Saturday, 24 May 2014

How Not to Measure Temperature Part 95 – New Temperature Record of 102° in Wichita, but Look Where they Measure it


"On Sunday, May 4th, the temperature peaked at 102° at Wichita’s official NWS site Mid-Continent International Airport. Not only was this the hottest temperature ever measured during the month of May (previous record was 100° set on four different occasions) but was the earliest 100° reading on record (previous was May 9, 2011) and hottest reading so early in the season by a whole month: since June 4, 1933 when 102° was also observed. The city is off to its 2nd driest start of the year on record as well. Temperature and precipitation records for Wichita date back to 1888."

But, as I’ve pointed out many times, airports are one of the absolute worst places to measure temperature for climatic records. Not just because the airports themselves are massive heat sinks of asphalt and concrete, but also because the ASOS weather station system is known to be highly unreliable and prone to giving false high temperature readings like in Honolulu, Tucson, and more recently in Tacoma, where after fixing the ASOS temperature sensor, temperatures dropped to normal levels.

Have a look at where this thermometer is located for Wichita:

I think the asphalt pad right next to the ASOS (just north) is a nice touch, don’t you?
http://wattsupwiththat.com/2014/05/05/how-not-to-measure-temperature-part-95-new-temperature-record-of-102-in-wichita-but-look-where-they-measure-it/

If you go to the link you can see that the airport today is now surrounded by houses, although not completely, when compared to 1956. There will be impacts from the Urban Heat Island Effect! And the difference between city and country temperatures can be measured in whole degrees!

This is a fundamental problem when it comes to tracking climate trends. When your raw data is unreliable then the conclusions you draw will most likely be faulty.

There is an argument that climate scientists compensate for the Urban Heat Island Effect but unless they are using adjustments that factor in degrees of temperature, rather than tenths of degrees, their temperature tracking estimates will be too high.

[Posted at the SpookyWeather blog, May 24th, 2014.]

No comments: