By Steve Goddard
h/t to reader “Phil.” who lead me to this discovery.
In a previous article, I discussed how UAH, RSS and HadCrut show 1998 to be the hottest year, while GISS shows 2010 and 2005 to be hotter.
But it wasn’t always like that. GISS used to show 1998 as 0.64 anomaly, which is higher than their current 2005 record of 0.61.
You can see this in Hansen’s graph below, which is dated August 25, 1999
But something “interesting” has happened to 1998 since then. It was given a demotion by GISS from 0.64 to 0.57.
http://data.giss.nasa.gov/gistemp/graphs/Fig.A2.lrg.gif
The video below shows the changes.
Note that not only was 1998 demoted, but also many other years since 1975 – the start of Tamino’s “modern warming period.” By demoting 1998, they are now able to show a continuous warming trend from 1975 to the present – which RSS, UAH and Had Crut do not show.
Now, here is the real kicker. The graph below appends the post 2000 portion of the current GISS graph to the August 25, 1999 GISS graph. Warming ended in 1998, just as UAH, RSS and Had Crut show.
‘
The image below superimposes Had Crut on the image above. Note that without the post-1999 gymnastics, GISS and Had Crut match quite closely, with warming ending in 1998.
Conclusion : GISS recently modified their pre-2000 historical data, and is now inconsistent with other temperature sets. GISS data now shows a steady warming from 1975-2010, which other data sets do not show. Had GISS not modified their historic data, they would still be consistent with other data sets and would not show warming post-1998. I’ll leave it to the readers to interpret further.
————————————————————————————————————-
BTW – I know that you can download some of the GISS code and data, and somebody checked it out and said that they couldn’t find any problems with it. No need to post that again.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
latitude says:
August 31, 2010 at 10:17 am
James Sexton says:
August 31, 2010 at 9:41 am
==========================
But James, we all know that 1998 (12 years ago) was the dark ages.
You can’t trust any of the temperature data from way back then……;-)
This “science” needs a much bigger shovel………
==========================
Naw, they’ll just code an algorithm to make the mess decrease to eventual nothing to where it never really happened……… retroactively, of course. Much like
19841998. Dang, now I’m confused about what work of fiction is what?Will the real 1998 please stand up? Is it the 1998 of the 1998 vintage or the 1998 of the 2005 vintage or the 1998 of the 2010 vintage? OMG!!! I just realized GISS has fixed the quantum time difficulty!!! It shouldn’t be long before they solve the “grandfather paradox”!!! Phi isn’t used after all!!!
“Steven Mosher
The 1998 temperature was adjusted downwards years later. That is called “adjusting past history.” You focus on minute details and completely miss the big picture.
You display a classic case of “can’t see the forest for the trees.” You also display a classic case of “The Emperor’s New Clothes.””
Not just that, but EVERY past year seems to get adjusted DOWN. Always. Look at the two graphs. Pick a year, any year out of the last 20. 1992. Higher in the 1998 graph than the 2010 graph (.17 vs. .12). Look at 1990. .48 (1998) to .38 (2010). They ALL go gown, always. It would seem to me that if what Mosher says is happening is really happening, some would go up a little, some would go down a little, but by and large they should bounce around the same point by a few hundreths. Not ALWAYS go one direction (down) by up to a tenth.
Something else is going on…
As an engineer, I have an internal BS meter, and this is pegging it on “10”…
Venter says:
August 31, 2010 at 9:50 am
there are two issues. Hansen and co have been trumpeting about their hottest ever 2010 based upon GISS data. They certainly did not come to Moshtemp or to Jeff ID for their data.
Actually I don’t believe that they are trumpeting that, they said that the last 12 months were the hottest in their timeframe which is not the same thing. You’d have to produce the 12 month running average for the other databases to do a comparison.
The fact that UAH, CRU and RSS do not sow as 2010 as hottest and GISS is the only dataset of the 4 currently in use which says 2010 is the hottest is itself proof of their statements.
Really, I’m fairly sure that RSS has indicated that 2010 is the hottest to July. Here’s a quote from Roy Spencer re UAH:
“As of Julian Day 212 (end of July), the race for warmest year in the 32-year satellite period of record is still too close to call with 1998 continuing its lead by only 0.07 C:
YEAR GL NH SH TRPCS
1998 +0.62 +0.73 +0.51 +0.90
2010 +0.55 +0.74 +0.36 +0.63”
Judging by the daily data on AMSU-A on NOAA-15 it’s still going to be neck and neck going into Sept.
