Changes in Irish mean temperatures in recent decades

M.T. Cranium

This report follows up on some great work done by forum member Deep Easterly who recently worked up a new statistic that we use in the Boards forecast contest, the "Irish mean temperature" (a name that was chosen after a poll) which averages the means at five representative inland stations, namely Shannon, Oak Park, Casement, Mullingar and Claremorris.

Those stations were not in each case always recording temperatures over the past fifty years, but nearby stations were in some cases. After studying any station variations and the question of the climatological day, DE came up with the following three-decade means starting with 1961-90 and ending with the very recent 1981-2010.

Using those data, we have a good record of recent changes in Irish mean temperatures over the past half century.

The figures are presented below:

Irish Mean Temperature (IMT) values for 30-year periods

PERIOD ..... JAN .. FEB .. MAR .. APR .. MAY .. JUN .. JUL .. AUG .. SEP .. OCT .. NOV .. DEC .. ANNUAL

1961-1990 .. 4.6 .. 4.7 .. 6.2 .. 7.9 .. 10.4 .. 13.3 .. 15.0 .. 14.7 .. 12.7 .. 10.2 .. 6.6 .. 5.3 ... 9.30
1971-2000 .. 5.0 .. 5.3 .. 6.6 .. 8.1 .. 10.7 .. 13.3 .. 15.5 .. 15.1 .. 12.9 .. 10.1 .. 7.0 .. 5.7 ... 9.59
1981-2010 .. 5.2 .. 5.3 .. 6.9 .. 8.5 .. 11.1 .. 13.7 .. 15.2 .. 15.3 .. 13.2 .. 10.2 .. 7.3 .. 5.4 ... 9.77


Note that the overall trend has been upward in these periods, with a few exceptions such as July which peaked with the 1971-2000 value, but I'm pretty sure that the four-year period 2007-10 has shown a general levelling off or even decline in the values in the most recent data, so in post-analysis it may turn out that perhaps 1977-2006 would be the warmest 30-year period in modern times.

Thanks to Deep Easterly for working on these data sets and providing them to us for general interest and specific use for our contest.

dirtyghettokid

it's very interesting to see the summer period more or less remaining the same, but the major differences being in the winter/spring period

Godge

This years figures (from the monthly contests, hope I go them right) in the first column compared to the 1981 -2010 figures.


Jan 3.3 - 5.2
Feb 6.8 - 5.3
Mar 6.6 - 6.9
Apr 11.0 - 8.5
May 11.2 - 11.1
Jun 12.2 - 13.7
Jul 14.4 - 15.2
Aug 13.8 - 15.3

So the cold winter from November and December carried on into January before a mild spring and a cool summer. To my uneducated eyes though, it looks as if this year is a good bit cooler overall so far than the 1981-2010 average.

Of course, by the standards of 1971-1990, it is a warm year, made so by the spring weather.

token56

It would be interesting to see the standard deviation for each month for each period too. Dont think they would vary significantly but would just be interesting to see if any did.

Pangea

Slightly off topic, I was just taking a look through these and noticed how temperature records go back to the 1800s.
http://met.ie/climate-ireland/extreme-mintemps.pdf

Since Northern Ireland was only created in 1921, Did Met Eireann have weather stations and recordings in the region now known as Northern Ireland?

Yahew

Jebus 13.8 for August Just 2.8 above April.

octo

Nice work, but just curious, why did you not include Malin Head? It has a stable and consistent record. Otherwise your selection excludes the northern third of the country.

Deep Easterly

octo wrote: »

why did you not include Malin Head? It has a stable and consistent record. Otherwise your selection excludes the northern third of the country.

M.T Cranium wrote:

the "Irish mean temperature" (a name that was chosen after a poll) which averages the means at five representative inland stations, namely Shannon, Oak Park, Casement, Mullingar and Claremorris.

The IMT was an index created for the boards competition. We did use all the reporting stations a few years back but due to the loss of Rosslare, Clones and Kilkenny in recent times it was decided to narrow down the number of stations used to make up a national average, mostly for practical purposes really as it was easier to do a review the daily means at each of these stations; it also seems that by using just the data from these 5 stations, we get a fairly good idea of how a more 'national' average would work out on a monthly basis.

