Summary: It was almost impossible for the temperatures seen recently in the Pacific North West heatwave to have occurred without global warming. And only improbable with it.
It’s been clear for at least a decade that global warming has been in general increasing the intensity of heat waves, with clear trends in observed maximum temperatures that match what climate models have been predicting. For the specific situation in the Pacific NorthWest at the end of June, we now have the first attribution analysis from the World Weather Attribution group – a consortium of climate experts from around the world working on extreme event attribution. Their preprint (Philip et al.) is available here.
In the paper, they show that this event was truly exceptional in the temperature statistics in the region and specifically in Vancouver, Seattle and Portland, but that the geopotential height anomaly (one measure of the ‘heat dome’ or the ridging) was not that far off from the underlying trends. Using various methods to construct the return time for this event, they show that even in today’s climate this was a one in 400 yrs or longer event. Using only data from before 2021, an anomaly this large is estimated to never occur!
Everyone is agreed that the specific synoptic situation is unusual. A large ‘omega’ pattern (so-called because of the resemblance to the Greek letter ) set up by a Rossby wave breaking event, intersecting with the topography and the warm temperatures to the south-west that were advected to the PNW. But the question is whether the temperature extremes are being made substantially more likely by underlying climate changes.
First, look the maximum annual daily temperature in the region (according to the ERA5 reanalysis for the historical data and the actual weather forecast analyses since June 1st). There is a real trend of about 4ºC over the last 70 years – roughly 3 to 4 times the trend in the global mean temperature. Nonetheless, the magnitude of the regional anomaly is more than 5ºC above the previous record. That is, literally, phenomenal.
The temperature anomaly at the local station level is similarly huge at SeaTac airport, Portland International Airport and New Westminster (nr. Vancouver). Note that while the cities might be affected by urban heat island effects that would exacerbate the temperature signal, that would not affect the regional analysis above, nor the situation in village of Lytton, BC which set a massive new all-time Canada-wide temperature record on Tuesday June 29th and promptly burnt down a day later.
However, the situation is a little different if you look at the geopotential height anomalies – these are affected by the synoptic situation as well as the integrated temperature anomalies. In that case, while still record-breaking, the anomaly is not totally beyond expectations. Indeed, the trend in z500 values is similar to the situation in Western Europe last year.
Together these analyses suggest a synoptic situation that is rare, but not inconceivable, but with temperature anomalies that are off the charts.
The way that attribution for extreme events works (as discussed previously on RealClimate here and here etc.) is that you look at the situation with and without the anthropogenic global warming signal and calculate the ratio of probabilities. If an event is say, twice as common with the GW, then one can give a fractional attribution of 50% to anthropogenic forcing and the return time is half what it used to be. If it is five times more likely, the attribution is 80% = 100*(5-1)/5 and the return time is a fifth of what it used to be. In this case, we are seeing probability ratios of 150 to 1000s, suggesting that these, improbable, temperatures can be almost entirely attributed to global warming. Without the anthropogenic signal, temperatures this extreme wouldn’t have happened in thousands to tens of thousands of years.
Rainfall and soil moisture deficits as a precursor?
In many previous extreme heat events, such as the 2003 European heatwave, rainfall deficits and dry soils the prior spring were shown to have made an important contribution to the temperature extremes, and so it’s worth looking at the same phenomena here. The IMERG data which are mostly based on satellite rainfall amounts do show a moderate deficit in the area over the last four months, but not so much of an effect that it could explain the anomaly on it’s own. The magnitude of this effect will be examined further in the months to come.
All models are wrong?
This kind of attribution is of course only as good as the models being used. In such a rapid attribution study, that means that the authors depend on an existing database – in this case, from CMIP5 and CMIP6 – and while they screen the models for fidelity in matching this genre of event, it’s possible that there are systematic issues with this class of model for a specific aspect of the situation. For instance, Mann et al., (2018) find that the CMIP5 models have a poor representation of a quasi-resonant (QR) phenomena in jet stream waves that are associated with the ‘omega’ pattern blocking event seen here. [Update: the specific claim in the paper relates to oscillations with wavenumber 6-8, while this event was more of a wavenumber 4 phenomena – see comment #9 below]. The expected trends in QR suggest an increase of about 30% in such events today over the situation in the pre-industrial. If models don’t capture this behaviour, it will make the event seem more unlikely than it really is. This might be resolved in higher resolution modeling specific to this event, but doesn’t really affect the broader conclusions.
Maybe it was just really, really, really unlikely?
Some people still reject these lines of argument, typical of this is Cliff Mass in this recent blog post. For them, the trends in max temperatures are (literally) ignored, and the fact that this phenomenon is being seen around the world is just a series of increasing unlikely combinations of factors that for some inexplicable reason keep happening. But this is really just a case of synoptic myopia – paying too much close attention to the series of specific events that lead to the specific situation, and not seeing the wood for the (burning) trees.
p.s. (8 July): In 2012 we published the highly relevant post Extremely hot, starting like this:
One claim frequently heard regarding extreme heat waves goes something like this: ”Since this heat wave broke the previous record by 5 °C, global warming can’t have much to do with it since that has been only 1 °C over the 20th century”. Here we explain why we find this logic doubly flawed.
Pretty much exactly what happened! And it ends thus:
So in summary: even in the most simple, linear case of a shift in the normal distribution, the probability for “outlandish” heat records increases greatly due to global warming. But the more outlandish a record is, the more would we suspect that non-linear feedbacks are at play – which could increase their likelihood even more.
- M.E. Mann, S. Rahmstorf, K. Kornhuber, B.A. Steinman, S.K. Miller, S. Petri, and D. Coumou, "Projected changes in persistent extreme summer weather events: The role of quasi-resonant amplification", Science Advances, vol. 4, 2018. http://dx.doi.org/10.1126/sciadv.aat3272
222 Responses to "Rapid attribution of PNW heatwave"
Jim Galasyn says
The expected Cliff Mass blog post:
Was Global Warming The Cause of the Great Northwest Heatwave? Science Says No.
