Dec 04 2011

Exposed: How Briffa/IPCC Produced False Error/Uncertainty In Hockey Stick

Update: Reader Blog Lurker at CA discovered my dyslexia – and the fact I have the wrong email number. Fixed below, but the number is #3468 – end update

Update: Jeff Id at Air Vent wanted something pithy to summarize this post.. So since I am watching endless Harry Potters today I have decided to preview this post with a summary of the problem and trick covered at the end. Here was the challenge facing the IPCC team in 2006:

It is not right to ignore uncertainty, but expressing this merely in an arbitrary way (and as a total range as before) allows the uncertainty to swamp the magnitude of the changes through time.

Without doing something, the IPCC claims of historic warming would sink into the noise of the measurements and be forever determined a fraud. So Briffa comes up with a plan:

the comparison of past and recent temperature levels is not as influenced by the outlier estimates.

Mischief Managed! – end update

I have been reading the Climategate  2 emails, repeatedly stunned at how bad it really is in the climate ‘science’ backwater. Talk about your amateur scientific method!

Just this morning though I tripped over an email from Keith Briffa at CRU that likely is another smoking gun that will completely up-end the Hockey Stick and all similar claims of unprecedented warming in the last half century.

Before we get to the email (and I request assistance from people more skilled in the use of time series), I want to reiterate my primary case against the IPCC claims of historic warming. The actual temperature measurements from 1850-1960 are sparse and uneven (to be kind). Apparently for the entire southern hemisphere there are only a handful of continuous records to cover this HALF of the world. From 1960 to 1990 the sample size fills in globally and the consistency increases, but it really is only in the satellite era (1990 onward for good temps)  that we get consistent global measurements of decent precision. 20 years is a very short time to make claims  covering centuries of time – even if we were using this data.

The IPCC is not using this data it turns out. The modern (and most accurate) global temperature record (1960 to present) diverges from the primary temperature proxies used to assess current temps against historic temps (i.e., tree rings). Tree rings show “a decline” in temperature in many parts of the Northern Hemisphere in the last 50 years.. Therefore we do not see in the Hockey Stick calibrated tree ring temperature signals from the last 50 years compared to previous ages like the Medieval and Roman Warming periods.

The reason we do not see this is because if they did compare the modern era to the past using tree rings calibrated to the modern record there would be no historic warming today.

This is known as hiding the decline.

My beef all along is how no one has defined the integrated error in the IPCC global temperature value. Starting from the errors of the temperature reading itself, to the errors that combine (i.e., expand)  by averaging and smearing these local measurements to 500×500 km grids over weeks, months and seasons, which then combine at the hemispheric and global level, we are looking at what has to be an error/precision level at the global values of +/- 2-5°C. And from this we find a 0.8° C warming???

The IPCC charts don’t show this formal error analysis. And now we might know why.

In email #3468 from 2006, Keith Briffa makes an astounding admission. He explains how he combined errors from multiple time series to ensure the modern temp numbers (already fudged by covering up the tree ring numbers with thermometer numbers the tree rings don’t agree with) do not fall inside the precision of the historic numbers. First he admits why he must fudge the data:

we are having trouble to express the real message of the reconstructions – being scientifically sound in representing uncertainty , while still getting the crux of the information across clearly. It is not right to ignore uncertainty, but expressing this merely in an arbitrary way (and as a total range as before) allows the uncertainty to swamp the magnitude of the changes through time.

Let this settle in for a moment. If the Team has to represent uncertainty as a total range (which is not just arbitrary, but the professional and scientific norm), then the uncertainty would swamp their claims and destroy the IPCC’s alarmist message.  As I predicted, the uncertainty in the measurements combined with the uncertainty of deriving a temperature value from tree rings results in an error bar so wide we don’t know if we warmer or colder than the Medieval Period.

