Apparently the Meteorological Service of Britain does it still differently and does not use automatic measurement but takes manual readings twice a day and creates the daily average by ( M a x + M i n ) / 2 {\displaystyle (Max+Min)/2} {\displaystyle (Max+Min)/2} and the argument is that comparison with observations from before the computer era must be made.
HansR - Revision as of 17:06, 10 April 2021
The text above is an incorrect summary. I present the full facts below, and leave you to decide how to summarise for your purposes.
The main problem is that you confuse 2 totally different topics.
- How measurements are made
- How information is presented to a world-wide audience
Britain's Meteorological Office has closed most of the weather stations it used to operate. It has fully automated the few sites it still operates directly.
Most of the sites providing measurements reported by the Met Office are operated by private organisations. These include METAR reporting from airfields, some of these are tiny and do rely on manual readings. Larger airports use automated instrumentation, but their METAR are supplemented by professional observers in critical weather such as fog. Other organisations reporting weather to the Met Office include botanical gardens and other research organisations. Some of these operate the traditional way, and still depend on manual measurements, others are automated.
If you get access to the daily records on the Met Office web site, you see tables by day, month, and year. These present temperature in four columns:
- Maximum for period
- Minimum for period
- Average based on measurements throughout day (when available), or throughout the extended period
- WHO average calculated from maximum and minimum for each day, (for longer periods, all maximums and all minimums are summed, and then divided by number of items)
Thus for anybody only interested in UK, what you call the true average is provided for those stations that can provide it. For all stations, the WMO formula is available.
For the global and the historic readers, the UK Met Office does what the WMO dictate, i.e. present averages in a way consistent with the past, and in a way that is comparable with manual measurements. That is what is needed for climate comparisons that are expressed comparing 30 year periods, so cover time before widespread automation. This basis is also required for scientists studying Climate Change where records included may have been made a century or more ago, or extrapolated from soil columns or tree rings, that do not convey the level of "trueness" that you strive for.
(max + min) / 2 average usage
I disagree with the continuous usage of the named average because:
- it is not internal to CumulusMX
- comparison with historic series should be accomplished by statisticians and not by maintaining ancient calculation methods. An example of this method exists in the Netherlands. It is is not up to Cumulus to maintain usage of the old calculation method.
Continued fall back in discussions for the usage of (Max + Min)/ 2 is confusing as if you are saying it is better then the Cumukus method. In the discussion on Temp Sum it even became the definition of Temp Sum which definitely is wrong.
We are talking about different estimators for the average of which the 'integral method' is native to CumukusMX. It is not the responsibility of CMX to supply comparison valid values.
I modified the fragment (removed @sfws striked text) and added a link to the scientific article by the KNMI on the homogenisation of different series of temperature measurement.
NOTE: 1) is it possible to send a message if a watch page is talked about? 2) is it possible to send a message when a watched page is changed by somebody not yourself?
I am beginning to understand how the wiki works and that i am supposed to read the watch list. There should be a short instruction for new users!