I would have rather included Ballyhaise in the IMT than Malin Head as it located well inland, but unfortunately there are no temperature means for this station available on the monthly data page While Malin Head does have consistent record, I don't think its daily/monthly/yearly data would represent the greater northern third of the country as it has one of the lowest diurnal temperature ranges in the country wheras Ballyhaise could well have one of the highest.

I suppose the IMT is our version of the CET in the UK but one advantage we have is that Ireland is fairly square in shape as opposed to the big long string of misery the geographical shape of the UK is so our IMT would represent a greater proportion of the country. It isn't perfect but as good as we can do with the data that is available to us to use.

Richard

Pangea wrote: »

Slightly off topic, I was just taking a look through these and noticed how temperature records go back to the 1800s.
http://met.ie/climate-ireland/extreme-mintemps.pdf

Since Northern Ireland was only created in 1921, Did Met Eireann have weather stations and recordings in the region now known as Northern Ireland?

All of Ireland was covered by the UK Met Office until the 30s. At that point, the Irish Free State established the "Irish Meteorological Service". The name "Met Eireann" only dates From 1996.

So no, neither Met Eireann, nor its predecessor, had recordings from NI. They probably have access to the historical UK Met Office data for the whole island, however.

http://www.met.ie/about/our-history.asp

octo

Deep Easterly wrote: »

The IMT was an index created for the boards competition. We did use all the reporting stations a few years back but due to the loss of Rosslare, Clones and Kilkenny in recent times it was decided to narrow down the number of stations used to make up a national average, mostly for practical purposes really as it was easier to do a review the daily means at each of these stations; it also seems that by using just the data from these 5 stations, we get a fairly good idea of how a more 'national' average would work out on a monthly basis.

I would have rather included Ballyhaise in the IMT than Malin Head as it located well inland, but unfortunately there are no temperature means for this station available on the monthly data page While Malin Head does have consistent record, I don't think its daily/monthly/yearly data would represent the greater northern third of the country as it has one of the lowest diurnal temperature ranges in the country wheras Ballyhaise could well have one of the highest.

I suppose the IMT is our version of the CET in the UK but one advantage we have is that Ireland is fairly square in shape as opposed to the big long string of misery the geographical shape of the UK is so our IMT would represent a greater proportion of the country. It isn't perfect but as good as we can do with the data that is available to us to use.

I see your point. As far as I know, Ballyhaise is only around since 2003, so there are no long term means available.

Its a shame that Clones, Birr and Kilkenny close down without replacement.... all those long term records.

Pangea

Richard wrote: »

All of Ireland was covered by the UK Met Office until the 30s. At that point, the Irish Free State established the "Irish Meteorological Service". The name "Met Eireann" only dates From 1996.

So no, neither Met Eireann, nor its predecessor, had recordings from NI. They probably have access to the historical UK Met Office data for the whole island, however.

http://www.met.ie/about/our-history.asp

Interesting, thanks for that Richard.

octo

Have you considered Armagh Observatory?

Deep Easterly

octo wrote: »

Have you considered Armagh Observatory?

Not a bad idea. I'd say it takes a few weeks for the previous month's data to be made available. It all depends if M.T.C is willing to wait a few days/weeks after the start of each month to finalize the IMT value.

M.T. Cranium

It might make for a better statistic in scientific terms, but it would be deadly for the contest to have to wait that long to score the previous month -- perhaps the solution would be to include Ballyhaise after comparing it to Armagh for some overlapping period to see if the two are similar, and getting a reasonable long-term average for Armagh to apply to Ballyhaise.

If this were done, I would change over the contest IMT means in January rather than some time later this year. My guess is that the IMT would fall slightly in the winter half year as Ballyhaise strikes me as being colder than Mullingar on average then, but the IMT would remain about the same in other months with this station added in. Does that sound right?

Let's say Ballyhaise was 1.2 C colder than the existing IMT (which seems a bit on the high side) then the IMT would drop by 0.2 if we included it for that particular month. So we're not really talking about a huge change in the numbers by adding one station.