Really Sciency

Visit my other blog 'Really Sciency' looking at Climate Science and its portrayal, misrepresentation and denial in the media.

Tuesday, 13 March 2012

Goddard: Tampering With Data All Over The Planet

Of course that's Steve Goddard, from the Orwellian named Real Science, and nothing to do with NASA’s Goddard space centre.

Unfortunately Mr. Goddard doesn’t do very well with graphs. We have seen it on Really Sciency before, (still waiting for the correction), and his latest howlers were when he posted the animated .gif below;

Goddard uses this to make the claim;
 “Hansen has been deleting cooling trends all over the planet.”, “He has erased historical cooling trends (blue) in North America, South America, Africa, the Arctic, Antarctica, and elsewhere.”
And very helpfully draws little red rings around the areas he mentions. What he has discovered is shocking! The chilly light blue areas have suddenly become a less chilly looking white. Oh the humanity! Surely this is proof of tampering, proof of erasing cooling trends?

Hang on a moment, what about those very angry looking dark red areas on the same image as the blue (the ones he rather conveniently doesn't draw little red rings around), – haven’t those become a less warm orange? Drat those pesky scientists, (Hansen specifically), they have gone and erased the hot, hot, hot areas, clearly they are erasing warming trends as well! Drat and double drat.

Below I have included the maps as separate images for easy comparison;

But wait – the maps use the same colours but at the bottom they use different scales - Actually Goddard knows this because almost as a foot note he does say;
The color scales are shifted, but the legends are located below the maps – and you can see that the values have changed.”

The first thing that should ring alarm bells with any true sceptic is that Goddard  thinks comparing two maps with different scales is a perfectly reasonable thing to do. Who with any inkling of science and an idea to communicate would even let such a notion cross their mind and publish such shoddy work as a credible evaluation? Well Steven Goddard obviously.

The second thing is his preoccupation with his own bias. If he is right to be concerned, and cooling trends are really being ‘erased’, then why does this concern not extend to warming trends being erased as well?

Well the truth is obvious, Goddard has his answer and is trying to get ‘facts’ to convince himself and others it is right, instead of looking at all the facts and drawing a rational conclusion from them. If he was an honest man, even believing he's right, he would be as concerned for the changes at both ends of the scale, not just the bits he thought made an anti-AGW case. That is why he will always be prone to cherry picking and making howlers such as the ones I will show below.

So the only question that remains; Is Goddard correct about trends (at either end) being erased?

Goddard actually gives a specific example to reinforce his case;
“The circle over Africa has changed from a -0.5C to -0.3C cooling trend, into a warming trend.”
I can’t see exactly where he got these figures from. The cool area circled over Africa in the first map ranges from around -0.5 to +0.1. In the second map the cool area ranges from around -0.2 to +0.5, so the values are slightly different even if they are not the values Goddard works it out at.

An easier area to work out is the circled U.S. area. It goes from -0.5 to -0.1 in the first map to -0.2 to +0.2 in the second map.

But doesn’t that make Goddard correct about erasing trends? They are different in each map after all. Well all that really depends on if these maps are really showing the same data.

Goddard gives links to where he got the maps from. The second is straight forward; it is the GISS Surface Temperature Analysis using GHCN v3.

The first map however is over 12 years old. Goddard gets if from a brief piece on the NASA, Goddard Institute for Space Studies, web site from August 1999 about the US climate at that time. It is worth a read if only to understand how real scientists view data, not as Steve Goddard has done above, but with guarded statements, based only on what the data, and all the data, supports. Of the drought that occurred in the Eastern U.S. in 1999, Hansen and his team have this to say;

“Was the heat wave and drought in the Eastern United States in 1999 a sign of global warming?
Empirical evidence does not lend much support to the notion that climate is headed precipitately toward more extreme heat and drought. The drought of 1999 covered a smaller area than the 1988 drought, when the Mississippi almost dried up. And 1988 was a temporary inconvenience as compared with repeated droughts during the 1930s "Dust Bowl" that caused an exodus from the prairies, as chronicled in Steinbeck's Grapes of Wrath.

Does this sound like a scientist prepared to ‘erase’ trends or one that admits the data does not support anthropogenic warming in the U.S. at that time?

