From Indiana University
Scientists detect seismic signals from tornado
BLOOMINGTON, Ind. — An Indiana University geophysical experiment detected unusual seismic signals associated with tornadoes that struck regions across the Midwest last week — information that may have value for meteorologists studying the atmospheric activity that precedes tornado disasters.
The experiment by IU researchers involves deployment of more than 100 state-of-the-art digital seismographs in a broad swath of the U.S. midcontinent. One of the twisters that struck southeastern Missouri and southern Illinois on Feb. 29 passed through the seismic detection array.
“In examining the seismograms, we recorded unusual seismic signals on three of our stations in southern Illinois,” said Michael Hamburger, professor in the department of geological sciences at IU Bloomington and one of the researchers conducting the experiment.
“The seismograms show a strong, low-frequency pulse beginning around 4:45 a.m. on Feb. 29. Our preliminary interpretation, based on other seismic records of tornadoes, suggests that we were recording not the tornado itself, but a large atmospheric pressure transient related to the large thunderstorms that spawned the tornadoes.”
The seismographs that detected the pulse are near Harrisburg, Ill., a town of 9,000 where a pre-dawn twister caused extensive damage, killed six people and injured about 100 more.
IU researchers initially feared that some of the instruments might be damaged by the storm, setting back a National Science Foundation-funded project that included the investment of hundreds of thousands of dollars and months of effort. But when principal investigator Gary Pavlis, an IU professor of geological sciences, checked the digital recordings of the Illinois stations on Feb. 29, he found they were still alive and streaming data. As he checked further, he discovered the strange “tornado seismograms” that were recorded on seismographs near Harrisburg.
Hamburger said a seismic pressure gradient associated with the tornado produced a slow, minute tilting of the seismograph that lasted for several minutes. He said this sort of pressure-related signal may help scientists better understand atmospheric activity that takes place right before tornadoes touch down. The IU researchers are working with colleagues at the University of California San Diego to try to compare recordings with other tornado-related signals and to dig deeper into the analysis.
While seismographs have been known to detect seismic activity related to tornadoes, it is highly unusual to have state-of-the-art digital instruments recording information in such close proximity to a tornado, the researchers say.
The IU seismic experiment, dubbed “OIINK” for its geographic coverage in parts of the Ozarks, Illinois, Indiana and Kentucky, includes the positioning of 120 seismometers to study earthquakes and geological structure in a key area of North America. Installation of the instruments began last summer. They are recording thousands of earthquakes from the study area and around the world, as well as nearby mining and quarry explosions.
The $1.3 million, four-year undertaking is part of the NSF’s EarthScope program, which seeks to cover the entire U.S. with a grid of detection devices for the purpose of better understanding seismic activity and predicting earthquakes. Researchers liken EarthScope to “an upside-down telescope” that allows them to look into the Earth and gain a better understanding of seismic forces.
More details on the experiment are available at newsinfo.iu.edu/news/page/normal/18612.html.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I’d say the Severe Storms Center already understands the “atmospheric activities” that take place before a tornado. They’ve been predicting with a half-hour lead time for MANY years. Nearly always enough time to reach shelter.
The real problem is poor response by broadcasters and other local communication in states outside Tornado Alley. (i.e. states that haven’t learned from Gary England.)
I think their numbers are skewed. There is a direct effect of the frequency and locality of Tornadoes by the……………….
Trailer Park Island Effect
Reblogged this on Kmareka.com and commented:
Interesting news about scientists discovering a connection between tornadoes and seismic activity.
Maybe they need to retune their equipments since the 10WALB posted news from heartland news at 4:35 PM CET about possible tornado that swept through the area killing 10 in souterhn Illinois and south east Missouri.
So it seems they picked up the signal a little too late.
I think some WUWT readers may be suffering from PAGWS (Post-AGW Syndrome)! Years of disputing silly, over-the-top claims by warmists have left us overly skeptical of science news in general.
I think this is an intriguing piece of information, the sort of accidental discovery that may lead to something useful. It may also lead to nothing at all, which is the nature of research.
Now if it turns out that the warmists link all this back to CO2 (the miracle gas that can do anything at all) I will retract my statement about being overly skeptical.
Hey! Here’s the beef!
(Hamburger, that is….)
Eartthscope has more like 400 sensors in it (“more than 100” is an understatement). Interestingly enough for the open-source people among us, anybody here can get access to and down load any of the relevant data—it’s all publicly available.
The component that they are probably using is called USARRAY and (slight /brag) contains infraosund sensors developed by me and manufactured under my control. That data is publicly available too.
The data are all transmitted back from the various array components using broad-band wireless (e.e., 3G) if anybody is interested in that.
I’d guess what they are seeing is the infrasound wave generated by the storm, and low frequency sound couples to the ground via coupling of the ground to the atmosphere. Seismic-only is less likely because generally you’re going to excite the surface wave with a storm like this, and they typically fall off pretty rapidly with distance (usually in the noise floor after about 5-km).
That might be useful for near-field warming. I had a poster on measuring the infrasound signal from tornadic storms in the fall 2011 AGU meeting.
Sorry if I’m hijacking this thread, but we’ve discussed the issue of the end-of-the-glacial period North American “extinction” several times on WUWT. Below is an interesting post at Universe today about a Mexican impact site:
http://www.universetoday.com/94001/mexican-lake-bears-witness-to-ancient-impact/
Alan the Brit says:
March 9, 2012 at 1:52 am
Jeff, (if I am reading you correctly) is there enough lead in time with these things, as somebody suggested earlier? With all these things, it’s the amount of time the warning gives you that is crucial
________________________
With large super cells warning times are decent and yes Gary England is my goto guy when the sky’s get dark. The tornado chasing craze has helped and hurt all at the same time. For the last few years it seems there are chase teams on almost every cloud that crosses the state. Little tornadoes some not much bigger than a dust devil are now being spotted and recorded on video. I have heard of grumbling from the research community that the amateurs/tour groups are getting in the way of them being able to get their equipment in place and in time to make measurements. The small storms can still pop one out with almost no notice.
