The National Climate Assessment report denies that siting and adjustments to the national temperature record has anything to do with increasing temperature trends. Note the newest hockey stick below.
h/t to Steve Milloy
Source: http://nca2014.globalchange.gov/system/files_force/downloads/low/NCA3_Climate_Change_Impacts_in_the_United%20States_LowRes.pdf?download=1
Yet as this simple comparison between raw and adjusted USHCN data makes clear…
…adjustments to the temperature record are increasing – dramatically. The present is getting warmer, the past is getting cooler, and it has nothing to do with real temperature data – only adjustments to temperature data. The climate reality our government is living in is little more than a self-serving construct.
Our findings show that trend is indeed affected, not only by siting, but also by adjustments:
The conclusions from the graph above (from Watts et al 2012 draft) still hold true today, though the numbers have changed a bit since we took all the previous criticisms to heart and worked through them. It has been a long, detailed rework, but now that the NCA has made this statement, it’s go time. (Note to Mosher, Zeke, and Stokes – please make your most outrageous comments below so we can point to them later and note them with some satisfaction.).
Janice Moore says:
May 6, 2014 at 2:34 pm
“There is MUCH more to Mr. M0sher’s words than meets the eye…”
The Mosh is now just a paid mouthpiece shill for Best . I’m sure he will enjoy it for a while.
Wonder if its also his job to turn on the red light in the evening.
Zeke Hausfather says:
May 6, 2014 at 12:35 pm
Anthony,
Methinks the last point in your raw vs adjusted USHCN graph is in error.
As far as the need for homogenization goes, we’ve been over this time and time again. There are certain network transitions (TOBs, CRS to MMTS, de-urbanization of stations post 1940s) that introduce some pretty significant biases into U.S. temperature records, most of which are (unfortunately) in the same direction.
_________________________________
Actually, the biases aren’t mostly in the same direction, are they? Seems they changed directions sometime after the year 2000 from making the raw data cooler to making it warmer… why is this?
@K Scott – actually, as the data ages, the adjustments start changing as well. Notice that 1998 was adjusted up, before it was adjusted down (to accommodate 05 and 10 looking more like records). I call 98 the kerry year.
After another lustrum, or perhaps two, of no warming that can no longer be concealed, even the slowest of politicians is going to back away from the panic stricken carbon strangulation policies that have become so Politically Correct these days.
We can only hope that their enlightenment occurs before they’ve inflicted irreparable harm on our economies.
Wow Anthony. Looking forward to reading the paper! This answers that question I posted last week about why you focus on the temperature record nicely, thank you.
That graph puzzles me. It shows an ‘adjustment’ of nearly 1 degree from 1979 to 2013 (the 2014 point looks spurious), yet the surface temperature record is in reasonable (i.e better than 1 degree) agreement with the satellite record over that period. How can that be the case? Has the satellite record been adjusted too?
Anthony, You know I love you. An affection I have demonstrated with my treasure.
That said, this is post hoc arm waving.
Your surface station work was groundbreaking and under appreciated by the establishment. Bravo for you!
May I suggest another group sourced experiment to “test” the adjustments. I believe there have to be high fidelity proxies to temperature in the historical that can be used to test the adjustments. Being a citified farm boy, I’ll go back to my roots and suggest agriculture. If the 30s really were comparatively cooler than they were a now, are there agricultural records we can consult? Do individual farms/farm families maintain records?
So my suggestion to you is to:
1. Call on your community to suggest the proxies.I wonder if tree farms might be a good source of data.
2. Design a statistically valid experiment.Run it and compare the results to the proxies.
3. Do it a couple of times.
4. Write another paper.
Very respectfully (and affectionately) yours
RobertInAz
I’ll followup on my tree farm suggestion. I mentioned it because somebody was writing about the difficulty of paleo temperature reconstructions to get temperature as an independent variable. IIRC, it was a most excellent series of 4 posts on the extraordinary difficulty of such an endeavor. He mentioned tree growth models. Given the agricultural importance of lumbar and the noncontroversial nature of local rainfall records, I wonder if it might be possible to back out temperature to a reasonable degree of precision using tree growth models, historical tree growth results and the historical precipitation record.
This may work for other crops. But tree rings are so very well established as treenometers to fractions of a degree.
jimmi_the_dalek,
Indeed, over the U.S. the agreement between satellite and surface records is quite good: http://rankexploits.com/musings/wp-content/uploads/2013/01/uah-lt-versus-ushcn-copy.png
How is the systematic difference in variability between CRN12 and CRN345 stations addressed?
