The difference is most likely because GISTemp uses one interpolation method, and only uses ground temperature measurements. Cowtan and Way's hybrid method uses satellite data, which can see more of the Arctic. If the short-term warming is happening further north than most of the thermometers, then of course the Cowtan and Way method will show more warming than GISTemp. As well as their global and near-Arctic cross-validation where they 'hide' data (Section 4 of the paper) they test against buoys in the Arctic (Section 5.2) which aren't included in GISTemp or HadCRUT4. According to the buoy data, the hybrid reconstruction including the satellites does the best job.
Big regional swings in temperature tend to only last 1-2 decades. That's why all of the methods agree well over a 30-40 year time period, you only see bigger differences over shorter times.
This study does not consider the Argo buoy measurements at all, so doesn't affect the finding that there has been increasing heat storage deeper in the oceans recently. This has been due to the recent natural changes in the Pacific mainly related to the El Nino Southern Oscillation. So the studies on heat storage, such as those by Meehl and Balmaseda aren't affected by this. Kosaka and Xie's paper from this year is particularly interesting regarding this, and it doesn't seem that Cowtan and Way's temperature record disagrees strongly with their model results either.
And the reason why they compared their hybrid with HadCRUT4 is because it's an apples-to-apples comparison; they were filling in HadCRUT4.
>>As you dig deeper into the study, it is apparent that by far the most infilled data occurred in the Arctic region, an area that HadCRUT4 has very little data.<<
As you dig deeper? I don’t see how the authors could be more explicit. It is all explained in the first paragraph of the abstract:
“Incomplete global coverage is a potential source of bias in global temperature reconstructions if the unsampled regions are not uniformly distributed over the planet's surface. The widely used HadCRUT4 dataset covers on average about 84% of the globe over recent decades, with the unsampled regions being concentrated at the poles and over Africa. Three existing reconstructions with near-global coverage are examined, each suggesting that HadCRUT4 is subject to bias due to its treatment of unobserved regions.”
If there is any legitimate cautionary tale to all of this, it is to not over-interpret short-term trends – but, no matter how many times it is explained to Deniers, they just cannot understand it.
Why does this confuse you, Ottawa Mike? (or is it that you only wish to confuse others?)
The Cowtan and Way paper highlights how the UK Met Office uses the HadCRUT4 data to determine the mean surface temperature and that this data only represents 84% of Earth's surface to determine the mean surface temperature. The methodology that Cowtan and Way used to cover the gaps in surface data is clearly explained. The time frame in reference is 1997 through 2012. ... So why do your next two links,nasa and woodfortrees, reference 2002 through 2012? Does the starting point of 1997 make any difference for the woodfortrees graphic? Well, let us take a look at what you linked for 2002 through 2012 - http://www.woodfortrees.org/plot/gistemp... - Now let us look at 1997 through 2012 from woodfortrees - http://www.woodfortrees.org/plot/gistemp... . This looks a bit different, does it not?
Cowtan and Way are saying that the UK Met Office did not even try to extrapolate for the 16% of Earth's surface that was not covered due to not enough local temperature stations to obtain the data from. When Cowtan and Way use the statistical method known as kriging to extrapolate temperature data for these poorly sourced areas they were able to show that the global mean warming is happening twice as fast as the UK Met Office graphic shows it to be happening.
What is going on with Cowtan and Way is that they published information you preferred was never published as it further shows that vour "skepticism" of our warming climate is based on ideology and not on science.
I am skeptical of their study. No surprise there I am sure. I think they likely set out to cherry pick data to show increased warming. Why do I think they didn't go out and try to debunk AGW?
I remember Hansen removing some stations in the Arctic since obviously less data is more accurate than more data. I was looking for links to refresh my memory on it and wasn't too successful but I found this one
http://wattsupwiththat.com/2011/06/09/cr...
It seems it is mostly models that suggest the extreme warming in the Arctic and not so many temperature records.
This site has some old ones (therefore maybe less susceptible to biased removal of inconvenient data)
http://www.john-daly.com/stations/statio...
Did they fill in the HadCRUT4 data with UAH data or with the GISSTemp data?
I thought that it was wattsupwiththat, or was it Faux News or the Dailymail, that never cherry picks.
Sarcasm aside, what, other than your denialist bias, would make you think that it is cherry picking and not expert scientific judgement?
I can do that too :
The Global average temperature cooled from February 1882 to February 1994 by 0.05C, but then it warmed by 0.77C (-0.01C to +0.76C) one year later in February 1995 and since then it has cooled by 0.15C to date. Current warming stands at 0.61C above the established normal temperature.
http://data.giss.nasa.gov/gistemp/tabled...
We can now deduce that Global Warming's most extreme temperature fluctuation happened between Feb 1994 and Feb 1995. An increase of 0.77C in 1 year which means Global Warming only happened over a one year time frame.
:-)
Recently, a paper from Cowtan and Way claimed that recent warming was really happening at least twice as fast as previously thought. Their study basically took HadCRUT4 and claimed that the missing locations could be filled in using UAH data to come up with a more complete global coverage. Of course, skepticalscience.com just about wet their pants so you can read the gory details here: http://www.skepticalscience.com/global-warming-since-1997-more-than-twice-as-fast.html
As you dig deeper into the study, it is apparent that by far the most infilled data occurred in the Arctic region, an area that HadCRUT4 has very little data. And since we all know the Arctic is warming faster than the rest of the Earth, Kowtan and Way (and skepticalscience.com in their wet pants) now claim the pause has been erased.
I'm not sure if there is still missing heat in the ocean now or not but apparently that's not important any more since the pause has been erased and we can all sleep well at night.
Hang on a second, let's look at GISTemp LOTI. Here is one of their recent graphs: http://data.giss.nasa.gov/cgi-bin/gistemp/nmaps.cgi?year_last=2013&month_last=10&sat=4&sst=3&type=trends&mean_gen=1212&year1=2002&year2=2012&base1=1981&base2=2010&radius=1200&pol=reg
Okay, so far so good. They show a lot of Arctic warming from 2002-2012. It all fits so far. So let's go to Wood for Trees to look at GISTEMP: http://www.woodfortrees.org/plot/gistemp/from:2002/to:2012/plot/gistemp/from:2002/to:2012/trend
Whoops. GISTEMP 2002-2012 clearly shows the Arctic warming faster than the rest of globe but the overall trend is basically zero. To be fair, I know the more astute readers will notice that Cowtan and Way used 1997-2012. Over that time, the GISTEMP trend in C/decade is 0.08, HadCRUT4 is 0.046 and Cowtan and Way gets either 0.11 or 0.12 (depending on hybrid or kriging).
GISS has the arctic warming clearly in its data set yet the trend is much less. So what's going on with Cowtan and Way?
___________________________________________________________________
Note to skeptics: Please don't mention that the skepticalscience.com web page title which includes "more than twice as fast" is a cherry pick of HadCRUT over GISS temperature sets because those guys never cherry pick.
