Greenland revisited..

[Derek]

Hi All,
Please excuse the rushed nature of this post.

Sometime ago I looked at the greenland ice core data sets and made this little film for you tube.

http://www.youtube.com/user/Weiun#p/a/u/0/VZMJg2Z8Xpw

OK, so it was a little OTT. Recently with the CRU emails and various other things happening at present
I have been thinking about going back and "revisiting" the greenland ice core data.
So, I have..

There are known problems with ice core data, particularly CO2 measurements,
but does this also apply to O18 temperature reconstructions. I am not so sure.
BUT it does mean that the problems of ice cores can be aired, one must be one's own biggest critic.
The only "issue" I can see with the ice core data as used is in the actual date of measurement assigned to it.
Rather than merely a year it seems to go down to date and month. I have ignored this and just used the year,
so I may have combined winter / summer / autumn / spring measurements (supposedly).. I do not think it matters.
BUT - it does lead into a good discussion of how questionable the ice core data is. Jawowoski, etc.
I think that it might be the case that temperature and CO2 reconstructions maybe of differing degrees of "causion".

The greenland ice core data used plots as below.


Furthermore, I think it reasonable to suggest that the ice core data dampens temperature variations, and possibly extremes,
so it is favouring the "alarmist" view point.
I bet regulars here never thought I'd do that......
Well, there is good reason as will become obvious later hopefully..

HADcrut - Using this data set gives a good oppotunity to question it's reliability as a source, open science, CRU emails, codes etc,
in the end the metric is almost certainly an exaggeration of the temp changes that did occur over the last centuary,
so again using the HADcrut data favours the "warmist" paradigm.
The HADcrut data plots as below.



I have spliced the greenland ice core and HADcrut data sets together. I know it is a no no, but..
I did not process the greenland data at all, I merely plotted it against year of measurement.
(getting the data to each year of measurement was the "donkey work" that I'm not aware anyone has done before.)
The HADcrut data - I added Jan to Dec anomolies up for a year and divided the result by 12.
On the spliced plots I reduced the number of years plotted for the HADcrut yearly anomolies to one every 8 years, to match the ice core data "frequency".
I think that they have plotted very well together.
They plot together as below.


I thought it reasonable to reduce the ice core data plotted to the last 10,000 years of the record, and then add the HADcrut data (post 1900).
ie 9025 BC to 2007 AD.
This plots as below.


Given the problems known regarding ice cores it seems reasonable to cross reference the record. But to what.
Human civilisations known history would seem reasonable.
This plots as below.


AND, for good measure, what is the linear temperature trend over the Holocene.
This plots as below.


I have also added on most of the plots a plus two degrees celcius upper "baseline".
I thought it worth putting into the context of the Holocene the possible effects of a 2 degrees celcius temperature rise globally.
Usually pretty good, if only it was happening.......

I would of liked to have written this up properly, but time has overtaken me,
so any offers gratefully accepted. Joint piece anyone...

[Derek]

Hi All,
In relation to the above post Questioning_Climate sent me the below links.

ftp://ftp.ncdc.noaa.gov/pub/data/paleo/i...spd18o.txt

http://www.gisp2.sr.unh.edu/DATA/Obrien.html

http://epic.awi.de/Publications/Wer2002e.pdf

Excerpt,
" This indicates that the year-to-year variability of d18O over central Greenland in the model
is solely influenced by the simulated internal variability of the atmosphere in higher northern latitudes,
and not by the variability of the prescribed SST.
Because of the stochastic nature of this internal atmospheric variability
it is therefore not possible to model d18O anomalies of any specific
calendar year in agreement with the observations,
as is clearly shown in Figure 1. Nevertheless, an influence of SST variations
on the isotope record on longer timescales might still exist and
cannot be ruled out by these findings. "

ftp://ftp.ncdc.noaa.gov/pub/data/paleo/i...tacked.txt

I take some heart from the suggestion that the northern latitudes winds and their variations may more effect the d18O than SST.

I didn't know "our" SST was that influencial - or have I misread that....

[Sunsettommy]

The splicing of HADCRUT data may indeed be inappropriate,but despite that it is obvious from reading the charts,that most of the last 9000 some years it was been ...... warmer than now in the Greenland area.

Now that we can safely consider the possibility that the HADCRUT temperature data you grafted on is biased to a higher temperature level than it should be.The problem is how much higher?

If an untrained layman Derek can make this basic presentation,that easily explodes the idea that TODAY it is unusually warm,or that the increase is unusually rapid,how come too many trained scientists fail to do the same?

Good work Derek!

[Derek]

Hi All,
Thank you SST - I try..

Re - inappropriate splicing
I would suggest that the central greenland temp record is "better" as a global metric than a few bristle cone pines in california...
I did think about using Loehle's global reconstruction for the last two thousand years (upto 1900) and then HADcrut.
BUT that would not of covered the period I wanted to cover, namely the Holocene.
Maybe we should do a Loehle / HADcrut version as well.
I think it would show much the same, but would be happy to let it plot "as is", as I have done so far.

NB - The "other" thing that occurred to me as I was doing this excel sheet was,
"what would the effect of removing the bias from the HADcrut data be. ? "
I think it would alter the "shape" of the end of the plot to become far flatter.
That would be interesting (read new) as that sort of "shape" is not shown
in the rest of the Holocene record that I have seen.
It could well have ramifications.