Anyway I digress. The map Goddard uses for his ‘before’ has been taken from the article mentioned, but it is actually from a 1999  paper; 'GISS analysis of surface temperature change'; by J. Hansen, R. Ruedy, J. Glascoe, and M. Sato.

There is a lot of mapped data in this 1999 paper but you will find the one used by Hansen in the US climate article, and subsequently used by Goddard, as the first map on Plate 4.

This map is GISS Surface Temperature Analysis using GHCN v2.

Any one spot the difference? Yes, the maps are not even displaying the same data! What a silly boy Steve Goddard is. The first shows GHCN v2 and the second GHCN v3.

In November 2011 NASA switched from GHCN v2 to GHCN v3, since GHCN v2 was no longer updated. I expect that Goddard and his minions may now try and claim that the change to the new data set is proof that they altered the data. But it is clear Steve Goddard had no idea that this was the case, and certainly never thought to check where the discrepancy was - a real sceptic would.

So before they make any such claims of data alteration, they should note that this was a well documented change. There is a whole page at NASA, comparing the differences between version 2 and version 3.
And you will recognize the same kind of differences that Steve Goddard has only just managed to figure out, but without the innuendo of conspiracy or malpractice – it is there for anyone who is truly interested in finding the truth to see.

One important thing to notice is that the actual global trends are not affected in any way. The global mean change is listed as 0.43 degrees Celsius in both GHCN v2 and the modern GHCN v3.

Reto Ruedy, co author of the 1999 paper has described the advantages of GHCN v3 thus;
“For GHCN v3, NCDC developed a homogenization that is used to combine different station records for the same location as well as deal with discontinuities created by station moves, changes in instrumentation, the urban heat island effect etc. which eliminates many known and documented discontinuities still present in the unadjusted data and caused us to no longer use their unadjusted data.”
A grey area

Since Goddard's 'analysis' has been nothing but shoddy up to now we should not hold up much hope for any great revelation when he says;
"He (Hansen) also has created data in the southern hemisphere which didn’t previously exist. Note how the area of grey has shrunk."
Should we really expect that data has been 'created' or just that Mr. Goddard doesn't understand the difference in the data and would rather insinuate a grand conspiracy instead of actually finding out the truth of the matter? Sorry, but there are no prizes for picking such an obvious answer.

The change in grey area is a separate issue from not knowing the difference between GHCN v2 and GHCN v3. In 1999 NASA used the Reynolds and Smith ocean data which started in 1950, later they switched to the Hadley data since it went back to 1880 and covered a larger area. This explains the shrinking of the grey area. Again, this is not a secret only known by Hansen and a shadowy cabal of climate scientists. In Hansen's paper mentioned above, the one that the 1999 map came from, it clearly states in the Introduction;
"we also illustrate results for a global surface temperature index formed by combining our land analysis with sea surface temperature data of Reynolds and Smith [1994] and Smith et al. [1996]"

In Steve Goddard eyes, challenging him on any of this apparently makes you a “complete idiot” and “MentallyChallenged”. I wonder what, not knowing that you are comparing different data sets and then making ludicrous conspiracy type claims of malpractice, makes you? Other than someone misleading themselves and gullible others with shoddily researched pseudo-scientific nonsense that is.

Based on what I have found here, I strongly suspect that all of the posts Mr Goddard does about data being 'erased', 'created', inappropriately manipulated etc. are simply down to his ignorance about what data sets are used and the fact that he doesn't really care to find out because the truth might get in the way of his theories of conspiracy where he can make scientists like Dr. Hansen out to be some sort of bogey man.

I also strongly suspect that even if Goddard admits he didn't know that the maps were displaying different data, his published ignorance will go uncorrected, because we have seen that even his admitted errors are nor corrected at source.

There is some good news in all this. I noticed the problem with the scales immediately but when I went to comment, some others already had pointed out the same thing. Even one commenter who regularly posts supporting Goddard’s’ nonsense and often flames me stated “Actually the warmists are correct, the graphs show the same thing”, even if he went on to imply that the underlying data was garbage. But it is really heartening to see some others being truly sceptical about Goddard’s unqualified posts. I do not claim to have influenced this but there is hope that his readership is attracting more truly sceptical people, which can only be a good thing.

Saturday, 10 March 2012

What Hansen et al got right decades ago.

Dr James Hansen gave a recent TED talk; ‘Why I must speak out about climate change’. Hansen like many scientists is not a natural orator. He comes across as quiet, dry and even shy and nervous, hiding under a hat and referring to his notes constantly.

Having said that, his talk is worth every minute it took to watch, he gives an overview of how he went from studying planets for NASA to Earth's climate, and why he has become an environmental activist. None of that however detracts from a good overview of his work, the science, and his concerns.

Early on in his talk he mentions a research paper published in Science from 1981; ‘Climate Impact of Increasing Atmospheric Carbon Dioxide’, J. Hansen, D. Johnson, A. Lacis, S. Lebedeff, P. Lee, D. Rind, G. Russell.

He says of the paper that he and his colleagues found that; “observed warming of 0.4 C in the prior century was consistent with the greenhouse effect of increasing CO2", and they also found, “that Earth would likely warm in the 1980s, and warming would exceed the noise level of random weather by the end of the century.”

I thought I’d take a closer look at the paper because it seems almost forgotten, with most deniers concentrating on Hansen in 1988, the year of his well-known Senate testimony, which included the three scenarios that deniers try to claim were hopelessly wrong predictions. They don't seem to know or want to know the difference between a scenario and a scientific prediction.

But having looked at this 1981 paper you realise that it must have formed a core part of his work that led him to being chosen to give testimony in front of the Senate. The whole paper seems uncanny in what it suggested the climate in the next few decades and into the 21st century would look like.

The first thing to notice is that there are a lot of caveats and even some statements that some could claim are contradictory. This isn’t fudge or hedge betting but an honest summary of the uncertainties that existed in the science over thirty years ago.

The paper states some beneficial effects of increased warming; “Beneficial effects of CO2 warming will include increased length of the growing season.” But with the caveats; “It is not obvious whether the world will be more or less able to feed its population. Major modifications of regional climate patterns will require efforts to readjust land use and crop characteristics and may cause large-scale human dislocations.

All of which is either occurring or seems certain to. For example, “According to the Alaska ClimateResearch Center, Fairbanks is 2 1/2 degrees Fahrenheit warmer and 11 percent drier than it was 100 years ago. The changes have stretched the growing season from 85 days in the early 20th century to 123 days.

 And on human dislocation; “Temporary migration as an adaptive response to climate stress is already apparent in many areas. But the picture is nuanced; the ability to migrate is a function of mobility and resources (both financial and social). In other words, the people most vulnerable to climate change are not necessarily the ones most likely to migrate.

However in 1981 the net impact is still not determined and needed further study; “Improved global climate models, reconstructions of past climate, and detailed analyses are needed before one can predict whether the net long-term impact will be beneficial or detrimental.

The papers states, “It is shown that the anthropogenic carbon dioxide warming should emerge from the noise level of natural climate variability by the end of the century, and there is a high probability of warming in the 1980’s.”

The IPCC, established in 1998, confirmed in its third assessment report in 2001 that ‘There is new and stronger evidence that most of the warming observed over the last 50 years is attributable to human activities’, and by the fourth, "warming of the climate system is unequivocal", and "most of the observed increase in global average temperatures since the mid-20th century is very likely due to the observed increase in anthropogenic greenhouse gas concentrations."

The team was also spot on about warming in the 80s and beyond. This was only a few years after the media frenzy in the 70s about a coming Ice Age and most people probably thought that was more likely even if the scientists knew better. For any climate deniers this must seem like a very lucky guess, but there is a reason why these people were picked to do research for NASA.

The paper also states; “Potential effects on climate in the 21st century include the creation of drought-prone regions in North America and central Asia as part of a shifting of climatic zones”. 

It still seems that these climatic zones are still in flux, but According to the U.S. Geological Survey, the current drought in the West of North America is the worst that has occurred in the last 500 years, with water flow levels at close to half those of the drought in the dirty thirties of the previous century, and  in Asia, the United Nations Development Programme has concerns about “The return of drought conditions to Central Asia" and "possible impact on food security” 

The paper states; “erosion of the West Antarctic ice sheet with a consequent worldwide rise in sea level,” confirmed in this paper from 2004;  

And sea level is on a rising trend;

Hansen et al, predicted the “opening of the fabled Northwest Passage.” This was a hell of a call at the time. “Floating polar sea ice responds rapidly to climate change. The 5° to 10°C warming expected at high northern latitudes for doubled CO2 should open the Northwest and Northeast passages along the borders of the American and Eurasian continents. Preliminary experiments with sea ice models suggest that all the sea ice may melt in summer, but part of it would refreeze in winter”.

To most people at the time this must have been seen as alarmist. Hansen states in his recent book 'Storms of My Grandchildren', (Which the video talk seems mostly lifted from), he believes he lost some funding in 1981 from the US Energy Department because this paper was considered too alarmist. Which flies in the face of 'skeptic' claims that scientists are being deliberately alarmist just to garner funding.

But the papers prediction is exactly what has happened, with that passage and it Eastern sister are now open at least part of the time in recent summers. At least one scheduled cruise liner (the MS Bremen in 2006) has successfully run the Northwest Passage.

Even more interesting due to recent research is the claim that “Even a partially ice-free Arctic will modify neighboring continental climates.

If the most recent research about lack of ice causing a shift to the jet stream, and making colder winters is accepted – this is another accurate prediction.

This papers also states that “Climate models predict the larger sensitivity at high latitudes and trace it to snow/ice albedo feedback and greater atmospheric stability, which magnifies the warming of near-surface layers. Since these mechanisms will operate even with the expected rapidity of CO2 warming, it can be anticipated that average high-latitude warming will be a few times greater than the global mean effect”.

This is exactly what has been detected with the Arctic warming several times faster than average. 

And it has this to say on the worlds ice sheets; 
Melting of the world’s ice sheets is another possible effect of CO2 warming. 

If they melted entirely, sea level would rise ~ 70 m. However, their natural response time is thousands of years, and it is not certain whether CO2 warming will cause the ice sheets to shrink or grow. 

For example, if the ocean warms but the air above the ice sheets remains below freezing, the effect could be increased snowfall, net ice sheet growth, and thus lowering of sea level.

This is another good call because recent research and data from the Grace Satellites do show that over all ice sheets and glaciers are melting but in a few areas, such as the internal of Antarctica and the Himalayas there is increased snowfall and ice build up. 

But what of possible lowering of sea level? The paper gets it spot on again; “If the West Antarctic ice sheet melts on such a time scale, it will temporarily overwhelm any sea level change due to growth or decay of land-based ice sheets.

Stunningly Hansen’s team states; “A 2°C global warming is exceeded in the 21st century in all the CO2 scenarios we considered, except no growth and coal phaseout.”

Over thirty years later, UN COPs discuss actions needed to try and keep temperature rise to under 2C, a temperature the paper suggests will cause 5C warming in Antarctica, but without much success so far and it seems that keeping future warming down to such a limit is becoming all but impossible and not without the end to massive coal burning.

In summary, Hansen et al;
  •  Was right about the 1980s warming, even though this was just after the media frenzy about a coming ice age.
  • Was right about erosion of ice sheets, rising sea levels and opening of the fabled Northwest Passage – a pretty good call.
  • Was right about warming at higher latitudes being greater than the global mean.
  • Was right about the increased growing season.
  • Was right about increased snow fall and net ice sheet growth – yet 'skeptics' today still use this as an argument against global warming.
  •  Was right about a partly ice free Arctic modifying neighbouring continental climates – if the most recent research about lack of ice causing a shift to the jet stream, and making colder winters is accepted – another good call.
  • Appears to be right about 2C warming being reached within a century and that temperature being the accepted limit before irreversible and detrimental effects occur.

There is nothing obviously incorrect in this paper.  All this was predicted using models that according to 'skeptics', are not supposed to work, over 30 years ago in 1981, the year IBM released it’s first PC with Microsoft MS DOS.

You have got to give this man and his team some credit for saying all this when most people thought we were still expecting an ice age. This is either good science or, if you reject AGW, they must have been very, very lucky. Personally I think I know why they were employed by NASA. 

Wednesday, 7 March 2012

Goddard Statisically Proves Hansen Correct

Goddard, if I understand his comments correctly, has a theory using using school boy statistics that in his mind proves the recent record climate extremes are nothing to be concerned about, in fact they are to be expected. His theory also proves that unqualified bloggers should be ignored by rational people when it comes to thier pseudo-scientific analysis.

