appropriate for Science. It was a very, very clear climate signal that had direct impact on people. It explained why Europe was getting warmer and warmer and the south was getting drier and drier. It also explained why Scandinavia was getting warmer and wetter, although it was still cold enough that the precipitation had fallen in the form of snow. So it explained why the glaciers in Scandinavia had been expanding in the 1980s and 1990s, as opposed to the glaciers elsewhere in the world, which had mostly been retreating because of global warming. That really caught the attention of the multidisciplinary scientific field. So biologists, for instance, looking at the way ecosystems were behaving, all of a sudden could say, this explains all the changes we've been seeing. And not just in fisheries, but in why birds were migrating differently and hatching eggs earlier, and why large mammals like reindeer seemed to be behaving differently, etc. And that multidisciplinary aspect of the impacts explains why the paper was so widely cited.

  What do you consider the primary challenge or obstacle in your research?

Well, just ignoring the issue of anthropogenic influence, we still don't know why the climate system behaves the way it does. Why do we see these changes and shifts? That's still a very big unanswered question and a very big challenge to understand. And then the anthropogenic component has to go on top of all this other uncertainty. I can give you one brief example, if it’s helpful. I just published a paper in Science that sort of deals with this issue. We have this North Atlantic Oscillation that has this interesting trend. But why is it there? It's not typical atmospheric behavior. We know the ocean in the North Atlantic is being forced by the atmosphere. What I did with some of my colleagues—the lead author is Marty Hoerling—is argue that this trend observed in the North Atlantic climate is due to a warming trend in the tropical oceans. This is another remote forcing. We know that El Niño changes on a year-to-year time scale. This paper is arguing that on a much longer time scale—30,40, maybe 50 years—the warming of the tropical oceans has caused this trend in the North Atlantic Oscillation. Now getting back to the issue of anthropogenic vs. natural climate change. There's the question. Why are the tropical oceans warming? It's very possible that warming is a result of anthropogenic increases in greenhouse gasses. If that's the case, then the change we're seeing in the North Atlantic Oscillation could be an anthropogenic signal.

  Are you satisfied with the rate of progress in global climate research?

I think we're making terrific progress. Meteorology and climate science are relatively new fields. In many ways, you can argue that we're only 60 or 70 years old. People obviously have been documenting and thinking about weather for centuries, but modern meteorology and climatology are much younger. I think we're making tremendous progress, and I think climate models are improving to the point that we now have some reasonable insight into how climate might change in the future.

  What are your research goals for the next five years?

I'm very much interested in working on this business of the North Atlantic Oscillation. Even though we've just written a paper claiming we understand why these trends occur, like with any research, it still needs a lot more testing and evaluation. I would very much like to work on that. The North Atlantic Oscillation is the dominant driver of weather and climate over much of the northern hemisphere. Even though El Niño gets a lot of press, the North Atlantic Oscillation is a very key player. We would like to continue to work on that and better understand the processes responsible for its variability.

Dr. James W. Hurrell
NCAR/Climate and Global Dynamics Division
Climate Analysis Section
Boulder, Colorado, USA