Climate change and severe tornado outbreaks

[Casuarina Head]

Severe Tornado Outbreaks
defined as outbreaks with:
≥ 2 long-tracking tornado families
each family w/ ≥ 2 (E)F3+ tornadoes
each of these w/ path lengths ≥ 25 mi
these families over wide geographic area

Examples:
19–20 February 1884
28 March 1890
29–30 April 1909
28 March 1920
9 May 1927
21–22 March 1932
16 March 1942
21–22 March 1952
2–3 April 1956
11–12 April 1965
3–4 April 1974
5–6 February 2008
27 April 2011

I think that the overall warming of the Pacific basin over the past decade, coupled with recent -AMO trends (weakening AMOC) in the North Atlantic, has made low-amplitude, progressive patterns less likely, even during -ENSO/-PDO setups, hence the dearth of major, widespread tornado outbreaks, for the most part, since 2011–12, especially over the Plains. So events like Palm Sunday 1920/‘65, Super Outbreaks ‘74/‘11, Super Tuesday ‘08, et al. will be harder to come by for the most part, due to AGW being reflected in the absorption of heat and fresh water by the major ocean basins, especially the Pacific and the North Atlantic.

Source

Climate change is leading to forces that make widespread, violent tornado activity—that is, outbreaks with multiple long-tracking supercell tornado families over a wide geographic area—less likely than it was prior to 1965 and especially prior to 2013. Certain models, including mesoscale such as the NAM/HRRR, still struggle to account for climate change’s impact on wavelengths vis-à-vis the warmer Pacific and the weaker AMOC signal in the Atlantic. Significant tornado outbreaks can still occur, but they are becoming less frequent, more limited in aerial coverage, and more confined to one (or two) big supercell that thrives more due to mesoscale quirks than large-scale synoptic factors vs. the past.

Source

 

[andyhb]

Why exactly are we talking about mesoscale models capturing climate signals here? There is a lot of basic meteorology/climatology that is being ignored. Number one that climate forcings are on entirely different time and spatial scales than models like the HRRR and NAM (or even the GFS and Euro) are meant to forecast for.

There’s a reason we use GCMs instead of operational NWP to forecast changes in the climate. Do climate signals feedback onto the shorter term? Absolutely, but operational NWP is effectively separate/independent from that because the initial conditions and length of the forecasts are already within the current climate regime.

Also, until you can show me proof that severe weather outbreaks are becoming less common with time, this is a lot of conjecture. As someone who has studied this stuff for years and read papers on it, I can tell you that the conclusions you draw here are at best conflicted in the scientific literature and at worst disproven by the scientific literature.

 

[Casuarina Head]

@CheeselandSkies @pohnpei Thoughts?

 

[pohnpei]

I was focus on this subject for such a long time and used to read literally any paper or findings that I can find. But I still find myself know so little about this topic. I may have many ideas want to share about but I also don't know how to start it or what kinds of conclusion I can make.
So, first I want to talk about the trend we have already know or quite sure about. That is even with the the development of detection technology, the trend of fewer and fewer tornado days every year was apparent. The reason is that as jet retreat early and early because of the golbal warming, tornado activity in summer was reducing rapidly through last 30 years. This conclusion was already back up by either obsevation or statistics calcuation or climate model. Also I think Grazulis noticed this change long time ago.
Some finds about this change:https://www.sciencedaily.com/releases/2018/08/180806104253.htm
But I know the most important question must be how about the change of tornado activity in spring? Even in relatively cold era, spring was the most prolific time to produce tornados, especially violent tornados.


I don't know if you have ever watched this, but Dr.harold had some insightful ideas about this topic. In his model, it seems that the tornado activity was largely increasing in spring. But I will also recommand you to read this paper last year:https://journals.ametsoc.org/view/journals/bams/aop/bamsD200004/bamsD200004.xml
This suggests that despite increasing instability, thunderstorms in a warming climate may be less likely to develop due to stronger convective inhibition and lower relative humidity. (also higher and higher LCL)
So they may give a hint to a opinion that fewer and fewer tornado days due to higher and higher CIN but more violent tornado if stroms succeed to fire up.




Another great paper I recommand to read and I believe you must already read before was this one:

Spatial trends in United States tornado frequency - npj Climate and Atmospheric Science

While the number of tornadoes has remained fairly static in the United States over the past 40 years, strong geographic contrasts are apparent. Tornadoes have the potential to cause severe damage, yet understanding their changes in time—particularly the impact of anthropogenic warming—has been...

www.nature.com

The spatial trends in tornado environments and tornado frequency showed in this findings was obvious, but something more interesting was this statistics they made. It seems that the Annual accumulated STP had quite good correlation with the annual tornado counts.




