News Section
Stories from Climate Central's Science Journalists and Content Partners

2012 Record Temperatures: Which States Led the Nation

Repost This

Summary

The summer of 2012 has been one for the record books in the lower 48 states. On the heels of the country’s warmest-ever spring, several record-breaking June and July heat waves kept the Southwest, Midwest and Atlantic Coast sweltering. July went on to become the all-time warmest month on record for the country. In fact, 2012 to-date has been the hottest year for the U.S. since instrument records began in 1895, and the summer was the third warmest summer on record.

The record-breaking heat has affected nearly every part of the country at some point this year, and so far there have been more than 28,000 daily high-temperature records broken or tied. This begs the question: which state was the biggest record-breaker in 2012? Or put another way: Which state had the most extreme heat?

Answering this question is not as simple as counting the number of records broken or tied in each state. If that were the case, the states with the most thermometers would almost always be the top record-breakers. When quantifying record-breaking heat by state, there are more factors to consider: How often do these stations set records? How long have records been kept for at each station? (It’s harder to break a record at a station with 100 years of data compared to one with 40.) How does this year compare to average? How many record-high temperatures compared to record lows were set in the state this year? To put it simply, finding the biggest record-breakers depends on how you look at the data.

Our top 10 record breakers list is based on two factors:

  1. Which states broke or tied the most high-temperature records (after accounting for the number of stations in each state, and the expected number of records based on the age of the stations), and;
     
  2. Which states had the largest disparities in the ratio between record high and record-low temperatures.

When weighing these factors, we considered the high-temperature records to be twice as important as the ratio of high-to-low temperatures in capturing the concept of record-breaking heat, although the results of all ways of calculating are presented in the interactive and are accessible below.

Number of Records | Records per Station | Observed Average vs. Expected | Record Highs vs. Record Lows

Our top 10 record-breaking states are:

1. Wisconsin
2. Iowa
3. Missouri
4. Indiana
5. Minnesota
6. North Dakota
    Illinois (tie)
8. Ohio
9. Colorado
10. Arkansas

Notably, the most blistering, record-breaking heat was concentrated in the middle of the country, an area that not coincidentally experienced record-setting drought this summer. The drought and heat interacted with each other in a positive feedback loop. The heat accelerated the drying of plants and soils, which sent temperatures climbing even higher and evaporated even more moisture from the land, thereby worsening the drought.

Extreme heat is closely tied to climate change, and this summer’s heat extremes left a global-warming signature in the data, particularly in the ratio of record-high to record-low temperatures.

Over the long-term, in the absence of climate change, the number of high-temperature records to low-temperature records would be, on average, close to 1:1. According to a 2009 study, as global average temperatures have increased in recent decades, the records’ ratio has shifted, too, with about two record highs set in the U.S. for every record low during the past decade. Numbers for individual years are usually more lopsided. But even accounting for such short-term swings, this year has been extremely unusual. So far this year:

  • 25 states had high to low-temperature record ratios of 10:1 or greater. Fourteen had a ratio of greater than 20:1, three states had greater than a 40:1 ratio, and none had less than a 1:1 ratio. Ohio topped the list with 49 high-temperature records for every record low. Nationally, through August 31, record highs were outpacing record lows by a 7 to 1 margin.

Although heat records tend to come in clusters when records fall during a heat wave, the average frequency of records broken so far this year is impressive.

  • Weather stations in the hottest states set records on average every two weeks from January through the end of August, a rate more than four times the historic average.

Number of Records

The simplest way of looking at the data is to see how many records were set in each state. Our Number of Records map illustrates the total number of high-temperature records broken this year in every state (through August 31).

Across the country, weather stations report the highest temperature reached each day. For a given calendar day, a daily “record” is set or tied if that temperature is higher than or equal to what it has ever been on that same day in any previous recorded year. By this measure, the state with the most records is simply the one where there were more stations with more days with record-high temperatures. In this basic counting method, states with more stations have an advantage.

The top 10 states with the most records set to date are shown in the interactive above under the Number of Records tab at the top. Texas has broken nearly 2,000 daily high-temperature records, and several Midwestern states have all broken more than 1,000 records this year.

Looking at the total number of records is a quick way to rank states, but it doesn’t provide an accurate analysis of which state saw the most unusual heat. The total number of records set in each state depends on how many weather stations exist in each state and also on the age of the stations involved (see below).

Records Per Station

In some states, there have been well over a thousand record-high temperatures set or tied this year. In others, there have been hundreds or fewer. While some of this difference depends on how extreme the temperatures were, the number of stations that keep records also matters. For example, Texas has 232 official weather stations that report daily high temperatures, but the small state of Rhode Island has only one. 

Our Records Per Station map illustrates the total number of high-temperature records broken or tied thus far in 2012 in each state (through the end of August) divided by the number of stations in the state, or the average number of records broken per station.

The top 10 states with the most records broken or tied per station are listed in the interactive above under the Records/Station tab at the top. The list is dominated by Midwestern states, and both Missouri and Illinois have seen an average of 16 records set per station this year. That’s roughly a new high-temperature record set every two weeks, from January through August, but the actual pace, of course, may well have been different, and likely so, since heat waves come in . . . waves, and it is likely that records get broken in consecutive days when those hot spells happen.

