On 27/06/2019 09:23, tony sayer wrote:

Much has been written re the long term temp variation of the climate but
as it appears were talking about around 1 deg C change over quite a long
time, pre industrial era, just how absolute accurate where the
thermometers in use?.

Probably pretty good if designed for the purpose.

Essentially temperatrurre was *defined* buy mercury thermometers.

That is 50C is the difference in length of a mercury column half way
beytween the boiling pint at seal level, and 1000mb, and the freezxing
point of water.

Unless the column was not constant diameter, this is a pretty good estimate.

Secondly averaging 100 thermometers all measuring the same things, nets
yuou a very much higher accuracy assuming te thermometers are all
randomly inaccurate.

HOWEVER the bad news is that often and to this day, the 'average' daily
temperature is talen as beiing the 'average' between the maximum and
minimum temperatures recorded in a single day.

This tends to be higher than the true average, as sunlight and
temperaure peaks quite sharply around midday and a little after. And
will peak higher in the vicinity of concrete, asphalt and other man made
materials.

Utbaisation also tends to incerase temperatires genuinely overall as
well as te peak.



Course these days there are some very accurate ones around but back then
in the 1800's how accurate was a standard one back then anyone any
idea?.

Ive got a mercutry thermometer in my victorian or edwardiam barometer
andd I'd say its accutrate to quarter of a degree or so, How well
calibarted is another matter, but it tends to agree will what the local
airfields are reporteing on the internet



Lets not take into account as to where the measurement was made either
in town out in the country at the sea side etc, but the gear in use to
measure?.

Well one should. Because the issue is less how inaccurate the
thermometers were, but how they were utilised.



--
Any fool can believe in principles - and most of them do!