Adam Lea says
36: “What’s causing all this heat? ”
I would guess primarily the synoptic setup, blocking highs positioned in places to persistently advect warm air north, then prolonged sunny days allow the temperature to keep rising and rising. AGW will act to boost that heat. In other parts of the northern hemisphere, cool air is being advected south. SE England, for example, has experienced a cloudy cool wet May, a very wet, cloudy and cool second half of June, and the first week of July has been cloudy and wet, with temperatures not even reaching 20C (in other words the summer here so far has been poor from the perspective of someone who enjoys outdoor activities). I think the effect of AGW is not just due to the increase in temperature from the general trend, it is increasing the frequency of blocking which causes stagnated weather patterns. It is these stagnated weather patterns that lead to monthly extremes both in temperature and rainfall. In the UK this seems to have become more common in the last decade or so.
Adam Lea says
38: “Was Global Warming the cause of the Great Northern Heatwave?”
I feel questions like these are the wrong questions to ask, as they don’t mathematically make sense. Global warming doesn’t “cause” anything directly. Global warming is related to climate change, and climate is the statistics of weather, so the question should be: Did Global Warming make the Great Northern Heatwave more likely? People shouldn’t be looking for deterministic answers from a probabilistic situation.
Clifford Mass says
This kind of ad hominem stuff is not professional (you did it before with the snowpack issue….and you folks were proved wrong on that!). Synoptic myopia? What is that or than name calling. Can you find any errors in my science or technical analysis? You suggest I ignored the trends in max temps? Nonsense….I have a figure showing the trend in record breaking max temps. Did you even bother to read my blog? The rushed attribution analysis you cite has major flaws and the internal text contradicts the headlines. Whoever wrote this should sit down with me and talk science and keep away from the name calling…cliff mass
[Response: Hi Cliff, might I suggest that instead of getting hot under the collar, you actually engage with these arguments? As for synoptic myopia, it’s inspired by comments like this: where despite there being an obvious trend in extreme temps you simply declare that it doesn’t exist with no calculation at all! Literally you appear to not be able to see what is right in front of you. Stay cool! – gavin ]
John Pollack says
I think that attribution of the PNW heat wave to global warming is more clearly established by taking a larger scale look at the event, in both space and time.
The exceptionally intense thermal ridge aloft which facilitated the PNW heat wave was building for several weeks beforehand. Although the position shifted, it was anchored in the severe to extreme drought area that covered much of the Southwest U.S., with an extension into the Dakotas. The diabatic heating which normally pumps up a seasonal ridge in the Southwest was enhanced by drought-reduced evapotranspiration and higher altitude snow cover. This pattern is fully consistent with global warming, as climate models and observed trends both show increasing Southwest drought. In addition, the overall poleward expansion of subtropical summer highs fits the same pattern.
The PNW heat wave was actually the third expansion of the hot upper ridge, and the most intense. The first expansion was around June 4, when early season heat records were set at Salt Lake City UT(100F) and International Falls MN (98F). The second expansion came in mid June, setting early season records at Grand Junction CO (105F) and Havre MT (104F), and tying all-time highs at Salt Lake City (107F) and Palm Springs CA (123F). Reaching all-time highs so early in the season was already an indicator of the exceptional intensity of the thermal ridge, although this was quickly overshadowed by the even more exceptional records in the PNW.
My point is that the heat wave was larger in scope and longer in development than the immediate PNW in late June, and is likely tied to the Southwest drought. This means that the larger scale lack of moisture should be examined for attribution, not just the preceding few months (which would normally be dry in the Southwest; their drought began or intensified over the winter.) This also undermines the Clifford Mass approach, which only looks at the moisture pattern in the PNW, and doesn’t consider the possibility that the extreme synoptic wave had been amplified by anomalous diabatic heating over a larger area and longer time period.
The small-scale circulation off the West Coast, which Mass suggests helped to enhance the hot easterly downslope flow west of the Cascades, was more likely to accompany an early-season heat wave. If I recall correctly, it developed from the remnants of a baroclinic wave in the eastern Pacific, and is more characteristic of a spring than a summer circulation. I don’t know if the climate models would have sufficient resolution to pick up on such a feature, which interacted with the larger scale synoptic pattern.
I am inclined to attribute this extreme event to interacting factors, as mentioned in the report: “The second option is that nonlinear interactions in the climate have substantially increased the probability of such extreme heat, much beyond the gradual increase in heat extremes that has been observed up to now. We need to investigate the second possibility further, although we note the climate models do not show it.” Mass dismisses this possibility, but it’s a real one.
Scott Slotterbeck says
‘tord records’. So global warming wasn’t responable for that earlier temperature, but is now. Magically.
You have zero proof or even observational data.
The record high in Sacramento CA is 114°F. That occurred quite recently…1925. Has global warming eased up?
John Pollack says
I have no idea what “tord records” refers to. I didn’t mention any earlier record that global warming wasn’t responsible for. I did mention observational data, lots of it. .
It’s not magic, it’s meteorology. You don’t know much about it.
Paul Pukite (@whut) says
Take a look at the wavenumber=4 standing-wave pattern that encircled the world for most of June (courtesy of @KKornhuber on Twitter)
The PNW was pinned to a persistent state of this pattern.
We still have a ways to go in getting the attribution of extremes more precise, as climate science still can’t predict the next ENSO (the most well-known standing wave pattern) with the same accuracy as tidal forecasts. Once that is improved we should be able to better compensate temperature readings for the impact of El Nino events, and also any related standing-wave patterns. For example, the PNW is geographically part of the Pacific North America (PNA) pattern, which shows a distinctly different standing-wave cycle than ENSO.
I wonder if the probabilistic explanation is just a loser in helping folks understand the relationship between global warming and an extreme event?
Take the case of the recent PNW heat wave: if the level, extent and probability of a particular heat wave depends to a fairly certain extent on the underlying increase in global temperature (which is almost certainly does), then the simple, generally correct and possibly more persuasive answer to the somewhat wrong-headed question of causation is, yes, the heat records being set arise from global warming.
