Ok, so why not just be specific? “On record” usually means since we started recording history , at least 5k years ago.
And have you looked into the records? satellite surface temps and high resolution recording have not been around for very long. 1880 methods were very crude and narrowly scoped.
> “On record” usually means since we started recording history , at least 5k years ago.
I'm a journalist who has published "highest/lowest on record" statistics tens, if not hundreds of times, and I've never heard of anyone thinking it means "since Herodotus" or anything like that.
How would readers know the reference point unless you inform them. Of course they will defer to colloquialisms . In some cases 5000 years , some 1000 years . With something as broad and impactful as this, they certainly assume more than 150 years .
You both have a point, reading further provides context as to which record is being talked about.
There are, of course, many records - newspaper records, human logged records of conditions that day, and human created records of proxy data - ice cores, dendrochronology, cosmic ray induced crystal formation in beach sand, etc.
and scientists edit the historical temperatures because of, and i hope you can see my eyeroll here "anomalous readings" - but they're overwhelmingly erroneous in only one direction. that's strange.
i'm literally in the middle of trying to parse a couple of papers that examine the methodology of at least the NOAA homogenization model.
did you know there's only eight sensors, globally, that we have data for >95% of the last 100 years, that are labelled as "fully rural"? so this means that 99.9% of the stations must therefore be, at least, more likely to be adjusted, doesn't it? The entire premise that UHI is irrelevant because they "normalized" the 99.9% and it showed it was irrelevant is... i don't know, it's something, though.
I agree it's suspicious. Rather than dismiss it altogether, I'm trying to understand the e2e process. i.e. divide 1880-present into Epochs and understand what %-age of coverage, resolution & how precise were the instruments
after you do all of that, see if you can find pictures of the stations, the little structure the old ones were in are called "CRS" which stands for Cotton Region Shelter. these were the old liquid-in-glass style thermometers. The new ones have multi-plate radiation shields to house the thermistor and protect it from direct sunlight while allowing airflow. i guess some people say they look like a beehive, but they're a cylinder.
so the CRS ones could be put anywhere. the new MMTS sensors had a pre-cut length of cable for installation, as the base unit was to be situated indoors.
So CRS could be out in the middle of a field, where the MMTS are generally within 30' of a building.
Maybe we should throw out all the MMTS data?
Now, though, someone finally realized, and they're installing wireless MMTS - minimum-maximum temperature sensor, tells the min and max over the last 24 hours, and you record those numbers once per day, at a specific Time-of-Observation. Changing the time-of-observation requires de-biasing the data, too!
i agree that the apparatus varies tremendously over the decades. Real science is hard work and I applaud environmental scientists for their efforts. We need to be careful about qualifying "raw data" since software engineers usually stop at a textfile and forget about the origins.
