Warming near Coal Plants

I was learning some Python over the weekend, and then afterwards decided to go for a jog at my favorite spot. As I was jogging, I thought about finding out what the temperature trend at this location was by extracting and plotting it with Python. This site seems to offer the best local 4km data for the continental US.

My first Python program:

import requests as req
import numpy as np
import matplotlib.pyplot as mp

loc = {
	'spares': '4km', 'call': 'pp/yearly_timeseries', 'proc': 'gridserv', 'units': 'si',
	'stats': 'tmean', 'start': '1895', 'end': '2019', 'lon': '-84.3922', 'lat': '34.0000'

res = req.post('https://prism.oregonstate.edu/explorer/dataexplorer/rpc.php', data=loc)

tmean = res.json()['result']['data']['tmean']

y = np.array(tmean)
x = np.arange(1895,2020)
m, b = np.polyfit(x, y, 1)
mp.plot(x, m*x+b)



Azalea Park, Roswell, GA

Well look at that: no warming where I like to jog. How about downtowns of some big cities?

OK. Definitely warming here.

Then I had a great idea. What if I chose spots with coal plants?

I got a list of the top 20 coal plants in the USA from here.


Bowden               -0.0076
Scherer              -0.0084
Gibson               -0.0098
Monroe               -0.0005
Amos                 -0.0020
Miller               -0.0071
Parish               +0.0044
Cumberland           -0.0059
Gavin                -0.0071
Rockport             -0.0030
Paradise             -0.0048
Roxboro              -0.0090
Sammis               -0.0005
Stuart               -0.0070
Navajo               +0.0116
Sherburne            +0.0032
Martin               -0.0048
Belews               -0.0045
Jeffrey              -0.0062
Gaston               -0.0057

Units are degrees Celcius per year, based off the linear regression.

Only 3 out of 20 coal plant locations experienced warming in the last century. 85% had cooling!

I suspect this fact will not cause climate alarmists to pause and think for one second.

What do skeptics think?

Enjoy 🙂 -Zoe


Above, Sammis image turned out to be a duplicate of Roxboro. Corrected.

Also, a typo: LogAngeles = Los Angeles


# Zoe Phin, 2021/05/09
# File: prism.py
# Setup: sudo apt-get install python3-numpy python3-matplotlib
# Run: python3 prism.py

import requests as req
import numpy as np
import matplotlib.pyplot as mp

locations = {
'NYC':{'spares':'4km','call':'pp/yearly_timeseries','proc':'gridserv','units':'si','stats':'tmean','start':'1920','end':'2019','lat':'40.7', 'lon':'-74'},
'LosAngeles':{'spares':'4km','call':'pp/yearly_timeseries','proc':'gridserv','units':'si','stats':'tmean','start':'1920','end':'2019','lat':'34.0543', 'lon':'-118.2438'},





for name in locations:
    res = req.post('https://prism.oregonstate.edu/explorer/dataexplorer/rpc.php',data=locations[name])

    y = np.array(res.json()['result']['data']['tmean'])
    x = np.arange(1920,2020)

    mp.plot(x, y)
    m, b = np.polyfit(x, y, 1)
    print("%-20s %+.4f" % (name, m))
    mp.plot(x, m*x+b)


Published by Zoe Phin


51 thoughts on “Warming near Coal Plants

  1. So where’s the ghg effect? CO2 concentrations at these locations should be relatively higher and thus warmer – if the theory is true.

    Liked by 1 person

      1. Bowden temperature trend, 1920-2020, with CO2 concentration held constant at preindustrial level: – 0.0098C/year

        Bowden temperature trend, 1920-2020, with elevated C02: – 0.0076C/year

        CO2 warming: + 0.0022C/year


        Obviously, we are not really able to compare Bowden at preindustrial and elevated CO2 levels. This means that what you believe to be a valid experiment is in fact inherently flawed.


        1. Sure. Local temperatures trends can vary with or without changes to CO2 levels, right?

          It follows that the locations that cooled may have cooled even more if not for elevated CO2. Or not.

          No way to tell for sure because we can’t compare the observed trends with what the trends would have been had CO2 remained at 1920 levels.