Briffa admits this behind the scenes in this email. But how to salvage the Hockey Stick? Briffa creates his own new math (in total violation of all professional rules of applied statistics):

By overlapping all reconstructions and giving a score of 2 to all areas within the 1 standard error range of the estimates for each reconstruction , and a score of 1 for the area between 1 and 2 standard errors, you build up a composite picture of the most likely or “concensus”  path that temperatures took over the last 1200 years (note – now with a linear time axis).

This is kludge is arbitrary and a completely nonsensical way to combine uncertainties. Basically he throws out uncertainty that is unique to a time series. The only uncertainty that is allowed to be shown is where it overlaps with another series. So if one series has huge error bars (like tree rings) that portion that is outside the others is lost. Another way it falsifies the integrated picture is if one series is showing cool temps while 2 others are showing warm. The cool side of the error is now lost, only the overlap between the cool one and warm two shows up. Basically, he is filtering data AGAIN to select only what he wishes was reality. And he freely admits it:

This still shows the outlier ranges, preserving all the information, but you see the central most likely area well, and the comparison of past and recent temperature levels is not as influenced by the outlier estimates.

Good Lord, if you are not going to integrate the data that disagrees with your expectations, why even pretend to be doing statistics or science? Just pick your data and plot it and get it over with!

I leave it to those who do know statistics to describe how bad this really is, but it looks to be a complete crock.

Addendum: To be clear here, I am not sure what Briffa did (I do not have the picture) or if it made it into any report. I assume it did. Maybe not a defensible assumption – but it seems everything else in these emails has a tangible product out there so I will stick with it. And while I am not sure what Briffa did mechanically, I know why he did it since he says it himself. He could not use standard error ranges, so he concocted something else. I’ll leave it to others to determine the severity of this action – but in my mind it simply was another way to hide the mathematical truth that the hockey stick is all smoke and mirrors.

20 responses so far

20 Responses to “Exposed: How Briffa/IPCC Produced False Error/Uncertainty In Hockey Stick”

  1. crosspatch says:

    Again the problem I have is using a summer growing conditions indicator as a proxy for *annual* temperatures. Summer temperatures can (and have in many places) trend exactly the opposite direction of annual temperatures. Tree rings can not proxy conditions occurring when tree rings are not being produced (fall, winter, early spring). At best they can *possibly* contain information on late spring conditions where a late snow melt could delay the onset of ring growth but a cold, dry winter will never be picked up if it is followed by a warm/wet spring and summer.

    In short, tree rings are NOT thermometers of *annual* average temperature and trying to use them as such is misleading at best and at worst incompetent or dishonest depending on the belief of the person presenting the information (incompetent if done in good faith, dishonest if they know better).

    A proxy that can possibly show aggregate growing conditions over a three to four month period can not possibly show total growing conditions.

    Here’s a novel idea for why they decline in recent years: reduced sulfate causing the soil to be sulfur depleted. As we “scrub” coal plants and switch to low sulfur fuel, the amount of sulfur deposited in the soil declines. This decline has already resulted in farmers around the world now having to add sulfur to their fields where they did not need to do so in past decades. I would imagine that forests might reflect the same decline in aggregate growing conditions as the amount of sulfur in the soil declines and reduces the ability of the trees to extract other nutrients from the soil.

  2. crosspatch says:

    And it is kind of ironic about that whole sulfur thing. We pay extra for fuel because the sulfur needs to be taken out, then the farmers have to buy the sulfur to put on their fields. We pay more for fuel and fertilizer in order to get something we used to get “free”.

  3. WWS says:

    “Again the problem I have is using a summer growing conditions indicator as a proxy for *annual* temperatures.”

    Now that the e-mails have conclusively revealed that the bad intent was there from the beginning, we know exactly why they used this weak indicator as a proxy.

    They used it because it was the easiest way to fake the data. Anything more “robust” might not have given them the numbers they had to have.

    They never gave a damn about actual data, the entire effort was dedicated to supporting “The Cause”. So a measurement that was easily manipulated, but obscure enough that few people would be able to cut through the gobbledygook, was just perfect.

