How not to measure temperature, part 40

Rounding out a review of California weather stations this week we visit Gilroy, CA, the garlic capital. This COOP station has an MMTS temperature sensor on a pole just a few feet from a concrete slab. We’ve seen a lot of that lately. But look closely – roasted garlic anyone?


Photo from NWS, San Franciso/Monterey CA

While it’s likely the BBQ grill is not used daily, one has to wonder just how much bias it’s proximity imparts into the temperature record. This station  COOP number is 04-3417 as is part of NOAA’s “A” network which reports climate to NCDC. It is located at the Fire station in Gilroy, seen below. Notice that is is also near a large parking lot and major intersection downtown. So much for NOAA’s 100 foot rule for station siting.

gilroy_firestation_aerial.jpg
Click on the picture for a larger interactive view

The recently released paper from LaDochy et al. showed that “urban” stations warmed at a rate of 0.20°C per decade while the “non-urban” stations warmed only 0.08°C per decade, with the lack of attention to the measuring environment such as we see here, is it any wonder?

About these ads

22 thoughts on “How not to measure temperature, part 40

  1. Another day, another How Not to Measure temperature!

    Chalk up another for the Rev! (And the good UHI cite.)

  2. Hey, the MMTS is installed over a nicely mowed grassy patch at the required 5-foot height. That should count for something! :-)

  3. Goto the aerial view and scroll to the right, or up, or down a few times. Plenty of better places to put the sensor that would remove any risk of bias. That’s the best they could come up with?

  4. My oh My, I just wonder how thier readings are on those nice days, you know how firefighters are they just love the flavor of smoke in their food. Just kidding but most fire fighters and policemen that I know love to bbq. I can invision the grill being used at least 2 days a week during the times of good weather. I would just wonder what a graph of a year would look like on a short average of say a week. bet it would be something to compare with a rural site for the same time and duration. ( I mean a real rural site)

    Bill

  5. I’m not sure what you think this means?

    Scientists are way ahead of you on the issues of validity of data, methods of collection, and adjustments, and have spent considerable time and effort in making sure that appropriate corrections are made. Please read this material

    http://www.realclimate.org/index.php/archives/2007/07/no-man-is-an-urban-heat-island/#more-454

    and check the links. If you have a meaningful criticism of these corrections, if you have any data or science, or can even link to any, put it out there for scrutiny.

    The issue of warming is really off the table and has been for some time. Even Fred Singer and Richard Lindzen now agree on that.

  6. Seesdifferent:

    Welcome and thank you for your comment.

    As to the “scientists are way ahead of you” comment, I assume you mean in their claim of being able to totally eliminate the microsite biases such have been outlined in this series.

    If that is the case then one phrase comes to mind: “lights=0″. As a learning exercise, I offer that you should bone up on that and get back to us and then explain exactly how that procedure accounts for all of the micosite biases we’ve seen here.

    There’s a new paper from NASA JPL on this very subject of near sensor bias, which will be posted up here as soon as I can obtain the original PDF copy.

    As to the “issue of warming” being “off the table” I’m not sure just what you are implying, but I can assure you there remains a quite vigorous debate in the scientific community on that issue, its true magnitude, and its cause.

  7. I’m wondering what the adjustment will be for when we find the MMTS actually in the fireplace. – you know it’s going to happen!

    Look here seesdifferent, even your heros at realclimate find that

    There is a possibility that these (micro-site) effects may change over time, putting in artifacts or jumps in the record. This is slightly different from the more often discussed ‘Urban Heat Island’ effect which is a function of the wider area (and so could be present even in a perfectly set up urban station). UHI effects will generally lead to long term trends in an affected station (relative to a rural counterpart), whereas micro-site changes could lead to jumps in the record (of any sign) – some of which can be very difficult to detect in the data after the fact.

  8. I picture one of the GISS crew pouring over a chart saying, ” We got a 350 degree jump in Fresno.”
    And then Hansen calls out from his office “That one’s a lights”.

  9. Anthony, on the 5 categories of weather station compliance, have you run a temperature series for each category on the 400+ stations already surveyed? I know you have the individual temperature series for each station. It would be highly instructive for the naysayers like Seedsdifferent to see the effect of inappropriate placement even after Hansen does his magical UHI adjustment.

  10. Yes he has, somewhere earlier in this blog. Just keep moving back until you hit a big pie chart.

    And yes, the results will knock your socks off.

  11. If I had more time, my “Q” would be getting fired up every night that it wasn’t pouring down rain. I was out there last year in the middle of the Big Freeze, ’twas about 27 F where I was (coast ranges, a bit NNW of Gilroy, upper 30s N latitude).

