Historic Pacific Northwest Heatwave of 2021

[Roger Smith]

Some highlights of the 12z 500 mb analysis (from EC) ...

598 dm high over s central BC. 

586 dm thickness max near Salem OR where sounding gives -2 C at 500 mb. 

==============================

Just a wait and see now, we got some relief overnight with most places dropping to the low 70s briefly. 

[GaWx]

As of Noon PDT vs 24 hours earlier, this is how much hotter it is:

Portland 7 F

SeaTac 6 F

Pendleton 6 F

Walla Walla 5 F

Spokane 4 F

Yakima (airport) 2 F

 

[Roger Smith]

On this question of the role of AGW in the appearance of such events, I agree with previous comments about the difficulty of disentangling the signal from the processes. Logically speaking, if we at least agree that AGW must be warming the atmosphere by some amount (1 C deg on average but closer to 2 C in subarctic regions, and also from some studies stronger in the n.h. than the s.h.), then we have a number of possible interactive scenarios:

(a) the atmosphere is doing what it would have done without the signal, but all components are warming up;

(b) the atmosphere is being forced to develop warm outcome patterns more frequently than it would do without the human input;

(c) some number of incompletely understood feedback loops are interacting with only partially predictable consequences. 

These are not necessarily completely independent or mutually exclusive outcomes. 

For example, I noticed that the singularity of the Nov-Dec 2015 warm signal across N America into Europe followed the northward track of remnants of Hurricane Patricia in late October 2015. Without that exceptional hurricane and its ability to move large amounts of tropical air north, would the full extent of that warm signal have developed? 

On the other side of the coin, certain statistical profiles argue for a general blanding of climate in some regions, with less extreme cold a certain factor, but also less frequent daytime heat as the increases seem to be falling mostly into the overnight low category. 

I have done an extensive study of Toronto and New York City temperature records. Both have an urban heat island factor to remove from the data but that should create more frequent warmth at all times of day, so even without addressing that factor, here's an interesting finding -- the frequency of hot days is not increasing in statistical terms since a peak around the 1944 to 1955 era. The frequency of warm nights is increasing. The main periods of stress on daytime high records is either end of winter, i.e., late Nov-Dec, and late Feb-early March. In those time windows, more than the expected number of record high maxima occur, at other times of year, the distribution is either random or it favors earlier decades. Both temperature series showed a natural warming signal around the 1890s with a relapse in the 1904-07 interval and further natural warming around 1911-22. The 1920s turned somewhat colder again. The 1930s are of course well known for hot summers but also had a number of very mild winters (and one very cold one, 1933-34, with 1935-36 known for its cold further west). 

Anyone who wants to take a deep dive into these stats can find links to my files (which are posted on the UK site net-weather, mainly because they have a larger download limit than here) -- the links are through a portal in the NYC subforum. The files are continuously updated. 

Right now I am working on warmest and coldest weeks at NYC. This is not completed but I can already see that the frequency of warm weeks by daytime highs is fairly well distributed and the frequency of warm weeks by overnight lows is dominated by more recent years. Since mean temperature is the average of both, the frequency of warm weeks by mean temp is skewed a bit towards the recent past but so far the impact of the higher mins is not that substantial as to overturn many of the earlier contenders which survive so far. 

December 2015 was quite a singularity in our recent climate. Britain for example had a mean temperature 1.6 C higher than previous monthly records (set in 1934, 1974). NYC and Toronto had means even more separated from second place. December has shown a much more robust recent warming than most months. We had 1982 which set new records, followed by a few more in Dec 1984, then 1990 and 1998 broke some of them and 2013 or 2015 broke most of the rest. I got far enough into the compilation of record warm weeks to find that Dec 2015 holds almost all the high values after mid-month. 

Even with the recent warming, some of the most impressive warm spells (even before adjusting for urban heat island) fall way back into the past. July 1868 was a very hot month at Toronto. Late May of 1880, late September of 1881, and two periods in 1895 (late May-early June, and mid-Sept) still hold on to records that are sustained over several days to a week. 

It's a mixed bag of signals in general in these historical climate records, and the details of modern or AGW warming are perhaps more subtle than many might imagine. When I look back on these historical warm spells, and even ones that as an older weather enthusiast I can remember (like the mid-April 1976 heat) I wonder if they would play out the same way (perhaps 1 deg warmer or so), or if they would evolve differently and hit further north than they did back then. 

The same thing applies to U.K. climate records. It remains true that the warmest May (by over 1 C deg) was in 1833, and the warmest June was in 1846. Until broken in 2006, the warmest September was 1729. Some months in the U.K. records have showed a greater tendency to modern warming than others. The process is not just some blanket warming of 1-2 degrees. Driving warmer air into the arctic is going to cause arctic air masses to try to flow out behind disturbances. Some think that in recent years, the climate of the Canadian prairies has become subject to more extremes with unusual out of season snowfalls and frosts like the snowstorm in Calgary in early September of 2014 (and the more recent Denver snowstorm and freeze in Sept). 

Without the luxury of a "control earth" that we could study for differences, the field is wide open to speculation, which in a way is causing a lot of unhelpful confusion since relatively low-information onlookers are quite free to take any climate occurrence and concoct an "explanation" for it that falls entirely into climate change, for example, the recent rather cold April they had in Europe was "explained" as a consequence of AGW, which of course begged the question, why were most of the years that had similar means back in the Maunder minimum or the coldest parts of the 19th century?  And why is every April not now like this? (most are quite warm in recent years). The answer is probably that the cold April was just "one of those things" and cannot realistically be explained by greenhouse gases, in fact one could argue that it was even less likely to occur than it might have been otherwise. 