Also some interesting trend about 500 Height and EF2/3 tornado numbers :


We know that global warming increase the 500 Height as time goes by.

After reading all those papers, I think we still can't make any certain conclusion about this topic. One thing that have been discussed these days in the March 17 severe weather trend was did high ceiling event bust more often now? It seems to be true in last few years and some of the bust of high ceiling conditions even find hard to figure out the reason of it. But was there really a clear statistics to show that we bust more often than before? I think the answer is no. SPC used to issue high risk more frequently. For example, 1990 and 1991 were all very active year, but did anyone remember what happened on 1990/2/1, 1990/5/9 1990/5/29, 1991/3/28, 1991/4/28? These were all high risk days and they all bust very hard. People don't seems to remember those bust 20 or 30 years ago. Also, what contribuate to the decresing high risk days over last 30 years? It seems that the criterion of the high risk day was largely on rise.
So there was many uncertainties about this topic. Like many people said before, the next super outbreak can happen 1000 years later or it can happen in next month. We‘ll ’never know, so keep you mind open as much as possible. Hope the answer is helpful to you.

 

[thundersnow]

For years, the thought was that climate change/global warming increases volatile/severe weather. Previous decades showing upticks in severe weather were supposedly bearing that out. Now that trends have gone quieter for a few years, it's climate change that's lessening it?

This shouldn't come across as being skeptical or outright denying climate change, but I think climate change has now become the perennial go-to scapegoat for just about ever weather occurrence... when really so much can be explained by cycles, and just the right ingredients coming together at the right time.

 

[warneagle]

For years, the thought was that climate change/global warming increases volatile/severe weather. Previous decades showing upticks in severe weather were supposedly bearing that out. Now that trends have gone quieter for a few years, it's climate change that's lessening it?

This shouldn't come across as being skeptical or outright denying climate change, but I think climate change has now become the perennial go-to scapegoat for just about ever weather occurrence... when really so much can be explained by cycles, and just the right ingredients coming together at the right time.

I think the issue is that it's difficult to suss out exactly how large-scale processes like climate change will affect small-scale events like tornadoes. It's relatively easy to figure out the deterministic and stochastic effects of climate change on bigger features (e.g. tropical cyclones and winter storms) but much harder to figure out whether there are deterministic effects on things like tornadoes which are more random, which also means that it's easier to attribute trends to climate change for those bigger processes than it is for tornadoes.

We might be able to draw some general conclusions about whether climate change makes the overall climatology and synoptic patterns more or less favorable for tornadoes (and even then it might be tough to separate what's resulting from anthropogenic climate change vs. the natural changes in the background state), but I don't think we can really say whether certain things that evolve on the mesoscale are being driven or altered by climate change at this point, especially since our understanding of the process of tornadogenesis is still incomplete. There are just too many confounding variables to draw firm conclusions at this point, imo.

 

[Casuarina Head]

The recent spate of inactivity could certainly lend some credence to this thread.

 

[Casuarina Head]

My take on climate change:
*Fears of a warmer climate are politically driven, not based on science.
*Despite this, all evidence points to ongoing, significant global warming.
*Historically, warmer climates lead to less storminess, better conditions for civilisation, and population growth.
*Decreases in tornado and hurricane activity are consistent with changes in the oceanic circulation that reflect global warming, not cooling.
*Claims that large-scale cooling or even a mini ice-age is about to commence have no solid basis in actual observations.







Thomas P. Grazulis’ view and my own seem to converge in regard to the impacts of climate change on the frequency of significant outbreaks.

 

[Casuarina Head]

Haven’t you been a broken record about this for the past couple years?

The events you are describing are extremely infrequent to begin with. Trying to ascribe one single reason why it’s been a few years since the last “super outbreak” is ridiculous.

I already defined significant outbreaks at the top of this thread. I wasn’t necessarily referring to “super” outbreaks exclusively.

It is ridiculous. Super outbreaks are often generational in nature, so guess we will know the answer in about 2045.

See above.

According to Grazulis, Easter can ranked No.7 of all outbreaks since 1950 and the overall sig tornado paths of Dec10 was even higher than Easter, even higher than 5/31/1985. If outbreaks of these magnitude can't satisfied your definition of truly major outbreak and was classified as geographicly restricted then how many outbreaks in history meet with your definition in the history? I can only assume that years like 2011 exerted a wrong impression on you.

I include data that go back to 1875, per Significant Tornadoes. The sample since 1950 is a bit limited.



