I think the mindset of the adjusters is that the current temperatures are by definition correct and that therefore it’s necessary to adjust the past to match the present. Since they won’t entertain the possibility that the current readings are biased, they have to change readings that normally would be considered good as gold. Since this is all done below the radar screen and they don’t have to justify their preconceptions, it hasn’t been much discussed except by “deniers” up to now. This is one reason surfacestations.org is so important. Regardless of what sort of bins are found to throw individual stations into, people will finally wake up to the fact that the surface record needs to be examined in detail and questionable procedures changed or justified.

I’m amazed that the largest overall adjustments have been made to the unlit stations, the largest category, and that about half of the 20th century temperature increase can be attributed to the adjustments. I see that the TOB and Max/Min and SHAP Data adjustments are larger for unlit stations than for peri-urban or urban stations. This magic just doesn’t make sense to me.

Hansen does this in his 2001 paper, I believe.

However, what you want to look at is each and every adjustment. We have to reverse engineer the
process, so a global report is of no consequence.

You can find his 2001 paper at Goddard. I wont link. too lazt. Just google Giss temp
click on publications or authors and select Hansen

browse around in the virtual stacks

#25 – The adjustment doesn’t seem completely unreasonable as a means of generating a corrected view of a contaminated sample, given a population of correlated samples. Take the mean trend, and encourage your target sample to fit the same trend.

That would be a good approach if what was required were an improved accuracy for the one site where the collected data had been contaminated. I believe that Steve is using a variation on this approach to identify specific stations which have been over fitted.

The problem is not that they have adjusted Marysville downward (in line with your suggestion), but that they have also adjusted Orland upwards … what’s up with that?

w.

The point is that Australian continental mean temperature did show an increase in temperature with time. But now I’m wondering – based on the apparent UHI effect we’re seeing here which is muddying up the waters, is it possible that the Australian data is being tweaked as ours are to show that increase? Particularly since I’ve read a number of times that the
Southern Hemisphere is not heating up. So why the temperature increase? Or could it be the temperature increase was accurate, at least for the Australian continent?

I’m going to try to tap back into that site again.

Not only are the graphs strikingly similar, the trend rates are nearly identical.

Was this a quote from a scientist or an alchemist? Sounds more like the later.

my bad earle. I thought I clipped the reference down to the abstract only.

SteveM can snip without complaint from me