  4. SteveGinIL says:

    I just began reading this (thx to Steve M). Probably another comment later…

    The actual temperature measurements from 1850-1960 are sparse and uneven (to be kind). Apparently for the entire southern hemisphere there are only a handful of continuous records to cover this HALF of the world. From 1960 to 1990 the sample size fills in globally and the consistency increases, but it really is only in the satellite era (1990 onward for good temps) that we get consistent global measurements of decent precision.

    This dating applies to the issue of the divergence problem, which was the underlying thing behind the “hide the decline” of CG1.0.

    It is precisely the 1850-1960 period in which the tree-ring proxies did NOT diverge. Perhaps 1850-1940 would be more precise, though, because the divergence isn’t really noticeable for the 1940-1960 period. Yet, if the good correlation was during a time of few instrument records, one can make the argument that the correlation was a false one, matching tree-rings to an inadequate instrument record.

    Then, in the “good” instrument era and the great (satellite) instrument era, the divergence appears.

    If it walks like a duck, and quacks like a duck…

    There does not actually appear to be a divergence problem after 1940. It is on their heads to prove that there was no non-divergence prior to that time. It is entirely possible (or even likely) that the divergence was never there and only – as the instrument record slowly got better and better – showed up as the instrument record stopped aligning with the earlier error.

    Biologists use tree-rings as proxies for precipitation. Thus, tree-rings are being used as proxies for two different forcings – which cannot be sound science.

    So, my point is, don’t accept the assertion that tree-rings are – or ever were – good proxies for temperature. The last 70 years argue that it is not. What happened during the 1850-1940 (or -1960) period seems as if it is a false correlation. I mean, a LOT of uncertainty has to exist in the fact that only five stations represented SH. Any correlation then has to be suspect.

  5. SteveGinIL says:

    Good! As I said, I had only begun reading.

    The reason we do not see this is because if they did compare the modern era to the past using tree rings calibrated to the modern record there would be no historic warming today.

    This is known as hiding the decline.

    Exactly! The global average today if based on the tree-ring proxy numbers would be close to what it was in 1900-1910. Meaning that there hadn’t been any warming – if the tree-ring>temperature conversion factor as used is valid.

    But it isn’t just the uncertainty. It is the stated correlation itself that is claimed as the basis of tree-rings as proxies for temperatures. I am glad you pointed out the uncertainty in the conversion, but it goes deeper than that.

    Several of the scientists in the CG2.0 emails bring up this questionable nature of that conversion. Sorry I can’t give email numbers, but I’ve read at least 4 or 5 in all my reading so far.

    BTW, if the divergence was in the opposite direction – in favor of alarmism – one would expect Briffa to have set the instrument numbers to ‘missing’ and inserted the proxy numbers, all in service to The Cause.

  6. SteveGinIL says:

    @crosspatch –

    All your points about tree rings are well taken.

    What would you say to the fact that biologists use tree-rings as proxies for precipitation? In fact, they were first used that way.

    How much extra uncertainty do you think that adds to the tree-rings>temp concept? Leaving out winter is bad enough, but now add in the very real issue of precip having a sizable affect on tree-rings, and where does that put the “science”? On pretty untenable ground, yes? Where is winter? And where is precip?

  7. crosspatch says:

    Oh, it’s even worse than that. Say, for example, there was a herd of ruminants that regularly migrated though the area and deposited, uhm, nitrates. Now lets say that herd no longer exists or is otherwise unable to migrate through that area due to the building of a major highway or railroad which they can’t cross. That will result in an impact, too. We could be measuring the change in deer poop for all we know.

    Precipitation is a really good point because we can have changes of circulation patterns in the atmosphere on multidecadal scales. We might have a storm track for 30 years that dumps some amount of precipitation more or less reliably but then changes and the precipitation moves someplace else. In that part of the world it is likely related to the AO.