        2. If I wasn’t born, the Earth would’ve imploded. Can you disprove that?

          If CO2 doesn’t warm at all, one can always claim it prevented cooling from happening, right?

          What’s the point of unfalsifiable statements?


        3. Most scientists agree that elevated CO2 is warming the planet.

          I doubt that any insist, like you do, that the warming effect must so strong that it overwhelms every other variable at every individual location, or else the above is falsified.


        4. And ALL engineers do not place CO2 between panes of glass to help keep your windows “trap heat”, because the result is: COOLING.

          You’re suggesting that 85% of places with high concentrations of CO2 (Coal Plants) were going to cool anyway? Don’t you find the odds low for that and this hypothesis self-serving?

          Tell me, when greenhouse operators pump CO2 into their warehouse, how much does the temperature go up?

          Surely such evidence would be useful to the AGW narrative. So where is it? Please think about it.


        5. – Double paned windows are intended to reduce conduction, not trap heat.

          – Off hand, I would have expected around 50% of the coal plant locations to have a cooling trend – in line with the lack of warming over the general area:

          I’m guessing most of the plants are in rural areas, where the UHI has had less impact than the nearby urban – perhaps helping explain the 85%

          BTW, it would be interesting to see a time series of CO2 levels near the coal plant’s in question compared to the county or state average. Maybe a significant difference, maybe not.

          – The glass in a greenhouse already acts like CO2. Allows visible in but absorbs the infrared emitted from other surfaces. More importantly though, the greenhouse is mostly warmed by preventing free convection, like a car in the sun with the windows rolled up. You should know this is different physics than the atmospheric GHE.


        6. “reducing conduction” is not “trapping heat”? That only goes for radiation? Strange how that works! Anyway, CO2 causes cooling when placed inside double window layer.

          Wait, so CO2 won’t warm a greenhouse because of the glass layer? That’s not in the theoretical model of GHGs. In that model there both a “glass” layer and absorbtivity between. Same as I’m talking about here.

          Why 50%? In your map negative is gray, not white. Look again. I feel your confusing white and gray.


        7. I should clarify this statement,

          “Double paned windows are intended to reduce conduction, not trap heat.”

          I was referring to the choice of gas, often Argon, used inside the two panes of glass. Argon is chosen mainly for its low conductivity, not for its ability to trap heat.


        8. “Anyway, CO2 causes cooling when placed inside double window layer.”

          Compared to what? Source please.

          “Wait, so CO2 won’t warm a greenhouse because of the glass layer? That’s not in the theoretical model of GHGs. In that model there both a “glass” layer and absorbtivity between. Same as I’m talking about here.”

          My thinking is the warming would be negligible and hard to notice.

          Think about it – CO2 concentration is measured in ppm, this compared to what a solid surface like glass would be able to absorb.

          And as mentioned, the lack of convection is the bigger player, dwarfing the warming effects of CO2 even more.

          Another problem is the potential lack of a significant lapse rate within the greenhouse (for various reasons). No lapse rate – no GHE if I understand correctly.

          “Why 50%? In your map negative is gray, not white. Look again. I feel your confusing white and gray.”

          Even though the white indicates no change, most locations will have either a warming or cooling trend if the value is taken to the nearest one hundredth. Guessing 50/50 as a ballpark.


        9. White doesn’t indicate no change, it indicates 0 to 0.1C per decade, or 0 to 1C per century.


          “No lapse rate – no GHE if I understand correctly.”
          Fair enough, that does logically follow from the math …
          But, it is also claimed that GHGs creates the lapse rate!

          “compared to what a solid surface like glass would be able to absorb.”
          But what does GHG math show? That you can stack all the varying density of air, and have its GHGs form average LAYERS. Without those LAYERS, there is no GHG math warming!

          Without a LAYER splitting IR beams, there is no GH effect.

          What does the glass provide? An IR beam splitter. Half goes up half goes down.

          But then when a beam splitter (glass) is provided, you say “My thinking is the warming would be negligible and hard to notice.”

          Do you now see why scientists are skeptical?


        10. Zoe
          “White doesn’t indicate no change, it indicates 0 to 0.1C per decade, or 0 to 1C per century.”

          No, the white indicates trends ranging from – 0.1 F/decade to + 0.1 F/decade, centered on no change at all.