If a person says, yeah, but was this heat wave caused by global warming, a good answer might be:
The relationship is complicated and causation is not the best way to understand it, but part of this situation is pretty simple, a hotter planet gives rise to more extreme heat waves. How could it be otherwise?
A simple glance at record temperature highs and lows shows quite clearly that record high temperatures are now much more common than record low temperatures. Connect the dots. Read the handwriting on the wall.
A heat wave like the recent PNW heat wave is simply a expected feature of a warmed and warming planet. A person who doesn’t grasp this situation is most likely to be a person who is driven by ideology rather than the numbers and the science. Scientific answers and statistical analysis are unlikely to overcome ideology on these questions. Time and direct experience of disastrous global warming may not even be sufficient to overcome anti-science, “alternative fact” ideology. I think it makes sense to recognize that scientifically solid explanations just don’t have much traction with a sizable percentage of the human population. I wish it were otherwise.
Reality Check says
“taking a common view; used chiefly in reference to the first three Gospels (Matthew, Mark, and Luke), from their similarity in content, order, and statement.”
From the Greek word synopsis meaning “a seeing together, a seeing all at once,” with roots syn- meaning “together, same” and opsis meaning “vision, sight, appearance.”
Nearsightedness, or myopia, as it is medically termed, is a vision condition in which people can see close objects clearly, but objects farther away appear blurred.
Things are not always how they first appear.
Doesn’t come close to being ad hominem.
Geert Jan van Oldenborgh says
Thanks for the clear summary. Regarding the resonance hypothesis, Mann et al, 2007, say that it operates in wavenumber 6-8 range at 30-45 ºN. The weekly mean Z500 anomaly during this event was a very clear wavenumber 4 pattern and the centre of the Ω block at 52 ºN, so as far as I can see the resonance hypothesis does not apply for his heat wave.
Chris Doyle says
As a retired meteorologist in Vancouver I watched this event unfold with fascination and trepidation. My red flag, almost literally, was the NAEFS ensemble situational awareness table. At one point, several days before maximum heating, the 850 mb temp was forecast to reach a mean departure of +9 sd from NAEFS climatology; inconceivable indeed. Both deterministic models and ensembles flagged this event with a reliable minimum of a week ahead. Nevertheless, several hundred folks, mainly elderly, died from heat stress in BC. “Novel” events have been the hallmark of the last few years and it seems in these cases that the social response is inadequate. Hopefully the conclusions of the report from World Weather Attribution will sink in where they need to, and help prevent a future failure of imagination on a similar scale.
Scott Slotterbeck says
Let me ask you a question: which kills more people each year, heat or cold?!
Cold, by far.
Hank Roberts says
This argument was debunked years ago. I recall someone — possibly in an old RC comment — who pointed out that during a cold snap, it’s the people in uninsulated houses who suffer the most. And in many areas there has not, in the past, been any urgent need to build or rebuild insulated housing. So during a cold snap there’s a large population who are newly at risk, and a disproportionate number of deaths.
In a hot snap, at least until the temperature-humidity reaches the death zone, cooling the body is possible.
Hank Roberts says
Here ya go:
cold-weather mortality exceeds hot-weather mortality in Brisbane, as it does elsewhere in Australia.17
“There’s generally an underappreciation of how cold winters are, and some people even believing that it’s healthy,” says Barnett. “People in Scandinavia would never expose themselves to [Brisbane’s lowest] temperatures inside their homes.” Indeed, in much the same way that heat has proven more deadly in Europe, studies have confirmed that excess mortality due to cold weather is more pronounced in temperate regions than in cold regions, a fact blamed on, among other things, houses improperly designed for lower temperatures.44,45
Erich Osterberg says
Important point, Scott. But also important to point out that increased deaths from higher heat leads to overall more deaths despite a decrease in cold-related deaths.
Same lead author as the paper you point to, but 2 years more recent:
Author’s words: “This study indicates that, in high-emission scenarios, most regions are projected to experience a steep rise in heat-related mortality that will not be equalled by a reduction in cold-related deaths, resulting in a substantial positive net increase in mortality.”
Large regional differences, of course. Warmest areas (SE Asia, S Europe, C/S America) hit hardest. Cold areas (N. Europe, and interestingly Australia) do see net benefit from reduced cold deaths,
Ray Ladbury says
Cliff, It is not so much the “science” in your analysis as it is the statistical reasoning. Indeed, I think the whole point of the WWA analysis is that it is wrongheaded to look for a single “cause” for such an extreme event.
Where I see your analysis falling short is that you do not note that the trends you investigate are affecting the expected temperature–the mean. Globally rising temperatures, drying soils…they are all moving the entire distribution toward higher temperature. That’s a couple fewer standard deviations–and probability of an event decreases quite nonlinearly with the number of standard deviations it represents. What climate change–through rising temperatures and drying conditions–did was take what would have been an effectively impossible event and make it merely a rare one.
One cannot say climate change was THE CAUSE, but it certainly was–by your own analysis–a cause.
Matthias Schürle says
Rainfall and soil moisture deficits as a precursor?
“In many previous extreme heat events, such as the 2003 European heatwave, rainfall deficits and dry soils the prior spring were shown to have made an important contribution to the temperature extremes, and so it’s worth looking at the same phenomena here.”
Sometimes it is worth asking: Was there the chicken or the egg first?
Are the constantly rising maximum temperatures (chicken) born of an (egg) from drought and absence of water? – or was it the other way around, that drought is a consequence of higher temperatures?
I am sure that, similar to the ice-albedo effect, we have set another vicious circle in motion, which leads to ever higher surface temperatures due to less water, evaporation and clouds, which then in many regions naturally reduces the water availability again.
The mistakes / crimes that humans have committed and continue to commit in dealing with the carbon cycle are largely known to humanity.