    Then there is the problem of “calibrating” it to a portion of the temperature record and then assuming that correlation is causation. For example, if I “calibrated” the number of pirate attacks to amount of CO2 in the atmosphere in ppm and ran those numbers forward in time, it won’t be long before I get some divergence because, in fact, the two numbers aren’t related.

    I will get good correlation when the main driver is SUMMER temperatures. So if I have a period where winter temperatures over some period are flat but summer temperatures have a significant trend, then yeah, my tree ring trends will correlate with average annual temperature trends. If the opposite happens (as is currently happening in North America with a flat summer temperature trend and rapidly declining winter temperatures), then there will be no correlation or even an inverse correlation is possible.

    Inverse correlations can be adjusted for with a change of sign, but when there is no trend at all in summer temperatures, there is no way to adjust.

    I would suggest that changes in treeline would be a more accurate measure of long term signal in climate data but you don’t get annual resolution that way.

  8. crosspatch says:

    A commenter over at Anthony Watts’ blog posted this link to the HadCRUT3 temperature series as some sort of evidence of recent AGW.

    http://www.cru.uea.ac.uk/cru/data/temperature/nhshgl.gif

    Interesting thing is, the coloration sort of plays tricks with you. Check out the “global” graph on the bottom. Notice the rise in temperatures from 1910 to about 1940. Now notice the rise in temperatures from about 1970 to around 2000. Both rises in temperature are of roughly the same magnitude and duration. But we are supposed to accept the first one as a normal, natural variation. The second one is supposed to be alarming and caused by humans and we must spend billions of dollars to counteract it.

    I say it’s a good thing nobody came up with the AGW hypothesis in 1900 else we would be even deeper in the hole than we are now!

  9. Redteam says:

    Crosspatch, first I agree with you, but you said:

    “So if I have a period where winter temperatures over some period are flat but summer temperatures have a significant trend, then yeah, my tree ring trends will correlate with average annual temperature trends.”
    That would appear to be true only if you wanted it to be and show select data to prove it. As Briffa did.

    If only really true data were used, (probably not possible) then I think it would prove unquestionably that tree ring growth is totally independent of ‘average daily temp’. Several good examples of reasons why have been given, (I like the one on deer poop just above) and so here is another scenario.

    Let’s say the area near the trees have a really unusually cold winter and huge amounts of ice accumulate. Then as the temp gets above freezing, the ice starts to melt. It melts all throughout the growing season (even though the daily temp is below normal) and supplies an unusual amount of water to make the trees grow. A situation such as this would ‘indicate’ higher than normal temps because the extra water made the rings larger when the true situation is that ‘average temps’ were truly lower than normal. Therefore the ring growth is totally independent of temperature.

    It is obvious that there is no correlation between humans inhabiting the earth and the average annual temp of the earth. One simple proof of this is that there is ‘no known average annual temp of the earth’ prior to the days of satellites measuring overall temp.
    The worst part of this whole deal is that this whole global warming thing is being put on us so some select crooked individuals can make a buck. They then spread this money to the researchers that are anxious to get the money to ‘prove’ that it’s true. All the palms get greased to fool the people.

  10. crosspatch says:

    hen as the temp gets above freezing, the ice starts to melt. It melts all throughout the growing season (even though the daily temp is below normal) and supplies an unusual amount of water to make the trees grow. A situation such as this would ‘indicate’ higher than normal temps because the extra water made the rings larger when the true situation is that ‘average temps’ were truly lower than normal. Therefore the ring growth is totally independent of temperature.

    Won’t work because a late melt delays the start of growth and retards growth until the soil warms up. These trees are pretty sensitive to soil temperature. They won’t start growing until the melt is complete, they’ll remain dormant until the soil gets high enough. If the snow doesn’t melt at all, they won’t grow that year, that simple. You won’t find too many trees growing in the summer snowline of most mountains. Until that snow thaws and allows the surface of the ground to thaw, the roots won’t do squat.