The above shows a clear downward trend in the frequency of significant events over time.

 

[pohnpei]

I already defined significant outbreaks at the top of this thread. I wasn’t necessarily referring to “super” outbreaks exclusively.


See above.


I include data that go back to 1875, per Significant Tornadoes. The sample since 1950 is a bit limited.



The above shows a clear downward trend in the frequency of significant events over time

I think your criteria is subjective at best and I can't see how Easter and Dec10 can't meet with the criteria from any perspective and it can surpass some of the examples you list like Super Tuesday from many perspective.

I used to say some of your observation seems reasonable but it isn't a simple, one-dimensional trend like you assume. One trend that did exist and probably few can debate is the tornado activity during summer season, especially June was not very active these years. But it actually can be related to a overall seasonal shift of the tornado season. With the less activity in June or July, was second season being more active and having more significant outbreaks in recent years? Outbreaks like 12/23/15, 12/16/19, 12/15/21 can all ranked pretty high in December's history, let alone the unprecedented 12/10/21 outbreak. 12/1/18 and 12/12-14/22 was also significant as well. Was this also affected by climate change?

And tropical cyclone activity is a different story and no observed trend can be noticed now in North Atlantic. What you can argue was obviously less activity in basin like south Pacific. But did this trend affected by climate change? Not a simple question here.

 

[Casuarina Head]

Also, until you can show me proof that severe weather outbreaks are becoming less common with time, this is a lot of conjecture. As someone who has studied this stuff for years and read papers on it, I can tell you that the conclusions you draw here are at best conflicted in the scientific literature and at worst disproven by the scientific literature.

In this case I would rely on sound meteorological sense that is based on an understanding of the dynamics at play. For instance, it is known that enhanced EF2+ tornado activity is correlated with a +AMOC and suppressed subtropical ridging, while a -AMOC (=melting Greenland ice → decreased salinity in the North Atlantic) and rising mid-latitude heights are correlated with global warming. Also, a decreased temperature gradient between the tropics and Arctic would tend to result in split rather than progressive flow, which would potentially decrease the incidence of large-scale outbreaks. Of course, climate change could also shift the main tornado season from the springtime to the wintertime, as @pohnpei mentioned, while leading to a decrease in the overall incidence of major outbreaks over time. Finally, the “scientific literature” that you mention is often unduly politicised by external forces such as the IPCC.

I think your criteria is subjective at best and I can't see how Easter and Dec10 can't meet with the criteria from any perspective and it can surpass some of the examples you list like Super Tuesday from many perspective.

I will concede that these events met the criteria, but my point about these kinds of outbreaks occurring less frequently overall still stands.

I used to say some of your observation seems reasonable but it isn't a simple, one-dimensional trend like you assume. One trend that did exist and probably few can debate is the tornado activity during summer season, especially June was not very active these years. But it actually can be related to a overall seasonal shift of the tornado season. With the less activity in June or July, was second season being more active and having more significant outbreaks in recent years? Outbreaks like 12/23/15, 12/16/19, 12/15/21 can all ranked pretty high in December's history, let alone the unprecedented 12/10/21 outbreak. 12/1/18 and 12/12-14/22 was also significant as well. Was this also affected by climate change?

I think so. Climate change could certainly shift the main tornado season into the winter rather than spring.

And tropical cyclone activity is a different story and no observed trend can be noticed now in North Atlantic. What you can argue was obviously less activity in basin like south Pacific. But did this trend affected by climate change? Not a simple question here.

The trend toward a -AMOC since 2012, along with the persistent Newfoundland warm pool, can be best explained by freshwater runoff from Greenland due to CC.

 

[andyhb]

In this case I would rely on sound meteorological sense that is based on an understanding of the dynamics at play. For instance, it is known that enhanced EF2+ tornado activity is correlated with a +AMOC and suppressed subtropical ridging, while a -AMOC (=melting Greenland ice → decreased salinity in the North Atlantic) and rising mid-latitude heights are correlated with global warming. Also, a decreased temperature gradient between the tropics and Arctic would tend to result in split rather than progressive flow, which would potentially decrease the incidence of large-scale outbreaks. Of course, climate change could also shift the main tornado season from the springtime to the wintertime, as @pohnpei mentioned, while leading to a decrease in the overall incidence of major outbreaks over time. Finally, the “scientific literature” that you mention is often unduly politicised by external forces such as the IPCC.


I will concede that these events met the criteria, but my point about these kinds of outbreaks occurring less frequently overall still stands.


I think so. Climate change could certainly shift the main tornado season into the winter rather than spring.