The warning times do help but when you hear the words you must be underground to survive offers no comfort when there is no ground to go to.
“Being located in tornado alley this is a pretty interesting piece for me.”
My concern is whether they’ll be able to distinguish between tornadoes and New Madrid getting ready to cut loose.
(New Madrid, MO — town motto, “It’s Our Fault!”)
According to the press release (often a scary source to use for reliable info), the transient was recorded beginning at 4:45am. From the NWS storm survey, the tornado began at 4:51am and struck Harrisburg at 4:56am. That’s not a lot of lead time but at least it seems to be positive.
I think this is more of a curiosity item rather than anything that will have predictive skill. There may be many other transients not associated with tornadoes. The storm in question might have to pass quite close to a seismometer to pick it up too.
u.k.(us) says:
March 8, 2012 at 10:33 pm
28 years in Aviation- Insturment Instructor, Flight Instructor Commercial Pilot, Multi and Single Engine Land. NOAA certified Weather Observer- had to take same test as the NOAA folks,
back in the 70’s . 11,500 hours, l worked for 91/2 years as an Airtanker Pliot. about 2800
hours in 4-Engine Douglas aircraft. Flew as a Co-pilot. Flew air freight and Commuter airliners.
worked as a Govn’t contract Pilot for Everone from the Department of Defence to the
Department of Energy. 7 years running a Charter service/Flight school.
My comment was concerning aviation safety and the idea of prediction.
form my own perspective as a particpant.
Jeez. What a bunch of soreheads
wayne Job says:
March 9, 2012 at 12:15 am
These instruments are a tad sensitive and a local backwoods farmer removing the odd tree stump or three with dynamite may cause a state wide panic.
Not to mention the odd mid western “amateur pyrotech” testing out this months batch of black powder.
I wonder if DHS helped fund the project?
” EarthScope Program scientists are deploying thousands of seismic, GPS, and other geophysical instruments” to study the structure and evolution of the North American continent and the processes that cause earthquakes and volcanic eruptions”…..
Maybe Earthscope should first figure out that “North America” doesn`t end North of the 49th Paralell…
DirkH says:
…and IF they recorded a signal BEFORE the formation of actual tornadoes, the next obvious question would be, what’s the false positive rate, and can they say exactly where the tornado will hit. Because otherwise you just end up alarming the whole country for naught, and the warnings would immediately be ignored by everyone…
So, what are you suggesting? It sounds a lot like you’re just dismissing this completely. Should they have not bothered reporting the observation, because they can’t immediately make use of the information?
Seems like you’re demanding quite a bit from a single observation of a phenomenon.
TonyG says:
March 9, 2012 at 11:50 am
DirkH says:
…and IF they recorded a signal BEFORE the formation of actual tornadoes, the next obvious question would be, what’s the false positive rate, and can they say exactly where the tornado will hit. Because otherwise you just end up alarming the whole country for naught, and the warnings would immediately be ignored by everyone…
So, what are you suggesting? It sounds a lot like you’re just dismissing this completely. Should they have not bothered reporting the observation, because they can’t immediately make use of the information?
Seems like you’re demanding quite a bit from a single observation of a phenomenon.
By the way they described the signal, it sounds like it is a long DC offset which is pretty much the opposite of an impact like explosives or periodic vibration from volcanic/ tectonic action. There are probably other phenomenon that cause offsets that could cause false readings.
What this study will more likely do, is help scientists in the field design new sensors that can more accurately measure this effect.
To my knowledge, it’s not typical that these stations are monitored real time other than to check that they are operational. Most of the data is recorded and stored for researchers. I suppose it’s possible it could be monitored and could contribute to early warning systems. I worked for one of the companies that make up EarthScope for many years. While it might be helpful, the NSF grants are not set up for this purpose. Time to write another proposal, I suppose. They could call it EWWS (Early Weather Warning System). Why not? The project discussed here was called OIINK. I like EWWS.
TG McCoy (Douglas DC) says:
March 9, 2012 at 10:43 am
==========
Is English your first language, your spelling gives me great concern.
As a former pilot, your posts don’t ring true.
A true pilot, would give an emotional (filled with curses) response to my inquiry.
Not a compilation of google searches.
One report covering Harrisburg Ili, at MSNBC is logged at 2:57 am on the 29th of febuary and according to witness’ tornadoes in southern Illinois was on at 12:25 am already. That would be central for US not for europe.
So, exactly how can they claim discovery to detect things going on at 4:45 am way after the fact for that area? It’s like stating after the crescendo of Katrina that there’s probably a storm having passed through here, and well, oh yeah, we kind of made that discovery.
It’s probably more probable they detected the whif of one of the earth quakes couple states over that same morning then discovering a storm already passed.
Now if they added temperature and humidity sensors, an automatic rain gauge, and an annemometer, maybe we could get some high quality weather/climate info as well!!
I’ve been through a few tornadoes in my 65 years, and don’t like them. I’ve seen them on weather radar and noticed the static interference on television screens (most notably on Channel 4). Tornadoes do weird things. I’d not be surprised that there would be a seismic signature for them any more than I’m not surprised there is an electrical and radar signature for them. This is something I’d be glad to follow for awhile. That’s the reason I love WUWT – you do it for me.