I do believe that the frequency of warm weather events of all kinds is bound to increase if we warm the atmosphere by 1-2 deg but it has to be said that the increase must be demonstrated to fall in a higher range than you could expect by relaxing criteria by 1-2 deg from past frequencies, in order to demonstrate a change in process, otherwise all you are left with is a realization of the shifting range of the bell curves. 

[Roger Smith]

23 minutes ago, GaWx said:

As of Noon PDT vs 24 hours earlier, this is how much hotter it is:

Portland 7 F

SeaTac 6 F

Pendleton 6 F

Walla Walla 5 F

Spokane 4 F

Yakima (airport) 2 F

 

you missed my living room 3 F 

[GaWx]

1 PM PDT vs 24 hours ago:

Portland 104 (+11)

SeaTac      96 (+4)

Pendleton 106 (+5)

Walla Walls 106 (+5)

Spokane 98 (+5)

Yakima 102 (+2)

[TimB]

New record at Portland, 109, 13 degrees warmer than same time yesterday...

[sojitodd]

1 hour ago, TimB84 said:

New record at Portland, 109, 13 degrees warmer than same time yesterday...

Now up to 111.

Seattle up to 102, one more degree to tie the record, two more degrees to make a new all time record high. 

[etudiant]

2 hours ago, Roger Smith said:

On this question of the role of AGW in the appearance of such events, I agree with previous comments about the difficulty of disentangling the signal from the processes. Logically speaking, if we at least agree that AGW must be warming the atmosphere by some amount (1 C deg on average but closer to 2 C in subarctic regions, and also from some studies stronger in the n.h. than the s.h.), then we have a number of possible interactive scenarios:

 

Without the luxury of a "control earth" that we could study for differences, the field is wide open to speculation, which in a way is causing a lot of unhelpful confusion since relatively low-information onlookers are quite free to take any climate occurrence and concoct an "explanation" for it that falls entirely into climate change, for example, the recent rather cold April they had in Europe was "explained" as a consequence of AGW, which of course begged the question, why were most of the years that had similar means back in the Maunder minimum or the coldest parts of the 19th century?  And why is every April not now like this? (most are quite warm in recent years). The answer is probably that the cold April was just "one of those things" and cannot realistically be explained by greenhouse gases, in fact one could argue that it was even less likely to occur than it might have been otherwise. 

I do believe that the frequency of warm weather events of all kinds is bound to increase if we warm the atmosphere by 1-2 deg but it has to be said that the increase must be demonstrated to fall in a higher range than you could expect by relaxing criteria by 1-2 deg from past frequencies, in order to demonstrate a change in process, otherwise all you are left with is a realization of the shifting range of the bell curves. 

Imho, Roger here is providing  a Masters Class on the topic, with emphasis on some real issues.

This stuff is tough, lots of unexpected secondary processes complicate developments, so inadequate models and simulations that fall well short of the needed resolutions produce unreliable results. 

Has there ever been an effort to really grid a suitable area, say 100 miles square, to verify the modeling parameters employed? 

It would not be cheap to do, requiring many tens of thousands of sensors at multiple altitudes, but the ground truths it would gather would be essential to constrain the models.

[donsutherland1]

Earlier, Environment Canada posted that Lytton, BC had reached: 46.1C/115F, on its website. However, the 3 pm PDT temperature now shows 44.1C, so the earlier number is in question.

[Roger Smith]

It's going over the top anyway Don, looking at the full range of reports I feel sure one of the official stations will top both the BC and national record values. I am going to post something I was working on but won't be able to finish parts two and three right away, for your general interest. This will be a list of the top 30 NYC weekly averages (max only in this post, will add min and mean later). That should be up shortly. 

The local station (Warfield) has been up to 41 C already and will likely top off at 42 C, it feels about that outside although I am up the hill 2 miles and 500 feet higher. However I was in Las Vegas in Aug 2011 when it was 46 C and that was noticeably hotter although the sky looks very similar.

[GaWx]

B.C. heat wave shatters Canadian record for highest temperature ever recorded | CBC News

 This also says new Canada record of 46.1C but could it also be wrong?

 

A little good news. The 4 PM Washington and Oregon readings that I've been following were all either the same or cooler than 3 PM.

[donsutherland1]

15 minutes ago, GaWx said:

B.C. heat wave shatters Canadian record for highest temperature ever recorded | CBC News

 This also says new Canada record of 46.1C but could it also be wrong?

 

A little good news. The 4 PM Washington and Oregon readings that I've been following were all either the same or cooler than 3 PM.

The hourly data and hot spots data are inconsistent today. We’ll know a little later what the final high temperature was.

[sojitodd]

Seattle tied it's all time record high of 103, will likely be higher tomorrow.

 

*Portland made it to 112. Damn!

[raindancewx]

It's still kind of funny to me that this heat wave, which has a chance at setting all time (national and state) heat records is being compared to the heat down here earlier in the month.  I think I had one record daily high in June on 6/14? I've got two potential all-time record low highs coming in the next four days. 120 in Phoenix is just very different from 110 in Portland Oregon. When you get the 600 decameter monsters down here, a lot of times they under-perform because any moisture in the air is destroyed, and you end up with a  67/102 day instead of a 75/105 day. In the Northwest you have actual oceans and bodies of water nearby to really juice the lows and allow for some nasty air-masses. When the map below happened last year, the lows were still around 70 each day, even with a ~2 degree urban heat island effect in Albuquerque for the lows on top of the 1.8F degree warming/century that is present in all other sites in NM. I can't even imagine a 112 degree day, +35F v. average, with a dew point in the 50s-60s all day like Portland had today, you just never have that oppressive stuff here. 104 with a dew point of 66 at 1pm is the type of the thing that would easily kill people not used to it. 

https://forecast.weather.gov/data/obhistory/KPDX.html

 

[GaWx]

52 minutes ago, GaWx said:

A little good news. The 4 PM Washington and Oregon readings that I've been following were all either the same or cooler than 3 PM.

