John, your graphs start at 1875, but you state in the article that the HADSET records began in 1775, 'with 1846 having the greatest mean daily temperatiure'. Should the graphs have started at 1775?
It is interesting that the trends you note of Tmax being constant over time, but Tmin increasing are repeated in the USA and Australia (and I suspect many other regions. Whilst UHI would be expected to increase Tmin, the same could be said for increasing GHGs. The only way of separating these (as far as I can see) is to contrast long time series records in very rural areas (few and far between) with those subject to urbanisation.
I very much look forward to your upcoming articles as set out 'In the kitchen'
They should have and now do - well spotted! Yes indeed, the only sure fire way of getting to the bottom of this is by using data from genuinely rural stations. There may well be a few dotted across the USA but identifying them will be time consuming. That being said it would make for a natty little project for somebody happy to fiddle with Google Earth.
Yes, I've come across various bits. BTW I've also revised the warming trend estimates for those mean series that now stretch back to 1772 - my stats module does this weird thing of truncating records when multiple time-based regressions are run, so I had to go back and run the 1772-2023 means separately from the 1878-2023 mean maxima.
Sorry if this is a stupid question, but do we have mean temps back to 1772, but only max (and presumable min temps) back tom1872 (hennce your max temp graphs starting at 1872)?
What is interesting me is the warm temps in the late 1700's, then the lower av temp at around 1812 (Napoleon's retreat from the Moscow in the bitter cold winter of that year). It really is a good illustration of how misleading short time series can be when data are truncated at say 1872 or worse, the 1970s.
Not stupid at all! Daily means start 1772 whereas daily max & min start 1878. Yes indeed, those very early figures on the expanded graphs certainly raise an eyebrow. The Hadley crew also offer monthly means back to 1659 - I am sore tempted to run a quick note revealing that series! At some point I shall be using this extra long series in conjunction with sunspot data to see what we can see.
John, your graphs start at 1875, but you state in the article that the HADSET records began in 1775, 'with 1846 having the greatest mean daily temperatiure'. Should the graphs have started at 1775?
It is interesting that the trends you note of Tmax being constant over time, but Tmin increasing are repeated in the USA and Australia (and I suspect many other regions. Whilst UHI would be expected to increase Tmin, the same could be said for increasing GHGs. The only way of separating these (as far as I can see) is to contrast long time series records in very rural areas (few and far between) with those subject to urbanisation.
I very much look forward to your upcoming articles as set out 'In the kitchen'
Many thanks for all the hard work.
They should have and now do - well spotted! Yes indeed, the only sure fire way of getting to the bottom of this is by using data from genuinely rural stations. There may well be a few dotted across the USA but identifying them will be time consuming. That being said it would make for a natty little project for somebody happy to fiddle with Google Earth.
Re rural stations, I think there has been some work done on this by Willie Soon and Michael Donelly
Yes, I've come across various bits. BTW I've also revised the warming trend estimates for those mean series that now stretch back to 1772 - my stats module does this weird thing of truncating records when multiple time-based regressions are run, so I had to go back and run the 1772-2023 means separately from the 1878-2023 mean maxima.
Sorry if this is a stupid question, but do we have mean temps back to 1772, but only max (and presumable min temps) back tom1872 (hennce your max temp graphs starting at 1872)?
What is interesting me is the warm temps in the late 1700's, then the lower av temp at around 1812 (Napoleon's retreat from the Moscow in the bitter cold winter of that year). It really is a good illustration of how misleading short time series can be when data are truncated at say 1872 or worse, the 1970s.
Not stupid at all! Daily means start 1772 whereas daily max & min start 1878. Yes indeed, those very early figures on the expanded graphs certainly raise an eyebrow. The Hadley crew also offer monthly means back to 1659 - I am sore tempted to run a quick note revealing that series! At some point I shall be using this extra long series in conjunction with sunspot data to see what we can see.