The accumulated overexploitation and worldwide looting of the groundwater reserves (water mafia), deforestation and land use which has also continued since industrialization, has brought a global rise in temperature with it, that is similar to the rising CO² concentrations. The most recent record temperatures in the north-west of the USA were foreseeable decades ago.
Greetings from Jack Nicholson / Chinatown – the man with the big bandage on his nose. Unfortunately, the most basic sense organ that is linked in our brain to the primordial of all instincts.
Language, Children, Language,
So what does “cause” mean??
I presented this thought experiment on a previous attribution post:
We have two parallel-universe Earths; we are Earth-A, and Earth-B differs in only one way. On B, carbon that has been fossilized forms CO2 that does not absorb long-wave radiation.
The same amount of CO2 is produced, cows still belch and fart methane, forests are destroyed… equally on A and B. The only difference is the amount of energy in the climate system; it is higher here on A because of that absorption of radiation.
Now, whether we use the term “chaotic” or not, I think we all agree that what we experience as weather occurs in a complex, non-linear system, so that what is popularly termed “butterfly effect” is real. Small changes in initial-values/inputs tend to result in large changes in outcomes/outputs.
So, would anyone like to offer their opinion as to what the probability of occurrence of that heatwave under discussion on Earth-B would be?
In this experiment, I’ve changed a single variable (system energy content) with as simple/small a change in input as is consistent with the scope of this question. But to me, the established magnitude of difference in system energy is way, way, more than the breeze from a butterfly’s wing.
So why are people afraid to say that there is essentially zero probability that this specific event would also be occurring on Earth-B? And in the common understanding of causality, why isn’t that the default answer to Denialists of various persuasions and backgrounds who question whether it was “caused” by the CO2 input of humans?
The headlines this am say the heat wave is “virtually impossible” without global warming. Anybody think that “virtually impossible” is going to end the arguments over causation?
I continue to wonder how to explain these things to folks who don’t want to accept the explanation because the topic is ideologically loaded.
I think as others have said that accepting the discussion framing of causation is not smart and will not help us expand understanding and address the climate crisis.
My current response to a question about causation and the heat wave might go like this:
Was the current heat wave caused by global warming?
my answer: do you ask if cars cause flat tires or engine failure or do you accept that this type of mishap is a component of operating a car? Does being born as a human cause you to have feet or hands or do you accept that feet or hands are a common component that is present at birth?
The answer if you persist with the causation question is: Yes, extreme events are a component of global warming. In a very simplistic manner, extreme events are caused by global warming, just like being born caused you to have hands and feet.
michael Sweet says
It is always pleasant to read a summary of scientific data written by experts who know what they are talking about. Thank you to all the Real Climate authors for taking the time to help inform the rest of us who want to discuss the science.
These 20 degree excursions may well provoke The Economist to declare victory on the Black Swan thick statistical tail arguments it has introduced into the climate conversation over the past decade.
Other journals have been less astute in their heat wave advice
CLIFFORD MASS says
Gavin… first…that plot is not a synoptic chart. More importantly, there is no trend during the past half century when GW is significant. If you considering the entire period and consider the variability, there is no significant trend. I can tell you how to do that analysis if you need it…cliff
Anyway, please skip the myopia business. And try to answer the science issues. Perhaps explain why there is no trend in record temperatures over the Northwest…..cliff
[Response: “That plot is not a synoptic chart”? Why should it be? It’s your plot of >99ºF occurrences by decade, and quite clearly there is positive trend. Similarly, the trend in the ERA5 annual max temperatures in the region seen above (fig 4 in the WWA preprint) is also trending up – and faster than the global mean is warming. Same is true at the stations they highlight. Not quite sure why you can’t acknowledge this. Plus when you get a chance, please explain what your “attribution” method will give as we (increasingly) see events that are outside of the bounds of the previous climate. Thanks. – gavin]
Kevin McKinney says
Yes, but also:
Cliff might do better praising the unprecedentedly large (3 cm!) Ranier cherries harvested in Washington just before the heat wave decimated his fellow pickers.
Would like to see more back and forth between Gavin and Cliff Mass, hopefully without the snark. Seems like there are differences in basic questions that should be relatively easy to answer: is the number of record hot temps in the PNW increasing? Is the number of very hot (e.g. 99+) days in the PNW increasing? What is the relevant time frame to answer these questions? Are increasing frequencies of very hot and/or record hot days a requirement for attribution of this event to AGW?
For me the event would not have been possible without global warming. The drying of the southwest is caused by global warming to a good extent. And if it was this air that was moved northward because of the global circulation becoming more meridional because of Arctic amplification both would not have been possible without global warming – drier and hotter tropic hot air being moved north where it descended.
But what i do not understand here is that the hot dry air that descends across the subtropics comes from the tropics, so the air that descended across the American Northwest should have been ascended air from the tropics which descended in this case not in the subtropics but in the mid- to high-latitudes. And with a zonal circulation (or a cold arctic with no regional large scale sea ice loss) this should have not been possible. – SO WHERE DID THIS DESCENDING AIR MASS CAME FROM IN THE FIRST PLACE?
And with the QR theory – the main point for me is not that the mid-latitude jet can disturb the high-latitude jet in a way that a 6-8 wave pattern is created but the possibility that the southern jet can disturb the northern jet in the first place with the outcome of a meridional circulation across the affected area and be it only one huge wave of the northern jet which eventually leads to more waves further downstream. So the QR theory opens up the possibility that the northern jet can also be disturbed by the southern jet during summer and not only by e.g. warm surface temperatures in the high latitudes. Thus the QR mechanism could have been a cause for the Omega block.
Scott Slotterbeck says
I love visiting state and national parks. It was on one such visit in the Sierra, where I saw an exhibit on droughts. Tree rings can give an indication of dry periods. This particular tree showed multi-year droughts that occurred hundreds of years ago, long before Europeans arrived on the West Coast, and long before the industrial revolution.
Climate is always changing. The hysteria accompanying the modern era is manufactured, and very harmful. Many people report their children do not want to bring children into a ‘doomed’ world.