          Which is why I guesstimated about a 50/50 toss up between positive and negative trends.

          (And notice the units are Fahrenheit, not Celsius.)


          From the link,

          “But in their study Berkeley laboratory test experts Reilly, Arasteh and Rubin…”

          And they would be horrified by the bonehead conclusions drawn from their work and then spread around social media.


        11. I suggest you use a screen color picker program to see the difference between gray and white, because your eyes are failing you.

          “And they would be horrified by the bonehead conclusions drawn from their work and then spread around social media.”

          It’s rhetoric like this that shows skeptics are right. If their work doesn’t show CO2 between glass causing cooling … then why isn’t industry making money selling a better insulator ? Where is the product?



        12. Zoe,

          – Ok, I’ll take your word for it regarding the white/grey issue, even though no discernible difference on my mobile device.

          – What you and the author of the article are doing, probably without realizing it, is constructing a strawman argument.

          As an analogy, imagine someone who’s convinced a blanket can’t really make you warmer, and to “prove” it, he sets a thermometer on a large rock and then covers the rock and thermometer with the blanket.

          “Look. if what you say about blankets is true, then the rock should have heated up, but as you can see the temperature is unchanged! Experiment proves I’m right.”

          You have read that CO2 is not the best fill for a centimeters wide thermal barrier (panes of glass separating two air filled spaces – the inside and outside of a house or building), and have taken this as proof that CO2 cannot possibly be a thermal barrier to infrared moving from Earth’s surface to the vacuum of space.

          A simpleton’s assumption, with little or no consideration for the vastly different scenarios – just like the guy with the rock and blanket.


        13. Blankets don’t warm you.

          At best, you warm the blanket.

          As the distance between glass was separated, more cooling resulted. At what distance do you suggest CO2 should warm by trapping RADIATION?

          The two layers of glass is like the surface and TOA. According to GH theory, CO2 in between should warm by reducing OLR at TOA.

          Guess what … there’s a temperature difference between inside and outside of window. It doesn’t matter if outside is not as cold as space, radiation can still go there.

          According to GH math … radiation should overwhelm conduction. Air conduction is pitiful compared to radiation.

          Again, a blatant attempt to not allow any experiment to verify or falsify theory. We have one Earth, one Venus, and no experiment can verify that GHGs make it hotter. None.

          I think I can design an experiment that NASA, ESA, etc can perform in space for very few $$$. Will they? Did they? Would they? Use your brain.

          They will not do it, because the results predicted by GH theory are false, and they don’t want to look bad.


        14. “Blankets don’t warm you.”

          Hah. I wonder how long it’d take you to change your mind on a cold night in Wisconsin without a blanket!


        15. Sure is nice to have warm air between you and the blanket … warmed by your body. Sure is a waste to constantly heat 10 km of replaced air above your head with your own body.


        16. Are you warmer, or a higher temperature, with a blanket than without it? Then it’s warming you!

          If you’re seriously saying “blankets don’t warm you,” then you’re just playing word games.


        17. You, from the other thread,

          “… there is no doubt that a blanket can make you warmer by blocking convection.”


          The first part explains what a blanket can do:

          “ there is no doubt that a blanket can make you warmer…”

          The second part explains how this is accomplished:

          “…..by blocking convection.”


          But when the guy in my analogy says a blanket can’t really make you warmer, you come back with this,

          “Blankets don’t warm you.”

          Why the word games, Zoe?


        18. I distinguish between poetic and scientific usage. But if you cherrypick that one sentence and miss the whole point of the article, then I suppose I could’ve worded it better. But if that’s all you have to say, I think we’re done here.


        19. CO2 is used in double glazing, where sound insulation properties are required, SF6 is also used. It’s the density of the gas which matters, not its radiative properties.


        20. Ulric,

          The study found more heat transfer through a double paned window with a CO2 fill than with ordinary air. If the bottom pane represents Earth’s surface and the top pane the tropopause then….. ?

          But my understanding is that radiative heat transfer was already greatly reduced because the bottom pane had a low emissivity coating. This negated much of CO2’s value.

          At the same time the greater weight on convective transfer exploited CO2’s weakness – low viscosity.

