Global warming and hurricanes seem like a catastrophic match made in heaven. Surely, there must be a connection between the two, right?
While Carl Ebeling doesn’t address this question directly, his research just might help climatologists and meteorologists give a better answer. A doctoral student with professor Seth Stein in the Department of Earth and Planetary Sciences at Northwestern University, Ebeling has been studying hurricane data since 1992.
More specifically, he is looking at whether there is a correlation between microseisms – seismic noise generated by storms in the ocean – and hurricane activity. Is there a specific microseism frequency range that coincides with a specific hurricane? Using Hurricane Andrew as a model, Ebeling was able to show that microseisms were generated with unusually high energy in the five- to seven-second period range while the hurricane was active. But he found that a methodology based only on energy in this range is probably not enough to detect a hurricane.
Ebeling sat down with Medill Reports to discuss microseisms, their role in hurricane detection, and the connection to rising sea temperatures and global warming (a term Ebeling is a little cautious in using due to its political implications).
What are microseisms? One definition says they are “the ambient seismic noise that continually bathes the Earth.” Could you put that into layman’s terms?
Think about earthquake waves – P waves, S waves, and so on. If you look at those squiggles on a seismograph, a lot of people might think about a paper roll recorder with the needle going back and forth. In the absence of an earthquake signal, there’s always this little back and forth that’s continually going on that travels over long distances and is generated by storms in the ocean. It’s hard to see it, but it does get in the way of processing if you’re interested in earthquakes. That’s how it came to be called noise. And I sort of flipped that on its head; I’m interested in the very small signals in the absence of earthquakes. Earthquakes for me have turned into noise.
How will your research help determine whether global warming has an effect on hurricane energy or frequency?
With Hurricane Katrina, there was a lot of discussion about global warming causing rising sea surface temperatures. Hurricanes draw their energy from the heat in surface waters. How does that translate into hurricane behavior changes- are they more frequent or intense? That’s where the question becomes hard to answer. Our really well-constrained database began only with the onset of satellite observations in the 1960’s. Well, it turns out we have ambient seismic noise records in the form of analog seismograms that go back much earlier than that. We can go back to those records and say “Well, there’s a hurricane here. This one corresponds to one we know about for 1938,” for instance, and compare the seismically-identified hurricane record with our meteorological and satellite observations.
Microseisms were used in weather forecasting in the 1940s, but research on microseisms effectively stopped in the latter part of that decade because of the start of satellite-based observations. Why did it take so long to get the ground moving on microseisms again?
I don’t know. Microseisms have been a topic of recent research, but the hurricane connection was investigated up to about 1950, and that was about it. It was interesting that the military was driving a lot of the research in the mid ‘40s. The Navy put together the project that ran during this time in the Caribbean, and they did a pretty good job. But I think between then and perhaps the last 15 or 20 years, in terms of using it, it was looked upon just purely as noise. I think [the] stigma [is] that it was named noise, and somehow it takes on a characteristic of being totally worthless. But why that is? I don’t know.
Your paper on microseisms was centered on Hurricane Andrew in Florida in August 1992. It was a big hurricane, and it was also the only hurricane in that month. Why were those factors so important?
This was our test case. If it works, it’s gotta work for this one. And it did work out well. But it turns out that, had we started with a different month with maybe multiple hurricanes, our approach might have been different. It turns out that non-hurricane storms can also generate energy in the same frequency pass band that we were looking at for hurricanes. So it becomes a little bit more of a difficult question to unravel. If I do this processing for a month and I see a spike on the 5th or the 6th that lasts for a few days and a spike on the 20th that last for a few days – are they both hurricanes? One hurricane? Two hurricanes? There are some additional tools that I’m working on right now that will hopefully answer that question.
What is the next step?
I’m going to continue working on the Category 5 storms. I’m also going to continue to try and develop different threshold measuring and processing algorithms and see if there’s something that I haven’t come up with yet that will be able to say “Yes, this is a hurricane. No, that’s not a hurricane,” and have it be right for every single case. I’m still hoping to find that. In the absence of that, then [the next step is] to concentrate on learning what we can about seismic noise and storms and the connection for Category 5s, 4s, 3s, and perhaps 2s and 1s.