Mal Adapted says
It looks like you’re a newcomer to RC, Scott. You’ll find most of the regular commenters agree with the following propositions:
1. The globe is warming, and it’s causing climate around the world to change. Climate is always changing, but the current rate and extent are unprecedented in the history of human civilization.
2. The warming is anthropogenic, caused by the economically-driven transfer of fossil carbon to the atmosphere by the gigatonnes annually, along with the damage we’ve done to the earth’s re-uptake capacity. Past climate changes were driven by natural forcings, but this time it’s all due to human activity.
3. It’s already taking a toll in money and tragedy around the world, to say nothing of biodiversity loss; and it’s hardly hysterical to point out that the final bill is open-ended, as long as the concentration of CO2 and other anthropogenic greenhouse gases in the atmosphere keeps rising.
If you disagree with any of those statements, Scott, I for one would like to know why.
Further what is very interesting to see is how the Earth system reacts to the warming of our planet. From the 70s to the end 90s we had a strong warming phase. Then from 1999 to 2012 the warming stopped due to increasing SOx emissions and leveling Methan increases (The intensifying walker circulation and the increased heat uptake of the oceans from 2007 onward was just a result of the hiatus). But in terms of extreme events the increases still kept coming during the hiatus – so the Earth system was still reacting to the increases in temperatures before the hiatus. But then from 2012 to 2016 we had an insane temperature increase of 0.41°C with extreme events increasing non-linear from 2012 onward, especially the last years. The cause: untypical circulation patterns and marine heat waves. So i guess what we just saw in the American Northwest is just the next kind of extreme events we will experience on an decadal basis because the synoptic circulation patterns being behind it that will be the outcome of a much more meriodional global circulation of a fast warming planet which produces ridges, throughs and blockings in a much higher frequency, and be it through marine heat waves and neigboring much cooler ocean surfaces. Funny thing is that we loose somewhat the meaning of subtropics, mid-latitudes and high-latitudes when it comes to air masses and temperatures – so my hypothetical question would be the following: what happens if an ascended tropical air mass gets advected far into the high latitudes – would that mean that ~45°C could also become possible in the arctic during summer be it in Canada or Sibiria?
Sorry for all the non-scientific writing stile – not a fan of it ;)
Reality Check says
@18 Good point. (smile)
Be that as it may, what Gavin said wasn’t ad hom.
While I cannot speak to the physics, I can speak to the stats. In general it is trivial to find distribution where a trend is found in an entire data set which is not found in each and every band when the data are disaggregated. There can be many physical reasons for this, but first one really ought to check out whether simply throwing away the bulk of the data (together with their degrees of freedom) reduces the power of any statistical test to be able to identify a true difference even if it exists.
This is a fundamental issue for extreme value stats analyses. This is why you see such analyses all the time on denier sites. They know this statistical fact and make use of it all the time.
The authors of the study Gavin linked to, claim that hot weather has gotten more frequent and more intense in the Pacific Northwest. They base their claim on analyzing data, specifically ERA5 daily Tmax for the Pacific Northwest, and GHCN-D daily Tmax for many stations, with special focus on Seattle WA, Portland OR, and Vancouver BC.
So I looked at the data they used (it’s easy to get), and they’re right. There’s a clear (and statistically significant) upward trend in a number of indicators, such as hottest-day-of-the-year, number of days above threshold, and mean summer temperature. There’s a clear (and statistically significant) difference in the distribution of high temperatures. And those results hold even when leaving out 2021, not including the remarkable heat we just endured. The trends were already clear — and statistically significant — before this year.
Meanwhile, Clifford Mass says “There is no trend during the past half century when GW is significant. If you considering the entire period and consider the variability, there is no significant trend. I can tell you how to do that analysis if you need it … cliff”
Clifford, I am personally very familiar with your statistical skill level. For you to offer to tell anyone “how to do that analysis” is a text-book example of Dunning-Kruger.
[Response: Thanks! – gavin]
Kevin McKinney says
Yes. Just to illustrate the point for any readers who might like such, a great example IMO is the fondness among denialati for the metric “major hurricanes making landfall in the continental US.” It even got the catchy but misleading name “hurricane drought”–making it a nomenclatural pars pro toto fallacy.
It enabled those who embraced it in uncritical fashion to more or less completely ignore events like 2012’s “Superstrom” Sandy, which although it peaked as a Cat 3 (and hence “major”) hurricane over Cuba, came ashore in New Jersey as a ‘mere’ Cat 1 extratropical storm. Nevertheless, Sandy killed 160 people in the US, affected 27 states, and cost an estimated $65 billion (good for 4th-costliest so far, after major hurricanes Katrina, Harvey and Maria.)
Some “drought”!–even accepting the above relentlessly US-centric focus as somehow appropriate. Sandy killed 73 more people outside the US–mostly in the Caribbean, although 2 Canadian deaths are also included in the grand total of 233.
Mal Adapted says
Excellent. Once again Ray, you’ve made the argument I wished to make, with greater clarity and concision than I could. Thank you.
hey, Cliff, give it up. Tamino and Gavin both say you are wrong and they are not alone. Even if they were alone, these are heavy hitters when it comes to crunching numbers and understanding the analyses. You are getting schooled on the topic of trend, are you learning anything?
A smart person would walk it back and think you are smart enough to do that now. It would be helpful for all of us if folks like you would learn on the fly and recognize when you have been getting it wrong because your ideas on these questions are persuasive with a lot of folks who can’t crunch the numbers. You are probably also persuasive with folks who don’t want to accept the trend and what it means for ideological reasons and when you fuel their ideology with your misunderstanding, you create a hotter planet in the future for my grandchildren. I don’t appreciate that.
You may think this is an ad hom attack since you are already on that track, but it’s really a plea to you to learn a lesson and be part of the solution.
Paul Pukite (@whut) says
For complex time-series, the antidote is to perform cross-validation by training a model on intervals of a subset of the time-series and then testing on “out-of-band” intervals.