          That’s the best I can come up with anyway. Thoughts?


  2. My computational skills do not match yours! I like your style too, always curious and with the skills to match the curiosity. Thanks and keep it up, as a skeptic I find much of what you uncover just adds to the uncertainty sometimes , like this one, counter to the prevailing theory.

    Liked by 1 person

  3. Hi Zoe,

    Great work as always. I remember looking at the old CRES CO2 data, that changed by the seasons, looking for CO2 hot spots around major capital cities and Australia’s coal fired power stations.

    I visually couldn’t find any pattern in fact just the opposite. I was wondering if you could do a similar analysis with CO2 data and major cities and coal fired power plants?

    Thank you.


    1. Thank you.
      I tried to do this, but only found some USA data. If you can find the proper data, I can analyze it. I will analyze it. Best regards, -Z


  4. Actually, the greenhouse effect is only said to work if you have an atmosphere with a lapse rate. No lapse rate, no greenhouse effect. William Happer et al. “Methane and Climate”, 2019. I believe the effect (if existent) is only theoretical. The myriad of other processes in the atmosphere will respond, and nobody can know that the net result is warming. Only guesswork. From 1941 to 1970 the temperature dropped along with exponentially rising co2-concentration.

    Liked by 1 person

    1. Alarmist Satire: “WWII produced so much black smoke that the sun was blocked for decades, so despite CO2 always warming, we experienced a cooling.” lol


    1. Jarle

      Nice article (mostly). Time will tell if the lukewarmers are right. I don’t really care one way or another, other than my ego would be bruised for betting against them.
      The science is interesting regardless.

      For Zoe, who likes to pretend that people from the “hard” sciences don’t believe in the GHE,

      “So, the wavelength of thermal radiation is 10 to 20 times longer than the wavelengths of sunlight. It turns out that the sun’s energy can get through the Earth’s atmosphere very easily. So essentially all sunlight or at least 90 percent, if there are no clouds, gets to the surface and warms it. But radiation cooling of the surface is less efficient because various greenhouse gases (most importantly water vapor, which is shown as the third panel down, and CO2, which is the fourth panel down) intercept a lot of that radiation and keep it from freely escaping to space. This keeps Earth’s surface temperature warmer than it would be (by about 20 or 30 degrees). The Earth would be an ice cube if it were not for water vapor and CO2.”


      1. It doesn’t matter what “hard” scientists “believe” about GHE, unless their evidence comes from science. Have “hard” scientists ever shown GHE via hard science? Still waiting …


        1. When it comes to outgoing longwave radiation and the “co2-ditch” as a proof, here is what Claes Johnson has to say:

          “The spectrum in the ditch with it’s flat bottom at 220 K, is constructed in a two-step procedure:

          1. Identification of a line spectrum from the presence of CO2 (Hitran, interferometer).

          2. Translation of the line spectrum to a continuous spectrum defining atmospheric irradiance (Modtran modeling).

          Step 1 can be performed as direct physical measurement (as shown in Radiation of Solid vs Gas), while Step 2 is based on a model of spectral line broadening which determines the irradiance supposedly resulting from the line spectrum.

          The scientific basis of CO2 alarmism thus rests on a mathematical model of broadening of the line spectrum of the trace gas CO2 and the mathematical model is very simple as it describes the line broadening in terms of the parameter p x L where p is the partial pressure of CO2 and L is the path length.

          Classical experiments by Hottel and Leckner concern the case p ~ 1 bar and L ~ 1 m, while for atmospheric CO2 warming p ~ 0.0001 bar and L ~ 10000 m, to which direct extrapolation may not be possible.

          The net result is that CO2 alarmism is based on a simple mathematical model without direct experimental support and thus is ready to collapse once this fact becomes known and acknowledged, first by climate skeptics then by the general public and finally by CO2 alarmists.”

          As you say, Zoe – no direct experimental support, or “hard science”, if we are to believe Claes Johnson.

          Liked by 1 person

        2. Jarle,

          Ideally there would be two identical time series except for different CO2 values:

          1) 1880 – 2020 with CO2 remaining constant at 280 ppm

          2) 1880 – 2020 with CO2 rising to 415 ppm.

          Then we could make a direct comparison and there would be little room for debate.