While looking at the number of records per stations levels the playing field among states with different numbers of stations, it does not account for the fact that in some states, official temperature stations have been operating for a lot longer than others. At these long-running stations it is harder to set a record. In the following section, we consider how the age of the stations influences how likely records were to be broken or tied this year.

Observed Average vs. Expected

Even if the climate weren’t changing we would still expect many new daily high-temperature records to be set each year in the U.S. That’s because there is always some probability that temperatures on a particular day will be as hot as or hotter than any of the other years they’ve been recorded at that station. In fact, the probability is 1/n, where n is the number of years that station has been recording temperatures. Thus the probability decreases with the age of the station: the longer temperature has been recorded at a station the more records have been broken over the years, making it more difficult for the current year to set a new record.

Since records are defined for each individual calendar day independently, the expected number of records broken thus far this year at a station is C/n, where C is the number of days that have already passed in 2012. We can add these values together for every station in a state to calculate how many records we would expect for a typical year (assuming that the climate has remained stationary). When we compare this expected number to the actual number of records set this year over all stations (see above), we get a ratio that shows how unusual the number of records is for this year. The higher the number is (i.e., greater than 1), the more records have been broken or tied this year compared to what we could expect if the climate weren’t changing. A state-by-state comparison of these differences in observed vs. expected values is illustrated in the Observed vs. Expected map.

Note that in this calculation we are not accounting for the spatial correlation among stations, rather, we are treating each station as independent of the others.

The interactive above shows the top 10 states with the most records set this year compared to what we should expect under the vs. Expected tab. Once again, Midwestern states make up the biggest record-breakers, and the top eight have all broken or tied at least three times as many records as would normally be expected.

Record Highs vs. Record Lows

The past three maps still only tell part of the story. Record highs only show the highest temperature a thermometer records in a day. In order to show that the heat has been consistently above average, we calculated the ratio of record highs to record lows.

Just as we expect there to be some record-high temperatures set or tied for every calendar day of the year (see above), we also expect there to be record-cold temperatures set. In an unchanging climate, the probability of a record-high temperature being set at a particular station is the same as a record-cold temperature. So, the expected ratio of record-high to record-low temperatures (including the average across a state or the entire country) is 1:1. In reality, when one considers a short time period, this expected 1:1 ratio could vary.

Click on the image for an expanded view. 

To see how unusual the number of record-high temperatures has been this year, we compared them to the number of record-low temperatures, which is shown in the Ratio To Low Temps map.

The interactive above shows the top 10 states with the biggest difference between the number of high-temperature and low-temperature records under the Ratio to Low tab at the top. Ohio has thus far seen 49 high-temperature records set for every one low-temperature record. Maryland and Wisconsin have both seen about 41 record highs for every one record low. All of the states in the top 10 have seen more than 20 record highs for every record low, and there hasn’t been a single state that has seen more low-temperature records than highs (or even a 1:1 ratio). Of note is Rhode Island, where there has not been a single low-temperature record set yet in 2012.

The data was retrieved on Sept. 4 2012 from the U.S. Records database maintained by the Climate Monitoring Branch of NOAA's National Climatic Data Center. 

Comments

By Jaelstruth (Albuquerque, NM 87121)
on September 11th, 2012

While US data is sufficient to tell a limited story, I’d like to know if this obvious heat belt in our Midwest extended up into Canada.  And, if so, how far.  All the way to the arctic?  I think seeing the heat records for all of North America would have far more impact on people as they start to realize, it’s planet wide, not just a hot summer in Wisconsin.

Reply to this comment

By Quentin (02481)
on September 12th, 2012

Mitt Romney thinks climate change is funny.  If you don’t go to :

www.notfunnymitt.com

Reply to this comment

By Peter Swinson (Loxley)
on September 12th, 2012

I am not convinced that man-caused global warming is settled. But let’s say for argument that it is. What is the next course of action? China and India will continue to grow and burn coal. The US already uses less than the 70’s and 80’s per capita. Anyone that works to change the behavior of the West is a liar or a fool. If you think it is going to get warmer, prepare for warmer weather. All the other Al Gore-solar panel-windmill-shit is stupid.

Reply to this comment

By V Antonelli (MA)
on September 14th, 2012

Good article!
FYI: You used “begs the question” incorrectly. It doesn’t mean that. Google it.

Reply to this comment

By Jay (Texas!)
on February 6th, 2013

“Global warming” has long since been proven a hoax.  This report may be, probably is, accurate as far as it goes. After all, careful selection can make statistics say anything you want.  For example, this report claims to be for record highs “this year” but, if you look at the fine print at the bottom of the chart it notes the period reported was only for “January through August 2012.  That only records 3 cold months to 5 warm months and actually, in the South February and especially March are very apt to produce higher temperatures.  That leaves out several months that might be expected to produce record lows temperatures.”  This report also leaves out reports of record lows in Europe and China.
So what we have here in a nut shell is yet another on the very many cases of typical liberal obfuscation or, in a word, “bald faced lies.” (OK, 3 words.) 
I am not going to get worked up about an issue that is proven false and that it’s supporters have to falsify the record to try and make it sound real.  The truly sad thing is there are enough over-medicated and uneducated mush brains in this country (USA) to buy into this guff.

Reply to this comment

Name (required):
Email (required):
City/State/Zip:
Enter the word "climate" in the box below:

[+] View our comment guidelines.

Please note: Comment moderation is enabled. Your comment will not appear until reviewed by Climate Central staff. Thank you for your patience.