Taking the temperature

I have a digital thermometer thingy bought for £10 from Lidl. It has an internal digital thermometer and a separate, battery-powered temperature sensor connected via a short range radio link.

I hang the external thermometer on a rusty nail in the brick outbuilding attached to the garage. As the place has a pitched, tiled roof, one smallish west-facing window glazed with Stippolyte and isn’t heated, I reckon it gives me a reasonable outside temperature measurement without being exposed to the elements, including direct sunlight. It agrees well with the Met Office anyway.

It’s not a complex device, but I get max and min recordings inside the house and outside, in the outbuilding.

I don’t record the temperatures in a log or anything nerdy like that, I’m just interested in how the temperature wanders up and down as seasons change. I’ve no idea how accurate my thermometer is apart from the crude Met Office checks, but as a former environmental scientist, I don’t really care – approximate will do. I’m not going in for NPL calibration for something so casual.

I also have an outside temperature display in the car and again I don’t know how accurate it is, but I suspect it’s near enough. It gives an anxious little ping and shows a snowflake symbol when the temperature outside the car drops to 4 °C or below. Presumably it’s warning me about the potential for icy roads.

Anyway, as I drive around the lanes of Derbyshire, particularly on early mornings when we’re setting off for a walk, I tend to notice how the temperature changes as we drive through the countryside. Warmer at low level and cooler if you climb into the hills, as you’d expect.

Sometimes it’s cooler in the hollows, especially during the winter months where conditions are right for inversions. The air temperature can easily change by 3°C over a short distance. In fact I’ve seen changes as high as 5°C over a few miles, driving out of a valley up into the hills.

So how would I measure the temperature of Derbyshire? Where would I put the thermometers and how many of them would I need? What would I measure – daily maximum and minimum or continuous recordings? How would I account for the differences between hills and valleys? How high off the ground should my thermometers be, or should I bury some of them to measure the temperature of the ground? Surely ground temperature is important?

How would I calibrate the thermometers and check that all readings had been taken correctly? How would I deal with faulty thermometers? How would I know when they were giving biased readings? How would I check for calibration drift?

And given I could sort out all of this, how would I check my network of Derbyshire thermometers? How would I compare them to another network which doesn’t actually exist? And how would I build a rationale to cope with the obvious fact that Derbyshire doesn’t have a temperature, that it all depends on how you measure it?

Because I’d need that rationale, wouldn’t I?

I’m not making an obscure scientific point here. I’m just pointing out some well-known problems behind the apparently simple issue of taking temperature measurements. Environmental field measurements, including temperature, used to be an important aspect of my job. I know how difficult it is – even with simple measures such as temperature.

I’m also interested in how climate moonbats (is that the correct term?) claim to measure global temperatures to a tenth of a degree, because I know it’s impossible. And by the way, the BBC won’t tell you this, but there is no such thing as a global temperature. That is to say, it could only ever be a convention, yet still there is no agreement as to what that convention might be.

Environmental temperatures are conventions.

20 comments for “Taking the temperature

  1. SteveW
    September 25, 2012 at 9:21 am

    Just to reinforce the idea, I was driving from Lancashire up to the Lakes to deliver a talk/seminar a couple of years ago, approaching the Keswick turn off on the M6 (J40?) the temperature on the car’s thermometer was reading -8C, when I got to the top of the slip road it had dropped to -12C.
    Which temperature would you smear across a grid cell for modelling purposes?

    • September 25, 2012 at 5:17 pm

      Often it seems to be the temperature recorded at the nearest airport they use to smear across the hills and valleys.

      • SteveW
        September 25, 2012 at 9:20 pm

        True, but to be fair that is probably the best way to ensure a marked increase in daily minimum temperature and why let such a lovely collection of heat sinks go to waste?

  2. September 25, 2012 at 9:44 am

    Anyway, as I drive around the lanes of Derbyshire, particularly on early mornings when we’re setting off for a walk, I tend to notice how the temperature changes as we drive through the countryside. Warmer at low level and cooler if you climb into the hills, as you’d expect.

    Sun also rises, AKH.

    • September 25, 2012 at 5:27 pm

      It does – night and day temperature changes are far greater than the supposed global temperature rise of the past century.

