Right.
So you say you don't connect the measurements as they could be considered two stations. Different places, non comparable. Didn't say it was only
altitude. But altitude is important? You move up a slope and what?
But then you say you shouldn't correct, because of QC blah.
If you want a continuous measure over a period of time what do you do? The methods being used here are used by all climatology organisations.
Do you connect the two stations taking no account of the change of siting? We know that moving up the slope is going to lead to colder measurements on
average. So the time series will have a serious systematic bias.
Or do you try to make an adjustment for the change of siting reducing the systematic bias as best as possible?
And when the methods are well-established, justified, and well-known where is the deceit?
ABE:
lets expand on this with data to illustrate. We have three sites. Three places. One is at 3m ( Thorndon), another at 4m (Airport), and one at 125m
(Kellburn). Non-adjusted data show:
Clearly obvious that the higher site is showing cooler temps. No doubt. It's right there. The sites at 3m and 4m are warmer. Comparing the overlap at
first glance for Airport (4m) and Kellburn (125m) shows very similar
trends over time.
Lets adjust - we know that 3m is more similar to 4m than 125m. What happens if we remove the differences between 4m to 125m, but also reduce the 3m
the same amount :
We can see that the trends are very similar between Airport and Kellburn. Very close indeed. Still minor differences. But much closer than before.
So the question is which series would be closer to the reality:
1. Just attach Thordon to Killburn as coherent time series without attempting to account for siting issues; or
2. Account and adjust for siting issues, then attach as coherent time series.
It's pretty obvious that altitude is very important and the moving of site must be adjusted for. Alternatively, you could be anal and say 'no, QC,
can't attach at all!'
Fair enough, but then the deception is still Treadgold's, he doesn't even think this is important. At least these adjustments are a well-known and
-established method.
And if we ignore Thorndon altogether, the temperature is still only going one way. Up.
[edit on 26-11-2009 by melatonin]