UI researchers link smoke from fires to tornado intensity

Can smoke from fires intensify tornadoes? Study finds smoke from Central America intensified 2011 tornadoes


The satellite image for April 27, 2011 shows the southeastern United States, Central America and the Gulf of Mexico — along with tornado tracks (red solid lines, with the thickest indicating a magnitude 5 tornado, descending to magnitude 1 for the thinnest — for the period April 26-28, 2011. Yellow markers indicate fires, and an iridescent overlay shows particulate matter in the air, with red showing highest amount and purple the lowest. Credit: Imagery courtesy of Brad Pierce, NOAA Satellite and Information Service Center for Satellite Applications and Research.

“Yes,” say University of Iowa researchers, who examined the effects of smoke–resulting from spring agricultural land-clearing fires in Central America–transported across the Gulf of Mexico and encountering tornado conditions already in process in the United States.

The UI study, published in the journal Geophysical Research Letters, examined the smoke impacts on a historic severe weather outbreak that occurred during the afternoon and evening of April 27, 2011. The weather event produced 122 tornadoes, resulted in 313 deaths across the southeastern United States, and is considered the most severe event of its kind since 1950.

The outbreak was caused mainly by environmental conditions leading to a large potential for tornado formation and conducive to supercells, a type of thunderstorm. However, smoke particles intensified these conditions, according to co-lead authors Gregory Carmichael, professor of chemical and biochemical engineering, and Pablo Saide, Center for Global and Regional Environmental Research (CGRER) postdoctoral fellow.

They say the smoke lowered the base of the clouds and increased wind shear, defined as wind speed variations with respect to altitude. Together, those two conditions increased the likelihood of more severe tornadoes. The effects of smoke on these conditions had not been previously described, and the study found a novel mechanism to explain these interactions.

“These results are of great importance, as it is the first study to show smoke influence on tornado severity in a real case scenario. Also, severe weather prediction centers do not include atmospheric particles and their effects in their models, and we show that they should at least consider it,” says Carmichael.

“We show the smoke influence for one tornado outbreak, so in the future we will analyze smoke effects for other outbreaks on the record to see if similar impacts are found and under which conditions they occur,” says Saide. “We also plan to work along with model developers and institutions in charge of forecasting to move forward in the implementation, testing and incorporation of these effects on operational weather prediction models.”

In order to make their findings, the researchers ran computer simulations based upon data recorded during the 2011 event. One type of simulation included smoke and its effect on solar radiation and clouds, while the other omitted smoke. In fact, the simulation including the smoke resulted in a lowered cloud base and greater wind shear.

Future studies will focus on gaining a better understanding of the impacts of smoke on near-storm environments and tornado occurrence, intensity, and longevity, adds Carmichael, who also serves as director of the Iowa Informatics Initiative and co-director of CGRER.


Paper co-authors are Scott Spak ofthe UI Departments of Urban and Regional Planning and Civil and Environmental Engineering; Bradley Pierce and Andrew Heidinger of National Oceanic and Atmospheric Administration Satellite and Information Service Center for Satellite Applications and Research; Jason Otkin and Todd Schaack of the Cooperative Institute for Meteorological Satellite Studies, University of Wisconsin-Madison; Arlindo da Silva of NASA Goddard Space Flight Center; and Meloë Kacenelenbogen and Jens Redemann of NASA.

The paper “Central American biomass burning smoke can increase tornado severity in the U.S.” can be found at: http://onlinelibrary.wiley.com/doi/10.1002/2014GL062826/abstract?campaign=wolacceptedarticle.

The research was funded by grants from NASA, U.S. Environmental Protection Agency, National Institutes of Health, National Oceanic and Atmospheric Administration, and the Fulbright-CONICYT scholarship program in Chile.


93 thoughts on “UI researchers link smoke from fires to tornado intensity

  1. Explain to me how the particles would increase wind shear? I thought wind was caused by either warm air rising and cold air rushing in, or the high pressure/low pressure difference (and those may be related somehow, but my physical science is a bit rusty). I can see how more particulates in the air might make a cloud heavier and therefore lower, but the wind shear thing doesn’t make sense to me.