 Here's something weird. After all of the 6 stations in Wash/Ore either dropped 1-3 F or were steady the previous hour, they all rose 1-2 F this most recent hour (5 PM PDT)! The hottest hourly of the day was reached at both Portland (112) and Yakima (107).

[donsutherland1]

Lytton, BC.

Data as per Environment Canada: 46.6C/116F.

https://twitter.com/ECCCWeatherBC/status/1409306438074310661?s=20

[psv88]

So it was 116 in Canada today?

[Roger Smith]

Yes but only 107F where I live, thinking of getting the parka out. 

[Roger Smith]

One of the big news stories here is that the raspberry crop, which usually matures in early July for a leisurely picking by thousands of ordinary citizens who pay for the privilege, has instantly appeared and is now baking to worthless quality on the plants. Next stop, blueberries, they are usually ready to pick in August. 

Have been seeing bears on a regular basis around town here, they are clearly distressed by the extreme heat which is probably throwing off their feeding habits too. 

[Roger Smith]

48 minutes ago, GaWx said:

 Here's something weird. After all of the 6 stations in Wash/Ore either dropped 1-3 F or were steady the previous hour, they all rose 1-2 F this most recent hour (5 PM PDT)! The hottest hourly of the day was reached at both Portland (112) and Yakima (107).

Here are some hourlies from one of the hottest locations, PSC ( Pasco Tri-Cities airport in se WA)

2 pm PDT __ 109 F

3 pm PDT __ 110 F

4 pm PDT __ 113 F

5 pm PDT __ 114 F

6 pm PDT __ 112 F

(max reported as 115  F)

This looks to be a fairly normal sequence for this abnormal heat. 

[donsutherland1]

Seattle finished with an all-time high of 104 today.

[Roger Smith]

This may illustrate the point I was making about daytime and overnight temperatures ... 

Top 91 NYC warm weeks 94.0 avg or higher (and top 48 ranked years for high 7d avg max)

_ equates to top 25 NYC warm spell years (since many entries have 3-5 consecutive warm 7d intervals overlapping)

_ (any qualifying value is highest of a number of consecutive values, those which belong in top 30 are added to entry, the entry is given the dates of the highest qualifier so the dates of the other entries can be determined by counting back or forward from that as needed -- all weeks that exceeded the indicated 30th place value or a higher ranked value with ties through 30th place are in the table. ... this table cuts off at 94.0 F) 

late June into July 1966 could be considered two separate hot spells but it was probably experienced as one uninterrupted month-long hot spell that was more about duration than extreme values.

A few notes are added about Toronto heat wave comparisons, in the 1953 heat wave below, not much space so will add here that averages around 95F were noted for Aug 27-Sep 2 and Aug 28-Sep 3. There were heat waves in late July, August 1916 and 1918 that made top ten for Toronto that failed to reach this level for NYC despite one or two notably hot days individually. But otherwise the Toronto list would not be very much different from this NYC list.

Heat waves in 1955, 1966 and 2002 could be divided into two apiece but with little cooling between them, they are shown as unified entries. A second heat wave for 1953 has the rank "2b" to keep the ranking oriented to years and not to specific heat waves. 

 

ranks of hottest mean daily max (with consecutive 94+) __ ranks of all noted weekly intervals (consecutive)

 1. July 15-21 1977 _ (96.86) (96.71) 98.43 (97.29) (95.71)(94.71) _ t10 t12 1 t5 t22 t50  ___ 12 days covered by 6 intervals

 2. Aug 27*-Sep 2* 1953 _ (94.57)(95.86)(96.71) 98.29 98.29 (97.14)(95.29) _ t53 21 t12 t2 t2 7 t31___ 13 days covered by 7 intervals 

 3. July 7-13, 1993 __(96.29)(96.57)(97.00) 98.00 (97.00)(95.57)(94.00) ___19 t14 t8 4 t8 28 t83____ 13 days covered by 7 intervals

 4. July 17-23, 1991 _ (94.71) (96.57) (96.43) 97.29 (96.43) __ t50 t14 t17 t5 t17 _____ 11 days covered by 5 intervals

 5. Aug 10-16 1944 _ (94.29) (95.71) 96.86 (96.57) (94.14) __t66 t22 t10 t14 t78 __ 11 days covered by 5 intervals

 6. July 17-23 and Aug 1-7 1955 __ (94.00) 95.43 ... ... (95.00) 96.14 (94.00) ___ two hot spells with only slight cooling between them

___________________________________ t83 t29 _____________t37, 20, t83  _________ 8d 2int, 13d separated then 9 d covered by 3 intervals

t7. June 27 - July 3 1966 _ 95.71 (95.29) (94.57) (94.43) (94.29) (94.86) _ t22^ t31 t53 t59 t67 t47 __ 4d 91-93 then July 7-9 ...

