August 2025 Discussion-OBS - cooler than normal first week but a big comeback to warmer than normal for the last 2-3 weeks

[uofmiami]

 

[Roger Smith]

Don's top ten summers

2010 1966 1993 1995 1999 1980 1988 1949 2011 1994

Their ranks in my table

__ 2 __ 3 __ 11 ___ 24 __ 21 __ 14 __ 22 __ 06 __ 36 __ 30

LB favorites 1944 and 1983 finished 8 and 12. 

I'm not sure whether Don considered summers before 1949 in his top ten listing, or where in time his list begins, would be interesting to find out where some of my top ten (1876, 1906, 1908) finished in his list. But in general I think cool starts to June are the main determining factor in placing a given summer lower on my list, that part of summer rarely produces much of the heat content that Don's method was measuring; also, if September heat was included, none of that would affect my rankings. Even without Sep 1-3, 1953 ranked 29th.

The four worst modern summers were 2009, 2000, 1992 and 1996. 

[winterwx21]

0.52" here. Not bad. A much-needed watering for the vegetable garden. 

[Wxoutlooksblog]

1 hour ago, Roger Smith said:

Don's top ten summers

2010 1966 1993 1995 1999 1980 1988 1949 2011 1994

Their ranks in my table

__ 2 __ 3 __ 11 ___ 24 __ 21 __ 14 __ 22 __ 06 __ 36 __ 30

LB favorites 1944 and 1983 finished 8 and 12. 

I'm not sure whether Don considered summers before 1949 in his top ten listing, or where in time his list begins, would be interesting to find out where some of my top ten (1876, 1906, 1908) finished in his list. But in general I think cool starts to June are the main determining factor in placing a given summer lower on my list, that part of summer rarely produces much of the heat content that Don's method was measuring; also, if September heat was included, none of that would affect my rankings. Even without Sep 1-3, 1953 ranked 29th.

The four worst modern summers were 2009, 2000, 1992 and 1996. 

Surprised not to see 2002 and 1991 mentioned somewhere in the top ones.

WX/PT

[Roger Smith]

1991 28th and 2002 33rd in my list; bearing in mind there really isn't much differential from about 20th to 40th, and the older years get a boost of 2 F compared to anything recent. Without that boost, about six fall out of the top 20 and all these 20-30 ranked years move up that many ranks at least. 

I am somewhat conflicted about applying any urban heat correction, the positive is that it allows a better assessment of climate change from other sources, the negative is that people alive in those earlier years experienced actual temperatures shown in my tables and people nowadays 2 F warmer than shown. In between, like around the 1930s, the differential was closer to 1 F. 

Perhaps the annual average is 2 F but seasonal differences are not all exactly the same, and I should retool the comparisons based on seasonal ranges. Summer is probably the season I would reduce the boost, and winter I might increase it, leaving it as is for spring and autumn.

That would likely move 1876 down a few rungs, as well as 1906, 1908 and 1870. Oddly, while urban heat island increase is supposed to give a larger increase in overnight minima, the years making it into the top ten are known mainly for warm nights back in the early phase of NYC's urbanization. These years may have resembled 2025 more than the really hot summers of recent decades (a steady dull warmth). 

[Roger Smith]

NYC high of 70F was not quite a record low max (68F 1990, 2000). Today (21st) won't break the record low max either, since it was 59F in 2007 (tied with Aug 31, 1911 as lowest of all maxima in August). 

Record low minima for 21st to 23rd are 53 (1922), 52 (1895) and 51 (1923). Highs associated with them are in the low 70s F and not record low maxima. This indicates they were all intrusions of fall-like cooler air from the northwest and not like 21st-22nd 2007 held down by cloud and northeast winds. (max on 22nd 2007 was also a record low 65F).