Monthly Temperature Records Perspective

My last post dealt with record high temperatures set on a daily basis. In Atlanta, for example, we tied the record high of 80 on the first of March, which would count as one tally in my daily records database. Meteorologically it would be much more difficult to tie or break the all-time record for March in Atlanta, which was 89 set on 3/23/95. Monthly records are the subject of this discussion. I will present my tallies of monthly record counts back to 1920 again using the NCEI database. I like the following Threadex site to look up records for individual stations:

As planetary averages warm due to carbon pollution it should become more difficult to set cold records, particularly monthly records, or some of the coldest of the cold records. Indeed, the decadal chart from my database for monthly records is showing this trend:


Notice that so far for this decade ratios have exceeded those of the 2000’s and yes the 1930’s. Contrarians often hark back to the 1930’s as a “warmer or just as warm” period in the United States as the current or last decade….. not true.

Also, as one might expect in a warming world, the decadal ratios for monthly records are now much  higher than for ratios of daily records depicted here:

Many of my meteorology and climate friends marveled at how many monthly maximum temperature records were tied or fell in February 2017.  For the first time over 700 monthly records were set across the lower fourth eight states while there were 0 monthly record minimums. The lone monthly record low was set at Cold Bay, Alaska of -6F on the 2nd. As far as averages go, February 2017 was in the top five anomously warm months according to Climate Central:

So, where, statistically, did last month fall as far as the shear number of monthly records set? After updating my “Records Genome” database for monthly records (shown at the bottom of this post) I endeavored to create the following chart similar to what I made in my prior post: Mathematically, monthly records can’t be judged too well using ratios since the warmest tallies have ratios approaching infinity. (1000 to 0 has the same ratio as 100 to 0, which is infinity.) So, I have opted to depict the top 15 months with the biggest differences in totals between set monthly high and low records. Drop me a note if you think my rankings are in error… under the crucible of science, changes often need to be made.

Somewhat surprisingly there were only three months from the Climate Central chart (starred on my chart) in which top averages lined up with the highest ranked months of the total difference of monthly maximum to monthly minimum records. Six out of the top 15 years occurred in the 1930’s mainly due to the intensity of the Dust Bowl drought. As expected, another six months occurred since 1990. What we see here (number of reports) from the 1930’s is impressive considering that there are many more reporting stations across the U.S. now in the 2010’s. February 2017 only came in at number twelve on this list, but you’ll notice that last month had the fewest number of reported record low minimums.

Also of note, intense heat waves don’t necessarily need to be over widespread areas of the U.S. to produce numerous monthly records. Such was the case in April 1980 when the Plains began to bake while the rest of the country had near average temperatures; thus, the departure of only +0.39F.

We will take a look at some brand new charts on my next post.

The Climate Guy


The rankings are for the lower 48 states with the warmest ranking since 1895 being 122 and 1 being the coldest. Blue colors represent cold months and red warm. Months that had near 1 to 1 ratios are colored black representing neutral months. Time stamps for when I updated each dataset are located in the upper left hand corner of each chart.




Leave a Reply

Your email address will not be published. Required fields are marked *