6 May 2017

Using Basic Algebra To Develop An Extreme Temperature Index

Posted by Dan Satterfield

See here for an explanation.

This is a cross-post from my friend and fellow meteorologist Guy Walton.

Guy is working on an extreme temperature index that will take into account not just the magnitude of the record but weight this by the length of the record. Breaking a record at a site with 50 years of data is not as important as doing the same in a place with 135 years of data. The next step is to do some statistical analysis to see how valid the equation is, and this is an excellent example of using maths to answer some basic questions about our environment and to quantify how it is changing.




In previous posts on this site and on Weather Underground I introduced the weather and climate world to the concept of an Extreme Temperature Index. Let’s apply the current index algorithm to reports from April 2017 in the National Center for Environmental Information’s surface records site and see what the most extreme event and location was. Keep in mind that these are not the hottest and coldest records, but the records that differed farthest from the norm, climatologically. Drop me a note if you think you have spied a more extreme record and its location than what is presented here for April.


and: http://www.guyonclimate.com/2017/03/24/extreme-temperature-index-some-calculations-and-comparisons/


The current Extreme Temperature Index is defined by:


In the ETI equation, the difference between the old and new record is the most predominant factor,  so I have selected a record low minimum and record high maximum with the highest T sub r.

The big cold winner was Cuthbert, GA, which broke their old record low minimum by 14 degrees. Oklahoma City, OK, also broke their record by 14 degrees, but had a much shorter period in which that station had been keeping records (Period of Record…POR or n). I took the norms, or average high and low for the date the new record was set, for April 29th, from the closest large major city, Columbus, GA:

Now for the hot record. Tully Lake, MA broke their record by 15 degrees. They got up to 90 on 4/12. I’m using the norms from Worchester, MA, the closest large city:


There is no surprise that the ETI top value for the two coldest and hot records based mainly on the difference between the new and old records goes to the hot record since April 2017 was a warm month, overall. So far, the total of record high maximums for April were 989 vs. only 213 record low minimums.

The top ETI prize goes to Tully Lake, MA for April 2017. I am sure that the residents there were shocked to see mid-summer warmth there on April 12th. Carbon pollution should lead to future high ETI values.

The Climate Guy

Note: Guy writes a blog on climate and also keeps track of the running totals of record highs vs record lows in the U.S. April was the 29th consecutive month with record highs exceeding record lows. Another indication of our warming climate. You can see the latest data he sent me below. Guy is also working on a kids book about thermodynamics and weather; Something that no one has really done before. I think it is sorely needed, and you can read more about that on his blog, along with some great artwork that will accompany it.

Note: that we had 989 record highs vs. 213 record lows in April 2017. An extreme month like this isn’t out of the realm of chance, but 29 months in a row? That’s something that should set off alarm bells. It has in the science community, but a large percentage of the public still just does not “get it”.