BPSDB
303 is the number of consecutive months we have now had with temperatures greater than the mean for the 20th Century.
303 months is a little over over 25 years, so it may sound impressive – but does it actually mean anything?
Well, yes. It means that the last 25 years have been warm, but then we already knew that. it tells us nothing about why it has been warm or what it is going to be like in the future. In isolation it is just a number, nothing more. And in a couple of weeks it will change to 304.
Some more records for the year so far:
- According to the NOAA, May 2010 was the hottest May on record. They also claim new records this year for:
- The warmest March and the warmest April
- The warmest January to April period
- The warmest January to May period
- The warmest March to May period
- In a paper submitted to Reviews of Geophysics Hansen et al claim that the last 12 months (to the end of April 2010) have been the hottest 12 month period on record.
- The NSIDC gives us the latest Arctic sea ice maximum on record, and
- The rate of Arctic sea ice decline through the month of May was the fastest in the satellite record.
All this has led to speculation about what the rest of the year might bring. Will we see a new record annual temperature, or a new sea ice minimum, or both?
On the temperature front, a lot depends on what ENSO gets up to in the rest of the year. We’re now in a neutral state; the models suggest that a La Niña is likely to develop in the latter half of the year which will cool things down a bit. I wouldn’t put money on a new record, but it is possible (note that both 1998 & 2007 went into a La Niña in the latter part of the year).
With sea ice, it’s anybody’s guess (and there’s no shortage of people prepared to guess).
ARCUS/SEARCH have a selection of predictions, nearly all of them serious. The one exception (although I’m sure the author is serious) is the one by Charles Wilson – Gareth has aleady covered this one on Hot Topic, well worth a read.
Of the others, about half place this year between 2008 and 2009 which would mean the third lowest extent on record. Only one (ignoring Wilson’s contribution) predicts a new record this year – and not by a very big margin.
I liked the approach taken by Polar Science Weekend, a nice way of involving the public.
Robert Grumbine discusses his submission on his blog, which is interesting reading. William Connolley also thinks that Grumbine’s approach is interesting, but wrong. He’s taking bets on the minimum (and betting that it will be more than 4.935 km2, again somewhere between 2008 & 2009).
In summary, then, it seems that the majority of the people who actually know about this stuff think that the downward trend will continue – but that we’re not very likely to see a new record this year.
There are other opinions, of course. Over at WUWT, on May 29th we had Watts & Goddard calculating that Arctic Ice Volume Has Increased 25% Since May, 2008. PIOMAS disagrees, but hey, who needs models when you’ve got a calculator?
On the 14th June Goddard followed this up with a confident prediction that this year will have the greatest extent since 2006.
My favourite bit is where someone asked him for a figure in sq. km. He replies that he doesn’t have a such a value, since his “measurements are in pixels”. 
Goddard followed this up with another post on the 25th June, much to Tamino’s amusement, and then kind of shot himself in the foot on the 2nd of July (though to be fair, he acknowledges that he got it wrong).
The real point is that when you have a system with a reasonable amount of variability (like climate or Arctic sea ice) then it doesn’t really mean a lot when a record gets broken. Pretty much any record will get broken eventually if you’re prepared to wait long enough. If there happens to be a trend then the probability of a new record (in the direction of the trend) increases with time, just as the chance of one in the opposite direction decreases.
Trying to claim that Arctic sea ice has been recovering since 2007 is probably even more stupid than Monckton’s claim (in 2008) of Global cooling by 0.4 °C/decade.
Of course those of you with a knowledge of statistics will understand this. For those of you that don’t, a couple of pointers:
- 2005 had (by far) the lowest Arctic sea ice extent on record at the time.
It is now the 4th lowest (and likely to become 5th this year) - 1997 was (at the time) the warmest year on record. It is now 12th. Every single year from 2001 onwards has been warmer than every year prior to 1998.
IMAGE CREDITS:
[1] – Wikimedia
[2] – NOAA
[3] – Polar Science Centre
[4] – Atmoz
(It should have been Dr. Roger Pielke Sr., but that site is down at the time of writing)
Comment Policy
Comments that are not relevant to the post that they appear under or the evolving discussion will simply be deleted, as will links to Denier spam known to be scientific gibberish
- The “Mostly” Open Thread”
is for general climate discussion that is not relevant to a particular post. Spam and abuse rules still apply; - The “Challenging the Core Science” Comment Thread is for comments that purport to challenge the core science of anthropogenic climate change.
For more on records set, see
http://climateprogress.org/2010/07/05/heat-wave-global-warming/
“New daily high temperature records beat new cold records by nearly 5 to 1 in June”
also record highs vs record lows over the past several decades – trending up
One could calculate the possibility on this 303 month period happening by chance, and compare it with poker hands or some such…
303 consecutive months. That “consecutive” brings home the message that the warming is here and now. Unfortunately over 25 years human memory fades and that more than anything is what is driving the denial chorus on the internet.
Trying to claim that Arctic sea ice has been recovering since 2007 is probably even more stupid than Monckton’s claim (in 2008) of Global cooling by 0.4 °C/decade.
