Thursday, 10 September 2026

Addendum to ‘How unusual were US winters in the Carter and early Reagan eras? An analysis using extended winter temperatures’

In ‘How unusual were US winters in the Carter and early Reagan eras? An analysis using extended winter temperatures’ I attempted to see if the period was as unusual as thought by authors Karl, Livezey and Epstein in their ‘Recent Unusual Mean Winter Temperatures Across the Contiguous United States’. As I suspected, when we either:
  1. extend the winter to a November to March or October to March period, or
  2. in the Midwest at least, extend the series back to the 1870s
we find that the period form 1975/1976 to 1983/1984 is less abnormal than made out by the article. Point 2) may not apply to the South or the West: data for both regions in the middle 1870s are very sparse. What data exists, illustrated by the table below for Galveston in Texas and Wilmington in North Carolina, suggest that the run of extreme winters over the Midwest in the late 1870s and 1880s extended to the West South Central states [Texas, Oklahoma, Arkansas, Louisiana] but not to the East South Central or South Atlantic states, nor to west of the Continental Divide.
 Galveston, TexasWilmington, North Carolina
Winter (˚C)σOctober to March (˚C)σWinter (˚C)σOctober to March (˚C)σ
1871/1872 6.867-1.15210.083-1.418
1872/18737.833-0.58610.400-1.139
1873/187413.23+0.1415.97+0.039.333+0.29311.567-0.111
1874/187512.00-0.6515.37-0.557.600-0.72211.017-0.596
1875/187615.53+1.6116.98+1.0111.233+1.40612.300+0.535
1876/187710.60-1.5514.17-1.707.200-0.95610.133-1.374
1877/187812.17-0.5515.22-0.699.167+0.19512.600+0.799
1878/187911.20-1.1715.38-0.537.267-0.91711.067-0.552
1879/188015.60+1.6617.57+1.5712.733+2.28414.167+2.179
1880/188110.53-1.6013.32-2.527.400-0.83910.350-1.183
1881/188215.23+1.4217.67+1.6711.533+1.58214.117+2.135
1882/188312.13-0.5715.67-0.269.833+0.58612.267+0.505
1883/188412.80-0.1416.45+0.5010.933+1.23013.5171.607
1884/188511.60-0.9115.00-0.909.300+0.27412.033+0.300
1885/188611.17-1.1914.43-1.447.367-0.85910.533-1.022
1886/188713.63+0.3916.22+0.279.000+0.09811.583-0.097
1887/188812.20-0.5315.13-0.779.067+0.13710.950-0.655
1888/188913.07+0.0315.65-0.278.367-0.27311.183-0.449
1889/189018.17+3.3118.12+2.1014.000+3.02613.950+1.988
1890/189113.67+0.4216.10+0.1610.167+0.78112.250+0.491
1891/189212.93-0.0515.35-0.569.267+0.25410.900-0.699
1892/189313.17+0.1016.32+0.377.933-0.52710.733-0.846
1893/189414.03+0.6516.53+0.5810.233+0.82012.817+0.990
1894/189510.87-1.3814.87-1.036.733-1.23010.467-1.080
1895/189612.87-0.1015.27-0.648.433-0.23411.083-0.537
1896/189712.63-0.2516.22+0.278.400-0.25411.867+0.153
1897/189813.97+0.6116.93+0.969.167+0.19512.567+0.770
1898/189910.37-1.7014.07-1.807.933-0.52711.217-0.420
1899/190011.70-0.8515.37-0.557.967-0.50711.133-0.493
1900/190113.33+0.2016.57+0.617.300-0.89811.483-0.185
1901/190211.20-1.1715.10-0.806.567-1.3279.883-1.594
1902/190312.27-0.4815.98+0.059.400+0.33213.067+1.210
1903/190412.67-0.2315.67-0.265.900-1.7189.633-1.815
1904/190510.30-1.7514.80-1.096.000-1.65910.000-1.492
1905/190611.43-1.0214.97-0.938.767-0.03911.317-0.332
1906/190716.23+2.0618.03+2.029.033+0.11712.217+0.461
1907/190813.07+0.0316.13+0.198.233-0.35111.317-0.332
1908/190914.43+0.9116.55+0.5910.967+1.25012.817+0.990
1909/191011.57-0.9316.32+0.377.800-0.60511.500-0.170
1910/191115.33+1.4917.62+1.629.033+0.11711.350-0.302
1911/191211.00-1.3014.07-1.807.467-0.80010.967-0.640
1912/191312.77-0.1615.53-0.3911.167+1.36713.050+1.196
1913/191412.83-0.1215.40-0.518.700-0.07810.867-0.728
1914/191511.53-0.9514.32-1.568.700-0.07810.883-0.713
1915/191614.57+0.9917.55+1.559.133+0.17612.000+0.270
1916/191713.97+0.6116.47+0.519.700+0.50812.350+0.579
1917/191811.40-1.0415.12-0.796.100-1.6019.867-1.609
1918/191912.33-0.4415.25-0.669.733+0.52712.383+0.608
1919/192013.67+0.4216.72+0.757.933-0.52712.100+0.359