  11. Redteam says:

    Crossp, sorry you didn’t read a little closer: “Let’s say the area near the trees” note the word near, as in the hills nearby. the soil the trees are in are at normal temp, but the ice is all uphill and water runs down as the ice melts.
    Now, I’ll admit I’m not an expert on frozen land (I live in the South) but growing season is generally defined by air temp, such as from the first day after 5 consecutive days above freezing til the first day after 5 consecutive freezing days, and not by the temperature of the soil. So I’m gonna say that just because Ice remains in the hills above the forest plain doesn’t keep trees from growing on the forest plain if the air temp is consistently above freezing.

    Your statement above implies that ring growth is related to ground temp and snow layer. I’d say it is absolutely and totally unrelated as most ring growth (probably all) takes place when there is no snow on the ground or freezing temp above ground.

    Your’s is but one more example of all the scenarios that can be painted to prove that the total work done by Briffa and IPCC is total fabrication. Had they intended to prove that humans make the average temp lower, I have no doubt they could have taken the same data and ‘proved’ that point.

  12. crosspatch says:

    but growing season is generally defined by air temp

    Nope, soil temperature determines growth, blossoming, germination of seeds, etc. Now to some extent soil temperature will follow air temperature, except when there is snow on the ground.

    This is why growing seasons can be so much shorter. We can ski at Lake Tahoe in March in our shirt sleeves when air temperatures in the middle of the day might be near 70 but we are skiing on 30 feet of snow that won’t melt until late June or July (last measurable snowfall n the Sierras last year was 1 July). The daffodils still won’t bloom until the snow is gone.

  13. Redteam says:

    Crosspatch, for some reason you are completely missing the point of my comment.
    Here is a definition of growing season from wikipedia, a link is provided below. There are likely other definitions as well, but as in Briffa’s case, I only select the one that supports my point of view. If Briffa and IPCC can do it, I can also…. chuckle…..
    “North America

    In the United States and Canada, the growing season usually means the days between last and first frost, or approximately the last and first occurrence of 32° F (freezing) overnight low temperature.

    In the northern regions this is roughly May to October, in southern-southwestern-Californian regions it is roughly March to November or longer. Proximity to maritime of extremes can extend the growing season”
    link: http://en.wikipedia.org/wiki/Growing_season

    did you just completely miss this:“Let’s say the area near the trees” note the word near, as in the hills nearby. the soil the trees are in are at normal temp, but the ice is all uphill and water runs down as the ice melts.”

    So while you are spending your time on the snow during the winter, I’m spending mine downhill, from the higher elevations with it’s snow and ice, where the air temp is 70 and the soil temp is 70 and since it is so nice and warm the ice is melting up on the hills and supplying plenty of water for the trees to grow in their warm climate.

    Let me ask you a basic question, since you live in a mountainous area. Have you ever been in a valley where the temp was about 70 and the snow is all gone? and up on the top of the hills nearby there is snow? That’s the place I’m talking about.
    I have been in Bend, Oregon when it is 100 in the valley and the tops of the mountains are covered in snow and ice and it is melting and water is running down the hills (all summer long) and the trees are all growing very nicely. Well, if it all sounds so simple, that’s what I said from the beginning, you just overlooked it in my original comment.

  14. Redteam says:

    Crossp, just to further support the definition, here is another one from the same link I provided above. In this case soil temp has absolutely nothing to do with growing season, the only real variable is water (and that is the case I was making above, that the presence or absence of water would control tree ring size without any regard to temperature. My whole scenario was to make global temp average completely separate from tree ring growth. They are clearly not solely dependent on each other, there are several factors that make the measurements meaningless.