  2. Why is it that I suddenly hear Billy Joel’s “We Didn’t Start the Fire”. Obviously the problem is not 1st world co2, but 3rd world slash and burn. What can stop that? Modern, gasoline/diesel powered equipment. Nuf said.

    • And here I was hearing Deep Purple – “smoke on the water, fire in the sky”
      Thank God for grant money, or we wouldn’t be able to enjoy all this simulation. Sadly, though, they actually take their effort seriously.
      I suppose someone got the idea to research smoke when they saw the smoke they blew out from the last inhale of the joint caused a “vortex” to form in the air(smoke ring) and said “wow, man, that’s like a flat tornado. You suppose smoke from Central America can intensify tornadoes? We gotta get us a grant and see if we can simulate that on a computer. Anyone got any more of this weed?”

  3. Did anyone measure the cloud base, and discern whether it was, in fact, significantly lower than other historical cloud base levels in otherwise similar storms?
    After all, scientists know that model predictions require verification in physical reality before they get assigned any meaning.

    • Pat Frank
      February 2, 2015 at 4:55 pm
      Did anyone measure the cloud base, and discern whether it was, in fact, significantly lower than other historical cloud base levels in otherwise similar storms?

      The same question entered my mind. Then I read:

      “We show the smoke influence for one tornado outbreak,…

      Then they write a paper based on “one tornado outbreak”, based on a model simulation. I’m not saying their conclusion is not right, I’d like to see OBSERVATIONS for future outbreaks, not simulations of past outbreaks. Just my 2 cents.

  4. All I know is, at this moment in time, it’s hard to keep focused on the comments while trying to keep my eyes from being drawn inexorably to the horror displayed in the sidebar.
    (Probably go to hell for saying that.)

      • Lest anyone think I was commenting about Oreskes’ looks, I don’t give a good damn about what she looks like. If she looked like Daryl Hannah, she would be equally repulsive. As a matter of fact, DH’s coziness to James Hansen and his grotesque political theater makes her repulsive, despite her looks.

    • Alan,
      I had the same reaction…. Perhaps it’s designed to create the internal reaction “What the hell is THAT?” and count on morbid curiosity to get the individual to ‘click on it’.
      Didn’t work, though. I don’t stop to look at dead animals along the highway either.

      • I came to my realization about this matter (comment at 6:07pm) after reflection and realized that the true turnoff about Naomi had nothing to do with any superficialities, such as appearance. If looks were what really mattered, none of my friends or associates would have any thing at all to do with me.

      • Alan,
        I didn’t recognize her at all. What I saw was a smirking and obviously well-to-do woman engaged in some sort of dumb protest over who-knows-what. I saw lots of that stupidity as a student at University of Wisconsin-Madison back in the early/mid ’80s, as I worked my way through a couple of engineering degrees. It has the same appeal to me as ‘road kill’.

  5. For more violent tornados you need more work to be done by the atmosphere. The work that can be done is proportional to the temperature difference. So unless pollution can increase the temperature differences in the system I don’t see how tornados can become more intense.

  6. I’ll take on the term “miscreant” gladly, if anyone would just set the rules of the game.
    We gonna play nice, or is it no holds barred ?

  7. The only thing that males any sense here is that man causes the smoke ergo man causes incresed tornado intensity. ATI. Anthropogenic Tornado Intensity. Alarm! Alarm!

    • So 2011 was the only year that they carried out slash and burn and caused worse tornadoes or was it the only year they looked at. I’m sure that there must be satellite DATA on the smoke from other years and plenty of DATA on tornadoes in other years so why rely on just one year and models unless they don’t make frightening reading?
      James Bull

  8. “They say the smoke lowered the base of the clouds”
    And the tops. And the breadth. And density. And timing of creation.
    Meh. This takes cherry picking to a whole new level. I think there is much more to cloud creation than CCN availability. So there is assumption built on assumption as a basis for this paper.

  9. This only tells us that particulate matter in the atmosphere may impact weather events. Something skeptics have been saying for a long while. In any event, the chances of this type of research providing any real insight into the formation and life of a totally chaotic system like a tornado is just about the square root of nothing,
    More useless busy work.