1920/192114.10+0.6916.62+0.669.567+0.43012.850+1.019
1921/192214.07+0.6716.78+0.829.400+0.33212.483+0.696
1922/192315.67+1.7017.08+1.119.867+0.60512.500+0.711
1923/192412.73-0.1815.05-0.859.533+0.41011.383-0.273
1924/192513.10+0.0516.83+0.8710.067+0.72312.450+0.667
1925/192611.80-0.7814.72-1.178.800-0.01911.000-0.611
1926/192715.47+1.5717.38+1.3910.900+1.21112.850+1.019
1927/192812.03-0.6316.47+0.519.367+0.31312.383+0.608
1928/192912.17-0.5515.92-0.029.233+0.23412.533+0.740
1929/193012.13-0.5714.60-1.2810.233+0.82012.267+0.505
1930/193112.87-0.1014.97-0.938.333-0.29310.533-1.022
1931/193215.70+1.7217.82+1.8113.667+2.83114.617+2.576
1932/193313.43+0.2715.63-0.2911.467+1.54313.117+1.254
1933/193415.43+1.5517.58+1.599.233+0.23411.683-0.009
1934/193514.00+0.6317.57+1.578.600-0.13612.333+0.564
1935/193611.43-1.0215.57-0.356.233-1.52310.933-0.669
1936/193713.90+0.5715.58-0.3411.467+1.54313.067+1.210
1937/193813.63+0.3916.45+0.509.133+0.17611.917+0.197
1938/193914.17+0.7416.78+0.8210.933+1.23013.283+1.401
1939/194011.30-1.1014.77-1.126.700-1.24910.300-1.227
1940/194113.73+0.4615.85-0.088.333-0.29310.900-0.699
1941/194212.57-0.2915.62-0.318.167-0.39012.350+0.579
1942/194313.87+0.5416.53+0.589.067+0.13712.133+0.388
1943/194413.47+0.2915.78-0.159.200+0.21511.667-0.023
1944/194513.43+0.2716.82+0.858.100-0.42911.950+0.226
1945/194612.87-0.1016.35+0.408.533-0.17612.317+0.549
1946/194712.30-0.4615.23-0.679.5330.41012.083+0.344
1947/194811.50-0.9815.38-0.537.733-0.64411.867+0.153
1948/194914.33+0.8416.60+0.6412.7332.28414.050+2.077
1949/195016.70+2.3618.00+1.9912.4672.12813.750+1.812
1950/195113.27+0.1616.42+0.468.200-0.37111.483-0.185
1951/195215.93+1.8717.27+1.2810.633+1.05412.617+0.814
1952/195313.57+0.3516.07+0.139.733+0.52711.933+0.212
1953/195414.07+0.6716.23+0.2910.067+0.72312.150+0.403
1954/195513.77+0.4816.58+0.627.633-0.70311.250-0.390
1955/195614.03+0.6516.48+0.537.867-0.56610.917-0.684
1956/195716.00+1.9117.70+1.7011.100+1.32812.733+0.916
1957/195812.30-0.4614.93-0.966.500-1.3669.717-1.741
1958/195911.97-0.6815.25-0.667.633-0.70310.983-0.625
1959/196011.93-0.7014.45-1.438.267-0.33210.567-0.992
1960/196111.70-0.8515.83-0.107.133-0.99611.317-0.332
1961/196213.73+0.4615.950.029.167+0.19511.367-0.288
1962/196311.00-1.3015.33-0.586.133-1.58110.500-1.051
1963/196410.53-1.6015.07-0.847.200-0.95610.800-0.787
1964/196513.23+0.1415.57-0.359.400+0.33211.817+0.109
1965/196612.67-0.2316.13+0.198.133-0.41011.433-0.229
1966/196712.57-0.2916.17+0.229.200+0.21512.450+0.667
1967/196812.00-0.6515.10-0.807.167-0.97610.667-0.904
1968/196913.27+0.1615.52-0.406.900-1.13210.283-1.242
1969/197012.63-0.2515.57-0.356.233-1.52310.117-1.389
1970/197114.80+1.1416.18+0.248.633-0.11711.417-0.243
1971/197215.43+1.5517.98+1.9711.033+1.28913.267+1.386
1972/197311.83-0.7615.27-0.648.867+0.02012.067+0.329
1973/197414.30+0.8217.93+1.9211.567+1.60114.050+2.077
1974/197514.37+0.8716.67+0.7010.500+0.97612.183+0.432
1975/197613.50+0.3116.42+0.469.967+0.66413.567+1.651
1976/197710.83-1.4013.40-2.445.900-1.7189.617-1.829
1977/197810.00-1.9414.28-1.595.933-1.69810.017-1.477
1978/197911.17-1.1915.37-0.557.467-0.80011.333-0.317
1979/198012.63-0.2515.22-0.697.533-0.76110.750-0.831
1980/198112.10-0.5914.83-1.066.133-1.5819.450-1.976
1981/198212.50-0.3316.22+0.277.533-0.76110.533-1.022
1982/198313.40+0.2415.97+0.038.200-0.37111.250-0.390
1983/198411.17-1.1915.57-0.358.067-0.44911.267-0.376
The table above also suggests that, like in the Midwest, extreme winters in the South may cluster over different periods than for the CONUS as a whole, and that fluctuating extreme winters over the wholse CONUS may in particular regions be only extremes of one type. Note the very warm winters in the South Atlantic, represented by Wilmington, in the early 1880s and the long run of cold winters in the South from 1976/1977 to 1983/1984.

