Graph by Anthony (click for larger image) text by Dr. Roy Spencer from his blog here
May 2009 Global Temperature Update +0.04 deg. C
June 4th, 2009 by Roy W. Spencer, Ph. D.
YR MON GLOBE NH SH TROPICS
2009 1 0.304 0.443 0.165 -0.036
2009 2 0.347 0.678 0.016 0.051
2009 3 0.206 0.310 0.103 -0.149
2009 4 0.090 0.124 0.056 -0.014
2009 5 0.043 0.043 0.043 -0.168
May 2009 saw another drop in the global average temperature anomaly, from +0.09 deg. C in April to +0.04 deg. C in May, originating mostly from the Northern Hemisphere and the tropics.
A reminder for those who are monitoring the daily progress of global-average temperatures here:
(1) Only use channel 5 (”ch05″), which is what we use for the lower troposphere and middle troposphere temperature products.
(2) Compare the current month to the same calendar month from the previous year (which is already plotted for you).
(3) The progress of daily temperatures should only be used as a rough guide for how the current month is shaping up because they come from the AMSU instrument on the NOAA-15 satellite, which has a substantial diurnal drift in the local time of the orbit. Our ‘official’ results presented above, in contrast, are from AMSU on NASA’s Aqua satellite, which carries extra fuel to keep it in a stable orbit. Therefore, there is no diurnal drift adjustment needed in our official product.
Flanagan, Is it not true that UAH/RSS cannot use 2-3% of the globes surface?
Is it not also true, that Landbased measurements like GISS and Hadcrut only has 30% land area to begin with, and then has to extrapolate over the ocean (Including the arctic ocean..).
But you think Landbased temperature measurements has better coverage of the globe?
Fine point, David Archibald. I note that the temperature anomalies continue to decline, and as Lucia has pointed out, every month that this happens it decreases the chances that the models are correct, as the autocorrelation works both ways.
The thing I crave, more than any other, is watching Tamino. He laid out that bet, in such specific terms (good for him) that if it goes wrong he will have nowhere to hide. I only wonder if he has the guts to man up and admit it if he is wrong. Given the arrogance and childishness he displays, it is unlikely, but it will be fun to watch if he can manage it.
wow, I had never been to Climateprogress.com… Their post about WUWT makes me want to start keeping an ad-hominem counter for both sites, and compare the two numbers.
Neven (10:01:13) :
I don’t know how important David Archibald is in the Global Warming discussion, but he predicted an anomaly of -0.4 degrees C for the May 2009 UAH MSU Global Temperature Result here: http://www.warwickhughes.com/blog/?p=197
It goes to show that it’s not that easy to do short-term predictions.
The easiest one to predict in UAH is that May will have the lowest anomaly and February the highest. There’s about a 0.3ºC variation through the year.
RE: Frank Lansner (17:04:13) :
. . .
“Is it not also true, that Landbased measurements like GISS and Hadcrut only has 30% land area to begin with, and then has to extrapolate over the ocean (Including the arctic ocean..).”
Your statement is more incorrect than insightful. GISS and Hadcrut do use non-land based estimates for the ocean. GISS does have a convoluted reference to satellite data for the oceans, but its algorithm borders on bizarre and will not be detailed in this post. Hadcrut by necessity does not have a consistent source of data for ocean temperatures, and it likely that Hadcrut over relies on the Northern Hemisphere oceans. Nevertheless, neither GISS nor Hadcrut simply extrapolate land temperatures to the ocean as your post suggests.
David Archibald (16:13:25) :
I see that the overall direction of anomaly is down on both UAH and RSS.
When you made your prediction of -.04C for May 2009, what is your reading of the noise/uncertainty factor (i.e the +/-? Even Leif has a +/- in his solar prediction.
You were off by .083 for the exact reading. How much of that .083 is outside your uncertainty?
RE: George DeBusk (08:11:28) :
. . .
“Where can I find the UAH Lower Tropospheric data separated by land and ocean? I have only seen global data and data segregated by latitude bands.”
Look at http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
It has 25 different separations of the UAH historic record. The May data has not yet been added.
The running mean UAH data for the first 5 months of each year
1998 0.65
1999 0.11*
2000 0.03*
2001 0.20*
2002 0.36
2003 0.31
2004 0.30
2005 0.38
2006 0.26
2007 0.38
2008 -0.02*
2009 0.20
So far, 2009 is cooler than 6 out of the last 7 years
0.13 degrees below the 2002/7 mean
Bill P (12:59:25) :
A 120,000 acre-foot Strategic
Do you have that measurements in firkins, sydharbs or perhaps cubic attoparsecs!
http://en.wikipedia.org/wiki/List_of_unusual_units_of_measurement
Come on – science went SI units years ago. It is impossible to tell if temperature is being discussed in deg F or deg C on this blog (it makes a difference!).
Flanagan (04:36:06) : One can clearly see that this reconstruction shows very positive anomalies at the level of the poles
Well, with record rate of ice formation at the poles, I’d not trust a “reconstruction” showing warmth as much as I’d trust the very thick ice formed and observed. But maybe that’s just me… I’m just a Joe Sixpack type at heart who believes what he can see sitting right in front of him more than someone’s manipulated reconstructed interpolated data food product…
Does anyone know wether GISS and HadCrut take some polar staiton data into account?
GISS uses Antarctic station data. For the technically inclined (ie. computer code geeks) the process of initial Antarctic data merging is described here:
http://chiefio.wordpress.com/2009/03/01/gistemp-step0-the-process/
Unfortunately, when creating, er, calculating anomalies they use a satellite homogenized pasteurized processed data food product “SST anomaly map” that is in part based on fabricated, oh, sorry, estimated interpolated synthesized arctic “temperatures” based on ice estimates which, IMHO, are suspect – especially given how the arctic ice sensors on some of the satellites have keeled over and given broken low ice readings…
A not so technical critique of this is here:
http://chiefio.wordpress.com/2009/03/05/illudium/
By the time GIStemp is done making it’s “adjustments” the anomalies have little to do with real temperatures. The canonical collection of my critiques, not-so-technical, are here:
http://chiefio.wordpress.com/category/agw-and-gistemp-issues/
(Sitting in a cold house with heavy clouds and cold outside, and waiting for the snow forecast in the Sierra Nevada mountains here in California… In JUNE …)
rbateman (22:34:07) :
David Archibald (16:13:25) :
I see that the overall direction of anomaly is down on both UAH and RSS.
When you made your prediction of -.04C for May 2009, what is your reading of the noise/uncertainty factor (i.e the +/-? Even Leif has a +/- in his solar prediction.
You were off by .083 for the exact reading. How much of that .083 is outside your uncertainty?
The prediction was for an anomaly of -0.4 not -0.04 . The prediction was off by ~0.44 degrees. There were those of us who said it was complete nonsense at the time but we tend to get ignored. Just as there are those of us who think that current relatively low satellite temperatures are a lagged response to the low SST which were evident a few months back.
Ozzie John (05:25:12) : Not sure how the maths ae calculated here but can someone explain how the global average is 0.043 when the tropics are -0.168 and both NH & SH anomolies are 0.043.
Because anything to the right of the decimal point is just playing in the error bands of the calculations and is entirely the product of False Precision:
http://chiefio.wordpress.com/2009/03/05/mr-mcguire-would-not-approve/
So any time you compare a satellite temperature to a historic “base line” you have temperature records in whole degrees F giving you a false sense of what happened with fantasy values in the fractional part. If the satellite data have accuracy of less than 1/100th degree, then they, too, are just playing in their error band. Somehow I don’t think that they have accuracy to 1/100th degree (NOT precision, accuracy…) with all the junk in the air between the satellite and what it purports to be measuring. Certainly putting up numbers with values out in the 1/1000 place, like 0.043 is just astounding wishful thinking.
Robert A Cook PE (06:44:42) : So, if we compare the above record to the ORIGINAL (pre-Hansen) surface temperature record, what do we find for rural and small town sites? Once UHI effects are removed, what id left? How much have we cooled since 1935-1945?
The unmolested data are directly downloaded from NOAA by GIStemp. The first steps just merge in some Antarctic data and not much else. Simply take the GHCN data and use it rather than the GIStemp “stuff”. (Input files listed in prior link).
“(2) Compare the current month to the same calendar month from the previous year”
Do these temperatures not cover an even portion of the globe? Why should we not compare to any month? Seasons should have nothing to do with global temperatures. — John M Reynolds
Neven (13:27:17) : But for me AGW is just part of a collection of problems that steadily grow more serious and irreversible, such as the peaking of resources (peak oil naturally being the most conspicuous one),
Neven, there is no shortage of resources:
http://chiefio.wordpress.com/2009/05/08/there-is-no-shortage-of-stuff/
and there is no shortage of energy, and there never will be:
http://chiefio.wordpress.com/2009/03/20/there-is-no-energy-shortage/
disruption and destruction of ecosystems
This is the real issue that needs addressing.
I think the underlying cause of all global problems (such as the present economic downturn) is the concept of exponential growth.
Real biological systems have an S shaped growth curve, not exponential. That is the basic flaw of Malthusian Catastrophe Theory. My dad was one of 12 siblings. I’m one of 4. I have 2 kids. etc
Please look at the demographics of Europe or Japan. Below replacement rate. Most of the US population growth is immigration, not births. The way to reduce population growth is: Give every woman on the planet a college education and every family a modern, prosperous, western lifestyle. Population growth drops below replacement.
The way to population catastrophe is to keep folks in poverty and deny them an advanced economy. To the extent the AGW alarmists succeed in stopping economic expansion via Cap and Trade and other economic growth limitations, they guarantee higher population growth rates. (This is not my speculation, this is the standard population dynamics and the impact of modernity stuff in University Economics degree programs).
I find it particularly hard to believe that this exponential growth can be infinite in a finite system. But that’s what they have us believe, or at least that’s what almost everybody is urging you not to think too much about.
I don’t know who “they” are, but the idea that population grows exponentially is broken. The early phase looks exponential, but it isn’t.
Not that there is anything wrong with growth, on the contrary. However, after a certain point growth increasingly takes on the form of a cancerous tumour.
Actually, this metaphor is broken too. Both population and economic growth tend to be like that of a tree. Rapid at first, then slowing with size. Probably for the same reasons, too. There is a too complicated and off topic discussion that applies here. I’ll just hint at it with this: All these processes have an interesting connection with the math of fractals. A recent breakthrough in biology has shown that the growth pattern slows due to the fractal nature of organism growth. It is literally a mathematical limit / effect…
http://www.ercim.org/publication/Ercim_News/enw29/vicsek.html
http://www.sciencemag.org/cgi/content/abstract/179/4079/1201
Mike Bryant (14:40:21) : IF the internet becomes a place where one’s views cannot be freely expressed this blog and many others will no longer exist. However freedom will not be denied forever.
Um, er, ah, some of us, I mean, of some other folks, have already made preparations toward that day…
Do a google search on “stunnel” and “tcfs file system” …
http://www.linuxjournal.com/article/2174
There has also been interesting work done on shared systems which only become visible once a quorum has signed in. You can compromise up to a configurable limit of individual passwords, but still get nothing. Only when a quorum presents credentials does the data become visible… (Invented in Italy and ideal for certain organizations that I’m sure had nothing to do with the development funding 😉
So rest assured that whenever the need arises “someone” can hide an infranet in plain sight…
The term Infranet is also used to refer to a proposed system of transferring sensitive data masquerading as anodyne data (an example of steganography) over the Internet. Its purpose is to enable people in territories whose régimes exercise strict control over information to have uncensored access to the web without risking censure. The system is currently under development by Hari Balakrishnan, David Karger and others.
Not that I would ever advocate it, of even know anything about it, in fact, I found this on a Cereal Box… gotta go…
Frank Lansner (17:04:13) : Is it not also true, that Landbased measurements like GISS and Hadcrut only has 30% land area to begin with, and then has to extrapolate over the ocean (Including the arctic ocean..).
It’s actually worse than that. While we have modest spatial coverage now, it does not have much depth in time, so there are extrapolations / interpolations / fabrications used to fill in most (by a very large degree) of space and time:
http://chiefio.wordpress.com/2009/02/24/so_many_thermometers_so_little_time/
It’s almost entirely the USA, Western Europe, and Japan…
http://chiefio.wordpress.com/2009/02/28/hansen-global-surface-air-temps-1995/
EM Smith,
I love that phrase. “…manipulated reconstructed interpolated data food product…”
Maybe the right word for all the new temperature, sea level, sea ice and innumerable other so called data is… VELVEETA DATA or Velveedata…
bill (02:59:15) :
Bill P (12:59:25) : Come on – science went SI units years ago.