So if the necessary and sufficient modeling information is within that training interval, the model will reveal that it works everywhere else as it’s able to capture the fundamental stationary behavior of the observed process. If it doesn’t then your model is likely an over-fit on the training interval leading to a bad fit elsewhere.
This is the only approach one can use on earth processes that only reveal patterns over many years, since we don’t have the time to make a prediction and then wait for decades to see how it pans out.
Unfortunately this is also likely why progress in earth sciences proceeds so slowly. Since the emphasis is mainly on “fresh” predictions (based on my experience), this precludes many potential models from getting considered as everyone is waiting in the queue for the results to slowly come in. And then as years go by, no one remembers the original model anyways.
This approach of cross-validation needs to be better known. I recently reviewed an open-review paper titled “101 Geodynamic modelling: How to design, carry out, and interpret numerical studies”, and not once was the concept of cross-validation mentioned. I raised this observation in the discussion, see the following link
michael Sweet says
Hey Tamino, it is great to hear from you. I hope you are doing well.
From your post I think I understand hottest day of year, days above threshold and mean summer temperature. Can you describe the statistically significant change in the distribution of high temperatures. Does that mean that there are more extra high temperatures so the temperature curve has broadened at the high end? Is there also a shift in low temperature distribution? I know there are less low temperature records set.
Jim Balter says
On his blog page, Cliff Mass says that his claims are demonstrably true because he demonstrated them. (The exact statement is “The headline speaks for itself, claiming the climate change drove the heat wave. This is demonstrably untrue as demonstrated in my blog”).
I guess that settles it.
Jim Balter says
“This kind of ad hominem stuff is not professional”
Does it have to be pointed out for the thousandth time that ad hominem is a form of informal fallacy in which an assertion is dismissed on the basis of an alleged trait of the person who made the assertion? The inference of a trait from someone’s behavior is not an ad hominem.
OTOH, responding to a lengthy factual post with a charge that the author is “not professional” is certainly somewhere in the vicinity of an ad hominem.
“Synoptic myopia? What is that or than name calling.”
It’s what he said it is: “paying too much close attention to the series of specific events that lead to the specific situation, and not seeing the wood for the (burning) trees”.
“Can you find any errors in my science or technical analysis? ”
This isn’t even relevant, because your conclusions do not follow from your analysis … especially statements like “Some of this is being done out of ignorance or laziness, but a few individuals are deceiving the public deliberately. Science journalism is only a shadow of what it was decades past, and a number of scientists now see social activism as more important than the determination and communication of truth” — talk about ad hominem! You attack the motives and competence of anyone who disagrees with you.
As for pointing out errors: “the trends in max temperatures are (literally) ignored, and the fact that this phenomenon is being seen around the world is just a series of increasing unlikely combinations of factors that for some inexplicable reason keep happening.”
Your response: “nonsense”, which could be taken as a description of what you write next: “I have a figure showing the trend in record breaking max temps” — a trend that you deny exists!
There’s a word for this sort of thing: projection.
Mal Adapted says
Also an excellent comment. The simplest definition of “climate” is “average weather over time”. Thanks to the international scientific enterprise, records are kept of hourly observations of multiple weather parameters, across an organized geographical network. It allows us to compute daily, monthly and annual values for mean, average and extremes of temperature in specific places. When statistics for those places show a trend over 30 years or longer, we say their climate is changing. There are domain-appropriate statistical methods that provide empirical justification for what our eyes tell us. For example, the apparent trend of decadal high temperatures for specific recording stations in the PNW, as in Cliff’s plot at #4. Cliff challenged Gavin to explain the absence of a trend. As Gavin replied inline:
AFAICT, it’s up to Cliff to see it or not. All Gavin and the other experts here can do, is try to show it to him as clearly as they can. Cliff should at least be more circumspect when replying to expert critics. There seems little hope he’ll acknowledge his errors, however. I agree with tamino:
Jim Balter says
“The headlines this am say the heat wave is “virtually impossible” without global warming. Anybody think that “virtually impossible” is going to end the arguments over causation?”
Ending all denial is not and cannot be the goal, and certainly not just with headlines … but the headlines are accompanied by articles that give supporting reasoning, and people who are motivated by their personally unpleasant experiences and their alarm over what is happening, rather than an ideological objection to the whole idea of AGW, are reading these articles and are learning things they didn’t know.
“I continue to wonder how to explain these things to folks who don’t want to accept the explanation because the topic is ideologically loaded.”
Don’t focus on those people … there are billions of others who are capable of accepting valid science-based explanations.
“do you ask if cars cause flat tires or engine failure”
I would take a different approach with cars. Imagine that a roofer drives by and a few nails fall off her truck. I then throw a a couple of handfuls of identical nails on the road. Several of the cars that drive by suffer flat tires. How many would have if I hadn’t added to the nails on the road? Which ones exactly? We don’t know, but there are certainly more flat tires because of what I did, and the more nails I throw, the higher the fraction of the flat tires that I’m responsible for.
Re: #30 (Michael Sweet)
I compared the distribution of temperatures before and after year 2000 (in data sets for the PNW from ERA5, for Seattle and for Portland from GCHN-D). First, I plotted histograms to compare their pdf’s (probability density functions), and it was rather striking how the whole pdf seems to be shifted to the right (toward higher temperature), indicating that low temperatures were more rare while high temperatures were more frequent.
But I never completely trust “seems to be” even when it’s “rather striking,” so I plotted their survival functions (1 minus the cdf, or cumulative distribution function), which is the probability of exceeding a given temperature, because I know how to compute uncertainty ranges for those. This confirmed, beyond doubt, that their distributions were indeed different — both on the high end (more and hotter hot weather these days) and the cold end (less and warmer cold weather these days).
P.S to last comment: the analysis compared data from before 2000, to that after 2000 but *before* 2021, so as to omit the extreme event of this year.
michael Sweet says
Thank you for the reply. I have a basic knowledge of statistics and can usually follow your explanations (although I cannot perform the calculations myself). It is interesting to me to read how you do these calculations.