The trend toward a -AMOC since 2012, along with the persistent Newfoundland warm pool, can be best explained by freshwater runoff from Greenland due to CC.

1. Where is it shown that EF2+ tornadoes are correlated with +AMOC? If this research is coming from AOML, I would be very suspicious of it, and I say this as a member of a research group that is attempting to work with them.

2. Split rather than progressive flow? Split flow can also be progressive. I believe the word you're looking for is blocking.

3. If you're going to claim that the literature is overly politicized, then what are your claims such as the tornado/AMOC connection derived from?

4. For Grazulis' plot, it's hard to derive any sort of trend there when tornado ratings have changed so much since the 1970s.

 

[Casuarina Head]

1. Where is it shown that EF2+ tornadoes are correlated with +AMOC? If this research is coming from AOML, I would be very suspicious of it, and I say this as a member of a research group that is attempting to work with them.

I actually can’t recall the source, but I do remember reading about this kind of correlation. Maybe it had to do with the AMO, which is tied to the AMOC.

2. Split rather than progressive flow? Split flow can also be progressive. I believe the word you're looking for is blocking.

Yes, blocking.

3. If you're going to claim that the literature is overly politicized, then what are your claims such as the tornado/AMOC connection derived from?

See above.

4. For Grazulis' plot, it's hard to derive any sort of trend there when tornado ratings have changed so much since the 1970s.

Given that many ratings were overestimated prior to the 1980s, I would admit that this is a fair argument. Still, I think that Thomas P. Grazulis is using his own reanalysis for pre-1980s tornadoes in the above chart, so I think his estimates are more accurate than the official data.

 

[Sawmaster]

I don't think we've got enough data to conclude anything just yet. Tornadoes and moreso outbreaks don't follow strict rules, and with the many variables involved you're bound to find something which correlates if you're looking for that.

The best research results come from looking for nothing while trying to see everything- that way you won't be misled into errors.

 

[Casuarina Head]

It does seem to be that "transitory" state that really gets things cooking, which may have been the "missing ingredient" so to speak over the course of these last few Nina years. Again we've had significant events here and there and this no way lessens the impact of what has occurred. Of course 2011 skewed perceptions of what the bar for a high-end event/season is, but even without that it feels like it's been a long time since a tornado season produced like some years in the 1990s and 2000s. I don't seem to be alone in this which is why we have seen posts/threads asking if there's something on the larger climatic scale that is reducing the chances of the already-rare higher end outbreaks.

The bolded section is precisely my point. Even portions of the 1970s and ‘80s have been more intense than the post-2012 period.

 

[Casuarina Head]

I don't think we've got enough data to conclude anything just yet. Tornadoes and moreso outbreaks don't follow strict rules, and with the many variables involved you're bound to find something which correlates if you're looking for that.

The best research results come from looking for nothing while trying to see everything- that way you won't be misled into errors.

Personally, I think that people have been brainwashed into ignoring certain effects of climate change. The problem is that the Establishment’s narrative is that global warming supposedly leads to more intense, frequent tornadoes or outbreaks, tropical cyclones, etc. When evidence comes along to indicate the opposite, people either a) start to claim, “See, global warming is a hoax!” or b) continue to support the Establishment’s narrative, in part by deploying excuses such as: “You’re being premature, you’re not an expert, etc.” I think there is already a lot of evidence that a warming climate is leading to less frequent high-end tornado outbreaks and weaker Atlantic hurricane seasons through various feedbacks, e.g., a warmer Pacific basin (the same factor that has promoted decadal drought over the West recently).

 

[tennessee storm chaser]

Personally, I think that people have been brainwashed into ignoring certain effects of climate change. The problem is that the Establishment’s narrative is that global warming supposedly leads to more intense, frequent tornadoes or outbreaks, tropical cyclones, etc. When evidence comes along to indicate the opposite, people either a) start to claim, “See, global warming is a hoax!” or b) continue to support the Establishment’s narrative, in part by deploying excuses such as: “You’re being premature, you’re not an expert, etc.” I think there is already a lot of evidence that a warming climate is leading to less frequent high-end tornado outbreaks and weaker Atlantic hurricane seasons through various feedbacks, e.g., a warmer Pacific basin (the same factor that has promoted decadal drought over the West recently).

I agree some with sawmaster, tornado outbreaks don’t go by certain rules n regulations … just matter time we see a high end tornado event .personally with a climate changing to warmer see more adverse extreme events eventually … major hurricanes , tornado outbreaks to even more snowstorms.

