Surface thermal balance, Central Southern England

Posted: April 2, 2014 by tchannon in Clouds, Measurement, radiative theory, weather

This is first art from access to the Chilbolton Observatory datasets.

Note: autoscaling has been allowed.

Data source Chilbolton Observatory and BADC archive [Ref1]

For comparison data acquired via the Chilbolton web site as part of an ongoing work [Ref2]. The CNR4 data confirms the computation used to approximate long wave heat flow, air temperature is used as a proxy for instrument body temperature.

These two days represent:

  • hot, clear summer day
  • typical English holiday weather warming nicely toward evening, ideal for hill walking, gardening and that favourite English pastime of complaining about the weather

I’ve very recently gained login to the BADC Chilbolton data archive. As I recall this previously prevented me so I assume something has changed at BADC.

Access reveals good planning by Chilbolton staff who installed a Kipp and Zonen CNR4 instrument for bridging shortly before the pyrgeometer and pyraneometer were removed for bi-annual service, away for several months. No mention of this in public so it came as a surprise.

CNR4 is a Net instrument with two pairs of each which provides full data for this kind of measurement.

These instruments are not as good as claimed WMO First Class would perhaps lead people to believe but provide a sensible idea on what is going on. (5% accuracy) There are various problems. I think of them as old fashioned based on technology from the mid 1800s instead of later electrical balance technology. I suppose the primary advantage is simplicity.

None of these instruments directly measure thermal flux, photons or EM, they are proxies. Kipp and Zonen published documents for the long wave instruments showing the usage of external maths for faking a real flux reading from instrument body temperature and that used by the Chilbolton instruments, no reflection on staff who would be out on a limb if they act unilaterally. This has no effect on so called “net” readings, added on flux cancels out.

If the claimed flux is real then throw a blanket over the instrument and harvest free electricity!

Figure 1 shows a defect highlighted by the computed albedo. I’d noticed the asymmetry in insolation. When I computed albedo my suspicion was confirmed, the ground facing instruments do not have a clear view of the ground there is at least a shadow late afternoon, hence ground absorbsion reads higher than it is. The pyrgeometer will suffer the same.

Nevertheless this is the only set of UK based instruments I have found where access is allowed. This must be highly commended.


  • download netcdf format data via FTP
  • decode and parse files using netcdf dump and a newly created parser library
  • if necessary decimate data and output for further use such as plotting

I’ll leave this article without more comment, wanted to get it published to clear the way for some important posts.

Chilbolton Observatory home page, links to Weather web and BADC.

BADC data is as NETCDF and for CNR4 is one second sampled data. I’ve decimated for plotting to 10 min sampling using signal processing.

Site location Bing map switch to aerial photo, cabin on stilts with meteorological and other sensors is left (west) side of site next to a field. Shadows can be seen. Zoom out and you can see the WWII airbase marks, aircraft dispersal stands.


This is produced from web published plots for the previous day and is now automated, critically via a new binary library which does OCR on the autoscaled plot Y-axis with regression to the true axis and then trace decoding.

Post by Tim

  1. Mike Flynn says:

    Did you really mean to to say BAD data, rather than BADC data? It would certainly seem so, although it would not be unknown for the Government to bury such a wonderful source of free energy.

    They would no longer be able to wield the long sword of warming fear, and have no excuse to redistribute wealth from the pockets of the poor into the pockets of the too clever by half recipients of an endless flow of Government subsidies, assistance, easing, kickstarting, pump priming, job creation and all the rest.

    Oh well, the promise of free energy will probably turn out to be more dreams of mathematically challenged jugglers, and itinerant hockey stick Mannufacturers.

    Live well and prosper,

    Mike Flynn.

  2. tchannon says:

    Added a link which I forgot, humour noted.

    I probably ought to link to the explanation, here we are

  3. Konrad says:

    It would be great if you could find a clear sunny day like day 1, but with stronger and more variable wind speed, so the effect (if any) on net IR out could be seen.

  4. Konrad says:

    only just read your link on Pyrgeometers, the instrument is looking the wrong way and cannot detect the actual IR from the ground. So it won’t see the correlation I was hoping for.

  5. tchannon says:

    Not so Konrad the CNR4 comprises two pairs, pyranometers (short wave) and pyrgeometers (long wave), one of a pair point up and one down, hence radiation balance.

    I’ve only decoded what you see but I have all the data here, 500MB compressed, GB on extraction.

    Doing anything with that en-masse means I have to write code to do it. Extract and decimate to a manageable size. Getting lucky is rare, things go wrong such as unexpected or corrupt data. I’ve already had to sidestep bad data, outages are not marked by the defined markers! Expect anything and then some.

    Any plotting I have to do, more code.


    Perhaps the largest problem is England, sunny days are rare, even then humidity burning off usually shows. Not without reason the CIA Worldbook describes us as “overcast more than 50% of the time”

    Water dominates. Today (yesterday) seemed a clear sunny day after slight rain, felt more like a winter day, chilly outside. The following appeared an hour or so ago.

    You can see there the effect of cloud on IR. Was sunny but not as clear as it might have seemed.

  6. Konrad says:

    My bad,
    should have web searched “Kipp and Zonen CNR4” 😦

    All that’s needed now is a clear day with highly variable wind.

    What I’m looking for is variable non-radiative transports (wind/conduction/evaporation) causing variation in radiation from surfaces (variation in the types 3 of energy transport into the atmosphere). I have detected a definite response in sun exposed black plate collectors that are also wind exposed using an IR thermometer, but I suspect the effect would be very different in materials able to cool by transpiration or evaporation, such as soil or vegetation.

    A further question – Does the station have an actual ground based soil temp thermometer?

  7. Joe Public says:

    Hi Tim

    “These instruments are not as good as claimed WMO First Class would perhaps lead people to believe but provide a sensible idea on what is going on. (5% accuracy) ”

    Accuracy or inaccuracy?

  8. vukcevic says:

    On CET (somewhat OT, but may be of interest)
    In the early November I commented :
    “Daily CET max & min temperatures suggest that forthcoming winter may not be as cold as the last one”
    while the Met Office forecast was of severe cold, NHS was getting ready for increased pressure, power companies were increasing fuel prices, there was talk of the gas supplies running out, etc..
    Now winter is well over, spring is in the air, question is: how did we do in the CET area?
    Nicely, thank you very much, winter on average was about 1.5 C above its 20 year average, and about 2-2.5C above the previous winter’s temps, see here:.
    No NHS crisis, gas supplies were adequate and the fuel prices are edging down.

  9. tchannon says:

    Joe Public, clarifying what I meant.

    The manufacturer states the instrument meets WMO standard First Class. I am stating that an official body using a phrase such as First Class leads casual consumers to assume very good means very good in absolute terms.

    I disagree with First Class meaning very good, an opinion from a position of technical experience.

    The field involved is Metrology, a word missing from many dictionaries.

  10. tchannon says:

    Konrad, the wind figures are in metres/sec, not that calm. The instruments generally are mounted in non-standard positions, this is not a standard weather station, which must be considered as a general context. We have what we have.

    Weather generally doesn’t provide conditions to order so you won’t find variable wind direction with fixed other conditions except near dead calm when local effects will be strong, and given the surroundings humidity is always involved.

    Can you explain what you would like to try and do?

  11. One of my pet peeves is the extent to which the public is excluded from accessing information produced by scientists funded via our tax dollars.

    One can understand secrecy when it comes to military research but how can anyone justify excluding the public from weather station data?