___ July 8-14 (94.29) 94.57 (94.14) ___ t67 t53 t78 ___w4 more d 90-93 _ 18 or 22 days covered by 9 intervals and 7 nearly as hot

____ ____ _____ _____ ______ _______ _____ _____ ____ ____ ___ ___ also six days before the first interval 91-92F avg (total 28d)

t7. Aug 28 - Sep 3 1973 __ (94.57) 95.71 (95.00) ______ t53 t22* t37 ________ 9 days covered by 3 intervals (93.86 values at each end)

t7. July 17/18-23/24 2011 (94.29) 95.71 95.71 ________t67 t22* t22* _______ 9 days covered by 3 intervals 

10. June 27 - July 3 1901 _ (95.00) 95.43 _____________ t37 t29^^ ___________ see note (extended but not as intense)

t11. Aug 9-15 1988 _(94.00) 95.29 (94.29) ____________ t83, t31, t67 __ an earlier hot spell peaked at 93.57 July 5-11

t11. Aug 4-10 2001 ________95.29 (94.14) _____________t31 t78 _____________ 8 days covered by 2 intervals

t13. Aug 5-11 1896 _ (94.29) 95.14 (95.00) (94.71) ____t67 t35 t37 t50 _____10 days covered by 4 intervals

t13. July 3-9 2010 __ (94.00) 95.14 (94.29) (94.00) _____t83 t35 t67 t83 _____10 days covered by 4 intervals

t15. June 1-7 1925 ______ 95.00 _______________________t37** _______________ see note

t15. July 28 - Aug 3 1933 _(94.43) 95.00 ______________ t60 t37 _________________8 days covered by 2 intervals 

t15. Aug 24-30 1948 _____(94.43) 95.00 (94.57) _______t60 t37 t53 _____________ 9 days covered by 3 intervals

___ ____ _____ ______ ________ _________ _____________ ________ ______ ____ ____ __  Toronto avg 95 F (Aug 23-29) 

t15. July 12-18 1983 _____ 95.00 (94.00) ______________ t37 t83 ________________ 8 days covered by 2 intervals

t15. July 23/24-29/30 1999 __ 95.00 95.00 (94.00) (94.14) _ t37 t37 t83 t78 ___ 10 days covered by 4 intervals 

t20. July 8-14 1936 _(94.29) (94.43) 94.86 ____________t67 t60 t47 _____________ this was Toronto's hottest week (99.29)

t20. July 16-22, 1980 _ (94.43) 94.86 _________________ t60 t47 ___ (a second hot spell averaged 93.57 Aug 3-9).

t22. July 29 - Aug 4, Aug 12-18 2002 _ (94.43) ... (94.29) 94.57 (94.29) _ t60 ... t67 t53 t67 __ entered as one (slight inter- cooling)

23. July 14-20 2013 _______ 94.57 (94.43) _____________t53 t60 __ note this is the most recent entry before 30th

 2b July 15-21 1953 _______ 94.43 ____________________ t60 ______ separate event from late Aug-Sept 1953, which ranked 2.

24. July 7-13 1981 ________ 94.14 _____________________t78

25. July 1-7 1911 _________ 94.00 _____________________ t83 ________________ Toronto avg 98 July 1-6

(after this point in the list, I have added years ranked 26 to 48 to provide a similar data base to the minimum temperature analysis posted later in the thread ... these are not entirely heat waves 24 to 48 in rank order because as noted above, a few years produced two or even three non-qualifying separate heat waves that would insert themselves into this list. But the maximum values above can be read in the same way as the rank ordered (1-25) highest seven-day average minima.) The numerous ties are created by the math of dividing whole number totals by seven, there can only be seven rounded two-decimal outcomes (.00, .14, .29, .43, .57, .71, .86). This list shows the second heat waves in 1988 and 1980 noted above but does not include them in the rank numbering system (if you wanted to convert to a list of heat waves then add one to each rank number below a second heat wave note). 

 xx. __ 93.86 ___ (none)

t26. __ 93.71 ___ 1940, 1949 ___ also 2nd 1988 heat wave July 5-11

(heat wave 30) __ 93.57 ___ (none) ____ also 2nd 1980 heat wave Aug 3-9

(add 3 to ranks if counting heat waves and not years, the additional ones include 1953 "2b" and 1988, 1980 above)

 28. __ 93.43 ___ 1937

t29. __ 93.29 ___ 1943, 1995

t31. __ 93.14 ___ 1872, 1892, 1917, 1952, 1964, 2016

 xx. __ 93.00 ___ none

t37. __ 92.86 ___ 1957, 2008

t39. __ 92.71 ___ 1885, 1984

t41. __ 92.57 ___ 1954, 1963, 2012

 xx. __ 92.43 ___ none

t44. __ 92.29 ___ 1923, 1924, 1941, 1961, 2005 

(also in top 50 ...)

t49. __ 92.14 ___ 1876, 1919, 1997, 2006

... note that in Aug 1918, seven days in a 10d span Aug 5-14 averaged 95, but a 71F max midway through left this two-part heat wave (with the all-time Aug record of 104F) on the sidelines. 

... ... ... the warmest week 2017 to 2021 so far is 91.57 June 29-July 5, 2018.

... ... ... the average for all years 1869 to 2020 is 90.8. For 1869-1900 it was 89.3, for 1901-60 it was also 90.8,

and for 1961-90, 1991-2020 both halves had identical 91.6 averages. (for clarity, this is the average of the warmest 7d interval of each year).

(note the coldest weeks have warmed up more, the same stats for that are 1869-2020 27.15,

1869-1900 24.2, 1901-60 28.0, 1961-90 26.7 1991-2020 29.2)

Just as a basis for discussion, these stats are consistent with an urban heat island increasing by 1 F deg in summer to the 1930s and another 1 F to present day assuming a levelling off after 1980. If you think that estimate of the u.h.i. for NYC is too large then the amount by which you estimate the u.h.i. leaves a trace of some other process at work (some combination of natural warming and AGW).

The winter or coldest week stats are consistent with an urban heat island increase about twice as large with evidence of a colder turn around the 1960s to 1980s and signs of a further warming recently that would more likely be from some factor unconnected to the heat island. 

_____________________________________________________________

* same value Aug 28 - Sep 3 1953 ... Aug 28 to Sep 3 1973 average was 95.71 ranked t22. 

^ This heat wave took top spot for all weeks from the one indicated to July 2-8, but values were lower than 20th (near 95 F).

^^ The 1901 value shown was second highest for that weekly interval but took top spots for the two previous intervals which were 92.57 and 95.00.

** The 1925 early June heat wave took top spots for weeks ending June 5 to 10, the other values were all in the range of 90 to 93. 

======================================================

ANALYSIS _ There were 25 qualifying heat waves with at least one seven day average of 94.0 F. That provided a total of 91 intervals with the average therefore being about 3.6 intervals, meaning that a nine- to ten-day span of three or four consecutive seven-day intervals was the core of the average heat wave. The more intense ones also had longer durations in the range of 10-15 days mostly, with the 1966 case extended with two separate qualifying peaks. Late July and early to mid August 2002 had two rather separate heat waves but they are grouped together since the study will concentrate on years rather than individual heat waves.  Of the 25 heat waves that qualified, one was in the interval 1869-1900, three were 1901-30, six were 1931-60, seven were 1961-90 and eight were 1991-2020. When the list was expanded to 48, this produced totals of 4, 6, 14, 11 and 13. The frequency has levelled off since around 1931. To give some idea of the degree of warming, for the 1869-1900 interval to provide six would require the minimum qualification to drop to 92 F, and for 1901-30 to double its total to six would require that we add these three near misses ... 93.14 in 1917 (late July early Aug), and 92.3 in 1923, 1924. ... Summer of 1930 was hot but only managed a peak weekly average of 91.86. 

One could postulate that NYC summer heat waves have remained steady state under an assumption of a 3F total increase in urban heat island that reduced to 2F by 1931 and rapidly decreased below 1F after 1940. That seems a bit contrived, 2F with a steady decrease is probably more accurate. So some of this change in frequency appears to be natural variability at work. There are only faint signs of any role of AGW in NYC heat wave daily high temperatures. I will add material in a separate post overnight (hopefully) to show that for overnight lows, a different conclusion can be drawn. Then we'll see how that blends together for daily mean temperatures. 

Some mention should be made of outliers, heat waves that were far enough from the peak of solar warming that they were at a disadvantage in reaching these criteria. The most prominent early outliers were five day heat waves in late May 1880 (91.6 May 24-28) and late May-early June 1895 (94.6 avg May 30-June 3, 93.17 May 29-June 3). These had high enough temperatures but could not sustain a seven-day average to qualify. Sept 20-23 1895 averaged 94.0, but seven-day averages were in the high 80s. In mid-September 1931 there was a mean of 91.0 for the week of Sept 9-15. The average for Sept 20-26 1895 was 89.3 F. Hot spells of about four days in April 1976 and 2002 are even greater outliers. These both averaged around 92 F for four days and had weekly averages near 86F. 

A second post will give a comparative account of the top 48 years for high seven day averages of summer minimum temperatures.

[Roger Smith]

Without going into quite as much detail, the rank order of highest seven-day average minimum values at NYC is as follows: 

 1. 2013 __ 79.29 _______t25 1876 __ 75.71

 2. 1896 __ 79.14 _______t25 2018 __ 75.71

 3. 1988 __ 78.14 _______t27 1972 __ 75.43

 4. 2011 __ 77.86 _______t27 2020 __ 75.43

 5. 1908 __ 77.71 _______t29 1892 __ 75.29

 6. 1999 __ 77.43 _______t29 1911 __ 75.29

 7. 2006 __ 77.29 _______t29 1917 __ 75.29

t8 1905 __ 77.14 _______t29 1948 __ 75.29

t8 1906 __ 77.14 _______t29 1955 __ 75.29

10. 1980 __77.00 _______t29 1961 __ 75.29

t11 1993 __76.86 ______ t29 1982 __ 75.29

t11 2016 __76.86 ______ t29 2005 __ 75.29

t13 1898 __76.71 ______ t37 1884 __ 75.14

t13 1981 __76.71 ______ t37 1944 __ 75.14

15. 1995 __76.57 ______ t37 1973 __ 75.14

t16 1885 __76.43 ______ t40 1890 __ 75.00

t16 1952 __76.43 ______ t40 2015 __ 75.00

t16 1953 __76.43 ______ t42 1887 __ 74.86

t16 2010 __76.43 ______ t42 1983 __ 74.86

t20 1977 __76.29 ______t42 1984 __ 74.86

t20 2002 __76.29 ______ 45. 1870 __ 74.71

t22 1909 __76.14 ______t46 1901 __ 74.57

t22 1979 __76.14 ______t46 1907 __ 74.57

24. 2001 __75.86 ______t46 1938 __ 74.57

____________________________________________________

Most of the years with the top average maxima are in this list, but quite a few rather average cases (for daytime max) are sprinkled in with them, 

indicating that some cloudier or more humid spells were included where average max may have been closer to 90F. The frequency distribution is

somewhat skewed towards two intervals, the decade of 1901-10, and the recent past. As to the decade of 1901-10, it is possible that site interference

was playing a role as the station was moved into Central Park from a more urban setting around 1919-20 (or so I am told). 

So with that in mind, here's the distribution of these top 48 years for average minimum in summer heat. 

The following chart shows the ranks of the distributed cases: 

1869-1900 __ 2, t13, t16, t25, t29, t37, t40, t42, 45

1901-30 _____5, t8, t8, t22, t29, t29, t46, t46

1931-60 ____ t16, t16, t29, t29, t37, t46

1961-90 ____ 3, 10, t13, t20, t22, t27, t29, t29, t37, t42, t42

1991-2020 __ 1, 4, 6, 7, t11, t11, 15, t16, t20, 24, t25, t27, t29, t40

________________________________

So unlike the average maxima, these warmer summer nights are increasing in frequency and severity (note the depth of high ranked cases as well as the frequency).