When you make a claim that is more stupid than Monckton’s claims, you’ve put in a hard day at the office. Time to crack open a beer and think about how you can lie about the science tomorrow. After all, it’s another day.
Top post – similar reasoning here too
http://www.abc.net.au/rn/scienceshow/stories/2010/2859986.htm
“..when was the last month that global temperatures were below average? Anyone remember? A global long-term average. The answer is February 1994 by three ten-thousandths of one degree, it was really close but it was below average. That’s 192 consecutive months. Those of you who know anything about flipping coins, will it be hotter or colder?.. know that basically what you have is a binomial random variable, and if you’re really good at doing the math you’ll know that to get heads 192 times in a row you have a probability with (if I’ve typed that correctly) 58 zeros before you get to the first one.
Now imagine! You’ve just flipped a coin that’s come up heads 192 times in a row and you’re being accused of having a coin that’s loaded to come up tails. That’s really pretty spectacular. By the way, if you don’t think that three ten-thousandths of one degree is enough, the last time when temperature was one one-hundredth of a degree Celsius below average, below the long-term average on a global scale was, guess what, a quarter of a century ago right now; 300 consecutive months. To do that you need about 90 zeros. The probability is less than…if you have 100 zeros does that make it a googolplex? It’s a bigger number than I know what to call it, but 90 zeros in a row. That’s the probability that the globe is really getting colder instead of hotter.”
I’m probably missing something but what do you mean by “being accused of having a coin that’s loaded to come up tails”? What is the real-world equivalent? Isn’t the accusation the opposite?
A googol is 10^100. A googolplex is 10^googol or 10^(10^100)
TS
I needed to go to the full Freudenberg interview to get the context. He uses a coin-toss example to illustrate that the scientific bias is in favour of conservative estimates by the IPCC, rather than over-estimations (never mind worse-case scenarios).
The dice is ‘loaded’ to show that things are better than they likely are — something we already understand.
Thanks Martha. That makes sense.
Thanks Martha and apologies TS for chopping off too much of that article.
The loony thing is to say it’s ‘cooling’ or ‘coin that’s loaded tails’ yet every monthly measurement for the last 303 months is above the average, or 303 head streak on a tail loaded coin.
The punters at 2-up would be asking questions for sure
Thanks, that was a good read.
I’ve just checked Freudenburg’s maths, and he’s pretty much spot on. In orders of magnitude he’s absolutely precise.
the argument on the link about ENSO causing runs of colder and hotter months than the average doesn’t hold since there has been la Nina periods during these 303 months. This is not a valid explanation.
“1997 was (at the time) the warmest year on record. It is now 12th. Every single year from 2001 onwards has been warmer than every year prior to 1998.”
Not according to the graph here:
http://www.co2science.org/articles/V12/N14/surfaceairtemp.jpg
from this article:
http://www.co2science.org/articles/V12/N14/EDIT.php
samspade
The site you deposit as a link is notorious for deliberately ignoring the longterm trend (among other things).
Since you are here on a credible science blog, and (unlike the site you link to) this is not Exxon-funded try:
http://greenfyre.wordpress.com/denier-vs-skeptic/denier-myths-debunked/myth-global-cooling-aka-climate-change-stopped-in-1998/
http://greenfyre.wordpress.com/2009/10/26/independent-statisticians-reject-global-cooling-fable/
Of course the HadCRUT data does not include many regions with the strongest current warming so if that is really all you ever wish to look at, it is to be expected that it shows the lower end of likely warming.
samspade:
To be fair, my comments were based on NASA’s GISTEMP record.
Had I used the Hadley record as favoured by co2science, then (because Hadley have 2008 as 0.02 °C cooler than 1997 as opposed to GISTEMP’s 0.04 °C warmer), I would have changed the above to:
Much less alarmist, as I’m sure you will agree.
—-
Martha – many thanks for the links.
I should probably have linked them myself, but I’m still kind of new to this.
…or here (though I may be reading the data incorrectly)…
http://www.drroyspencer.com/latest-global-temperatures/
samspade
http://climateprogress.org/2010/03/17/global-cooling-hottest-january-february-march-uah-satellite-data/
To understand Roy Spencer’s mistreatment of this data (among other things).
The 2 satellite datasets show similar results. Every year after 2000, except 2008, is warmer than every year before 1998, and even 2008 is warmer than most years before 1998.
Let’s use Spencer’s current favourite, the 13-month rolling average. The result is the same. This looks like Spencer’s graph to me.
The HADCRUT and GISTEMP series show a much smaller drop in 2008 so even that year is above any before 1998.
Here’s the 13-month rolling average version.
Not according to the graph here
(..)
from this article:
(..)
…or here.
Well that settles it then.
NOAA are clearly a bunch of liars!
You’ve discovered the conspiracy.
(…awkward silence…)
Send them a strongly-worded email and make sure that they get those valuable links too.
That will set them straight.
After all, if there are two sources of information then it stands to reason that one should be given just as much attention as another.
Right?
Get in the sack.