  3. Greg Tingey
    September 25, 2012 at 12:37 pm

    The “global” temperatures are AVERAGES taken over many hundreds or even thousands of readings ……

    • SteveW
      September 25, 2012 at 3:27 pm

      With no account taken of the relative humidity at the time the individual readings were taken. The effect this has on the amount of energy involved cannot be overstated, which is why an ‘average global temperature’ is a relatively meaningless number.

      • September 25, 2012 at 5:19 pm

        Good point – temperature is not energy.

  4. microdave
    September 25, 2012 at 3:18 pm

    I trust you are aware of Anthony Watts “Surface Stations” project?
    http://www.surfacestations.org/

    You are right to be concerned about the difficulties involved with taking global temperatures…

  5. Radical Rodent
    September 25, 2012 at 5:12 pm

    Yes, Greg. I think you have entirely missed the point of AKH’s post – in the tiny portion of the globe that is Derbyshire, there can be quite a wide range of temperatures; what is the location to get a suitably indicative temperature for the area? (Any area – try putting a few thermometers around your front room; even there, you will find differences, albeit small. Then decide which is the “accurate” thermometer…) At what altitude in the area should it be – the highest, the lowest, the average or the median? Location decided, at what height above the ground should the thermometer be? How are any thermometers used to be calibrated? How often should the calibration be checked?

    I think the point that is trying to be made is that, as yet, there are not sufficient stations collecting this data, and what we already have may be subject to many other factors affecting their reliability to really be able to make such sweeping claims as are being made by the AGW/ACC alarmists.

    • September 25, 2012 at 5:25 pm

      “I think the point that is trying to be made is that, as yet, there are not sufficient stations collecting this data, and what we already have may be subject to many other factors affecting their reliability to really be able to make such sweeping claims as are being made by the AGW/ACC alarmists.”

      Yes I am – in as simple a way as I can.

    • microdave
      September 25, 2012 at 9:05 pm

      “As yet, there are not sufficient stations collecting this data”

      Further to my earlier comment, on a global scale there ARE many more than are currently used to provide data for the 3 main temperature series. Anthony Watts work has queried why some 6000 stations have been whittled down to less than 1500, and “corrections” applied to those in order to fill the gaps thus created!

      And it just so happens that lots of these remaining sites are the ones suffering from the “Urban Heat Island” effect. This has resulted in the situation Dave_G notes below.

      • September 26, 2012 at 6:57 pm

        “corrections” applied to those in order to fill the gaps thus created!

        Exactly. Who else “corrects” data before doing the stats?

  6. Dave_G
    September 25, 2012 at 8:10 pm

    It doesn’t really matter (if you are an AGW proponent) – so long as the temperature readings show a steady upward trend, the reading is ‘correct’.

  7. Rossa
    September 26, 2012 at 8:21 am

    The difference between weather and climate often gets confused by supporters of AGW. I live half an hour or so from Leeds and it can be a lovely, dry, sunny day there and snowing and below zero here at home which is a semi-rural location. The locations are in different valleys and I live in the lea of a bluff in the foothills of the Pennines so we’re sheltered from the worst of the bad weather on open ground (Moors). Driving from the city over the moors and down into our valley the difference can be as much as 8-10 degrees.

    I drove through a hailstorm a couple of weeks ago and saw the outside temp gauge on my car drop from 15 to 7c in seconds. Completely localised as once through it the temp rose again.

    That’s the weather for you.

    • September 26, 2012 at 6:55 pm

      “That’s the weather for you.”

      And this phenomenal complexity is important – it affects local agriculture for one thing and in practical terms, that’s what matters.
      8-10 degrees is a big difference, but I can see how it comes about.

  8. Furor Teutonicus
    September 26, 2012 at 1:32 pm

    XX Environmental temperatures are conventions. XX

    Or convections, whichever.

    • September 26, 2012 at 6:51 pm

      😀

  9. Mark in Mayenne
    September 27, 2012 at 7:07 am

    Oh no! I never realised.
    The temperature of my house is a construct that depends on which room I am in, what floor I am on, how wndy it is, and whether or not I have sealed the leaky window frame.

    How will I ever know if I need to light the fire, turn on the heating, or open the window to let a cool breeze in?

Comments are closed.