He blogs;

"If you have 3,000 weather stations and a 100 year long temperature record, you would expect about 30 of them to break their all-time record in the current year. The math is a bit tricky for climate scientists : 3,000 / 100 = 30."
He is right when he says;
"It is very basic statistics. If you have 100 random numbers, each number has one chance out of one hundred of being the largest."
So far so good for the school boy Math. But weather and therefore the data from weather stations isn't a simple case of random number generation. Only a portion of any result can be down to unknowns and uncertainty. The results ultimately depend on the laws of physics and chemistry; meteorological cycles, input conditions, natural forcing, changes in local environmental over both short and long times scales etc. If this was not true then weather could not be predicted even in the short term. All these factors and of course any anthropogenic forcing there may be ensures that any result is far from random.

But suppose Goddard is correct or he is just using an analogy - Totally randomly any weather station can be expected to break their long time record according to basic statistics - a roll of a many sided dice if you like. Then we would expect that as many record high temperatures as low temperatures. Goddard's 'very basic statistics' where each number has a chance of being the largest also means it has the same chance of being the smallest. Is that what we see?

The loaded Dice

So it looks like something has been biasing the random numbers in favour of warm records - I wonder what that could be? Something has been loading Goddard's climate dice - where have I heard that analogy before? Oh yes I remember, a paper by James Hansen et. al. "Climate Variability and Climate Change: The New Climate Dice", 10 November 2011;

"The "climate dice" describing the chance of an unusually warm or cool season, relative to the climatology of 1951-1980, have progressively become more "loaded" during the past 30 years, coincident with increased global warming. The most dramatic and important change of the climate dice is the appearance of a new category of extreme climate outliers. These extremes were practically absent in the period of climatology, covering much less than 1% of Earth's surface. Now summertime extremely hot outliers, more than three standard deviations (σ) warmer than climatology, typically cover about 10% of the land area."
Hansen has used the analogy before of a climate dice. From 1951 to 1980 the climate could be represented as a dice with two sides hotter than average, two cooler, and two around average. Currently Hansen's climate dice are loaded; four sides hot, and one side each for cool and average.

So as Goddard's  'very basic statistics' shows, what should be random has a warm bias, adding proof to Hansen's loaded climate dice analogy - Good Job!

Tuesday, 6 March 2012

More means Less!

There is no doubt that globally February was cooler than what is becoming the new norm, but Goddard claims that not only was it the coolest since 1993, 'most' of the other cool years were  'due to either Pinatubo or El Chichon' volcanic eruptions. It is well know that Volcanic eruptions, if large enough, can cool the planet for a year or so, so if 'most' of the other cool or colder years were due to volcanoes then this February without a volcanic forcing must be exceptionally cool - which appears to be the point Goddard is making.

As evidence for his claims he posts the graph above that has a suspiciously smooth sine type curve. Did he draw this on himself to suggest a natural fluctuation and that everything is just peachy? So far he isn't saying.

But a sceptical eye cast over the graph and Goddard's volcanic claims show that he isn't being honest with his readers. Looking at the years that are cooler or about as cool as Feb 2012, we have 1979, 1984, 1985, 1986, 1989, 1990, 1992 and 1993. That is 8 years that might be included in Goddard's 'most'.

El Chichon erupted March 1982, so we can forget about 1997. 1984 was two years after El Chichon so may have been affected a little but it is doubtful as 1983 was actually warmer than the average. We can then drop 1985, 1986, 1989, and 1990.

Pinatubo erupted April 1991 so it is fairly certain that temperatures in 1992 and even to some extent 1993 were affected.

Even being generous, is that 'most' years or less? Perhaps in Goddardland more means less.

Friday, 2 March 2012

Strong And Growing Evidence Of Stupidity And Dishonesty

Steve Goddard claims "Temperatures haven’t risen for at least 15 years". Fifteen years of temperature records would be 1996 - 2011.

 Yes I'd have to agree with Mr. Goddard on this, there is Strong And Growing Evidence Of Stupidity And Dishonesty.

Snowiest four year period on record in North America ?

On Mr Goddard's blog he presented this graph and said;“This winter, global snow cover has been far above normal.

The most notable things here is that Goddard's statement may - be true but so what?, I think it needs challenged on three accounts. First; because of his myopic views that the USA, representing a few percent at most of the entire worlds surface, can in this case be used to justify anything meaningful about global warming.

