There’s a site with population density maps here:
http://sedac.ciesin.columbia.edu/data/set/grump-v1-population-density/maps
Someone pointed me at it some time ago and I’ve lost track of who and where, so if it was you, take a bow.
They require a login to get the maps. I generally studiously avoid doing anything that leaves identity information laying around or gives me Yet Another Account to manage or Yet Another Password to track or (HORRORS!) more email… but I did find one image that wasn’t locked down that is good enough for my purposes From the “do you want to download a detailed map of this” image:
http://sedac.ciesin.columbia.edu/downloads/maps/grump-v1/grump-v1-population-density/globaldens.jpg
Global Population Density
I’d been wondering why there was a stripe of thermometers just under the Sahara. Looks like that’s where there’s a stripe of people.
GHCN v3.3 NOW in red 50% transparent over Baseline in Blue
There are exceptions. The USA has thermometers all over even out in the big empty. Australia HAD a bunch of thermometers (the blue) but now have only a few in GHCN (the red) China, too, had a lot, mostly where there are the people, but now a lot fewer. India has a lot of people, but not a lot of thermometers. Still, the general trend is that places empty of people have poor coverage.
It is my opinion that this fact, by the very nature of it, causes a fundamental bias in the temperature data. We’ve increased population a lot since 1850 and all of them burning a LOT of gasoline, kerosene, Diesel, coal along with electrical appliances. You can’t dump a few TW of waste heat into the places people live and have them not be a bit warmer than the big empty places. I would be very surprised if there were not a 1/2 C to a few C just from that in the unadjusted data. (Then we’ve seen evidence that the UHI Urban Heat Island adjustment that is done is not right and sometimes backwards).
But at least I know why there is a line of thermometers under the Sahara, and why Canada had so many on the US Border during the baseline period (as that’s where all the Canadians go to get as close to warm as possible ;-)
Does make for an interesting question though. How do you get historical data to compare to current data from places without many people?
Musings from the Chiefio 
Many thanks for this … how about Agriculture vs Temps ?
Processes that lose humus from the land (heavy machinery, inorganic fertilisers etc) (upwards of 60 tonnes per Ha p.a. in my area http://www.water21.org.uk/wp-content/uploads/2011/09/Stroud-Soil-Erosion-Risk_Water21_170311.pdf ), causing compaction and aridification are all too often the effects of industrial arable farming; these must have significant implications for temps; this goes well beyond Urban Heat Islands … ?
I sat through a presentation a few years back where a UK Met Office speaker showed temperature rises in the land area around the old Met Office offices in Bracknell during past century – significant increases – that could have been wholly accounted for by changes in farming practices (in this area from changing small horticuture to ‘prairie arable’ and housing). The Met Office speaker was strangely undable to discuss …
@EM: Your comment about heat islands reminded me about an aspect of large city temps: Thermometers out-of-doors in a large city do not measure the average temp of the city… they ignore entirely the temp inside the buildings.
In the big empty proxies are needed. Isotope ratios in snow, pollen in lake sediments, plant wax D/H ratios, 4 thousand year old stumps in the permafrost at the arctic coast, the list goes on. But I would suppose that this is not news to you or your readers. Cheers and enjoy the interglacial!
The post reminded me of this item.
http://joannenova.com.au/2018/10/first-audit-of-global-temperature-data-finds-freezing-tropical-islands-boiling-towns-boats-on-land/
Pingback: GHCN v3.3 vs v4 – Top Level Entry Point | Musings from the Chiefio