The average for this statistic is 73.34. In the intervals above, the means are 73.4, 72.7, 72.7, 73.2, and 74.4 which is fairly consistent with urban heat island growth combined with site alterations to reduce direct effects. So this suggests that the warmer summer overnight spells are contrasting with recent cases of cooler summers at other times, such as 1992, 2000, 2003, 2004, 2007, 2009 and 2014. 

After some rest and reflection I will tackle the question of how overall means for these summer heat waves compare. I think it will be a blend of the two different patterns seen. 

Some further analysis possible now that I have created a similar list of 48 hottest years for weekly averages (not quite the same as 48 top heat waves, 1988 for example had two in the range, and there are two cases in my list that could be taken as two separate heat waves, but as we are studying the frequency of events, a yearly count seems appropriate). 

Of the above list of top 48 warm overnight low averages, the following ranks do not match a high max top 48, and from inspection I could say that when a year produces both, they are usually in the same heat wave if not exactly the same seven days. So I have listed the 15 of 48 that do not overlap with the mean maximum peak for that year.

5th 1908 _ 89.00

t8 1905 __ 90.43

t8 1906 __ 88.29

t13 1898 __ 91.43 

t22 1909 __ 89.71

t22 1979 __ 89.86

t25 2018 __ 91.57

t27 2020 __ 90.14

t29 1982 __ 92.00

t37 1884 __ 89.57

t40 1890 __ 88.14

t42 1887 __ 89.29

 45 1870 __ 89.71

t46 1907 __ 87.71

t46 1938 __ 90.86

__________________________________________

These include all five summers in the period 1905-09, and two recent ones (2018, 2020).

The somewhat reduced heat of the daytime components may have been due to either cloudiness holding highs down, or the flow pattern favoring sea breezes that would limit the peak warming on at least some days. 

This means of course that 15 of the 48 hot daytime contenders did not make the list of warmest minimum averages. Those include 1933, 1936, and 1966. Clear skies during those heat waves probably allowed for just enough cooling at night, and this observation is supported by the lack of daily records set, it wasn't a case where several nights were very oppressive and one or two brought the average down, more that the base level of overnight lows was set in the mid rather than high 70s. 

Another top heat wave (1944) did qualify here but was well down the list from its daytime rank. 

_______________________

the final segment will be posted by July 2nd, as I compare the top 48 mean seven-day peaks with these two lists.  By mathematics you would expect about a 40-8 outcome in each case but we'll see how that develops. 

 

[donsutherland1]

British Columbia Records courtesy of Environment Canada:

Weather summary
for British Columbia
issued by Environment Canada
at 12:05 a.m. PDT Monday 28 June 2021.

Discussion.

The following area sets the Canadian maximum temperature record on 
Sunday June 27, 2021: 

Lytton Area (Lytton RCS) 
New Canadian record of 46.6 
Old record of 45.0 set in July 5, 1937 in Yellow Grass and Midale, 
SK. 

The following areas set a daily maximum temperature record on June 
27, 2021: 

Abbotsford Area (Abbotsford A) 
New record of 41.5 
Old record of 32.3 set in 2015 
Records in this area have been kept since 1944 

Agassiz Area (Agassiz RCS) 
New record of 40.4 
Old record of 33.7 set in 2015 
Records in this area have been kept since 1889 

Bella Bella Area (Bella Bella) 
New record of 34.2 
Old record of 24.3 set in 1995 
Records in this area have been kept since 1977 

Bella Coola Area (Bella Coola Airport) 
New record of 36.0 
Old record of 35.0 set in 1895 
Records in this area have been kept since 1895 

Blue River Area (Blue River CS) 
New record of 37.3 
Old record of 34.7 set in 2015 
Records in this area have been kept since 1946 

Burns Lake Area (Burns Lake Decker Lake) 
New record of 38.1 
Old record of 29.5 set in 1992 
Records in this area have been kept since 1949 

Cache Creek Area (Ashcroft) 
New record of 45.0 
Old record of 40.1 set in 2015 
Records in this area have been kept since 1944 

Clearwater Area (Clearwater Auto) 
New record of 42.3 
Old record of 36.9 set in 2015 
Records in this area have been kept since 1913 

Clinton Area (Clinton RCS) 
New record of 39.0 
Old record of 32.2 set in 2015 
Records in this area have been kept since 1974 

Comox Area (Comox A) 
New record of 38.0 
Old record of 31.1 set in 2015 
Records in this area have been kept since 1914 

Courtenay Area (Comox A) 
New record of 38.0 
Old record of 31.1 set in 2015 
Records in this area have been kept since 1914 

Creston Area (Creston Campbell Scientific) 
New record of 39.0 
Old record of 37.9 set in 2015 
Records in this area have been kept since 1912 

Dawson Creek Area (Dawson Creek A) 
New record of 36.1 
Old record of 28.2 set in 2000 
Records in this area have been kept since 1926 

Dease Lake Area (Dease Lake (AUT)) 
New record of 32.0 
Old record of 27.6 set in 1982 
Records in this area have been kept since 1944 

Esquimalt Area (Victoria Gonzales CS) 
New record of 38.3 
Old record of 27.8 set in 1951 
Records in this area have been kept since 1874 

Fort Nelson Area (Fort Nelson) 
New record of 35.8 
Old record of 30.4 set in 2004 
Records in this area have been kept since 1937 

