How not to measure temperature, part 39

One of the most surprising things I’ve learned from the project is that for some odd reason, there are a number of climate monitoring stations of record in the USA at sewage treatment plants. If you’ve ever driven by one of these in the wintertime, they tend to look like steam saunas. They are localized heat bubbles from the waste-water processing.

At Orangeburg, SC not only is the official USHCN climate station of record at a sewage treatment plant, it’s also a nonstandard thermometer (a climate station normally looks like this), and strapped to the side of a telephone pole. I don’t know about you, but experience with creosote treated telephone poles tells me that they’d tend to create a hotter local measurement environment.


Photo by volunteer surveyor Don Kostuch. See the complete image gallery here.

Then there’s the brick building to radiate heat at night, the asphalt parking lot, the effluent channel running nearby, and the overall sewage treatment plant waste heat to consider. I doubt there is an easily applicable set of equations which can untangle the myriad of potential microsite biases.

Then there’s sewage. As population growth occurs, sewage plants add more vats and equipment to handle the increased volume. The increased volume of effluent loses some of it’s heat at this location during the purification process.

Click graph for larger version, data from NASA GISS

The real question is: What are we actually measuring at this location? Are we measuring temperature as an indicator of climate change or are we measuring waste heat from increases in sewage processing that mirrors local population growth?


I made an error, this is not a sewage treatment plant. It does treat water, and the description in the site survey from the surveyor was “water filtration plant” which I mistook to mean “sewage treatment” since so many other locations have been at sewage treatment facilities. For example, one of the worst is Titusville, FL, which has been highlighted in this blog in Part 31 and also surveyed by the same volunteer. I looked at the photos he provided, and did not discern initially that the tanks were not for sewage treatment.  Some sharp eyed readers have pointed out the identification problem, which I’m happy to correct.

The questions about the validity of the temperature measurement environment in the midst of a sewage treatment plant are still valid, but do not apply to this location.  So the question we now have for this location is; do the large water filtration pools on this site provide an evaporative cooling effect or do they release heat?

The water vapor impacts at the facility are likely a factor, possibly for Tmin overnight, which is more prone to such effects. Note that there has been new construction at this location, and given the apparently new water filtration pools added on site, there may still be an effect of measuring the local population increases by proxy.


14 thoughts on “How not to measure temperature, part 39

  1. That pole is too weathered to still have any thermal effect, but try to predict the airflow characteristics around the pole, conduit and cable which appears to not be fastened very well. I must say, these sites are lovely, just lovely.

  2. Hi George,

    That may be true, but at one time that pole was loaded with creosote and darker. So a long term diminishing bias has been applied. The view above is looking east, so there is also a period of time in the day, which varies through the year, when the sensor is shaded.

    The idea of taking continuous measurements is to eliminate as much entropy as possible. That doesn’t seem to be a consideration at all here, or in many other NWS COOP sites.

  3. Isn’t the pole, itself, a heat sink?

    BTW, Rev, I’ve been reading on other sites about how all these-here station observations completely confirm all the original data and that it is just amazing how it was all being done right all along. And how trifling the station biases really are. And how THANKS to the NOBLE efforts of Anthoiny Watts and the WORTHY Steve McIntyre, NASA, NOAA, and the IPCC have been COMPLETELY vindicated. On to solutions.

    Watts Up With That?

    (My position is that when the Rev and St. Mac, whose honesty I trust in these matters, tell me that this is so, I’ll believe it. Not a day before.)

    However, I was also reading about a western bias on the CRN 1 & 2 rural stations (and as we well know, the West is the place where the most warming actually occurred) and the data has not been properly gridded. Mosh mentioned it somewhere, IIRC.

    Watts Up With That?

    And I also read that the Stratosphere doesn’t “count”. Those nasty CFCs are cooling the strat, so dumper that data (why that wouldn’t be a homeostasis effect and highly relevant escapes me). So keep your warming trend, thank you.

    Watts Up With That?

  4. I also note we are in a cooling area of the country, yet we have a definite increase. Was that plant always there?

  5. How are these ‘official’ sites funded? As so many seem to break the most basic rules, wouldn’t the people funding them have cause to ask for a refund? Wouldn’t that make a few headlines?

    Surely like any other product or service, they are clearly “Not fit for purpose”.

  6. I’d like to know on just what basis George is making the claim that the “pole is too weathered to still have any thermal effect”.

    The entire problem with the environmental left is making spurious, non-provable, or patently false claims (DDT is bad, more trees is good, atmospheric CO2 content preceeds temperature increases).

    Two falsehoods is NOT better than one!!!

  7. In the UK these meteorological observation stations are sited at places that ensure ease of maintenance, like beside main roads and motorways. A very good way to ensure high biased temperature readings.

  8. The shape of the temp graph looks remarkably like the graph for the US over that time period: Hot in the 1930s-40s, cooling until the 1970s, warming up after that.

    Got another station in the area to compare it to?

  9. I thought it was policy to have these stations out in the grass away from any type of man-made sources of heat… isn’t there a standard distance they should be posted away from these places – like 100 feet or something?

    I should think that stations that do not meet the stated criteria would have to have all their data disqualified and removed from the body of data for this study since they do not reflect the true temps for that area…

  10. James:

    That pole was originally black and exuding creosote fumes like crazy. Back then it was a definite thermal and contamination problem. In its weathered state, wood is pretty benign thermally, and the bias effects have been reduced. Producing, as Anthony correctly noted, a changing bias during the life of the pole which would be almost impossible to crank into your model. You missed my point, which is that given the location, the weathered pole is likely the least of their bias contributors. Not that multiple wrongs make a right.

  11. John:

    The state of California does a very similar thing, placing remote monitors at most major Interstate Highway interchanges, or at least that was what I saw in 1976 when I was involved in the installation of a similar meterological monitor package at the sewage treatment plant in Pleasanton, CA. (Aha! I’m one of the bad guys.) I have wondered who got the data from those highway monitors, and exactly what all they were measuring ever since. We were concerned principally with local temperature, humidity and wind speed and direction at the facility and its effect on dispersion of viruses which might be released from the ponds and aerators.

  12. Pingback: Top Posts «

  13. Which reminds me, Anthony, no one commented on your leading statement about locations. It has to do with what facility in your town has to be manned 24/7/365? Police, fire and sewage. The police activity is a moving target as it were, but the firemen (also well represented in ss.o) stay in the station a large percentage of the time, and some staff [is] actually [used to be] always present at the sewer plant. In recent years, and I think we discussed this about 2 months ago, these are mostly automated, and require attention only when a pipe gets plugged. So, again, it is [was] availability of someone to read the instruments. These days, with digital SCADA, really remote, unattended, accurately reported sites are possible, but AFIK, none have ever been fielded. We proposed some for nuke plants back in the 70s, but no contracts ever materialized. And from what I see in current advertising, instrumentation is light years better than what we had back then.

  14. Unless all instrumentation is NBS traceble, with accompanying calibration records, the data is useless. You might as well use a wet forefinger. Even if recordings were conducted within a sq Km. of each other, variations will be expreienced. Its not the temperature we want to follow, it is the entropy. If we examine the amount of air vs. contaminents, the amount of contaminates is so miniscule that no logical thinking individual can suggest that there can be any correlation.

    Get real! and get a life!

Comments are closed.