Arguments
Wave goodbye to the stadium wave - global warming still caused by humans
Posted on 24 April 2014 by John Abraham
Many people have spent a lot of time trying to show that much of our recent climate change is just natural. So far, these studies have died as fast as they've been born.
A recent attempt was made to liken our climate to fans in a stadium, you know those annoying "waves" that fans make? Could our climate just be like that? Not likely; let me explain.
The so-called "stadium wave", as described by the scientists who coined the phrase, is "a hypothesized multi-decadally varying climate signal that propagates across the Northern Hemisphere." It is basically a signal that travels throughout the Earth's climate, in the ocean and atmosphere, and can be sensed by measurements. From author Dr. Marcia Wyatt's own website, you can find a more detailed description.
So, could our recent changes in climate be somehow associated with these "waves"? The great thing about science is these hypotheses can be checked. In fact, a very recent study did just that and found the stadium wave hypothesis lacking.
In the study, authors Michael Mann, Byron Steinman, and Sonya Miller estimate the low-frequency internal variability of the Northern Hemisphere (the stadium waves) by evaluating observed signals. These observations include the impacts of both forced (mainly human-caused) variations and natural variability (energy naturally "sloshing around" the Earth). They generated a set of alternative histories based on the statistics of past observations to show that the recent Northern Hemisphere variations are within the range of expected uncertainty.
The authors then show that past analyses which have been used to estimate internal variability have failed to find appropriate variability when it was known ahead of time. These methods errantly show natural variability which is too high and which has a biased phase. As a result, the claims of stadium waves are made based on an incorrectly methodology; they are likely an "artifact of this flawed procedure."
To get into the weeds a bit, the primary natural fluctuation that was focused on is called the Atlantic Multi-decadal Oscillation (AMO for short). The recent studies promoting large natural climate variability deduced the AMO by detrending the time series of North Atlantic sea surface temperatures, and then interpreting the left-over low-frequency variation as the AMO. This means they removed a linear variation. Prior works (here and here for instance) have shown that this method causes artifacts to appear. For instance, if you aren't careful, you might conclude that atmospheric aerosol cooling was actually a natural climate fluctuation. That is, you might confuse human-caused cooling effects with natural fluctuations.
Northern Hemisphere variability comparing differenced, detrended, and forced temperature anomalies
The researchers Mann, Steinman and Miller, further showed that the detrended AMO method yields a climatic variability that is approximately twice as large as prior estimates and outside the 95% error range. They also found that the detrending method got the timing wrong (biasing error in the phase). They compared the true AMO signal and the detrended AMO signal. The detrended signals have amplitudes which are too large by a factor of 2 and they are all in phase. As the authors write, this suggests, "an artifact of the common forced signal masquerading as coherent low-frequency noise."
As I've written about before, studies which purport to show that significant recent climate variability is just natural are taken seriously by the scientific community. In fact, I think we would all like to conclude that the current climate change is just natural. During the course of scientific investigation, these claims are given their fair hearing. Unfortunately, they have not borne scientific scrutiny well. Distinguished Professor of meteorology Michael E. Mann told me,
"Some researchers have in the past attributed a portion of Northern Hemispheric warming to a warm phase of the AMO. The true AMO signal, instead, appears likely to have been in a cooling phase in recent decades, offsetting some of the anthropogenic warming temporarily.
We conclude that the AMO played at least a modest role in the apparent slowing of warming during the past decade. As the AMO is an oscillation, this cooling effect is likely fleeting, and when it reverses, the rate of warming will increase.
Initial investigations into the multidecadal climate oscillation in the North Atlantic were hampered by the short length of the instrumental climate record which was only about a century long. And some of the calculations were contaminated by long-term climate trends driven or "forced" by human factors such as greenhouse gasses as well as pollutants known as 'sulphate aerosols.' These trends masqueraded as an apparent oscillation."
Although the stadium wave is undoubtedly an incorrect hypothesis - I consider the counterintuitive result of the recent Mann et al (2014) study to require greater scrutiny. In particular this result does not
The issues with the method are related to the input parameters of the energy balance model he uses, the accuracy of the forced components used and finally the lack of any spatial figures. IF this method is appropriate then he should be showing a spatial amplitude map and it should have the same spatial pattern as would be expected based on theory behind the mechanisms. This is somewhat a glaring omission. I think he provides a compelling case that the detrended AMO is inappropriate but I think his solution is theoretically appropriate but in practice is not sufficiently justified based on the paper. I also did not like that he cited Booth and other aerosol forcing AMO studies without citing their rebuttals which were compelling. The argument that the AMO was positive during the 1990s and is negative currently is at odds with the spatial distribution of temperature changes over that period - particularly in the Labrador Sea. In this area the temperatures are warming faster than projected by GCMs and were faster during the mid-century and cooler during the 1970-1995 section. This temperature history for one of the main nodes of the "amo" is at odds with the history implied by Mann's version. I suspect many of the experts on the physical mechanisms behind the AMO will disagree strongly with his new reconstruction of this index.
I think any "new definition" of an AMO needs to be supported by more than just time series analysis - there needs to be a physical understanding of the underlying mechanism. A point made in Climate Dynamics last year. Did they check to make sure these results made sense with respect to the underlying mechanism? Did they relate it to salinity and sea ice ? As a mode of NH temp variation it is possible there is some relation to this index - however the AMO which is traditionally referred to by authors was not cooling over the past 15 years.