Fort St. John Area (Fort St. John A) 
New record of 35.1 
Old record of 30.0 set in 1928 
Records in this area have been kept since 1910 

Gibsons Area (Sechelt Aut) 
New record of 39.6 
Old record of 30.6 set in 2015 
Records in this area have been kept since 1949 

Golden Area (Golden Airport) 
New record of 36.7 
Old record of 35.0 set in 1925 
Records in this area have been kept since 1902 

Gonzales Point Area (Victoria Gonzales CS) 
New record of 38.3 
Old record of 27.8 set in 1951 
Records in this area have been kept since 1874 

Hope Area (Hope Airport) 
New record of 39.9 
Old record of 33.4 set in 2015 
Records in this area have been kept since 1936 

Kamloops Area (Kamloops A) 
New record of 44.0 
Old record of 38.2 set in 2015 
Records in this area have been kept since 1890 

Kelowna Area (Kelowna UBCO) 
New record of 41.5 
Old record of 38.1 set in 2015 
Records in this area have been kept since 1899 

Lillooet Area (Lillooet) 
New record of 44.8 
Old record of 39.8 set in 2015 
Records in this area have been kept since 1917 

Lytton Area (Lytton RCS) 
New record of 46.6 
Old record of 40.2 set in 2015 
Records in this area have been kept since 1921 

Mackenzie Area (Mackenzie Airport Auto) 
New record of 37.3 
Old record of 28.4 set in 2015 
Records in this area have been kept since 1971 

Malahat Area (Malahat) 
New record of 38.2 
Old record of 32.6 set in 2015 
Records in this area have been kept since 1986 

Merritt Area (Merritt) 
New record of 42.2 
Old record of 38.0 set in 2015 
Records in this area have been kept since 1918 

Nakusp Area (Nakusp CS) 
New record of 36.1 
Old record of 35.3 set in 2015 
Records in this area have been kept since 1966 

Osoyoos Area (Osoyoos CS) 
New record of 42.0 
Old record of 40.9 set in 2015 
Records in this area have been kept since 1954 

Pemberton Area (Pemberton Airport CS) 
New record of 40.9 
Old record of 38.0 set in 2015 
Records in this area have been kept since 1908 

Penticton Area (Penticton A) 
New record of 40.5 
Old record of 37.8 set in 1925 
Records in this area have been kept since 1907 

Pitt Meadows Area (Pitt Meadows CS) 
New record of 40.0 
Old record of 33.9 set in 2015 
Records in this area have been kept since 1874 

Port Alberni Area (Port Alberni (AUT)) 
New record of 41.3 
Old record of 37.0 set in 2015 
Records in this area have been kept since 1900 

Port Hardy Area (Port Hardy A) 
New record of 30.2 
Old record of 21.3 set in 1995 
Records in this area have been kept since 1944 

Powell River Area (Powell River) 
New record of 36.6 
Old record of 30.0 set in 2015 
Records in this area have been kept since 1924 

Prince George Area (Prince George Airport Auto) 
New record of 36.4 
Old record of 29.4 set in 1928 
Records in this area have been kept since 1912 

Prince Rupert Area (Prince Rupert) 
New record of 30.0 
Old record of 20.4 set in 2013 
Records in this area have been kept since 1908 

Princeton Area (Princeton CS) 
New record of 41.4 
Old record of 38.5 set in 2015 
Records in this area have been kept since 1893 

Puntzi Mountain Area (Puntzi Mountain (AUT)) 
New record of 38.1 
Old record of 31.7 set in 2015 
Records in this area have been kept since 1959 

Quesnel Area (Quesnel Airport Auto) 
New record of 39.9 
Old record of 36.7 set in 1925 
Records in this area have been kept since 1893 

Sandspit Area (Sandspit Airport Auto) 
New record of 23.8 
Old record of 20.6 set in 2016 
Records in this area have been kept since 1945 

Sechelt Area (Sechelt Aut) 
New record of 39.6 
Old record of 30.6 set in 2015 
Records in this area have been kept since 1956 

Smithers Area (Smithers Airport Auto) 
New record of 37.1 
Old record of 29.9 set in 1992 
Records in this area have been kept since 1938 

Squamish Area (Squamish Airport) 
New record of 41.2 
Old record of 32.3 set in 2015 
Records in this area have been kept since 1960 

Summerland Area (Summerland CS) 
New record of 38.8 
Old record of 36.5 set in 2015 
Records in this area have been kept since 1907 

Tatlayoko Lake Area (Tatlayoko Lake RCS) 
New record of 37.1 
Old record of 31.3 set in 2015 
Records in this area have been kept since 1930 

Terrace Area (Terrace A) 
New record of 35.6 
Old record of 30.1 set in 2016 
Records in this area have been kept since 1912 

Trail Area (Warfield RCS) 
New record of 42.3 
Old record of 40.6 set in 2015 
Records in this area have been kept since 1928 

Vancouver Area (Vancouver Intl A) 
New record of 31.1 
Old record of 27.2 set in 1935 
Records in this area have been kept since 1896 

Vernon Area (Vernon Auto) 
New record of 41.5 
Old record of 37.3 set in 2015 
Records in this area have been kept since 1900 

Victoria Area (Victoria Intl A) 
New record of 37.7 
Old record of 29.5 set in 1995 
Records in this area have been kept since 1914 

Victoria Harbour Area (Victoria Gonzales CS) 
New record of 38.3 
Old record of 27.8 set in 1951 
Records in this area have been kept since 1874 

Victoria (Hartland) Area (Victoria Gonzales CS) 
New record of 38.3 
Old record of 27.8 set in 1951 
Records in this area have been kept since 1874 

Victoria (University of) Area (Victoria Gonzales CS) 
New record of 38.3 
Old record of 27.8 set in 1951 
Records in this area have been kept since 1874 

Whistler Area (Whistler - Nesters) 
New record of 40.2 
Old record of 34.8 set in 2015 
Records in this area have been kept since 1950 

White Rock Area (White Rock Campbell Scientific) 
New record of 36.9 
Old record of 29.2 set in 1995 
Records in this area have been kept since 1929 

Williams Lake Area (Williams Lake A) 
New record of 37.2 
Old record of 30.3 set in 2015 
Records in this area have been kept since 1960 

Yoho (National Park) Area (Yoho Park) 
New record of 33.3 
Old record of 31.1 set in 1925 
Records in this area have been kept since 1923 

Note: the temperature records reported here have been derived from a 
selection of historical stations in each geographic area that were 
active during the period of record.