  10. It is time for them to go back, research tornado history, find the many examples of massive tornado outbreaks, and then tell us how many of these occurred during months when no burning was evident. Computer geeks have no interest in history!

  11. So, in 2011 we had 122 tornadoes, but back 1950 we also had a high number of tornadoes during “tornadoes season”. From article; “We show the smoke influence for one tornado outbreak,” (out of 122 tornadoes). All because of spring agricultural land-clearing fires in Central America.
    Question: Did the spring agricultural land-clearing fires in Central America stop sometime back in the 50’s and then start again in 2011? .

  12. ” In fact, the simulation including the smoke resulted in a lowered cloud base and greater wind shear.”
    Because that’s how it was programmed. Was there an actual lowering of the cloud base at the time and location of the tornadoes? No-one knows. And wind shear is not air moving at different velocities it is air moving in different directions; the velocities may or may not be different.
    From the map it looks like the Gulf of Mexico was engulfed in fires that spring. How odd!

      • The research was funded by grants from NASA, U.S. Environmental Protection Agency, National Institutes of Health, National Oceanic and Atmospheric Administration, and the Fulbright-CONICYT scholarship program in Chile

  13. I think, if there is a correlation, they totally missed the point.
    Water molecules form around airborne particles: pollen, dust, microbes and yes Soot! No new revelation there. Point is that with high temperature differences between the two air masses the storm intensifies. If more particles are introduced will the storm intensity increase? The height of the storm causes the intensity to increase, known fact. Will a few hundred feet down increase the intensity or does several thousand feet on top increase it more?
    The study seems to throw every fact known about thunderstorms out the window and insert a heaping helping of AGWBS

  14. That old canard comes to mind: “correlation does not equal causation”.
    Those researchers all have impressive titles… and it all escapes me, but I’m the dumbest guy that I know.

  15. OMG! Give anyone within reach a computer, two or more data sets and enough time, they will inevitably find co-relation. A quick wave of the hands and “Presto!”, it becomes causation. What’s not to like when you are shopping for funding?

  16. Okay. It’s plausible. But what is the MAGNITUDE of the effect? Important, trivial, negligible? The press release is silent. And did they find any real-world evidence to shore up the theory, like measured parameters of cloud bases, tops, liquid water contents, etc.? If not, it’s just children playing in a virtual sandbox. Reporting on model results without any real-world evidence should be frowned upon. Instead they will probably get funding to continue playing in their sandbox.

  17. You can read more about the extremely active severe weather during April/may 2011 at the link below.
    It includes the outbreak on April 27th that was supposedly worsened by smoke from Central America.
    Referring to that day:
    “The second wave developed as moisture surged out of the Gulf of Mexico, maximum daytime
    heating occurred, a cold front advanced eastward wind shear veered at different levels of the
    atmosphere and a very active jet stream was enhanced by an upper level disturbance all of
    which led to the atmosphere becoming extremely unstable and conducive for severe weather. The atmospheric conditions spawned numerous supercell thunderstorms and produced long lived violent tornadoes, many of which devastated the northern two thirds of central Alabama.
    In advance of the event, the SPC issued a rare High Risk warning area for the second consecutive day in
    parts of Mississippi, Alabama, Tennessee and Georgia”
    There was an upper level trough approaching and with an upper level jet stream on the order of 150 miles/hour from the southwest that day(with dry air aloft) and a strong low level jet from the south feeding in surface dew points above 70 degrees(from the Gulf of Mexico) and winds veering with height ahead of a powerful cold front, that featured a wave of low pressure riding along it that day.
    We had every element for violent tornadoes present to the max. Probably a big mesocyclone rotating from the base of the clouds to the mid levels too.
    I don’t see how smoke particles from Central America could have been a factor here. How much could this factor have lowered the base of the clouds and even if the cloud base was slightly lower, so what?
    Wind direction and strength would have likely been the same at the base of the clouds whether they were a slightly higher or lower, so wind sheer would not have changed.
    Smoke from Central America being present was just coincidental. I couldn’t read the paper, just the abstract.

    • I’m sure their computer simulation accounted for all natural variability – we can be sure it was the smoke.