    Tropics and deserts

    In some warm climates, such as the subtropical savanna and Sonoran Deserts or in the drier Mediterranean climates, the growing season is limited by the availability of water, with little growth in the dry season. Unlike in cooler climates where snow or soil freezing is a generally insurmountable obstacle to plant growth, it is often possible to greatly extend the growing season in hot climates by irrigation using water from cooler and/or wetter regions. This can in fact go so far as to allow year-round growth in areas that without irrigation could only support xerophytic plants

  15. crosspatch says:

    In this case soil temp has absolutely nothing to do with growing season

    It may have nothing to do with the official designation of it, but a seed will not germinate until the soil temperature is right no matter what the air temperature is. A tree will not run sap until the ground thaws.

    For dendrochronological studies I would refer you to:

    http://academic.engr.arizona.edu/HWR/Brooks/GC572-2004/readings/vaganov-nature-siberia-tree-snow.pdf

  16. crosspatch says:

    And RT, I really don’t care what the official definition of “growing season” is, that is quite a different topic. The point is, tree rings show only summer temperatures with the POSSIBLE EXCEPTION of late spring conditions concerning snow melt. The fact remains that tree rings will not in any way reflect conditions from September through May in the places where these trees that are used for these studies are found.

    They generally use trees that are either at high altitude near the treeline that get a lot of snow (so water is not an issue) or very far North near the tree line at tundra (get adequate snow and moisture is not an issue) where temperature is the primary growth constraint.

    The only issue outside of summer temperatures that can impact their growth is late snow pack delaying the thaw of the soil. The air temperature can be 100 degrees but if the ground is still frozen solid, the tree will not run sap or put on growth.

  17. crosspatch says:

    From the paper I linked:

    “Thus, the main controlling factors of seasonal growth and tree-ring structure formation in northern timberline trees are early summer temperature and the date of snow melt, which influences the date of cambial initiation.”

  18. Redteam says:

    Crosspatch, you’re finally getting around to recognizing what I said and basically what you alluded to back in your comment at 12/5 at 4:57.
    “And RT, I really don’t care what the official definition of “growing season” is, that is quite a different topic. The point is, tree rings show only summer temperatures with the POSSIBLE EXCEPTION of late spring conditions concerning snow melt. ”

    The point is that tree rings DO NOT show summer temperatures. If it is 120 degrees and there is not a drop of water the tree WILL NOT grow at all. If it is 120 and there is a little bit of water, it might show a very small growth. If it is 120 and there is abundant water (as in snow and ice melting and running down the mountain to where it is 120 degrees) then it will show a very large growth ring. three different cases, all same temp, very different results in tree ring growth. in other words, it is NOT related to temp only. All conditions have to be considered, as in “there was a herd of ruminants that regularly migrated though the area and deposited, uhm, nitrates. ” as you said above.

    This whole about tree rings is just about showing that Briffa, etc used bogus data (and they knew that tree ring data was bogus) as proof of what they wanted to prove.

    As you said above: “Inverse correlations can be adjusted for with a change of sign, but when there is no trend at all in summer temperatures, there is no way to adjust.

    I would suggest that changes in treeline would be a more accurate measure of long term signal in climate data but you don’t get annual resolution that way.” basically you are saying that there is no correllation.

  19. crosspatch says:

    “The point is that tree rings DO NOT show summer temperatures. If it is 120 degrees and there is not a drop of water the tree WILL NOT grow at all.”

    While what you say is certainly true, the trees that are generally selected for this sort of time series live in areas that get a lot of snow or live in fairly swampy ground. So sure, they could have a dry year, but the trees in these locations generally get a lot of snow in the winter. They are specifically selected for places that get a lot of water or that live in desert-like conditions anyway during their growing season (bristlecones for example). In other words, they are generally selected for areas that have stable amounts of water either in abundance or perpetual drought.

    They aren’t using regular flat land forest trees for this research. They are using either high mountain edge of tree line / desert trees or edge of tundra trees where. In the edge of tundra, the ground is generally pretty swampy.

  20. […] to the realization that the unaltered, clearly represented data and results from the CRU actually cannot prove today’s climate is any warmer or cooler than previous period, to the shoddy code exposed in Climategate I, the mathematical confidence underlying the alarmists […]