Chilbolton Observatory unofficial dataset, numeric and plots

Posted: April 21, 2013 by tchannon in Analysis, atmosphere, climate, Clouds, Dataset, Measurement, methodology, Natural Variation, Surfacestation, weather, wind

Example data. Lidar is useful in identifying cloud layers, clear sky, it cannot though see multiple layers, where radar images if available are more useful. Click to open full size.

[UPDATE 2013/05/05,  archive release with corrections] An unofficial high resolution dataset has been derived from web published graphs running from 16th February 2013 to 20th April 2013 (less 23rd Feb), just over 60 days of data at about 2.3 minutes resolution for about 23 hours a day. The data includes atmospheric water parameters and thermal radiation parameters, all measured from surface level. Archive period includes the exceptionally cold March 2013. The weather hereabouts was cold but unlike most of the rest of the UK there was little snow, nor was the temperature very low. I live 30km from Chilbolton Observatory so the weather here is very similar, I can sometimes add knowledge, hence my particular interest. Yesterday for example, 20th April 2013 was the first really sunny day of the year, with just a few fleeting clouds middle of the day. We rarely get true clear sky here, what I tend to call mountain weather, so it was sunny but not blue clear like that. Early afternoon and then after dark I pointed a spot IR thermometer roughly at zenith, both cases giving a reading of -31C. Caveat: this data is derived from public web site plots, is unofficial, should not be relied upon but is supplied in good faith. Please report any errors found but do not trouble nor criticise Chilbolton Observatory. Any usage, please cite Chilbolton Observatory as the data source with caveat on unofficial. This might assist continued funding of one of the few cloud research stations in the world. Data I do not intend to provide regular updates.

This comprises PKZIP archives containing copies of the web images and where practical derived numeric data, plus a plot of some data. The numeric datafile is tab delimited. It can be imported to spreadsheet or used directly by gnuplot to create plots. More details are in the readme files inside each archive. I hope it is portable, accessible and self explanatory. Two perhaps contentious computed data are added.

  1. Assumed blackbody emission at air temperature, but this is not ground temperature nor pyrgeometer body temperature.
  2. Subtract of inbound and (1) giving a notional net radiation in Watt sqm, omitting solar input.

Back calculating equivalent temperature for pyrgeometer data could be added, similar for solar. A matter for discussion. Additional data Some additional data for the area can be accessed.

Known issues

  • Derived data and plot of wind direction does not yet handle data wrapping around north. I know how, is part coded, have to finish it.
  • 35 GHz radar plots are included but are rarely valid for the day in question, beware. An unresolved issue, not sure what to do.
  • Thumbnail sky images are of little use. Should they be removed?
  • No independent checking for errors nor detailed checks by me. Please report any you find.
  • Plots do not include units. The practicality to be investigated.
  • Computing and adding in wind gust, perhaps minima is under consideration.

Code There is a fair amount of software code involved, currently not a pretty sight, never is when you don’t really know where the end result will lead.


Chilbolton Observatory weather facility RAL Space STFC Rutherford Appleton Laboratory Harwell Oxford Didcot OX11 0QX Tel: +44 (0)1235 445 000 Image

Talkshop links to past items involving Chilbolton, let me know if I have missed any.

Please only comment on this article, not on what the data means. New thread provided link here.

Article by Tim Channon, co-moderator and contributor

  1. […] Chilbolton Observatory unofficial dataset, numeric and plots […]

  2. A C Osborn says:

    What is air_SB_Unity?
    Why does it track the Air Temp, are they related?

  3. tchannon says:

    Simply SB applied to air temperature, which gives an approximation of thermal emission from the ground.

    One of the major problems with all these installations is the lack of hard information on pyrgeometer body temperature and highly critically the actual emission temperature of the ground, trees, vegetation, everything. In my view this is all over the place on known. Even a downward pointing pyrgeometer for differential measurement will still get it wrong except over a huge area of flat uniform expanse.

    Emissivity is also unknown but at these wavelengths most things are close to unity.

    I did think a little about how valid it was to then combine this computed emission with the pyrgeometer figure given acceptance and emissivity are unknown. However, these things almost always cancel to unity so it probably isn’t too unsafe doing the math.

    It’s still incomplete, missing solar. Add that in and an apparently silly result appears, or does it?

    There will be to-space radiation from what is insulating as well reflecting back down. Then there is physical removal via water phase change and via air movement.

    I think this means we know very little of the whole.

  4. tchannon says:

    I’ve found some mistakes, several png and data are missing, sorted out but not uploaded yet. Two more days data there too.

  5. tchannon says:

    Okay, I’ve made a lot of software progress, workable management of the Chilbolton data. This is kind of version control, each day has to be eyeballed and in variable scaled plots configured for the day, just edit a few numbers. Now I can invoke software which will look and figure all the changes needed, files created or shunted around, including any edits of older stuff.

    More to do yet. Then I can start to look at what we have.

    I hate this sort of bitty stuff.

    Wind direction processing works most of the time. Not a surprise, direction goes nuts if wind speed is low. It must be an ultrasonic direction gauge, or similar.
    Why bother? It enables conversion to a complex number, and that means an XY vector plot can be done. I’ll post some sometime, gives a better feel for how the wind shifts around.

    And yay, a new data set arrived, few minutes and all processed. :-).

    Probably take a break from this and do something else. A little at a time.

  6. tallbloke says:

    Tim: A terrific and valuable effort, well done!

  7. […] The data archive is here. (143k zip)  I’ll upload a new version of the entire archive when I am ready, main data article here. […]