CRN345 stations are poorly rated mostly because of asphalt; concrete; buildings etc; too close to the instrument. In part the effect of this material is to buffer temperature and diminish the extent of abrupt change. In a cold snap the CRN3/4/5 instruments sitting on their warm concrete pads near their brick walls will react more slowly to the temperature change than the less buffered rural CRN1/2 instruments. Will this not lead to spurious detection of a step change in the CRN1/2 stations causing them to be adjusted upwards to match the nearby poorer quality thermally buffered ones?
Furthermore the immediate environment around a CRN5 instrument sited in a concrete yard near buildings while far from ideal, may actually be more stable than that of a CRN 1/2 instrument in a rural setting where grass can be cut, there are going to be seasonal changes in vegetation, and where the effects of agriculture – ploughing a field or cutting trees – on nearby land may lead to step changes which should not be adjusted for because they are temporary and will be reversed as grass grows, vegetation grows back and so on. Such changes are natural and do not bias the record over the long term. Because of this natural variability more step changes are likely to be detected at CRN1/2 stations. Each step change results in the CRN1/2 station being adjusted to match its CRN3/4/5 neighbours which are subject to the more gradual effect of UHI.
A step change detection algorithm is useful to flag possible bad instruments or changes to the measurement environment which should prompt a reevaluation of the site. But it should not be used in the absence of other evidence to automatically adjust the temperature record. Step changes reflect changes to the instrument or its close environment. These changes may be a cause for reclassifying the station. But changes which do not lead to reclassification are not expected to bias the temperature record and shouldn’t be adjusted for. If the effect of the step change detection algorithm is to change the measured rate of warming then it is spuriously introducing a bias into the temperature record.
Systematic UHI warming bias will not be picked up with a step change detector. The problem is the slow creeping growth of cities resulting in a systematic warming bias, not step changes to the immediate environment of the instrument.
k scott denison,
The zero line on the y-axis is an artifact of the baseline period chosen for calculating anomalies in individual station records and isn’t really physically meaningful. In general, the NCDC’s approach to homogenization assumes that the present records are correct, and will adjust past records up in down to align everything when removing detected breakpoints that occur at one at one station but none of the surrounding stations.
White House report on Recent US Temperature Trends cites Anthony Watts’ paper here:
http://nca2014.globalchange.gov/report/our-changing-climate/recent-us-temperature-trends
Fall, S., A. Watts, J. Nielsen-Gammon, E. Jones, D. Niyogi, J. R. Christy, and R. A. Pielke, Sr., 2011: Analysis of the impacts of station exposure on the US Historical Climatology Network temperatures and temperature trends. Journal of Geophysical Research, 116, D14120, doi:10.1029/2010JD015146.↩
Maybe not so fringe after all.
Zeke, are you saying they just detected a 1 degree increase step change in the present records…
so they adjusted the past down 1 1/2 degrees?
There’s no way to look at that graph and not realize they were detecting present day increases…
…and adjusting the past down (decrease) each time
But then getting rid of truly rural stations…and homogenizing the rest would get the same result
…and results look like what they are after
If I was a computer games modeler….I would be really pissed
No wonder they can’t even get a trend right….they are working with numbers that have been worked to show a trend that doesn’t even exist
Even if they designed the perfect climate computer game….it would never be right…and they wouldn’t even know it
“Better yet, why not simply throw out all the questionable data for questionable stations, by giving up on trying to “salvage” obviously polluted data with various corrections and methodologies, and just go with data you know to be free of such problems? Why is there this persistent idea that you have to use all the data, no matter how corrupted, because of a belief it can be “fixed””
Well cynical me thinks that if they discounted the bad, stations, or established realistic error bands, or stopped twerking around with adjustments, then the results wouldn’t be scary enough to keep the grant money flowing.
I’ve had the pleasure of running a statistical control system for critical components for a tier 1 auto supplier. (Critical in our lingo means that if you stuff it up people are maimed or killed) BEST admit to 70% of US stations being between +-2C and +-5C out. We are looking to detect a mean temperature shift of 0.6C. You cannot hope to find one with the other. Were you to take a similar proposition to an Auto OEM, even concerning a purely decorative component you’d be told to f*ck off out the room, they wouldn’t be coy about using that language either.
Zeke, with regard to your statement about UAH Satellite temperatures being in ‘quite good’ agreement with USHCN numbers… Where are you getting those UAH values..? I decided to go over to Dr. Spencer’s page to see what they have, being purveyors of that data, and they’ve got.
http://www.drroyspencer.com/wp-content/uploads/UAH_LT_1979_thru_April_2014_v5.png
Are you parsing out just USA data from the satellite record? To me, the data sets (your UAH and Spencer’s UAH) look quite different, and without a provided explanation.
You can’t just adjust data based on random numerical algorithms. All that will do is adjust in the direction of the mathematicians bias. Any adjustments would have to take into account the equipment (for example MMS equipment fails hot) (paint failure creates added heat) etc. It would also have to take into account the geography, as in nearby water will have a limiting effect, people create additional heat, and its not uncommon for elevation differences to create mathematically detectable anomalies due to air mass temperature changes. In fact the geography can cause legitimate differences. Homogenization improperly done just causes bias. Interestingly we have the satellite records that are near bias free, though short, this offers us a sort of limit for the mathematicians. Any trend from a land data set that is significantly different than satellite is probably biased one way or another; given those bias’s were probably better off with unadjusted data. At least with unadjusted data were not fooling ourselves.
Salamano, they are using adjusted UAH data obviously.
In a land where constants aren’t and variables won’t, the Fairfax Law strikes again.
The Fairfax Law clearly states that any facts which support the outcome you want are fair facts for the discussion. This especially seems to apply to CAGW, where made-up facts trump actual measurements. We HAVE TO HAVE a man-made catastrophe, so any facts that support that outcome are fair facts for the argument.
Could Zeke explain why the adjustments are on a nice slope? If major changes took place in the 1940’s like Zeke said then we should see no adjustments UNTIL the 1940’s then no adjustments again. Sorry Zeke but the fan is on and you are throwing a lot of …. into it and I’m just glad I’m on the right side of the fan because it’s all falling back in your face. Try the truth next time.
Dumb question: Is there a plot of the USHCN raw temp data over the time span of the corrections applied – 1880 – present? Cheers –
As popular as this site is, if you really want to reach and communicate with people who aren’t already convinced global warming is a scam, you have to do a better job of explaining stories like this. Who is the USHCN and what temperature data are they adjusting? Is it US data? Global data? Is it a data set nobody looks at? How can they make this much of an adjustment to one set of temperature data, won’t it look ridiculous when compared to other measurement data sets of the same thing? I’m sure your loyal climate gurus get it and catch on quickly and trust any data you show, and I trust you too, but many people won’t. And I can just see anyone who visits this site who isn’t already convinced global warming is a scam, seeing this graph, not believing it, asking those questions above, clicking on your data links and not seeing the graph, and not seeing enough explanation of where the data is that was used to make the graph, and especially seeing that 1.3 F bump above all previous data points, and just saying its BS. I can’t forward this story to anyone because it is so spectacular anyone I send it to will surely ask me “Where did he get that, show me the data, I don’t see it in his links” and I wouldn’t be able to tell them. If the data is in your links its too hard to find. So they won’t believe it, its too easy to dismiss. Reaching your fellow climate gurus is one thing, reaching the other 96% is another, and I don’t know how you would do it, but this isn’t doing it. Maybe its just too complicated.
I do not understand how government employees and government funded scientists can clearly admit they have fudged the data since 1960 (that is what the “final minus raw” graph shows) and not be charged with fraud. If a company sent in an SEC, EPA or IRS form with fraud as blatant as this they would be charged with a criminal offense. These people really do belong in prison.
USHCN doesn’t make the adjustments, the people there make them. Who are these people?
agimarc says:
May 6, 2014 at 5:30 pm
Dumb question: Is there a plot of the USHCN raw temp data over the time span of the corrections applied – 1880 – present? Cheers –
===========
yep
http://stevengoddard.wordpress.com/tracking-us-temperature-xxxxx/
substitute f…….r…….a…….u…….d for the X’s…….wordpress don’t like that word
Well if I had ever “adjustified” ANY report of ANY parameter in ANY R&D work, I ever did, for ANY of my former employers (all first line companies); or worst yet changed someone else’s reported measured results; they would have fired my arse faster than I could take the last swallow from my coffee cup.
I have always assumed that you could get thrown in jail for adjustifying any public reports of anything; erasing e-mails, destroying records; any of that stuff.
One of my employers; a big one, but not the biggest, was VERY big on lab bench data taken from standard production line samples of supposedly “final version” products, to verify, that products shipped to customers conformed to the written specs guaranteed in the company product catalog.
Falsifying any observed results, was grounds for immediate dismissal.
Archives of past information are vital for when problems show up , and need to be researched for causes.
The whole idea of altering public records of taxpayer funded information; just makes me puke.
If some past result is believed to have had some sort of anomaly; you don’t change the record, you add to the information, with a statement, of what possible sources of error might have been in play at the time.
Sounds like government climate science is akin to having a hook and ladder fire truck with three drivers, front, middle and rear, each driving what (s)he thinks is the better route; with nobody knowing exactly where the fire is.