          Obviously not possible so we’re stuck with models and uncertainty. It is what it is.

          OTOH, a recent study claims to have found direct evidence. Behind a paywall and probably over my head anyway so I have no comment.



    1. You are using raw data? I’m using adjusted data. I checked a random sample of 100 latitudes+longitudes, over and over again, and found warming in nearly all.


  5. Oh wow, you’re in Roswell GA? (I noticed the mention of Azalea Park in the illustration caption) I live in Woodstock, often go to Roswell for dinner. Small world! (I dunno why, I somehow thought you were overseas somewhere.)

    Huh, I’m not sure I would have expected to see higher temps around power plants. I’d think that prevailing winds would just blow the CO2 elsewhere, so there’d be little local effect from that. – OTOH, I could imagine other effects, depending on where the weather station was located. I’d tend to think that a few gigawatts of heat energy would do something.

    A friend told me one time that I could cook up a theory to match any set of data 😉 Here’s one mechanism that occurs to me might have some sort of impact: If a plant is cranking out thousands of megawatts of thermal energy, I’d expect that to create updrafts – which in turn would tend to create a net surface-level airflow towards the plant, drawing from surrounding areas. Depending on what’s around the plant, this could be almost an “anti-UHI” effect. (Not that it would make the immediate area cooler, just that it would tend to hide any local temp increases due to the plant itself.)

    I absolutely love what you do; it’s what should be the standard for *any* scientific discussion: “Here’s the data I looked at and where you can find it for yourself, here’s exactly how I analyzed it (the actual code used), and here’s my results.” It’s the only honest way to do analytical science; it’s depressing that you stand out as such an anomaly for doing it :-/

    Liked by 1 person

    1. Nah, I live further east, in Gwinnett. I drive to jog in different places.

      My thinking was that power plants exhaust high concentration of CO2, then convection and wind disperses it. If high concentration doesn’t warm, why would lower concentration?

      Thank you for the compliments. Yes, why don’t more people, especially the PAID ones do science so openly?


      1. Ah; we’re in the SE corner of Cherokee, 10-15 min the other side of Roswell from you.

        Hah, maybe the answer as to why the PAID researchers don’t follow your example is *because* they’re paid. You don’t want anyone upsetting the applecart and discrediting you to the funding agencies. (Not that your peer-review buddies would be so rude as to point something like that out, though >:-0)

        Liked by 1 person

  6. Sorry for the multiple posts, but it occurred to me wonder how the order of magnitude of plant thermal output would compare to insolation. Maybe not such a great perturbation as I was thinking, OTOH, it would be a very concentrated source vs the sun. Here’s some numbers:

    Average daily insolation here in GA is ~~5 kWh/m^2
    (From here, although it assumes a southern tilt for a solar panel: http://www.solarpanelsplus.com/industry-professionals/insolation-charts/)

    3 MW plant = 72 MM kWh/day, about 14.4MM x average daily insolation for 1 meter.

    1 kM x 1 kM = 10e6 m^2

    So the 3MW plant is putting out about 14.4 square kilometers of equivalent incoming solar energy.

    Errors/confounding factors:
    – Insolation number is based on tilted panels, so actual insolation normal to the surface would be lower
    – Solar input isn’t constant; none at night, lots at mid-day
    – Actual thermal output of the plant will be quite a bit more than the electrical output rating.

    Overall, I’m not sure how big an effect there’ll be, but I would think there’d be some degree of updraft generated by the plant that’d tend to contribute to pulling in air from the surrounding area.

    – And this is all a rabbit-hole with little point to it 🙂 Interesting to think about though, and your article prompted me to idle away 30 minutes or so thinking about it. Thanks! (I think :-0)


    1. There’s no evidence I can find that would suggest that a 4 km^2 grid (what my source uses) with a coal plant has a higher average wind speed than other places.


      1. Yeah, I doubt it’d be a big enough effect to show up in wind speed, especially on a 4km^2 grid. It’d be more of a subtle thing, and almost certainly quite local to the plant. (I wasn’t thinking of the grid size you’d probably be working with; most plants would represent only a fraction of a grid square, and any thermally-induced circulation would be a very minor delta over what was going on naturally in the first place.)

        Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: