.

Saturday, February 20, 2016

Anthony Watts finds another squirrel in Arizona when he knows there is no bias

Sou | 1:35 PM Go to the first of 10 comments. Add a comment
Anthony Watts is busy looking for squirrels in Arizona to distract from all the record high temperatures. He's found another poorly sited weather station (archived here). Without acknowledging that his previous find wasn't used in any dataset, national or global, he is delighted to find that this one is. It's located at the small town of Parker in north-western Arizona near the border of California.

Parker 26250 is included in the GISTemp global data set up until 2007. It is also included in the NOAA USHCN dataset. There are quite a few quality problems with it, as shown at Berkeley Earth. There is missing data, it's very poorly sited, and it's had several replacements of an MMTS sensor.

The issue is not that there are poorly sited weather stations in the data sets. The question is: are they biasing the record?

The answer would be a resounding NO! They are not.

NOAA now uses a lot of data in its official CONUS dataset, nClimDiv, and uses sophisticated algorithms to correct wonky data and present the information on temperature as close as it can. It must be quite good, because there is very little difference between its pristine record of its Climate Reference Network (CRN) and the larger dataset (ClimDiv) as you can see below. I've highlighted where there are the biggest differences and they are miniscule:

US CRN and US ClimDiv January to December annual data from 2000 to 2015 Source: NOAA

What about Arizona? Here are some charts of Parker and stations close by, from GISTemp:

Parker only goes to 2007 at GISTemp:




The nearest neighbour, Blyth, is 67 km away and goes to 2008:



Needles is 75 km away and goes to the present:



Wickenburg is 144 km away and goes to 2015:



Searchlight is 158 km away and goes to 2015:





Prescott is 177 km from Parker and goes to the present:




You can look at the others yourself.

NOAA has this chart for north west Arizona:



And it has this chart for the whole state of Arizona:



Berkeley Earth has this chart for all of Arizona:



What seems fairly clear is the this century, the temperature has been warmer in that part of the world.

Anthony Watts has yet to show that his claim that poorly sited weather stations are biasing the record for the USA or anywhere else. Many people have looked into this and have published papers showing they do not bias the record in any substantive way, including Anthony Watts himself. His paper found that while there were differences between maximum and minimum taken separately, there was no significant difference between the mean temperatures, with the differences cancelling out:
The opposite-signed differences in maximum and minimum temperature trends at poorly sited stations compared to well-sited stations were of similar magnitude, so that average temperature trends were statistically indistinguishable across classes. For 30 year trends based on time-of-observation corrections, differences across classes were less than 0.05°C/decade, and the difference between the trend estimated using the full network and the trend estimated using the best-sited stations was less than 0.01°C/decade.




From the WUWT comments


Anthony Watts has successfully dog-whistled his resident conspiracy theorists:

daveandrews723 seems to think that NOAA has "made up" the data reported by the COOP volunteers:
February 19, 2016 at 4:25 pm
fascinating and important info. NOAA/NASA have an agenda. It is obvious. They can make the temperatures, past and present, anything they want them to be, and of course the want the past to be cooler than the present. Urban heat islands, be damned. They don’t care. They fill in whatever numbers they thinnk they can get away with to try to support their nonsensical AGW hypothesis…. and to keep the money rolling in. It is all a travesty and a dark period in science.
FJ Shepherd didn't bother reading how Anthony wrote it was poor siting, not UHI effect, that is possibly affecting the Parker weather station. (Parker has a population of around 3000 people.) As the pristine record shows (see up top), it's not UHI effect that's causing the warming.
February 19, 2016 at 5:03 pm
Considering that even with NOAA’s inclusion of recording stations with the UHI effect, the overall temperature of the US is not showing much warming occurring. This is not a good sign.

JohnWho is learning (not French, he's learning climate):
February 19, 2016 at 5:25 pm
Que Stokes explaining that NOAA really doesn’t use that station and Mosher explaining how his “best” algorithm corrects that station to within .01 degree F. 

4TimesAYear could be referring to melting ice, hotter heat waves, worse floods etc. Or it could be that like Anthony Watts, he or she just doesn't want to face the bleeding obvious:
February 19, 2016 at 5:45 pm
You know, I don’t think the argument against CAGW is going to be won with equations – there are too many ways to fudge numbers. The argument is going to be won with common sense arguments. Those things that make you reconsider whether the numbers are worth anything to start with. Surface station temps don’t seem to be worth a whole lot (and for even more reasons than presented here). Keep up the great work, Anthony! 


References and further reading


Hausfather, Zeke, Kevin Cowtan, Matthew J. Menne, and Claude N. Williams. "Evaluating the impact of US Historical Climatology Network homogenization using the US Climate Reference Network." Geophysical Research Letters (2016). DOI: 10.1002/2015GL067640

Hausfather, Zeke, Matthew J. Menne, Claude N. Williams, Troy Masters, Ronald Broberg, and David Jones. "Quantifying the effect of urbanization on US Historical Climatology Network temperature records." Journal of Geophysical Research: Atmospheres 118, no. 2 (2013): 481-494. DOI: 10.1029/2012JD018509 (open access)

Menne, M. J., C. N. Williams, Jr., and M. A. Palecki (2010), "On the reliability of the U.S. surface temperature record". J. Geophys. Res., 115, D11108, doi:10.1029/2009JD013094. (open access)

Fall, Souleymane, Anthony Watts, John Nielsen‐Gammon, Evan Jones, Dev Niyogi, John R. Christy, and Roger A. Pielke. "Analysis of the impacts of station exposure on the US Historical Climatology Network temperatures and temperature trends." Journal of Geophysical Research: Atmospheres 116, no. D14 (2011). DOI: 10.1029/2010JD015146

From the HotWhopper archives

10 comments:

  1. Watts should start a surface stations project and publish his results.

    Oh...

    ReplyDelete
  2. I'm not seeing the graphs from Blyth to Prescott inclusive, in either Chrome or Firefox ...

    ReplyDelete
    Replies
    1. Sorry about that MWS. I've now copied the charts into the blog instead of just linking to them. I think that fixes the problem. Let me know if not.

      Delete
  3. I note still deliberately looking at squirrels rather than http://onlinelibrary.wiley.com/doi/10.1002/2015GL067640/full. Anyone would think he didn't like the results Zeke and colleagues found.

    ReplyDelete
    Replies
    1. Thanks Peter. I had at the back of my mind that I was missing an article. I've added it to the "further reading" section above, together with a link to Kevin Cowtan's blog.

      Delete
  4. He displayed a different parker

    "NASA GISS keeps a plot of Parker 6NE data, and it seems missing data has been a hallmark of this station for quite some time. Notice all the gaps:"

    The Parker he displayed was a different parker

    Here is 6E. Note how we correct it.

    http://berkeleyearth.lbl.gov/stations/34568

    Here is the Parker he pulled from GISS

    note the lat lon

    Note how we correct its trend

    http://berkeleyearth.lbl.gov/stations/29136


    Yes bad stations exist. And yes they can be detected and IMPROVED ( not corrected precisely, but we move the answer toward ground truth)

    ReplyDelete
    Replies
    1. arrg two different Parkers.. neither named 6NE..

      but both corrected. nice

      Delete
    2. Thanks Steven.

      There were three Parkers in the NOAA database. One was in Colorado, so it didn't fit. The other two were close together. One in California and the other just over the border in Arizona.

      26250 in Arizona, which is in GISS and in USHCN

      46699 in California, which is not USHCN or GISS

      I figured it was the first one, when I searched for Parker 6NE and found this:

      Did Anthony mix them up? I couldn't see it, but could be wrong. (I didn't check the "Steve Goddard" stuff. He's usually wrong.)

      Delete
  5. I don't think he mixed them up but it was cool to see that we fixed both. I'll look at the third. We had 4-5 parkers.

    ReplyDelete

Instead of commenting as "Anonymous", please comment using "Name/URL" and your name, initials or pseudonym or whatever. You can leave the "URL" box blank. This isn't mandatory. You can also sign in using your Google ID, Wordpress ID etc as indicated. NOTE: Some Wordpress users are having trouble signing in. If that's you, try signing in using Name/URL or OpenID. Details here.

Click here to read the HotWhopper comment policy.