Skip to content

Daily Record Temperatures Affected By UHI – Part II

March 16, 2014
tags: ,

By Paul Homewood


I ran a post the other day on daily record temperatures in the Southeast of the USA.

Just to recap:

A record daily high means that temperatures were warmer on a given day than on that same date throughout a weather station’s history.


In other words, it does NOT include records set, but subsequently beaten. Also note that subsequent ties are included.


My exercise showed that, in the last decade, across rural stations there were more new daily low records than highs, but at urban stations, the opposite was true. This seemed to be clearcut evidence of UHI effect.

For my selection of rural sites, I used population data on the GISS database, (itself sourced from GHCN), which classifies all sites with less than 10,000 as rural. As I pointed out at the time, this still left small towns in the rural category, and I therefore suspected I was underestimating the UHI effect.

Steven Mosher correctly points out that the GISS data is, at best, crude, and Ronan Connolly suggested I also use the “brightness index” that GISS use to indicate whether a site is truly rural. This is based on satellite data, and allocates three categories – “A” for dark, “B” for dim, and “C” for bright.


So, from the original 18 stations I had identified as rural, using the brightness index, I have filtered the list down to seven sites. While this is a small number, they are well spread geographically:


Highland Home –AL

Lincoln – VA

Highlands – NC

Little Mountain – SC

Santuck – SC

Yemassee – SC

Lake City – FL    See UPDATE



The results are startling. The ratio of highs to lows, at rural stations in the last decade, is 0.45, but at non rural sites, the ratio is 1.48. It is clear evidence that the UHI effect is keeping night time temperatures higher.






As the table below makes clear, the big difference between rural and urban is in night time lows, rather than daytime highs.

  2001-10 Decadal Average
1911 to 2010
% of
Record Highs      
Urban 1046 1494 70
Rural 176 279 63
Total 1222 1773 69
Record Lows      
Urban 705 1344 52
Rural 389 261 149
Total 1094 1604 68




These are only daily records, so do they tell us anything about overall temperature trends? Well, CDIAC certainly think they do.In their introduction of the new interface, from which my data is drawn, they have this to say:


Changes through time in record high and low temperatures (extremes) are also an important manifestation of climate change (Sect. 3.8 in Trenberth et al. 2007; Peterson et al. 2008; Peterson et al. 2012). Meehl et al. (2009) found that currently, about twice as many high temperature records are being set as low temperature records over the conterminous U.S. (lower 48 states) as a whole. As the climate warms further, this ratio is expected to multiply, mainly because when the whole temperature distribution for a location or region shifts, it changes the "tails" of the distribution (in the case of warming this means fewer extreme cold temperatures and more extreme hot temperatures.


And back in 2009, NCAR’s news release on Gerald Meehl’s paper on daily record temperatures stated:


Spurred by a warming climate, daily record high temperatures occurred twice as often as record lows over the last decade across the continental United States, new research shows. The ratio of record highs to lows is likely to increase dramatically in coming decades if emissions of greenhouse gases continue to climb.

"Climate change is making itself felt in terms of day-to-day weather in the United States," says Gerald Meehl, the lead author and a senior scientist at the National Center for Atmospheric Research (NCAR). "The ways these records are being broken show how our climate is already shifting.


Significantly, NCAR also go on to say:


The study also found that the two-to-one ratio across the country as a whole could be attributed more to a comparatively small number of record lows than to a large number of record highs. This indicates that much of the nation’s warming is occurring at night, when temperatures are dipping less often to record lows. This finding is consistent with years of climate model research showing that higher overnight lows should be expected with climate change


The evidence, at least as far as the Southeast is concerned, is that the small number of record lows is a result of UHI, and not climate change.





GISS have an error on their database, showing Lake City as <10000. According to the 2010 census, it should be 12046.

I have therefore excluded it from my rural section, and amended the graph and table accordingly.

  1. John F. Hultquist permalink
    March 16, 2014 4:59 pm

    I seldom see “water” mentioned in the UHI context. The 7 sites you use here for rural all seem to be in locations where summer rain is common. However, many cities have grown dramatically in places where the high sun season has a lack of precipitation. Even Seattle, WA is dry in summer and there is extensive watering of lawns and other green spaces. See the chart here:
    Here is something similar for a station in VA:

    UHI effects are mostly said to be from the storage of energy in the built environment (concrete and such). Water has a higher specific heat than the solids of the built structures. Maybe this is not a significant contributor to the UHI effect but my point is that I haven’t seen it demonstrated one way or the other.

    Related to temperature records: Luboš Motl of “the reference frame” wrote on this topic about 2 years ago.

    He includes the same chart you used yesterday plus some other diagrams and a few equations. Math helps but it is readable.

  2. March 17, 2014 3:54 am

    Thanks, Paul. Good follow-up article.
    As an amateur astronomer I have noticed that light pollution is farther-ranging than the UHI effect. This is; temperatures drop faster than sky glow as you drive away from a city.
    See “Luminic Map of Florida (1996-97)”, at

  3. Brian H permalink
    March 17, 2014 5:42 pm

    Your criterion for “rural” needs to be dropped drastically. Stations in villages <1,000 show UHI.

    • March 17, 2014 5:53 pm

      I am hoping the Brightness Index will filter out anything not truly rural. The other problem is that GISS/GHCN don’t show pop if it is less than 10,000, and I am reluctant to rely on Wiki, etc.

      Even with the bar set high, the contrast between “rural” and “urban” is startling.

Comments are closed.

%d bloggers like this: