At the time I emailed Phil Jones about this – he wriggled out of an explanation by saying there was a new paper coming out which explained all. For the IPCC Glitterati there is always a new paper coming out. Other standard Phil fend-offs were, “just arrived back from somewhere” – and “just rushing off to somewhere”.

Can empty grid points similarly affect the anomalies? The argument against this, as discussed above, is that we measure just the changes in temperature and these should be independent of any location bias i.e. CO2 concentrations rise the same everywhere ! However it is still possible that the monthly averaging itself introduces biases. To look into this I calculated a new set of monthly normals and then recalculated all the global anomalies. The new monthly normals are calculated by taking the monthly averages of all the stations within the same (lat,lon) grid point. These represent the local means of monthly temperatures over the full period, and each station then contributes to its near neighbours. The anomalies are area-weighted and averaged in the same way as before. The new results are shown below and compared to the standard CRUTEM3 result.

Fig5: Comparison of standard CRUTEM3 anomalies(BLACK) and anomalies calculated using monthly normals averaged per grid point rather than averaged per station (BLUE).

http://wattsupwiththat.files.wordpress.com/2012/02/image_thumb13.png?w=644&h=275

The anomalies are significantly warmer for early years (before about 1920), changing the apparent trend. Therefore systematic errors due to the normalisation method for temperature anomalies are of the order of 0.4 degrees in the 19th century. The origin of these errors is due to the poor geographic coverage in early station data and the method used to normalise the monthly dependences. Using monthly normals averaged per lat,lon grid point instead of per station causes the resultant temperature anomalies to be warmer before 1920. Early stations are concentrated in Europe and North America, with poor coverage in Africa and the tropics. After about 1920 these systematic effects disappear. My conclusion is that anomaly measurements before 1920 are unreliable, while those after 1920 are reliable and independent of normalisation method. This reduces evidence of AGW since 1850 from a quoted 0.8 +- 0.1 degrees to about 0.4 +- 0.2 degrees

http://wattsupwiththat.com/2012/02/07/how-reliable-are-global-temperature-anomalies/#more-56153

I’ve looked at peer-reviewed trend analysis (linear, polynomial, EMD, intrinsic/extrinsic etc) of land-SST HadCRUT3 and pragmatically accepted that dataset because of all the work done on it and because the SST component negates much of the UHI and adjustment problem. Compare HadSST2 and HadCRUT3 side-by-side:-

http://www.woodfortrees.org/plot/hadsst2gl/from:1850/to:2012

http://www.woodfortrees.org/plot/hadcrut3gl/to:1850/to:2012

Basically the same series and the 1940s warming is still intact. Not so in the (I assume) land-only CRUT3. The CRUT3 1940s “adjustment” is highlighted in an ICECAP post by Joseph D’Aleo, ‘NASA Games’:-

http://icecap.us/index.php/go/joes-blog/nasa_games/

The incriminating CRU plot being this one:-

http://icecap.us/images/uploads/CRU_Temps.jpg

The 1940s anomaly drops about 0.2 C. I’m thinking that the NIWA “adjustments” made it through to the land-only CRUT3 (along with lots of others) but not to the land-ocean HadCRUT3.

BTW, HadSST2 and HadCRUT3 are topical in comments under this post in terms of recent OHC build-up – or not:-

http://www.climateconversation.wordshine.co.nz/2012/01/sceptics-query-our-truth-we-shall-besmirch-and-slander-them/

“Fig. 3. New Zealand temperature response areas and stations used in this study (after Salinger, 1979).”

That’s for the T3 region, so the data up to 1979 would not have changed between 1979 and 1995 presumably.

What date was the CRUTEM2 to CRUTEM3 changeover?

I regret I can scarcely keep up with my inbox at the moment, much less research anything, so I must leave this to others for now.

But I note: “This one would be good enough. I don’t know if it is the right one.”

A cavalier attitude to science, indeed.

There are obviously several missing in between; maybe in the encrypted annex to Climategate 2?

date: Mon, 14 Sep 2009 16:14:30 +0100 (BST)
from: P.Jones@uea.ac.uk
subject: Re: new zealand temps: crutem2 to crutem3
to: “Tim Osborn”

Tim,
Maybe you should see David if he can recall making any changes to land
stations over NZ – and if so when? There might have been changes
immedaiately after Jones and Moberg (2003) that Harry wasn’t aware of.
Another thought is to check whether your programs work with CRUTEM2 as
of now. CRUTEM2 data are on the CRU site, go to the temp page and they
are near the bottom.

Cheers
Phil

> Phil, I had a look at figure 1 of Brohan et al. showing land station
> coverage and coloured dots for new/deleted/edited stations. All in
> New Zealand are black, implying no change from CRUTEM2! I’ll check
> with programs again in case I screwed up, but fairly confident I
> didn’t — so its rather confusing as to why CRUTEM2 and 3 are so
> different over New Z. in summer. Tim
>
> At 21:27 11/09/2009, you wrote:
>> Tim,
>> This one would be good enough. I don’t know if it is the right one.
>> We also got some additional NZ stations as well.
>> It is more about exposures in Australia – that cause the pre-1910 data
>> to be less good.

>> For NZ, it is more work on the early NZ data. Exposures would be a
>> problem, but before about 1880. There was a gap in much NZ data during
>> the 1870s. It may be that we had only a couple of sites for V2, but got
>> several more for V3.
>>
>> Cheers
>> Phil
>>
>> > Hi Phil,
>> >
>> > I’m wondering how best to explain the cause of the change in New Zealand
>> > summer temps from crutem2 to crutem3. I could just say that crutem3 uses
>> > differently (or better?) homogenised stations records from some locations
>> > (could I say that the homogenisation adjusted for 19th century exposures
>> > being different?), with a pers. comm. to you. But if there’s a reference
>> > for homogenised New Zealand data, that would be better. I saw this one,
>> > by Jim Salinger that mentions “newly homogenised SW Pacific data” and
>> > includes New Zealand. Would this be what you now use?
>> >
>> > Salinger MJ (1995)
>> > Southwest Pacific temperatures: trends in maximum and minimum temperatures
>> > Atmospheric Research
>> > Volume 37, Issues 1-3, July 1995, Pages 87-99
>> > Minimax Workshop
>> >
>> > doi:10.1016/0169-8095(94)00071-K
>> >
>> >
>>
>> >
>> > Though I can’t download the PDF from home to see what it says.
>> >
>> > Tim
>> >
>> > –
>> > Dr. Tim Osborn
>> > RCUK Academic Fellow
>> > Climatic Research Unit
>> > School of Environmental Sciences
>> > University of East Anglia
>> > Norwich NR4 7TJ, UK
>> > http://www.cru.uea.ac.uk/~timo/
>> >
>> >
>> >
>> >
>
> Dr Timothy J Osborn, Academic Fellow
> Climatic Research Unit
> School of Environmental Sciences
> University of East Anglia
> Norwich NR4 7TJ, UK
>
> e-mail: t.osborn@uea.ac.uk
> phone: +44 1603 592089
> fax: +44 1603 507784
> web: http://www.cru.uea.ac.uk/~timo/
> sunclock: http://www.cru.uea.ac.uk/~timo/sunclock.htm