Please note that this summary may contain preliminary or unofficial 
information and does not constitute a complete or final report.

End/PSPC

[rclab]

3 hours ago, Roger Smith said:

Without going into quite as much detail, the rank order of highest seven-day average minimum values at NYC is as follows: 

 1. 2013 __ 79.29 _______t25 1876 __ 75.71

 2. 1896 __ 79.14 _______t25 2018 __ 75.71

 3. 1988 __ 78.14 _______t27 1972 __ 75.43

 4. 2011 __ 77.86 _______t27 2020 __ 75.43

 5. 1908 __ 77.71 _______t29 1892 __ 75.29

 6. 1999 __ 77.43 _______t29 1911 __ 75.29

 7. 2006 __ 77.29 _______t29 1917 __ 75.29

t8 1905 __ 77.14 _______t29 1948 __ 75.29

t8 1906 __ 77.14 _______t29 1955 __ 75.29

10. 1980 __77.00 _______t29 1961 __ 75.29

t11 1993 __76.86 ______ t29 1982 __ 75.29

t11 2016 __76.86 ______ t29 2005 __ 75.29

t13 1898 __76.71 ______ t37 1884 __ 75.14

t13 1981 __76.71 ______ t37 1944 __ 75.14

15. 1995 __76.57 ______ t37 1973 __ 75.14

t16 1885 __76.43 ______ t40 1890 __ 75.00

t16 1952 __76.43 ______ t40 2015 __ 75.00

t16 1953 __76.43 ______ t42 1887 __ 74.86

t16 2010 __76.43 ______ t42 1983 __ 74.86

t20 1977 __76.29 ______t42 1984 __ 74.86

t20 2002 __76.29 ______ 45. 1870 __ 74.71

t22 1909 __76.14 ______t46 1901 __ 74.57

t22 1979 __76.14 ______t46 1907 __ 74.57

24. 2001 __75.86 ______t46 1938 __ 74.57

____________________________________________________

Most of the years with the top average maxima are in this list, but quite a few rather average cases (for daytime max) are sprinkled in with them, 

indicating that some cloudier or more humid spells were included where average max may have been closer to 90F. The frequency distribution is

somewhat skewed towards two intervals, the decade of 1901-10, and the recent past. As to the decade of 1901-10, it is possible that site interference

was playing a role as the station was moved into Central Park from a more urban setting around 1919-20 (or so I am told). 

So with that in mind, here's the distribution of these top 48 years for average minimum in summer heat. 

The following chart shows the ranks of the distributed cases: 

1869-1900 __ 2, t13, t16, t25, t29, t37, t40, t42, 45

1901-30 _____5, t8, t8, t22, t29, t29, t46, t46

1931-60 ____ t16, t16, t29, t29, t37, t46

1961-90 ____ 3, 10, t13, t20, t22, t27, t29, t29, t37, t42, t42

1991-2020 __ 1, 4, 6, 7, t11, t11, 15, t16, t20, 24, t25, t27, t29, t40

________________________________

So unlike the average maxima, these warmer summer nights are increasing in frequency and severity (note the depth of high ranked cases as well as the frequency).

The average for this statistic is 73.34. In the intervals above, the means are 73.4, 72.7, 72.7, 73.2, and 74.4 which is fairly consistent with urban heat island growth combined with site alterations to reduce direct effects. So this suggests that the warmer summer overnight spells are contrasting with recent cases of cooler summers at other times, such as 1992, 2000, 2003, 2004, 2007, 2009 and 2014. 

After some rest and reflection I will tackle the question of how overall means for these summer heat waves compare. I think it will be a blend of the two different patterns seen. 

 

Good morning, Don. Scanning the 24 official/unofficial temperature comparative list you posted I see an amazing 2 to  almost 19 degree (f) difference in the old and new records. With increased occurrence of heat wave events, permafrost melt awareness and frozen ground methane release, its past time to dial 911 Houston. Holding humanities feet to the fire will be too late to prove innocence three days later. As always ….

[LibertyBell]

I dont regard high min temps as "hot"- most of my most memorable heatwaves had high temps of 100 or above like  1991, 1993,  1995, 1999, 2010 and 2011.

[bluewave]

 

 

[donsutherland1]

1 hour ago, bluewave said:

 

 

Yesterday’s 104-degree reading in Seattle was more than 3.6 sigma above the mean June-August high temperature (1991-2020 base). It would be the equivalent of 108 degrees in Central Park or 109 degrees in Chicago.

[TimB]

SEA is currently running 7 degrees ahead of yesterday at the same time. PDX is +5.

[Vice-Regent]

The combination of longer days and peak sun angle seems to be driving this anomalous statistical anomaly. Keep in mind days are only roughly 12-13 hours long in the tropics. Not to discount warming elsewhere but this really shows the futility of thinking you can escape the heat so to speak or that north equals colder.