I live in Tampa, Florida, near the line where there is frost north of me and no frost south of me. I have a collection of tropical fruit trees that are killed by frost. I like to read your analysis to get a better idea of the chances of my trees survival. (Our last hard frost was about 6 years ago. Hard frosts are becoming rarer and rarer). I am about to harvest a jackfruit which would have died 20 years ago.
Reality Check says
US west heatwave: 31 million people brace for record-breaking temperatures
Las Vegas could surpass its record-high of 117F as residents of US west face very high risk of heat-related illness
It’s been quite a flurry of Record temperatures in parts of the US, Canada, Siberia, Finland, New Zealand, Russia, Eastern Europe – and I suspect there are too many places to list this summer.
Arctic Sea Ice is at its lowest point in the records for this time of year. Canada’s heatwave was unprecedented. Parts of Pakistan and the Middle East exceeded 50C for a short time. You’d almost think there was a climate crisis or something.
Scorching Heat in Siberia and Europe – Record Low Ice Coverage in Arctic Ocean’s Laptev Sea
According to Jennifer Francis, a scientist at Woodwell Climate Research Center, the heatwave is the result of a persistent northward bulge in the polar jet stream. “This is associated with a blocking pattern in the jet stream that has been prevalent over Scandinavia this year and contributed to unusually warm conditions there, especially in Finland,” Francis said.
A second region of warm surface temperatures is visible toward the east, along the Arctic coast in Siberia. According to James Overland, of NOAA’s Pacific Marine Environmental Laboratory, a low-pressure zone just west of the hot spot produced strong, warm winds from the south that kept away the colder Arctic air.
Mike (14) “I continue to wonder how to explain these things to folks who don’t want to accept the explanation because the topic is ideologically loaded.”
Mike, I wrote on my approach to this question in Unforced Variations (21):
Global warming increases the frequency of heat waves in two ways:
a) effect of the higher mean: shifting the Gaussian distribution to the right:
heat waves are stronger, because the same departure starts from a higher mean. However, this effect is asymmetric – it weakens cold spells, since they depart downward from a higher mean.
b) effect of the increased departures from the mean – thicker both tails of the distribution – climate change, through affecting the movement and/or blocking of air masses, may increase the local departures from the current mean in both directions
So for heat waves – the a) and b) ADD to each other
for cold spells – a) substracts from b)
[in Tamino data (35) for winter in NPW effect of a) > b) => the net effect: “less and warmer cold weather”.
And a related question: “Haven’t you just said that “a local weather is not global climate“, then how could the LOCAL extreme weather be caused by global warming?”
Local weather extremes, being local and short compared to the time-scale of climate (say, 30 years), have negligible contribution to the value of the long-term global average temperature
say, effect of 1 week of 20C colder than usual weather in Texas (0.7mln km^2):
(-20C)*(0.7mln km2/510mln/km2)*(7d/(365d/yr*30yr))= – 0.00002 C from the 30-yr average, or -0.0006C from the global avg. temp that year
So while local weather extremes have negligible contribution to the value of the long-term global average temperature, it is not true the other way around – the global warming does affect local weather extremes (the 2 distribution effects discussed above).
An analogy: that individual soldier does not have a significant impact on the outcome of WWII, does not mean that WWII could not have a dramatic impact on that individual soldier. So although a given man could lose his leg also in peace-time, if you suddenly see an unusually high number of young men who lost their legs recently – a war becomes a distinct possibility .
“You (Cliff) are probably also persuasive with folks who don’t want to accept the trend”
A smart person would not be telling Cliff (or any denialist) he is ‘persuasive’.
28 Mike says:
hey, Cliff, give it up. Tamino and Gavin both say you are wrong and they are not alone.
A smart person would walk it back and think you are smart enough to do that now.
It is my considered opinion a completely overlooked aspect of why society is how it is now is quite simply down to complexity. It is too difficult because of the sheer volume of data and the ease with which any voice can be amplified for most people to confirm clearly whom they should listen to. So, we get massive ignorance spouted constantly without the tiniest hint of irony.
It would be helpful for all of us if folks like you would learn on the fly and recognize when you have been getting it wrong because your ideas on these questions are persuasive with a lot of folks who can’t crunch the numbers.
And how many people does that apply to here? The glaring example is nigel. He must be beat about the head and shoulders for days, weeks or even years before he shifts his stance. Even then, he claims it was always his stance.
It is far too difficult to police the idiocy. There is too much info and zero mechanisms for people to have a clear sense of what is and is not bull. And this also is connected to issues of bastardized 1st Am. rights wherein the vast majority of nigels and Cliffs don’t understand their right to an opinion does not validate that opinion.
The nigels and Cliffs don’t change because there is no mechanism for that change to happen.
I have called and again call for Truth in News, Truth in Politics and Truth in Advertising laws. The former existed because of ethics, but then TV newsrooms were twisted to the profit motive. There is no 4th Estate. The 2nd has never existed. The last used to exist.
Want change, demand truth.
Barton Paul Levenson says
tamino, welcome back! We missed you.
Are you going to start up your web site again?
J Doug Swallow says
“In this case, we are seeing probability ratios of 150 to 1000s, suggesting that these, improbable, temperatures can be almost entirely attributed to global warming. Without the anthropogenic signal, temperatures this extreme wouldn’t have happened in thousands to tens of thousands of years.”
Could what is written above also be applied to these historical heat waves on the East Coast?
The deadly 1896 and 1911 New England heat waves
On May 10, 1896 most of the Eastern US was over 90 degrees. New Bedford, Massachusetts was 96 degrees, which was 43 degrees warmer than the previous day’s forecast high, but the worst was yet to come. A heat wave during July and August, 1896 was at that time the worst weather-related tragedy in American history. By the time it ended in mid-August, 1500 deaths from the Midwest to New York to New England had been recorded.
