Of course the devil is in the details. It was brought to my attention that this paper shows that the number of record high days outnumber the record low days 2:1.
It was highly critized in both WUWT and World Climate Report.
The relative increase of record high maximum temperatures compared to record low minimum temperatures in the U.S
Gerald Meehl, NCAR, Boulder, CO; and C. Tebaldi, G. Walton, D. R. Easterling, and L. R. McDaniel
Summed over the first decade of the 21st century and averaged over the U.S., the observed ratio of record high maximum temperatures to record low minimum temperatures is about two to one. That there are more record highs than lows is not surprising since observed U.S. average temperatures have been increasing over this time period, and a shift in the distribution of temperatures would dictate a greater number of record high temperatures being set than record low minimum temperatures. But is the two to one ratio particularly unique or characteristic of a warming climate? It is shown that the two to one ratio of record highs to record lows is not unique but happens to be a product of the recent time period in the non-stationary warming climate. Records that were declining uniformly earlier in the 20th century following roughly a decay of 1/n have been declining less slowly for record highs than record lows since the late 1970s, and for the recent decade the ratio has grown to about two to one. A multi-member ensemble of model simulations of U.S. climate of the 20th and early 21st centuries shows a greater ratio of about four to one for this recent time period due to more uniform warming across the U.S. in contrast to observations that have been characterized by relatively greater warming in the western U.S. compared to the eastern U.S. Following an A1B emission scenario for the 21st century, the ratio of record high maximum to record low minimum temperatures, averaged over the U.S., is projected to continue to increase, with ratios of about 20 to 1 by mid-century, and roughly 50 to 1 by the end of the century.
This is nothing more than an accounting issue and has nothing to do with it getting hotter as you will see in this example.
Using Station 4333 I took the highest TMax for each month for each year and made a chart with months as columns and years as rows. Each cell contained the highest TMax for that month and that year. I then scanned down each month and looked for what would have been recorded as record temps. For each of the months this is what I found:
Record Hot Days
Total record breakers
Record Cold Days
Total record breakers
Here is what it means. In Jan, the first highest Tmax was in 1906 at 8.9C. Which was then broken in 1932 at 11.7C, and later broken again in 2005 at 11.8C.
Each month is like that. In all there are 37 record breakers. But notice all but 6 are before 1950. 17 are in the first decade of 1900. In spite of the claim that 2010 was the hottest on record, the record breaking summer temp hasn't been broken since 1913.
The same procedure was done for the lowest of TMin for record breaking lowest temps which you can see in the above chart.
Counting the number of record breaking temps in the month you can see the record cold ended before 1952. None since. But there have been record hot days since. So the record hot days have go on longer than the record cold days. But the record temps are all still way in the past. With TMin increasing, this is what would be expected. We just have not gotten as cold in the winters as we did at the beginning of the last century. But this does not mean we are getting hotter in the summer months. Not even close.
Now this was just on monthly record breakers, if this were done on a daily basis we would see the counts much higher, but the trend would be the same.
Yes, Meehl's paper dealt only with the last ten years, but since TMin has been increasing, and TMax decreasing, there will still be more record breaking TMax days than TMin days, especially in the winter months. This would be a consequence of the rising winter temps.
The other issue is the accounting, which plays a much bigger role in these "record breakers". As we start to keep records, the number of record breaking days starts to drop as more data is accumulated. But with 365 days, and 100 years, that 36,500 potential record breaking days for temps to go into. Thus record breakers will still occur due to random temperature variation even in a temperature regeme that is not increasing. Couple that with winter TMax and Tmin getting less cold, and you will be bound to hit record breakers for Tmax long after there are no more for Tmin. Even with summer Tmax dropping, random temperature "noise" will hit the odd record breaker simply because there isn't enough records kept so far to cover all the possible temperature ranges that can happen.
If we had 5,000 years of daily temperature data, it can be argued that in the last 1000 years there wouldn't be a single record broken, winter or summer, as all the possible temps that can be achieved would have been in the previous 4000 years.
Hence Meehl et al's paper doesn't mean anything by itself. The rest of the data trends must be also considered. And that trend is that summer TMax is dropping since 1900. And that is an inconvenient truth that the AGW faithful are desperately trying to ignore.