Hell, look at 1940!!!
~.16 in 1998, ~.05 in 2010. What the hell “new data” did we get in the last 12 years on data 60 years old, that adjusted it (not shockingly) DOWN?!?!
Yeah, not buying it at ALL….
“Really, I’m fairly sure that RSS has indicated that 2010 is the hottest to July.”
Incorrect. 1998 leads in RSS as well.
RSS data: http://www.remss.com/data/msu/monthly_time_series/RSS_Monthly_MSU_AMSU_Channel_TLT_Anomalies_Land_and_Ocean_v03_2.txt
1998: .635C
2010: .552C
The only group that has 2010 hotter than 1998 is GISS. UAH, RSS, and Hadley all have 1998 as the leader through July.
Oops, my bad, grabbed the wrong numbers.
These are RSS numbers for 1998 and 2010 throught July. The ones above are UAH. Appologies…
RSS
1998: .653C
2010: .593C
NASA GIS station temperatures could earlier go back to about 1800. I have a print from 2003.
Today all temperatures earlier than 1880 are deleted!!! As far I know mean temperatures were higher before 1880. To get a constant rise in temperature they had to cut just there. Is that the reason?
I have never seen so much tap-dancing in my entire life, just to say “yeah, they backcasted and adjusted it down”
Which still means that any “records” they say about today’s temperatures are not worth the paper they were printed on……….
From: Steven Mosher on August 31, 2010 at 8:57 am
Actually people don’t consider it a “mistake”…
If they buy your arguments, then this is deliberate as it is part of the algorithms, such revisions of rankings and amounts of anomaly are normal and should be expected.
If they examine the directions these adjustments take and conclude there is a consistent pattern to these revisions that continually yields a certain result, they will view this as deliberate.
A mistake would be an inadvertent error, just notify GISS of the problem and it’ll get fixed. Does anyone here think that’ll happen?
Either way of seeing it as deliberate, especially the latter, leads one to question the science of (A)GW, especially the “definitive” version of “AGW science” espoused by Hansen the Activist.
BTW, talk of how the “BEST” method shows more warming, actually talk of the accuracy of any of those particular methods, really doesn’t cut it. I am really agreeing with the view that total Ocean Heat Content is what we need to be looking at. Far away from Urban Heat Islands and airport tarmac lies the largest surface area and what is soaking up the largest amount of solar radiation. If you were looking for changes based on atmospheric CO2 concentrations, the sought-after AGW signal, that is where it would be found. Air temperatures over land, which don’t tell us of the actual heat content without humidity measurements, with the atmosphere having much less heat content compared to the oceans anyway, with the taking and recording of said measurements filled with potentials for error and bias, with the processing showing such variance, just ain’t telling us what we really need to know.
Well, Goddard gets it wrong again. The first graph he shows, from a 2007 paper BTW, is a land only graph. The second one is (as stated clearly in the heading) is the land-ocean index. Big oops there.
kadaka (KD Knoebel) says:
August 30, 2010 at 2:25 pm
http://www.swcs.com.au/ansi.htm
Seven of the eight bits (which control the first 128 characters) have been standardized world wide and are known as ASCII – American Standard Code for Information Interchange. In the all encompassing Unicode, this set is referred to as Basic Latin. However, with the next 128 characters — known as the extended set — a number of variations have arisen.
Below is the character set sometimes referred to as the ANSI code page, or the ISO-8859-1 character set. This character set is made up of two groups: printable characters and control characters. Because of the additional control characters that ANSI (and ISO) then substituted for the printable characters in codes 128-159, it is now more correct to refer to it as the Windows-1252 code page. However, it retains the name ANSI code page in Microsoft’s Notepad for historical reasons.
And the first 8 items of the ANSI code page:
0 NUL 1 ☺ 2 ☻ 3 ♥ 4 ♦ 5 ♣ 6 ♠ 7 BELL
Please avoid trying to rewrite the history of those of us who lived through it. And enjoy this ANSI smiley.
☺
You’re still confused.
The fact that Bill Gates decided to map the byte value 01 (hex) to a smiley face in some software/hardware situations does not affect the ANSI standard:
http://tinyurl.com/2ejp8qo
The fact that some Windows programmers mistakenly call this character mapping “ANSI” does not make it so.
01 is the SOH control character: Start of Heading. It is not a graphic character that shows up as a glyph.