  12. Pingback: Top Posts « WordPress.com

  13. Yes Steve, I even put my Thanksgiving turkey in the Q around midnight and cooked it . I bbq year round so I know that if that station were on my back porch the data would be skewed. One of my friends from another blog that also posts on rc keeps saying higher temps dosen’t mean rising temps high temps dosen’t mean rising temps over and over. Some times I wonder how I could ever have managed before their help. I wonder how that would adjust as I am in a very rural setting……They make it so that my head hurts sometimes.
    I’m sure glad I’m not a scientist.

    I very much enjoy this site.
    Bill

  14. for dscott, thanks for asking.

    I have not personally run a series, preferring to wait until more of the network is surveyed. However one person who started frequenting Climate Audit just after I released the first batch of stations with 33% of the USHNC surveyed, wrote a program to analyse it, but I don’t think the data was complete enough to give meaningful results.

    I’m going to wait until we get more stations in before I attempt any detailed time series analysis. One of the problems is station moves, and we have to dig into the station history more to determine if a station that rates a 5 now, rated a 5 10 years ago or 20 years ago.

    The pie chart that Even mentioned is available on the main page of http://www.surfacestations.org and shows the percentage of stations as they rank in the CRN rating scheme.

  15. Anthony,
    Vigo is a small farming community that 75 years ago had one filling station (gone now) 2 saloons (gone now) and one general store (gone now) and one church still here. I am located in N. E. Alabama near the town of Piedmont on the edge of the Taladega National Forest. In the extreme north of Calhoun county. As we say “Square dab in the middle of the drought” It is still beautiful here

    Bill

    REPLY: Thanks Bill, just curious.

  16. “One of my friends from another blog that also posts on rc keeps saying higher temps dosen’t mean rising temps high temps dosen’t mean rising temps over and over. ”

    The trouble is that when sites don’t–start–with violations and then are slowly overcome by them, their biases accumulate in the–historical–record and affect the rate of change measurements.

    E.g., If there’s a clean CRN-1 type site in 1900 and by the year 2000 there is a parking lot, an AC, and a BBQ next to it, the 100-year history of all those violations are now part of the historical increase, with offsets for when each violation first occurred, when no actual temperature change may have occurred at all. So maybe there’s a 1 or 2 or even 5-degree increase over the 100-year record that is–completely–spurious.

    On top of that, say there’s the same half-a-degree C violation bias all the way from 1900 to 2000. Even though it would not affect the rate of change, it still increases the margin of error of the station, making the data that much more unreliable on account of the statistical snow.

  17. And on top of THAT, note the California article. It would seem that heat sinks (such as UHI or microsite violation) may exaggerate small increases as they occur. Or, in other words, “over and over”, if La Dochy’s premise is correct.

    Note that in that article, it claims that heavy urban sites increased at 0.2 degrees C PER DECADE [sic] while the rural sites (who knows how “clean”?) increased at only .08C PER DECADE. Or, in other words, “over and over”.

    Of course, if there were no increase, this would not show up at all, and if there were an actual decrease, the DECREASE would be magnified “over and over”.

  18. Anthony, If I remember correctly, some NOAA or GISS people talked with you from the conference you attended a few months ago. What amount of concern have these people expressed over the quality of the temperature measurements? Are they prepared to revise the temperature data to account for your findings? More importantly, have they begun any kind of UHI correction of the sites you already indentified on the order of 4 and 5 in being compromised so badly??? To your knowledge does the Data Quality Act cover these types of revisions?

    Let us know if you are meeting resistance, we are more than happy to give some extra public exposure to “encourage” the appropriate people to do the right thing.

  19. Thanks Evan,
    I had kinda figured that out all ready. I really don’t know why the warmers
    tend to think that we common folk are slow thinking. we may not have the degrees that they have but common sense viewing the findings atthe weather measurement sites would tell you that something is amiss. For me it would be a more intellegent thing to repair or relocate the faulty stations than to have to have so many possibly erroneous compilations of data correcting programs to correct to an extent we do know because we have no true readings from that site. It would also make good reasoning to follow your own rules about sitting of stations.

    2 cents worth
    Bill

  20. “I really don’t know why the warmers tend to think that we common folk are slow thinking. ”

    I guess for the same reason they thought we common people were slow thinking when we questioned the assurances that all the resources would run out by the year 2000 and that we were beyond the possibility of avoiding massive worldwide famine by 1980.

    Like anyone would bother to waste wealth searching for a mineral of which there was 20 years’ proven reserve! Sheesh! Talk about your slow thinking! Of course, by the time 20 years rolls around there’s bound to be 30 years’ proven reserve (which is, of course, what happened and what continues to happen).

    But noooooo, anyone who “denied” Limits for Growth was an idiot. (And a hateful, selfish one, at that.) No thought at all for the suffering “Limits for Growth” inflicts on the poor! Same kind of elitist (and sublimely unliberal) thoughtlessness at work in the GW-activist community, I think!

Comments are closed.