Second; the apparent cherry pick of just the winter values. The amount of snow in winter is far from the only consideration. Snow extent in the spring, which may partly determine the amount of snow melt through the year is also important. That snow will feed into rivers to be used for irrigation, drinking water etc in the coming year. A more meaningful question to ask might be; How is snow and ice holding up over all?

Third; can the amount of snow actually be used as a measure against the early stages of climate change? A warmer atmosphere holds more moisture, which will eventually lead to more precipitation. If cold enough, this will actually mean more snow. A concept climate deniers refuse to get in their quest to confuse and misinform themselves and others.

In answer to the first two points a look at the other graphs from the very site Goddard got his from reveals much.

For Autumn (Fall) and Winter, there is a mix of slight increases, decreases or essentially no change in the record. Springs shows a far more dramatic decrease across all areas. Why would Goddard only pick the graph that shows the largest increase and ignore all others?

A selection of  the graphed data can be seen on an animated gif at Skeptical Science;
snow extent
Graphic from Skeptical Science

the long-term trend in spring, summer, and annual snow cover is one of rapid decline. As a result, the planet as a whole is becoming less reflective and absorbing more sunlight, which is accelerating global warming.
For point 3, I have already suggested that one of the climate changes that global warming has predicted is heavier snow falls. This isn’t just something scientists made up on the spot to try and account for recent heavier snow falls across parts of the world. It is a long standing prediction,  using models that, if you belief the ‘skeptics’, don’t work.

A quick search on 'snow and global warming' with google Scholar finds a paper;  POLAR SNOW COVER CHANGES AND GLOBAL WARMING, from way back in 1998. From the Abstract;

“Many general circulation models suggest that current precipitation amounts in polar latitudes will increase under double CO2 scenarios. Even though temperatures in such high-latitude regions should also increase under a doubling of CO2, as long as those temperatures remain below freezing, the increased precipitation should accumulate as snow.”

But very recent research also confirm this effect. Even while the world has been calculated to be loosing 150 cubic miles of ice every year, Himalayan Glaciers were found to have lost little ice because the increased melting at the lower altitudes were mostly matched by increased snow fall at higher altitudes where it was too cold to melt.

And just this week new research suggests that Arctic sea ice loss could be weakening the Jet Stream allowing colder Arctic air lower over Europe and America actually causing colder winters, including heavier snow, even while the Arctic itself is warmer than normal.

So as much as Goddard would like to claim that increased winter snow extent is evidence against global warming and cry foul when the science suggests it is actually evidence for it, it was predicted by models around a generation ago and is being confirmed by empirical evidence today.

Snowiest four year period

If you have managed to read this far and are starting to wonder what the title has to do with all this, all will now be revealed. When I challenged Goddard about his cherry picking he replied with his usual eloquence and what might just be his standard greeting to anyone actually sceptical enough to query his claims;

The last four years was the snowiest four year period on record in North America. Cut the crap, please.
Snowiest? The obvious problem here is that this is a very specific pronouncement using a very unspecific term. I asked him “How are you defining ‘snowiest’ and what data set are you using for evidence?” Clearly from the graph he supplied he was not referring to snow extent. At least ten previous years had higher records than some in the last four. He has yet to reply.

Snow is such a transient phenomena. Would he be referring to depth? It is relatively easy to measure precipitation but the depth of snow depends largely on how wet it is. Sleet, a mixture of rain and snow, will have little depth, the fluffy stuff many times more depth than the equivalent amount of water. Perhaps he was referring to the number of days it snowed? How long the snow lasted?

So there is the problem. 'Snowiest' isn’t really a scientific term that can be used in a specific claim as Goddard has done. It doesn’t mean it can’t be used as long as it is defined. In my attempts to collaborated the ‘snowiest’ years I found that NOAA do use it to refer to depth in specified cities. However I can’t find any where all these figures are collated  in a way that would indicate the depth of snow that fell over the whole area of the USA, but judging form the years that are in the top ten for the cities I’d be very surprised if the 'snowiest' added up to be 2007 through 2011.

So claiming that the 'last four years was the snowiest four year period on record in North America’ appears to be an unsubstantiated and unsubstantiable claim from Goddard.

I invite him to clarify, otherwise perhaps it is he that should cut the crap.