I'm a professor at U Michigan and lead a course on climate change problem solving. These articles often come from and contribute to the course.
By: Dr. Ricky Rood , 8:13 PM GMT on May 29, 2014
Tracking El Niño: Underlying Models
El Niño and La Niña are names given to frequently occurring patterns of variation that are concentrated in the tropical Pacific Ocean, but that change the average temperature of Earth for about a year. When there is an El Niño the globe is warmer and when there is a La Niña the globe is cooler.
In the last blog I wrote about predictions of a 2014 El Niño and why it is of such interest to climate and climate change. In this blog, I want to write about models that predict El Niño and relation of this type of modeling to climate change and climate modeling. Reaching very far back, I have written a bunch of blogs about modeling. In this blog from 2007, I write about types of models: intuitive or heuristic, statistical and physical. For this blog I will focus on physical models. I have also written about the difference between weather predictions and climate projections, with a simplistic explanation of internal variability versus forced behavior. Finally, I wrote a series to introduce models and modeling to nonscientists and here is a link to a late article in that series.
Here, I focus on the modeling of El Niño and a set of issues that are potentially related to climate change over the next decades and centuries.
Basic Information on El Niño Predictability: It has been recognized from earlier than the 1990s that El Niño might be predictable. El Niño is often stated as the largest source of natural variability, though such a statement depends on how long a time period you are talking about. Someone, who studies ice ages might provide a convincing counterexample of large natural variability. However, if we look at the recent weather and climate that is relevant to humans and society, El Niño is a major cause of variability in, for example, the global average temperature. This is especially obvious when looking at the hot years, for example 1997 and 1998, and even a moderate El Niño this year is likely to lead to the hottest year on record. El Niño also has a big impact on the weather. For the U.S., El Niño is known to affect the precipitation on the West Coast and in the Southeast. There is also a reliable impact on Atlantic hurricanes. Therefore, with an El Niño forecast, we can say something about the characteristics of weather. This is an example of how a forecast or projection that focuses on climate variability provides usable information for planning – a strong El Niño will matter a lot to water managers and emergency managers in California.
Seasonal prediction is feasible if there are slow and predictable variations in measures such as sea-surface temperature, sea-ice, snow cover and soil moisture. The atmosphere and, therefore, the weather is sensitive to these changes. Especially for El Niño, which is first described by a change of sea-surface temperature in the eastern Pacific, there is an atmospheric response. This response has a pattern, which includes more precipitation in California in the winter and fewer hurricanes in the North Atlantic in the summer. Here are a couple of references I use in class if you want to read more (Seasonal Prediction in 2001) and (Seasonal Prediction 12 years later).
One of the interesting pieces of information that comes from these papers is the “springtime barrier.” That is, if the forecast extends through the springtime of the northern hemisphere, then the skill of the forecast declines. One explanation of this characteristic is that the northern springtime signal of El Niño is relatively small, therefore variability that might be construed as noise to an El Niño forecast dominates the projection. There is an obvious consequence of a springtime barrier, a limit of forecasting of about six months. There is an interesting paper Very Early Warning of Next El Niño, which “indicated (in September 2013 already) the return of El Niño in late 2014 with a 3-in-4 likelihood.”
Some Changes in the Climate?: I wrote a long, some would say tedious, series of blogs on the Arctic Oscillation, changes in sea ice and atmospheric blocking (all of those terms defined in that series). Whether or not the changes in the Arctic are having large impacts on weather in middle latitudes or the tropics remains an open question subject to scientific investigation. From the point of view of predicting El Niño, during this prediction cycle we have levels of sea ice that are far lower than in previous El Niño cycles. This changes the heat exchange between the atmosphere and ocean in the Arctic. This is outside of the range of previous variability, which intrinsically increases the uncertainty in the forecast. The same could be said for springtime snow cover. In short, our background environment, on which we have developed what forecast skill we have, is changing. Also in my mind is a project that I participated in back in 2011 and 2012, where we were concerned about La Niña and flooding in the Upper Missouri River Basin. In that project, any sensitivity to La Niña was overwhelmed by the Arctic Oscillation being in its negative phase.
We Can’t Predict Beyond Two Weeks: One of the sacred utterances of the community of the skeptic is that the the description of weather as chaos means that we cannot predict climate. Another statement is that errors in weather forecasting mean that we cannot predict climate. The skill that we have established in seasonal prediction stands as a concrete example of why these utterances are merely rhetorical diversions. A talking point would be that ocean sea-surface temperature can be predicted. We both predict and observe organized patterns of warm and cool regions of the sea. The atmosphere responds to these patterns. The atmospheric response is not random, for example, during an El Niño in winter, California can expect major rainstorms. In the summer, hurricanes are likely to be less frequent in the North Atlantic. The people who place real money on these predictions, emergency managers, insurance companies, farmers and water managers, ultimately, win.
Comments will take a few seconds to appear.