Errors in IPCC climate science

While working on my benchmarking study on April 5th, 2012 – I noticed a clearcut step change in 1997 between CRUT3 and CRUT4 over much of north America.
I have portrayed this in map form to understand better what the CRU team has done in the various grid boxes.

It seems that for most of the USA grid boxes CRUT4 is warmer by one division on the temperature anomaly scale (0.3 degrees C). But note the grid box for Los Angeles has warmed by 3 divisions or ~1 degree C.
I am intrigued that the CRUT team makes such large adjustments over the USA – home to so many climate researchers. In 2006 I found amazing agreement over the USA between HadCRUT2 and satellite lower troposphere temperature data. I assumed that was one area of the globe where the CRUT team had little “wriggle-room” to adjust data. Not so it seems.
Anyway, I will do more on the subject later but wanted to get this surprising result out there. Who was it said something like – “…the future is known – it is the past that changes…”

This global study at the European Institute for Climate and Energy (EIKE) – How Natural is the Recent Centennial Warming? – is the first that jumps to mind in over two decades by a group sceptical of AGW – that applies statistical analysis to thousands of temperature station data and concludes there is a standout UHI signal.
Andrew Bolt has an article on this – “Climate study: er, what man-made warming?”
I just hope this is the start of a determined research effort that might in the years ahead blunt the harm done to science by over two decades of the UHI infested various CRUTem versions.

The ACT despite being only the population (300K) of a medium city has an Independent Competition and Regulatory Commission (ICRC). There is no limit to bureaucracy here. Reading between the lines and noting the whingeing from ACTEW I suspect increases will not stop at 17%. I wonder what price hikes are due out in the States.
The Canberra blog RiotACT has an article – they also ran on staff cuts at Department of Climate Change and Energy Efficiency as they need to downsize from 900 to about 600 staff to meet the demands of its budget. Comments do not show much sympathy for staff there.

After comparing land only CRUT3 vs the shiny new CRUT4 for some Australian grid boxes I examined again grid box areas in East China, Russia, Africa and also North America which I have checked in past years for trend differences between CRUT data and AMSU satellite data from UAH. For map of grid box areas. Data from KNMI Climate Explorer
For East China the differences are large and have increased in CRUT4.

Over Russia differences are less and have increased in CRUT4.
In the case of the grid box over Africa the trend per decade differences between CRUT3 and AMSU are huge and have eased slightly in CRUT4. These results imply large errors in surface data – unless some other reason can be found.
In 2006 I found close agreement between satellites and CRUT2 over the USA. Now we find that CRUT4 finds more warming trend in the USA than did CRUT3 – this chart shows evidence of huge adjustments in 1850′s data – and also CRUT4 has been tweaked warmer in recent years. The North American grid box area still shows the closest agreement with satellites – but CRUT4 finds increasing warmth by a variety of adjustments. An example is this 0.25 degree jump in the CRUT4 minus AMSU trace at 1997. I wonder if this tweak was “peer reviewed” ?
Perhaps the CRUT team thinks USA climate scientists are not watching – we are told the science is settled – not where I am looking.

The maximum temperature Outlook must be a comprehensive failure over eastern Australia extending to the 50% line inside WA. Maps of actual weather averages can be made here.
The cool success in WA has to be limited by the warmth of the western WA coastal strip which would lower the score there. Overall surely a serious fail mark.

The minimum temperature Outlook must be an even worse failure – with not one square centimetre of the continent predicted to be even as cool as average. If anyone can spot the BoM 50% line – let me know. I see the “National night-time HotSpot” shining through – you can not hold down BoM errors.

Even the rainfall Outlook fails overall despite a win or two here and there. You have to look at the huge areal extent of the predicted dry inside the 50% line – note the majority of the NSW highest ever rain areas were predicted dryer than average.

And our gullible policymakers and politicians are convinced the BoM can predict the future – what a sick joke.

Many long term Australian temperature stations have data well back into the 1800′s. Often these 19C data show warmth on a scale with modern decades. See my two graphics from 1991.
Although the Jones et al 1986 data included very few of the long term Australian non-Capital city data – in 1994 many of these stations were included. The situation stayed like this up to the closing of CRUT2 in 2005 – included many Australian stations with relatively warm data a century or more ago – see my article in Coolwire Issue 5: Jan 2003 – “Global Warming” and the Australian Drought. Note the contrasting trends from the adjoining Murray-Darling Basin (MDB) and Sydney grid cells which IMHO are due to urban warming effects in the Sydney grid cell CRUT2 data.
In 2006 the UK Met Office (UKMO) took over the reins and rebuilt CRUT2 into CRUT3. In recent weeks Jones has announced his latest version CRUT4 and the point of this article is to highlight the huge extra warming in the trend for the MDB grid cell in CRUT4 compared to CRUT3. The MDB has been “corrected” to now agree with the urban affected UKMO version of their Sydney Airport station (947670) trend.

All the grid box data has been downloaded from KNMI Climate Explorer.
Some background on the warm 19C data. The BoM does not publish it “high quality” data further back than 1910 because the have a view that most data before that might not have been collected in a Stevenson screen. I disagree with that and have a 1995 paper with pages scanned here. Here are two studies into the temperature changes produced when thermometer exposures have changed from older more exposed stands to the louvre sided box – the Stevenson screen.
The first – Effects of changing exposure of thermometers at land stations D. E. Parker – International Journal of Climatology – Volume 14, Issue 1, pages 1-31, January/February 1994
In his abstract Parker found – “…little overall bias in land surface air temperature has accumulated since the late nineteenth century:” He speaks of a maximum bias of 0.2 deg but says summer and winter changes may have cancelled out.
Then in 1996 Nicholls, N., R. Tapp, K. Burrows, and D. Richards. Historical thermometer exposures in Australia. Int. J. Climatology, 16, 705-710 (downloadable at above link – note Nicholls et al are BoM).
Nicholls et al compared a long running experiment at Adelaide comparing readings in Stevenson screens and a Glaisher stand found that – “Over the year, the mean temperatures were about 0.2 deg C warmer in the Glaisher stand, relative to the Stevenson screen.”
So looking back to the main issue of the massive increase in the CRUT4 trend over the MDB grid cell – I have this chart of CRUT4 minus CRUT3 which shows massive adjustments of 1 to 1.5 degrees – seems out of all proportion to the 0.2 degrees adjustment justified by Nicholls et al 1996.
If anybody knows of greater warming adjustments in CRUT4 grid cell data – please let me know.