 

[Casuarina Head]

I agree some with sawmaster, tornado outbreaks don’t go by certain rules n regulations … just matter time we see a high end tornado event .personally with a climate changing to warmer see more adverse extreme events eventually … major hurricanes , tornado outbreaks to even more snowstorms.

An isolated event doesn’t, but collectively episodes run in cycles that are clearly modulated by large-scale climatic factors. The 1960s and earlier were active, the ‘70s and ‘80s were tame, and the 1990s–early 2000s were active. Since 2012 or so we have entered a downturn that can be linked to climatic factors. Climate change is cyclical, after all. At this stage no one can really credibly argue that we have not entered a decade-long downturn. Your statement about climate also isn‘t really true, as warm periods historically have been associated with tamer weather, as was the case during the Medieval Warm Period.

 

[warneagle]

Personally, I think that people have been brainwashed into ignoring certain effects of climate change. The problem is that the Establishment’s narrative is that global warming supposedly leads to more intense, frequent tornadoes or outbreaks, tropical cyclones, etc. When evidence comes along to indicate the opposite, people either a) start to claim, “See, global warming is a hoax!” or b) continue to support the Establishment’s narrative, in part by deploying excuses such as: “You’re being premature, you’re not an expert, etc.” I think there is already a lot of evidence that a warming climate is leading to less frequent high-end tornado outbreaks and weaker Atlantic hurricane seasons through various feedbacks, e.g., a warmer Pacific basin (the same factor that has promoted decadal drought over the West recently).

I don't think we have enough data to say that climate change has a deterministic effect on tornadoes or tornado outbreaks. Things like tropical cyclones are more straightforward since they're large-scale atmospheric processes and the deterministic effects (warmer SSTs = more intense TCs) are pretty obvious. Tornadoes are small-scale processes with a lot more variables to control for, so it's harder to demonstrate deterministic effects empirically, much less the kind of stochastic effects that are likely to be more significant in tornado events than tropical cyclones.

This is further complicated by the differences in data-gathering processes between tropical cyclones (which we can measure directly in a consistent way) and tornadoes (which we measure based on estimates that aren't always applied consistently). The fact that accurate data on the number/intensity of tornadoes only goes back about 50 years is also a confounding problem, since that roughly coincides with the last negative global temperature anomalies in the late 1970s. I'd be curious to know what you consider a "high end" tornado outbreak, because I'm skeptical that there's actually been a consistent downward trend in those events. Obviously the last couple of spring seasons in the plains have been quiet, but there have been substantial events in the south during that period and some potent plains setups (e.g. 20 May 2019) that have underperformed for reasons that I think would be difficult to empirically link to climate change. I'm not saying it's not a possibility, but I don't think we have enough data to say that there's a causal relationship--assuming there's even an effect, which I'm not totally convinced of.

I'm also not sure what you mean about "weaker" Atlantic hurricane seasons; six of the top ten seasons for total accumulated ACE have occurred within the last 27 years, along with four of the top ten individual hurricanes in terms of ACE. In that timespan, we've only had six below average seasons (none since 2015) vs. 17 above average seasons, 10 of which were highly active; last year was the first "near normal" season since 2015. Obviously there's some recency bias in that data as well since we're undoubtedly missing some storms from before the satellite era, but I don't know how you look at that data and come to the conclusion that Atlantic hurricane seasons are getting weaker. You would expect hurricanes to get more intense with warming SSTs, all other things being equal.

 

[TH2002]

Personally, I think that people have been brainwashed into ignoring certain effects of climate change. The problem is that the Establishment’s narrative is that global warming supposedly leads to more intense, frequent tornadoes or outbreaks, tropical cyclones, etc. When evidence comes along to indicate the opposite, people either a) start to claim, “See, global warming is a hoax!” or b) continue to support the Establishment’s narrative, in part by deploying excuses such as: “You’re being premature, you’re not an expert, etc.” I think there is already a lot of evidence that a warming climate is leading to less frequent high-end tornado outbreaks and weaker Atlantic hurricane seasons through various feedbacks, e.g., a warmer Pacific basin (the same factor that has promoted decadal drought over the West recently).

I honestly agree with you to a point... but it's really not unfair to say that quickly concluding global climate change is causing a significant downturn in tornado outbreaks is, well, premature.

Do Grazulis' charts show a clear downtrend in tornado activity since 2013? Absolutely. There may be evidence that a warming climate is leading to less frequent intense tornado outbreaks. But evidence is not proof, and data that spans less than a decade isn't anywhere near enough to conclusively prove your hypothesis (especially since many tornadoes went undocumented up to the 1990's or so).