Your own citation shows Bill Gate’s motivation in assigning graphic characters to these byte values that, in ANSI, are invisible control characters:
Now look at this page in more detail – your browser is probably automatically choosing the “Unicode” encoding for the bytes within this browser page. Force it to use Western (ISO-8859-1) encoding, and see what happens to your “ANSI smiley”.
Or, like I said, use a hex editor to look at the actual bytes encoding your “ANSI smiley” in this page – it is 263a (hex). Two bytes. Not ANSI.
As we used to say at MIT – RTFM.
And the SST’s are showing constant downward trend currently, as per Dr.Roy Spencer.
Smokey says:
August 30, 2010 at 1:28 pm
Knew it! No cojones — from the guy who pretends to be knowledgeable.☺
Anu says he might make a prediction , maybe [I predict he’ll chicken out again],
Again ?
I made a precise prediction about this summer’s Arctic sea ice extent minimum in early June:
http://wattsupwiththat.com/2010/06/06/wuwt-arctic-sea-ice-news-8/#comment-404240
http://wattsupwiththat.com/2010/06/06/wuwt-arctic-sea-ice-news-8/#comment-404352
IARC-JAXA numbers, less than 2009 – which was 5,249,844 sq km.
So did stevengoddard, which he clarified as to “error bars” more recently:
http://wattsupwiththat.com/2010/08/22/sea-ice-news-19/
IARC-JAXA numbers, 5.5 million sq km, finishing above 2009 and below 2006.
Good – now we know the prediction was not 5.5 million sq km ± 3 million sq km. It’s an actual prediction which can be proven wrong – if it finishes below 2009.
In which case my prediction will have been right. There is no need for me to whine about the “wind” not cooperating – it’s just simple annual variability (winds, clouds, ocean currents, forest fires, etc) superimposed on a strong climate trend.
Whether I make another prediction here at WUWT is the “maybe” part. I’ll see how the WUWT Arctic sea ice skeptic-expert handles being wrong first. I hope he sticks around long enough to accept defeat. The summer melt season post-mortem could be a nice learning moment for WUWT – since it’s founding in late 2006, the Arctic summer minimums have been 2007 (way down), then 2008 and 2009 (up and up). The “up and up” has been given lots of significance here – a new “down” might generate
funnyinteresting responses.but then he says: “Of course, any prediction I make would lie somewhere between young molten Earth and snowball Earth,” the ultimate Sharpshooter’s Fallacy, making Hansen’s giant error bars look puny by comparison. Very brave prediction… not.
Try to read more closely. That wasn’t my “prediction” – I was reminding you that you consider all future climate change to be within “natural variability” – a worthless concept.
Which moderator is censoring my posts?
[REPLY — They’re not deleted. Just set aside temporarily. Look, I know you’re het up, but Anthony’s going through a very bad time. We all need to tone it down a bit. (It doesn’t matter who started it.) And fear not, others will suffer a few snips, too. ~Evan]
Anu,
It is a standard betting trick to bet after, and just below or above other competitors. You have a 50/50 shot [snip].
[REPLY – Please, guys. Stop taking shots at Steve (et al.). EVERYONE, please stop taking shots. I really, really hate censoring posts, but it’s all gone far enough. CEASE FIRE. Our host is going through a bad time not of his own doing. So ramp it back. ~ Evan]
Evan,
Don’t let other people post ad homs about me if you won’t let me respond. That is a pretty ridiculous policy.
[Fair enough. ~Evan]
Rattus Norvegicus
Your claim about the land-only graph does not appear to have any basis.
http://www.giss.nasa.gov/research/briefs/hansen_07/
I have a new Arctic post up at
http://stevengoddard.wordpress.com/2010/09/01/nsidc-increases-their-arctic-forecast/
stevengoddard says:
August 31, 2010 at 10:48 pm
Rattus Norvegicus
Your claim about the land-only graph does not appear to have any basis.
http://www.giss.nasa.gov/research/briefs/hansen_07/
Actually it does as clearly shown by the reference you cited.
http://climateinsiders.files.wordpress.com/2010/08/paintimage1726.jpg
is land only, here’s the missing legend:
“Fig. 1: Annual and 5-year mean surface temperature for (a) the contiguous 48 United States and (b) the globe, relative to 1951-80, based on measurements at meteorological stations.”
The graph you compared it with is clearly labelled as global land-ocean temperature index:
http://data.giss.nasa.gov/gistemp/graphs/Fig.A2.lrg.gif
Phil. and Rattus
You mean Steven made the kind of error that might occur looking at graphs and
counting pixelscomparing them instead of reviewing the underlying data that generated them? Who’d have thought?I knew Steven was writing a post without merit because the issue of the changing estimates over time was long ago put to bed, known not to change the trend significantly over time and the people at GISS must have been sighing in disbelief over his exposition and what an ill-informed waste of time it was.
