Wyatt on Earth

by Marcia Glaze Wyatt, Ph.D.

Evolution of the Hypothesis

A Note to Reader:

The following story is one of how this stadium-wave idea evolved. The telling of that tale may be a source of encouragement for someone embarking on his or her own journey into the unknown. Regardless of whether or not the hypothesis stands the test of time, the unfolding of the stadium-wave idea, from seed to blossom, tells a story about not conforming and not giving up.


Evolution of the stadium-wave idea:

As with any work in science, many hands work to bring an idea to life. The stadium-wave hypothesis is no exception.  I first presented my hypothesis - the precursor to the stadium-wave  model - to my CU dissertation committee, circa 2007. This, after a year or two of the idea’s development.  I was unprepared for the committee's reaction: It was flat, at best.

I was humbled; yet undeterred.


In the months to come, a paper (Tsonis et al. 2007) caught my attention . Similarities in our ideas – climate as a network - seemed apparent. I contacted Dr. Anastasios Tsonis of University of Wisconsin, Milwaukee. He listened. That spontaneous encounter launched a trail of discovery! The ‘wave’ came alive. Sergey Kravtsov, a UW colleague of Dr. Tsonis, and Dr. Tsonis  joined the CU dissertation team. I went through 'comps' successfully, and was off and running - finally! Sergey Kravtsov coached me remotely. He designed a rigorous statistical approach with which to document the signal,  and taught me the coding with which to do this. The CU-UW ‘merger’ involved years of work on the ‘wave’, ultimately resulting in the stadium wave’s debut in the literature in early 2011. (Climate Dynamics (online April 2011; published in journal format 2012) : Wyatt, Kravtsov, and Tsonis (2012)). 


Work continued; this next step with John Peters. At the time John was a UW graduate student of Kravtsov. The goal of our research was to determine whether or not the computer general-circulation-climate models (GCMs) - the CMIP3 suite of models - used widely in climate research, successfully captured  dynamics that, in the real world, generated our hypothesized stadium-wave behavior - i.e. a  low-frequency, hemispherically propagating climate signal. John was the genius behind the necessary Matlab coding used to reconstruct simulated stadium-wave indices from model-generated data.  This study was completed in mid-2011. Our results failed to produce a propagating signal. In short, our results were negative. These findings suggested that the model designs in the CMIP3 ensemble lacked the full collection of dynamics that interact in a way to choreograph ocean and atmospheric indices into a multidecadally varying, hemispherically propagating climate signal. Our paper was published by Springer Publishing Company: (Wyatt and Peters 2012

The UW team temporarily diverted energy to projects other than the stadium wave. I continued my stadium-wave investigations alone, expanding the stadium wave temporally (through proxy indices) and spatially - this latter step incorporating indices for Eurasian Arctic shelf sea ice and the Intertropical Convergence Zone.


In late 2011, one of my co-advisors – Roger Pielke, Sr – brought to my attention a posting on Judith Curry's  blog site regarding spatio-temporal chaos by Tomas Milanovic. It was a fascinating essay, the content of which seemed relevant to my work. I contacted Dr. Curry, then the department chair of ocean and atmospheric sciences at Georgia Tech. We discussed the Milanovic piece . Then we discussed my work.


I sent her my ‘project-under-construction’. She read it and expressed interest. With that encouragement, I went back to work.  In January of 2012, with more progress under my belt, I again contacted Dr. Curry. Her interest was still keen. We discussed her joining the team. She accepted, becoming the newest, and last, addition to my geographically diverse dissertation committee.  The concluding topic, my expanded-stadium-wave study, took final shape  and was written up as chapter three of the Wyatt dissertation. Successful defense followed in early May, 2012.


Post-degree, I worked on converting my dissertation's final chapter into a paper - the only chapter remaining unpublished. Submission was followed by rejection. The journal's editor offered an option: to re-submit an abridged version. Which I did. I culled out material on proxy analysis and potential solar influence on the temporal pace of the stadium wave's variability - two aspects still of great interest to me. By late December 2012, things finally seemed to be coming together. In the meantime, I had heard through Dr. Pielke that Dr. Curry was interested in seeing this paper. She wanted to blog about it once it was published. I sent it to her just prior to submission of the newest manuscript version. She offered editing suggestions. Over the next several days, her extensive and well-thought-out edits tightened the paper’s message considerably. I invited her to be co-author. She accepted. After a grueling journal review, which enhanced the paper’s delivery further, the Wyatt and Curry 2014 paper was accepted by Climate Dynamics (online: September 2013) and added to the slowly growing list of published stadium-wave research papers: Wyatt and Curry 2014.

Two additional papers have been added to the stadium-wave team's collection since the Wyatt and Curry paper. Sergey Kravtsov was lead author on both. One was published in Geophysical Research Letters (Kravtsov et al. 2014); the other in Science (Kravtsov et al. 2015). Both papers were written in response to challenges on the hypothesis. The challenges were based on model-generated data and modeled outcomes. Our study results  exposed differences in features captured by the models versus those captured in observational data. Hence, much as with the Wyatt and Peters study, the Kravtsov et al. studies suggest that the current collection of climate models fail to fully capture dynamics embedded within the observational data base.


From the time (~2007) the seed was planted by the pioneering works of Klyashtorin and Lyubushin (1998; 2007), to the most recent paper (2015), eight years have passed. Many great minds have contributed to this effort: My co-advisors Peter Molnar and Dr. Roger Pielke, Sr., who supported me and remained patient through the wave’s development; Drs. Kravtsov and Tsonis, without whom the stadium-wave would have remained an undocumented and unknown idea; John Peters, who diverted his focus from his own studies to work on code for the modeling project; and Judith Curry, who showed interest and provided extensive editorial feedback,  enhancing the packaging of the message. 



The stadium-wave hypothesis gained considerable publicity after publication of Wyatt and Curry. This is both good and bad. The good is that a study's exposure allows its new ideas to invite interest and spawn related research efforts. The bad is that sometimes an idea, especially a somewhat controversial one, is morphed into being what it is not, and misconceptions about it can be hard to shake. Sometimes content is misinterpreted. People can have the tendency to interpret content according to individual filters, with the end-result sometimes exaggerating what a research finding is or is not; or worse, sometimes completely misrepresenting it. Thus, an idea can become something it was not intended to be. This stands as a cautionary note regarding citations. They do not always accurately portray the paper cited!


No hypothesis can be proven to be true. Surviving hypotheses have not been proven false, but there is always that possibility. Science is a slow process. New technologies and/or longer data sets constantly cast a hypothesis in new light. The stadium-wave hypothesis has been rigorously tested, and, at the time being, appears consistent with observation. Only time and further research will tell if  the hypothesis of a hemispherically propagating  climate signal survives. If it does, its greatest value may lie in identifying what dynamics are missing or poorly represented in current models, allowing for further improvements in simulated scenarios.