Taxpayer-funded climate alarmist scientists are forever pursuing the holy grail of global warming politics: eliminating prior warmer periods (i.e. Medieval, Roman and Minoan warmings) and thus making modern warming "unprecedented" - if any group of scientists can accomplish this, they are guaranteed to get future government funding for their research ... unfortunately, this political/funding agenda leads to a plethora of bogus claims and bad faux-science, which the new Marcott et al. study seems embodied with
(click on images to enlarge - CET & HadCRUT data sources)
Recently, a new simulation of reconstructed proxy temperature records was released with great fanfare and acclaim from the mainstream press. Without any due diligence, the science "journalists" shouted that this was the new and improved "super" hockey stick that proves modern warming is "unprecedented."
Based on the above/adjacent chart, mainstream headlines trumpeted the study with various claims of the "fastest" warming and overtones of end-of-the-world as we know it from dangerous global warming. Some typical brainless MSM examples:
These headlines are similar to the ones that greeted the infamous first hockey stick, which later was proven to be statistically bogus; and so discredited, that the IPCC was forced to jettison it. Based on that last experience, one would think (and hope) the mainstream press would be a little more cautious with hockey stick representations and claims of "unprecedented" but apparently not - gee, what a surprise.
As a collector of graphical paleo-climate reconstructions, it is obvious this study's hockey stick graph is another politically-driven, anti-science, robustly bogus representation of past global temperatures. Per the Marcott et al. scientists' own admissions, the above bogus hockey stick was literally fabricated from a thousand Monte Carlo simulations - figuratively, non-empirical "evidence" produced from a statistical 'smoothie' blender.
Objectively, their above 2,000 y.o. hockey stick simply does not comport with the major, 2,000 y.o., high resolution paleo-reconstructions from around the world.
Similar to the previously discredited, first hockey stick study, the primary purpose of this study's statistical fabrication is to eliminate the known actual temperature peaks of the Minoan, Roman and Medieval periods - well documented periods of high temperatures over extended time spans. How a "scientist" accomplishes this is easily illustrated using an example - a hockey stick production anatomy lesson.
An anatomy of a bogus hockey stick
For this example, the non-hockey stick Central England Temperature (CET) dataset will be used to create an imaginary hockey stick of CET temperatures, similar looking to the one presented by the Marcott et al. study.
1. How to construct the "handle" of the hockey stick
it is necessary to take higher resolution raw data, and then by using
any number of acceptable (and some questionable) "variance suppression" statistical
techniques, transform it into a blended proxy 'smoothie' for the stick's handle.
In essence, higher resolution input needs to be transformed to lower resolution output in order to suppress the temperature peaks and valleys (in the Marcott 11,000+ year analysis, suppressing the Minoan, the Roman and Medieval warming periods is required).
- The CET chart on the below-left is a plot of all monthly thermometer (instrumental) measurements since 1659 through 2012 - 4,248 observations, which represents the longest available instrument record.
- The monthly CET data is of extremely high resolution with a very wide variance (from -3.1 to 19.7, a 22.8 degree spread) - for a hockey stick, this large variance has to be suppressed in order to create a smoother handle
- Thus, hockey stick handles require low resolution datapoints
- The CET plot on the below-right graph depicts a relatively smooth hockey stick handle. This 'smoothie' handle (low resolution) creation was done by averaging 100 years of monthly CET data; and then having Excel apply a 36-period average to the lower resolution 100-year moving average datapoints
2. How to construct the blade
- The blade of the climate hockey stick always has to be modern high resolution data, not smoothed, low resolution historical data
- Instead of suppressing the variance, the blade needs the large variance of high resolution data to exaggerate the blade relative to the handle (by doing so, then the climate-liar perps can exaggerate it even more)
- For this example (the above-right simulated CET chart) the low resolution data used to fabricate the CET handle was truncated at the end of 1949
- High resolution annual temperature data was then spliced to the low resolution data (remember, the 100-year average of monthly observations)
- Then the built-in Excel averaging tool was applied to the entire spliced dataset
- Voila, the needed very scary, spiked blade is produced and seamlessly "attached" to the smoothie-handle
So, how do you think the blade of the simulated CET chart on the right compares to reality? Clearly, from the chart on the right, it appears that temperatures are relentlessly accelerating, soon to take most of England to their own version of physical climate hell, no? But note that the actual monthly data chart on the left does not portray such acceleration.
Now for real world reality: the actual annual CET temperatures have been declining at a minus 4.4 degree per century trend - that's right, a -4.4° C/century rate over the last 15 years that the simulated blade distorts.
As can easily be seen, the simulated CET's hockey stick's blade is a gross distortion of the climate reality; and the stick's handle completely misrepresents (disappears) the known extremes of the historical CET climate. This is what faux-science simulations do.
Likewise, the same is absolutely true for the Marcott et al. hockey stick - an extreme and gross misrepresentation of climate history and the current temperature trend.
(And btw, for the intellectually curious who don't reside/work in the liberal MSM echo chamber, the Marcott et al. hockey stick stops at 1950 with the spiked 'blade' increase starting at year 1900, which indicates a +0.75°C increase over that 50 years - however, according to the IPCC's own gold standard HadCRUT3 dataset, global annual temperatures during that same period declined by a -0.11°C...what's "screwed" is this blatantly bogus hockey stick, eh.)
In review, the Marcott 'blade' is a farcical depiction of modern temperatures that was entirely fabricated to encourage hysterical fear-mongering by mainstream reporters; and, the low resolution handle, stripped of all climate reality via 'smoothie' blender statistics, was designed to disappear previous significant warming eras.
Finally, Marcott et al. used over 70 temperature datasets that they threw into their statistical 'smoothie' blender. Adjacent is a typical pre-blender dataset they used that exhibits historical climate reality, with all its high and low extremes.
But they needed a hockey stick. So instead of reporting actual climate realities, like this one from New Zealand, they gave the mainstream media a highly manipulated and fabricated low resolution smoothie with a tall, high resolution synthetic straw, as represented by hockey stick chart at the top of this posting.
Needless to say, the money quote from their study's abstract - "Current global temperatures of the past decade have not yet exceeded peak interglacial values but are warmer than during ~75% of the Holocene temperature history" - is solely based on their statistical blender for climate-temperature smoothies - and, for the record, that quote is not true, it's not reality, and it's most definitely not actual empirical evidence.