Phil. and Rattus’s observations made me go check the underlying data.
Here is the current anomaly data for every month of 1998, as of July 30th 2010,
55 101 68 67 82 87 84 75 47 57 53 64Source
Here is the monthly anomaly data for 1998 from 2000 (file dated January 21st 2001) which I scraped for John Goetz back in 2008. It is no longer available at archive.org as GISS has stepped in to disable archiving. Remember I never said they were nice guys.
59 93 69 80 76 71 80 71 43 53 53 69Source which I pulled two years ago. John Goetz can attest to its authenticity.
If you look at the information at the top, you obtain the anomaly data in degrees celsius by diving by 100.
Averaging the year 2010 data for 1998 and diving by 100 I get: 0.070° C
Averaging the year 1999 data for 1998 and diving by 100 I get: 0.068° C
So the estimation algorithm has actually made the 1998 anomaly 0.02° C higher in 2010 than the the value calculated and used in 1999/2000.
Or in other words exactly the opposite of what Steven Goddard has claimed has occurred. The GISS algorithm has not cooled the past in this case of the crucially important 1998. It has warmed it.
Checkmate.
Game Set Match.
The Fat Lady has sung.
No, the people at GISS are not sighing in disbelief over Steven’s Post, they are rolling on the floor laughing so hard they are pissing themselves.
And you cheerleaders for Steven wonder why Mosh and I say Posts like this reduce the credibility of WUWT. At least on WUWT a dissection is allowed which lets us get the root of the problems. That is something. A little more honest skepticism from the cheerleaders would be an improvement.
From: Anu on August 31, 2010 at 8:53 pm
Tried that with my “back-up” browser, Epiphany, comes with GNOME, set it for Western (ISO-8859-1) encoding. Guess what? Smiley is still there.
Didn’t need a hex editor, just a simple ctrl-u from Epiphany to look at the page source. Found the 263a coding for the smiley.
You know why? Apparently not. See, at the top of the source it says “(left-arrow) meta charset=”UTF-8″ / (right-arrow)”, which is HTML 5 notation. WordPress served up a UTF-8 (Unicode) document.
My browsers are not choosing to display Unicode, they have properly recognized they were given a UTF-8 document so Unicode is what is displayed. I can’t “force” them to do otherwise, I can only set the default encoding and the document has specified what encoding to use.
BTW (emphasis added):
So by that draft HTML 5 specification, which is already incorporated into modern browsers as possible and practical, I can tell the browser to use ISO-8859-1, the document can declare itself to be ISO-8859-1, and the MS ANSI smiley would still be there. The test you proposed was improperly designed.
And you are still have not addressed the point I was originally making.
How do you know, back when Smokey was typing his comment, if he inputted a Unicode or an ANSI smiley? Can you tell if he used the long ctrl-shift-u 263a sequence or the short alt-1(numeric keypad) sequence?
The answer, of course, is that you can’t know. You stated he used the Unicode smiley, after all your hex editor gave you the 263a sequence. But you went off the info at your end, processed through browsers, ISP’s, and servers, that served to you the smiley as Unicode. Thus you never had the info to determine what smiley he used to begin with. Stating what smiley it was, was perhaps a 50/50 guess. At least it would be if it wasn’t for the prevalence of Windoze boxes. By trusting the info you did, you likely picked the losing end of worse than 1 in 10 odds.
And after MIT they can be saying “Would you like fries with that?” But that could just be the economy, of course. ☺
Reply: ι ∂ση’т тнιηк ι’νє ∂єαℓт ωιтн ѕυ¢н αη 域 тσρι¢ ¢σηνєяѕαтιση вєƒσяє. ~ ¢тм
jeez, rattus, Phil.: Brilliant.
MattN
Your BS meter was pegged at 10?
I’ll suggest that it was measuring what you and others were saying.
especially Goddard.
That’s what this engineer thinks.
Oh, and Steven, just so you can compare apples to apples, the GISS anomaly for Land only for 2005 is 0.77 ° C not 0.61° C as you wrote above, since you are using the Land only graph for 1999.
I don’t have data to compare Land + Sea Surface 1999 data to 2010 data, but from what I know of the code, perhaps Mosh can confirm, the direction of any change in estimations would be consistent with the Land only index.