My wife, who is from northern Indiana where winters are fierce, gets very nervous every winter when the Mid-South has near-record warm weather like today with high temperatures flirting with 70 degrees. She always wonders if these spring-like days are due to global warming. I did a little research today and found that not only is 70 degrees in January not that unusual for south-central KY, it actually occurred twice as often during the first half of the 20th century as it did in the second half of the 20th century.
When most non-climatologists think about winter climate they often compare the current winter climate to that of their youth. For most older Kentuckians today, they remember the cold and snowy winters of the 1960s while many younger Kentuckians may remember the big snowstorms of the middle 1990s. So it’s no wonder whenever we have a stretch of 65+ weather in January that many people automatically assume that this is unnatural and must be related to global warming. However, when you look at the numbers, it becomes apparent that winters with high temperatures greater than 65 are actually quite common.
Using Bowling Green, KY data from 1894-2007, I first generated a list of all of the days with a high temperature greater than or equal to 65 degrees. After looking at a graph of these data, I noticed that most of these warm January days occurred prior to 1950. Since there are 114 years in the period of record, I then split the data by 1950 to compare the frequency of these warm days. Here is what I found.
Avg. # days with MaxT >64 F
1894-1950: 2.54/month
1951-2007: 1.58/month
This represents a decrease of nearly 40% since 1950. A t-test shows that there is a 97% chance that this result is not due to chance. I found similar results when I subdivided these warm days by bidecadal periods.
Avg. # days with MaxT >64 F
1901-1920: 2.70/month
1921-1940: 2.30/month
1941-1960: 2.75/month
1961-1980: 1.60/month
1981-2000: 0.85/month
2001-2007: 2.86/month
Again, the frequency of these warm January days (MaxT >64 degrees) since 2000 is consistent with what occurred during the first half of the 20th century.
When I did a similar analysis using only very warm days where the high temperature was 70 degree or greater the results were even more impressive.
Avg. # days with MaxT >69 F
1894-1950: 0.75/month
1951-2007: 0.32/month
Avg. # days with MaxT >69 F
1901-1920: 1.15/month
1921-1940: 0.45/month
1941-1960: 0.65/month
1961-1980: 0.30/month
1981-2000: 0.10/month
2001-2007: 0.86/month
The results show that 70 degree January days occur roughly half as often since 1950. Even the uptick in 70+ degrees days experienced since 2000 is less than that of the 20-year average from 1901-1920, long before any attributions to global warming can be made.
Finally, since January is only 1/3 of meteorological winter, I decided to repeat the analysis of 65+ and 70+ days for all of winter (Dec-Feb).
Avg. # days with MaxT >64 F
1894-1950: 8.18/winter
1951-2007: 6.35/winter
Avg. # days with MaxT >64 F
1901-1920: 7.80/winter
1921-1940: 9.65/winter
1941-1960: 7.70/winter
1961-1980: 6.75/winter
1981-2000: 5.30/winter
2001-2007: 6.71/winter
Surprisingly, the frequency of warm winter days (MaxT >64F) since 2000 is roughly the same as during the 1960s and 1970s, which is remembered by most people as a cold and snowy period.
Avg. # days with MaxT >69 F
1894-1950: 2.56/winter
1951-2007: 1.88/winter
Avg. # days with MaxT >69 F
1901-1920: 2.85/winter
1921-1940: 2.75/winter
1941-1960: 2.20/winter
1961-1980: 1.70/winter
1981-2000: 1.75/winter
2001-2007: 1.86/winter
Once again, the frequency of very warm winter days (MaxT > 69F) is much less than that of the pre-1950 period.
Finally, an interesting but not surprising outcome of this analysis is that I also calculated the averaged of warm (MaxT >64F) and very warm (MaxT >69F) winter days by month.
Avg. # days with MaxT >64 F (1894-2007)
December: 2.06
January: 3.11
February: 2.09
Avg. # days with MaxT >69 F (1894-2007)
December: 0.54
January: 1.22
February: 0.46
Warm winter days (MaxT >64 F) are 50% more likely to occur in January than either December or February even though climatologically, January is the coldest month of the year. Very warm winter days (MaxT >69 F) are more than twice (126%) as likely to occur in January than in the other winter months. This puts some statistical muscle behind the idea of the “January thaw” that is often discussed in weather folklore.