  18. Plausible, maybe. I’m assuming they would be serious about tornado modeling in Iowa. The simulation needed to model a tornado is a different class of model than GCM’s. I assume a lot of data has been collected making the initial conditions less of an issue. Smoke particles could act to increase ice nucleation. What would that would do the tornado? I don’t know. They suggest further study is needed, you can hardly blame a researcher for that. They may have found a correlation that is validated by their model. I don’t know enough to reject this out of hand.

  19. I think this group of authors were simply trying to see what outlandish connections they could make in a scientific paper and see who would take the bait and publish it. In the process, could they get additional grant money to help the mechanical engineer lead author fund his studies or the chemical and bioengineering professor to fund his department. Slap the right words on the title and sprinkle in the text and Viola!!! – free money.

  20. Eight authors studying one, count’em ONE, tornado outbreak and a hypothetical accelerant from smoke from Yucatan.
    Gee, was that the only source of soot in the atmosphere all decade? Western states forest fires have little or negative effect on tornado outbreaks?
    Willis’s rule of thumb hypothesis about the importance of the paper is correlated with inverse square of the number of authors gets another data point of confirmation.

  21. Why do I have this sneaking suspicion that they decided what they wanted to find, then went inside and teased the computer with a cattle prod until it came up with the answer they were looking for?
    Was it the smoke from Central America? Or was it smokes from Mexico? I have another sneaking suspicion that a follow-up paper will report on the effect of hashish brownies on sensations of spinning counterclockwise in the northern hemisphere.

  22. I see lawsuits ahead now, for anyone who burns something around the time a tornado forms…no more campfires or fireplaces….wood stoves, etc.

  23. This whole Climate Change and the thousands (literally) of models, papers, studies, etc, etc are becoming nauseating, if any of these researchers could ever come out with a viable result they have now undermined their own outlook to the point nobody pays any attention or they are ridiculed and in a way that is a sad, sad description of our society. Is there a way out of this mess? Anyone? Can’t we get to some consensus on this? Can’t anyone say they were wrong and try to fix this? or am I being just naive?

  24. Paper co-authors are Scott Spak ofthe UI Departments of Urban and Regional Planning and Civil and Environmental Engineering; Bradley Pierce and Andrew Heidinger of National Oceanic and Atmospheric Administration Satellite and Information Service Center for Satellite Applications and Research; Jason Otkin and Todd Schaack of the Cooperative Institute for Meteorological Satellite Studies, University of Wisconsin-Madison; Arlindo da Silva of NASA Goddard Space Flight Center; and Meloë Kacenelenbogen and Jens Redemann of NASA.
    The research was funded by grants from NASA, U.S. Environmental Protection Agency, National Institutes of Health, National Oceanic and Atmospheric Administration, and the Fulbright-CONICYT scholarship program in Chile.

    Follow the money. (The Big Government institutions are going to pay (beforehand) ONLY for “research” that promises to create the results that Big Government needs. Or wants. Big Government bureaucrats are going to authorize their underliars (er, under-liens) to follow up research time and only apply for grants that will be successful – ei, create bigger budgewts and more funded research.)
    Follow the power. (This enables the Obama White House and its minions to continue their CAGW power and control.)
    Follow the paper. (Publish or perish. If a Big Government “researcher” or “scientist” has NOT proposed and edited and published papers that will GET APPROVED by the editors and his anonymous peer-reviewed Star Panel, she dies. )

    • Yes I did and followed “the money” and that is what made me comment the way I did. Orwell did write and predicted all of this as did many others, we should be learning, history repeating itself, as I said nauseating.

  25. I don’t suppose they used water to put the fire out .( sarcasm off ) How much water evaporated and mixed with the smoke and soot ? And was salt water used ?

  26. I can see how adding particulates to a nascent thunderstorm would affect it’s development. For a start, more particulates mean more phase change of water vapour into liquid droplets – I can see how that could accelerate the growing of a pressure difference. It seems plausible.
    Look, one storm is not going to be conclusive or even quantify the effect.
    But this does seem like an interesting first step.
    Give they guys a break.

  27. Anyone who has ever flown over Mexico would know that fires are constantly burning. Every where you go around the country side there is the palor of smoke hanging in the air. Every single day, unless there is a strong wind, the smoke from fires is present.
    This has nothing, NOTHING, to do with agricultural. It is the burning of their trash heaps, which you can see behind every single shanty and house you travel past.

  28. I can see where its certainly possible for smoke to intensify a thunderstorm. Modeling this scenario would require true knowledge of cloud mechanisms and its been said repeatedly in these forums that cloud mechanisms cannot be simulated in climate models. If clouds cannot be modeled, how did they do it in this study and why can’t they do it in GCM’s. You would have to make more assumptions than my mother-in-law.

  29. The two largest tornado outbreaks were in 1974 and 2011. One during the global cooling scare and the other during the global warming scare. The one thing that both 1974 and 2011 have in common were strong La Nina’s. Data also shows that the number of violent tornadoes trended downward after the mid 1970s, the end of the natural cooling period.
    I have attended the National Weather Service’s Skywarn class multiple times for storm spotting and weather nets on ham radio. It teaches that there are three things needed for severe weather to form. Cold air aloft, warm moist air below and wind shear.
    To this layman it is obvious that cold is the key to severe weather. Those strong La Nina’s kept cold fronts stronger (cooler) as they moved across the US to collide with warm moist tropical air. Once we entered the warming half of the natural cycle the temperature difference wasn’t as great, violent tornadoes trended down.
    I just don’t see how they can link smoke to tornado strength unless they can collect data from 1974 as a comparison.

  30. Here in Oklahoma, the heart of tornado alley, 2014 was a bad year for grass fires. Lots of homes destroyed. Nearly every day during spring you could smell the soot from the fires. 2014 was also notable for record low number of tornadoes in the state.

    • Ranchers now make a common practice of burning pastures in late Winter/early Spring. They’ve let the fires get away from them at times and have caused real damage. An incident a few years ago in my native Osage County prairies was the subject of much mirth, as local people joked about certain ranchers who’d burned more than they should have and then whined to the Feds to get disaster relief for loss of grazing pastures, when apart from a few dead steers trapped at a few burned fence sections, the fire- renewed pastures were the goal, anyway.
      Last year’s low soil moisture levels didn’t help matters, nor did arsonous firebugs, but I think fewer homes were burned down last year than in some previous years.
      Texas really had a problem with burned homes, not only because of even worse drought conditions than here in OK, but because of so much building encroachment into areas with accumulated natural tinder, resulting in greater fire loss, as has been typical in Western fires of recent years.

    • If last years grass fires had anything to do with our low number of tornadoes, then burn (carefully,) baby burn.

  31. We should start a petition to ban environmental assault matches!
    (Or least limit the number of matches in a box to 10.)

  32. John Isaacs once suggested that driving on the right side of the road increases tornado numbers and intensity–I think that suggestion went exactly nowhere, but he had observations on his side since they drive on the left in England, and England has fewer tornadoes than do we.
    Smoke particles are hydrphobic, and maybe they suppress cloud formation and storminess (reduce tornado number and intensity) or maybe by suppressing cloudiness they allow the atmosphere to become more unstable (thus increasing tornado number and intensity). Playing science by just modeling and making more hypotheses gets one exactly nowhere. Please, make some testable predictions, and allow a credible experimental test.

  33. such claims are NONSENSE.
    shore there’ll be more of that.
    – When there’s LOCALLY concentrated heavy pollution
    + continent wide, distribution of condensation / crystallisation aerosols is restricticted / EPA /:
    local WEATHER can get accellarated, triggered by that local concentrations.
    1. every kind of WEATHER events gets accellerated by LOCAL concentrations of pollutants
    2. since those aggregations of pollutants / versus continent wide, time spread, distribution /, stay sparse
    that leaves not much to tell for the CAGW believers:
    see: 1 event in 2012 reported in 2015.
    three years later: another local weather phenomenon, legend, to be pushed?
    Regards – Hans

  34. So how does this compare and contrast to the 2005 massive fires in Guatemala and Mexico?
    Duplication of results would be invaluable as well as accounting for air vapor and ocean currents.

Comments are closed.