[Richard111]

May I add my thanks to Derek for all the work he has shown on this thread. An inspiration to all us laymen of what can be achieved if you put your mind to it. The problem is time! Where do you find the time?

[Derek]

Thank you all.
Richard111 - What is sleep. ??

As some know I can not do much during the week at present, so
this weekend I'll try to "correct" (or find the "best" corrected to date) HADcrut data and replot the orange line on the plots.
I think it will show a "flat" period in our planets climate history.
Not no climate variation, but far less than was previously the case.
I suspect at present that the temp variation has been far less over the last 800 (ish) years,
ie since the end of the Vikings, than over the rest of the Holocene.
If that is a possibility, or even "the case" then that does change
what we see in the temp records.

Both present "sides" of the discussions are right and wrong.
But one "side" has been shown to be fraudulently altering the data as well.

To my mind there are obvious people here who have suffered from this debacle directly,
I hope in some small way this thread can be developed into something that helps.

[Derek]

(01-03-2010 10:20 AM)Sunsettommy Wrote:  The splicing of HADCRUT data may indeed be inappropriate,but despite that it is obvious from reading the charts,that most of the last 9000 some years it was been ...... warmer than now in the Greenland area.

How about we collect Holocene temperature records and do a "Loehle" type plot for the Holocene (finishing in 1850 AD). ?
Then add the "corrected" HADcrut temperature record since 1850.

[Sunsettommy]

(01-09-2010 01:07 AM)Derek Wrote:  

(01-03-2010 10:20 AM)Sunsettommy Wrote:  The splicing of HADCRUT data may indeed be inappropriate,but despite that it is obvious from reading the charts,that most of the last 9000 some years it was been ...... warmer than now in the Greenland area.

How about we collect Holocene temperature records and do a "Loehle" type plot for the Holocene (finishing in 1850 AD). ?
Then add the "corrected" HADcrut temperature record since 1850.

POST #6

In the CHARTS thread might be of some help.

[HarpoSpoke]

Nice work Derek!

I hope to have some time to dive into it this weekend.

[Derek]

Please move into the charts if needed SST.
My sincere apologies, but time is very, very short for me, I have to go before the snow arrives..
I will post the charts I have done and add text later.
In short I added by eye (and very roughly) what I thought was missing from the plots "as is".
Namely Medieval Warm Period, Little Ice Age - Greenland (both may be wrong or too little magnitude / duration - apologies all very rushed.
The 1930s nearly as warm as the 1990s, the post 2000 temp decline, and the temp decline between 1940 and the mid 1970s - HADcrut.

I think it puts a different "view" on the temp record, but it is all only done by "eye", so it is merely what I think might be the case, not what has been shown.
That could come with comparisons to human historical records perhaps...

Here are the plots.

This first plot is the HADcrut "as is" plot.
Several things strike me straight away.
What is that 1901 spike. ?
The 1930s were warmer than that surely. ?
Was the 1940 to 1975 cooling more pronounced than that. ?
Is the post 2000 global temperature decline more pronounced than that. ?



So I merely added "by eye" the missing "things" mentioned above.
The plot now looks like below.
Maybe I have over egged the 1930s, maybe not. ?
But I think the plot looks more like what we experienced than the original "as is" plot.
Any comments. ?


I simply overlaid the two above plots to show what and where I moved things about.
Whilst writing this I have also noticed and am wondering what and why is there a small late 1870s spike. ?



I then looked at the Greenland ice core plot. I immediately thought it has been "Hockey sticked".
Where is the medieval warm period, and the little ice age. ?
The last 800 years of the plot is peculiarly flat compared to the rest of the record.


I have added by eye (in probably too small a way) a MWP, and a LIA.
Both could probably be done better.



I decided to plot from 1000AD to 2008 to see the difference from
the "as is" plots, and the altered "by eye" plots.
Firstly the "as is".



And the "corrected by eye" 1000AD to 2008 plot.



The full "as is" plot from 8005BC to 2008AD.


The full "corrected by eye" plot fromm 8005BC to 2008AD.


I am genuinely very surprised that the "corrected by eye" plots actually look "better" than the "as is" plots.
Although I think I have not "corrected" enough for the MWP.
None of the above is scientific, and that is the rub. It still looks more realistic than the "science".
I chose the two data sets here because they are both so questionable. That is the point.
So, my "corrections" are argueably no better or worse (at least I am being open..) than the actual records,
and my "corrections" fit in far better with known human history.

I am left with the distinct notion that we would get better temperature reconstructions from
known human historical records and archeology
(and maybe some animal populations, ie fish, particularly salmon combined with ocean phases)
rather than modern climate science, and any of the major metrics of the present "consensus".

Is modern climate science really that broken - it appears so.

Any suggestions, or alterations to the "by eye" corrections gratefully recieved and considered.
I have sent a copy of the excel sheet to SST, I'll try to forward on a copy to anyone who wants one, or the pp version of the slides above,
but no promises from me as I am "away" a lot at the moment.

Can we please put more days in the week, then I might be able to get to the is there a greenhouse effect thread,
but I will sooner or later..