On Tuesday, August 11, 1896 the Boston Globe reported ten fatalities in the city from the heat, twenty the following day, and fifteen more on the 13th when the heat wave finally subsided.
Fifteen years later, the record for heat-related fatalities was broken. In June and July, 1911 an eleven-day heat wave recorded temperatures above 100 degrees Fahrenheit in which two thousand people died, some from drowning trying to cool off. On July 11, crowds gathered in the shade at Hartford City Hall watched the Thermograph fluctuate between 110 and 112 degrees.
File this article under Earlier Than Expected. That file is bulging.
In California’s interior, there’s no escape from the desperate heat: ‘Why are we even here?’
“Researchers have been warning of such extreme heatwaves for decades, he said, but the barrage of heat surges that California and the western US have been alarming, he said. Temperature records are being broken earlier than expected or predicted.
“We are breaking temperature records this summer. And are going to keep breaking temperature records, as long as we keep burning fossil fuels,” said Ortiz, who lives in the Valley. “It’s infuriating, it’s tiring and it’s emotionally draining to see.”
If people ask whether global warming caused this heat wave, most of us commenting here understand that causation is not a sensible question in such a complicated system, but the answer needs to be simple. Keep it simple.
Yup, this is global warming. This is what happens on a warmed planet.
you say “Don’t focus on those people … there are billions of others who are capable of accepting valid science-based explanations.”
I expect that is true, but in the US, a major political party that essentially denies global warming is a problem continues to win lots of elections and it is impossible to enact meaningful public policy to address global warming in the US until that party agrees that the warming is real, caused by humans and is dangerous.
I am always looking for a snappy response that might be persuasive in the long-term and does not require a lot of my time in the short-term.
I would love to be able to ignore these folks and watch the US respond to global warming despite the intransigent anti-science position that these folks embrace.
programmers use to encourage each other to kiss the problems: Kiss, of course, meant keep it simple, stupid. I am inclined to shorten that up and encourage that we keep the message simple. It’s not easy with a complex subject, but getting dragged into the weeds to argue climate minutia with ideological deniers is not a smart move in my opinion.
Philippe Chantreau says
Good to see Tamino back. I live in the PNW, very much concerned by the recent heatwave. I find Cliff Mass’ assertions that there is no significant trend produced by AGW in the region difficult to believe. My perception (purely anecdotal, from experiencing the weather year after year) leads me to believe the following:
-Average weekly precipitation decreases earlier in the year.
-The date of first hay harvest is earlier, at least in the Western part of the state.
-First day at/above 85 deg. F happens earlier in the year.
-Number of days per year reaching 85 deg F is increasing.
-Average high temp during the summer is increasing.
-I suspect the average temperature/dew-point spread during the summer is also increasing. The spread where I live during the heat wave reached 36 deg C, some places in Canada had relative humidity of close to 10%.
-Last day at/above 85 deg F is getting later.
Then there is also probably a host of measures related to snow cover and its evolution during the year, but that’s a whole different story.
Nonetheless, if this has been compiled by the state climatology office, I’ll be thankful for anyone to give me a pointer. If not, then perhaps someone should do it and publish the results.
“He must be beat about the head and shoulders for days, weeks or even years before he shifts his stance.
Complete fabrication. I’ve changed my mind in response to a couple of peoples comments and pretty much straight away. I’m not a stubborn fool. I’ve never changed my mind due to the nonsense you post.
“Even then, he claims it was always his stance.”
The things I claim have always been my stance, HAVE always been my stance. You either don’t read for comprehension, or you just make things up.
“I am always looking for a snappy response that might be persuasive in the long-term and does not require a lot of my time in the short-term.”
I’ve said the following several times already. Best response to denialists and fence sitting sceptics and republicans is often give them the bare facts concisely stated. For example if they say the sun is causing climate change (or words to that effect) say its not causing climate change in recent decades because the energy output of the sun hasn’t increased (this simplifies things into language they would get). Show them a graph. Leave it at that. Don’t get into huge details unless someone sounds very receptive or there’s a very good reason.
Now obviously this wont convince hard core denialists or only maybe 1 in 100 on a very good day, but it may convince a few republicans and fence sitting sceptics. They will maybe go away and think about the solar issue and with a bit of luck it may penetrate the concrete eventually.
One of our centre right leaning prime ministers John Key became convinced we were causing climate change when he saw a graph plotting recent solar irradiance against temperatures and saw they didn’t correlate very well.
Now I know I sound a bit hypocritical because I’ve engaged in some lengthy responses to a couple of the denialists here. Sometimes its just fun or because you get a bit of interesting discussion coming out of these things, but on other websites Im mostly much more concise and I avoid lengthy debates.
It’s The Physics Stupid,
But apparently many here can’t explain the physics, so they engage in weird analogies and supernatural mind-control explanations, much like the Denialists.
Piotr (and others): You can’t “shift the distribution”, because you are not God. The distribution moves to the right because the individual measurements are higher, not the other way around.
You wonder why you can’t counteract the Denialist propaganda… it’s because you start out accepting their framing in the first place.
And perhaps because you yourselves are indeed out of touch with the physical nature of physics… we’re not doing QM or GR here, people, where the distinctions between cause and effect get blurry, and the country becomes the map. Chaotic systems are just complicated but still deterministic everyday physics.
This is from Piotr, but he is certainly not alone in this error:
No, Piotr, the “long-term global average temperature” does not affect anything, because that value is an effect, not a cause.
To use your pointless analogy, it is like you are saying the soldier lost his leg because the newspapers printed stories about the war.
I looked at the Cliff Mass blog and it is a very slick bit of “science theater”, although the errors in logic are easy to see. If we want to counteract it, we have to do a lot better with our communication/education presentations.
Clifford Mass says
zebra,,,, can you tell me what the are “errors in logic” in my blogs. I want to get the science right…..particularly as I am now writing the papers on this event..cliff mass
William Jackson says
43 Comparing apples and oranges are you not JDS, 1896 was in a time of little or no AC not to mention that Boston is not the Pacific North West! You seem confused.