Bill, science does not depend on what units you use, only on how you use them. Some of us are comfortably multilingual for units. Get over it.
BTW, an acre is simply 1 chain by 10 chains of area. Very metric, in a way…
And the foot has a very long and surprisingly rational history and physical basis that has nothing to do with the size of a king’s shoes…
http://chiefio.wordpress.com/2009/06/06/chasing-the-greek-foot/
Who does among us all feel a 0.5°C difference?.
Here in WUWT it has been demonstrated that temperature records are wrong up to 5.0°C! (surface stations)…and satellites are “adjusted” according to surface stations…a dog biting its tail.
The fact that the tropics are cool and the higher latitudes are warmer may be telling us something relevant about CO2 warming. If this is how CO2 warming will pan out over time then it demonstrates that AGW will be more beneficial than anything. Areas with warmer climates won’t change much (or get slightly cooler) and areas that are cold will see warmer (=nicer) weather.
Another possibility is that what we are seeing is negative feedback in action. More rain and clouds in the tropics acting to lower the global temps.
“Leif Svalgaard (11:29:28) :
DavidsBSD (09:42:54) :
Any one concerned that all three metrics, G, SH, NH are the same? How likely is that?
I would be concerned if G would be different :-)”
Leif, your question reminds me of this Dutch joke:
Question: What’s the difference between a dead bird?
Answer: His one wing has the same length.
So my question is: Different from what, Leif?
G is almost always different from either SH or NH, hence all three are likely to be different from each other. So why would you be concerned if this is not the case or are you trying to be overly semantic?
Adam Soereg (11:31:22) :
Ozzie John (05:25:12) :
Not sure how the maths ae calculated here but can someone explain how the global average is 0.043 when the tropics are -0.168 and both NH & SH anomolies are 0.043.
Am I reading this correctly ?
Northern Hemisphere means all area to the North from the Equator, not just the northern exratropics. You can get the global average temperature anomaly relative to the 1979-98 period by averaging the NH and SH anomalies.
The tropics data covers the whole area between 20°N and 20°S, but it is also partly covered by the Northern and Southern hemispheres.
The current tropical anomaly is negative, when the NH and SH are both standing on 0.043. It means that we can find the larger positive anomaly in extratropical areas.
Not only is it surprising that the SH and NH anomalies are equal (as I pointed out above; a quick glance through the UAH data seems to show that it is unprecedented), but also the discrepancy between the tropics and hemispheric data. If my calculus is correct the area between 20ºN and 20ºS is 34% of the surface so if the anomaly there is -.168 then the extratropical anomaly for May should be about +0.15. It’ll be interesting to see the actual data on the UAH site.
REPLY: I mentally noted this 0.043 for NH and SH and wondered also, perhaps it is a typo made on Roy’s blog. When the actual UAH data is posted we’ll know. – Anthony
Chris Schoneveld (09:27:59) :
G is almost always different from either SH or NH, hence all three are likely to be different from each other. So why would you be concerned if this is not the case or are you trying to be overly semantic?
If SH and NH are the same, then G will be the same, so now we have not three coincidences, but only two. And there is a difference between a priori and a posteriori probability. If I consider a 13-card bridge hand that I just got and ask “what is the probability that I should get precisely this hand?” then my answer would be 1 in 635013559600 and should I question if there is something wrong with the deck or the dealer since I was dealt such an improbable hand? This is a priory probability. After having got the hand, the probability is 1. So coincidences are only improbable before they happen.
Furthermore the two hemispheres are correlated: if one is warm there is a good chance that the other one is too, in fact the correlation coefficient is 0.56.
Since 1880, there has been 20 cases before this one where the monthly anomaly has been exactly the same to 1/1000 of a degree in both hemispheres, so I don’t think I’m being overly semantic or pedantic.
David Archibald (16:13:25) :
John Finn corrects my 4am can’t sleep again question.
But I am still interested in your estimate of uncertainty in your prediction.