It can also be seen that abnormally cold or warm winters over the entire contiguous US do not necessarily affect the South at all. This occurred in 1916/1917, 1928/1929 and 1978/1979 (cold) and 1952/1953, 1953/1954 and 1975/1976 (warm).

Another point I wished to study in this addendum is the extended winter of 1919/1920. According to the table in my previous blog post the October to March period of 1919/1920 is 1.21 standard deviations colder than the 1896 to 1984 mean, although the actual winter from December to February was only about 0.6 standard deviations colder.


It can also be noted that April 1920 is the coolest over the contiguous US since at least 1895. The map however shows that, unlike 1917/1918 and 1976/1977, the extreme cold was over Canadian border regions, with the result that coal demand was not nearly so great as in 1917/1918. The main reason why this season appears abnormal only in the October to March season is that October 1919 was record cool over large areas of the Northwest, and record heat over the South could not compensate at a national level:

Standardised temperature anomalies for October 1919. Note the massive (over three standard deviations!) opposite-sign anomalies over the South and the Northern Great Plains

Another case omitted from the main section was the winter of 1906/1907. When extended to a November to March period, it is anomalously warm overall in spite of severe cold in the far north and New England, due to an exceptionally warm March. Between 1890 and 2016, the Southern Plains and Gulf Coast never experienced so warm an extended winter.


This winter is better remembered for the extremely persistent and heavy precipitation over most of the West, culminating in major flooding in March 1907 and a record annual flow on the Colorado River. This extremely mild winter outside the northern tier was followed by the “year without a summer” during the ensuing six months. In fact, there were large areas in the Midwest and South where March 1907 was hotter than April 1907. A similar anomaly occurred in the UK where in some places March was the sunniest month of the entire year.

No comments: