These are the news starting 01 Jan 2008
Click here for the news of the previous years.
|31 Dec 08||
This will be the last news for the year. The meteoLCD crew
wishes you a very good New Year! May climate Zen
wisdom be with you... Please remember that the 7days
graphs will be corrupt during the coming 7 days, as the logger
file does not include the year in the date-stamp, and this makes
life difficult for our old trusty Gnuplot.
There was a hickup with the Automate software yesterday. No data have been lost.
|18 Dec 08||Uploaded nov08.dat file into the data archive. Swapped usual logo for traditional Xmas logo (designed by Colette Heirendt).|
|12 Dec 08||Radioactivity sensor healthy again. Thanks for help to Claude Baumann, the developer and builder of the outstanding serial to voltage interface used for this sensor.|
|08 Dec 08||
NOx sensor bad since last calibration on 5th. Dec. I am not sure
if I will be able to repair this...
Gamma counter not online due to broken down computer. I am working on this.
|27 Nov 08||Since a couple of days the radioactive sensor is broken. Will be looked after asap.|
|SAR Grand-Duke Henri visited the Lycée Classique de Diekirch today, and also spent some time in the meteoLCD facilities. Pictures will be posted when available on this highest level visit since the start meteoLCD. The Grand Duke was in company of the President of the Chamber of Deputies Lucien Weiler, the Minister Marie-Josée Jacobs and a couple of other higher ranked personalities. Look here.|
|12 Nov 08||I participated today to Henvi2008, the International Conference on Health Aspects of Indoor and Outdoor Air Pollution, Luxembourg, and presented a poster session on 10 years ozone measurements at meteoLCD. See at the papers section.|
|09 Nov 08||Uploaded oct08.dat file to data archive|
|31 Oct 08||Recalibration of CO2 and NOx sensor.|
|26 Oct 08||I added a Bezier smoothing to the plot of the total ozone column. Easy to do with new Gnuplot 4.2|
|24 Oct 08||
1. Maintenance by Envitec
2. We installed a new Zero Air generator to correct a problem with the API CO2 sensor (sensor reacts on ambient room CO2 concentration, which is a problem if work is going on in the meteoLCD facilities. The new generator will also deliver zero air for the periodic zero calibrations of the O3 and NOx sensors.
|22 Oct 08||
1. O3 pump repaired
2. There is a memory leakage problem with the lightning computer.
|19 Oct 08||Ozone sensor malfunction... probably pump defect. We will receive a new pump next Friday.|
|02 Oct 08||
1. Uploaded sep08.dat file to data archive
2. Rain gauge was blocked by bird droppings. Cleaned up the 03Oct08. Will try to insert good values of the Davis backup station in the oct08.dat file, when available.
3. The proxy computer driving the webcam is out of order. Will be repaired asap (but this is not an imergency!)
4. The afternoon CO2 peaks at 29 and 30 Sep. 2008 should be discarded. They are caused by human respiration. A supplementary zero-air device will be installed soon, to correct this problem that is a regrettable feature of the new API 360EM CO2 sensor (the optical block is not gas-proof to ambient air).
|28 Sep 08||I started today a blog at http://meteolcd.wordpress.com. It's just a try... if there is some interest, it will live on, else it will vanish.|
|16 Sep 08||Recalibration and maintenance of CO2 sensor|
|15 Sep 08||The main lightning computer had a fatal breakdown the 12th Sep. The whole system has to be reinstalled on a new computer. I am working on it...|
|14 Sep 08||Uploaded Aug08 data to the archive. The NO/NO2 data are wrongly given in PPB instead in ug/m3 for the last 3 weeks. This will be more precisely commented in the file header asap.|
|10 Sep 08||
1.The pump of the ozone sensor stopped operating in the evening
of 7th Sep. Has been replaced today at 14:45.
2. List of ozone column measurements done in Aug08 has been completed.
|08 Aug 08||Maintenance and recalibration work yesterday and today.|
|05 Aug 08||
The Dobson measurements are made by Marcel Thilmany when I am
abroad. He uses instrument #3012 from his home at Insenborn,
some 15 km from Diekirch, practically same coordinates, but at a
slightly higher altitude (340m asl instead of 218m). A prior
comparison done at Diekirch with the 2 instruments suggests a
multiplier of 1.044 to apply to the #3012 instrument to make
readings similar to #5375 (the one in use at Diekirch). The
Dobson table holds these adjusted readings.
The following measurements have been made at Insenborn:
July08: 16, 23, 25,27
|04 Aug 08||
1. The live data transmission stopped the 23th July, due to a
problem with our main Catalyst 6509 switch. As I was abroad, the
problem could not be repaired, and I would apologize for this
inconvenience. As usual no data are lost; the July data will be
uploaded asap to the archive.
2. July data upload done (17:00 UTC)
|12 Jul 08||
1. I replaced the old faithful Win98 computer servicing the
logger... and this went wrong. The new mini machine is not
reliable and hang during the aftrenoon. So I went back today to
the trusted workhorse. I had to restart the logger 2 times, but
there is no data loss.
2. Maintenance and recalibration work on all gas sensors. Also some further housekeeping.
|05 Jul 08||Uploaded Jun08 data to archive.|
|04 Jul 08||Maintenance work by Envitec.|
|03 Jul 08||
1. The logger-computer had a breakdown yesterday (02Jul08). This
has been repaired. No data loss.
2. We had a severe thunderstorm yesterday evening. For reasons still unknown, several evening hours are missing in the movie file 20080702.mpg of the lightning page. I'm doing detective work to find out what went wrong.
3. Lightning/2000 software has been updated from 3.x to 4.6.3 version (12:16 UTC)
4. The detective work is successfull (with the help of Florence Massen): A movie is only made if the stroke-count read from a text file is > 60. If this number is very high (> 1000) it is written in US comma-separated fashion: 16143 -> 16,143. This broke the string to number conversion in the Perl script making the movie and hours with huge stroke counts were left out. The time given in the movie screens is UTC !!!
|18 Jun 08||
1. The afternoon CO2 peaks of today and yesterday are an
artefact: the sample-filter enclosure went loose, and due to
human presence. ambient CO2 levels were high in the
intrumentation closet. This will be corrected (by
interpolation) in the monthly data file of June.
2. Maintenance and recalibration work on ozone sensor
|11 Jun 08||Calibration and maintenance of API CO2 sensor.|
|08Jun 08||Uploaded May08 data to archive|
|07 Jun 08||I managed to repair the NO/NO2 sensor this morning. It is working again since 08:00.|
|02 Jun 08||
1. The NO/NO2 sensor has a malfunction since 2 days. I am unable
to find the fault up until now.
2. There was also a problem with the lightning sensor: the screen was not correctly updated anymore. Seems to work again (18:50 UTC)
|24 May 08||
1. Ozone pump broke down yesterday. Replaced by a spare this
2. Maintenance and calibration work on Ozone and CO2 sensors.
3. Cleaning of main inlet mesh filter; it was very grimy. Watch for possible higher O3 readings.
|15 May 08||Just checked the "movie" part of the lightning facility, and horror... all movies are empty! Probably comes from a missing mpeg2encode as this package is (for obscure license reasons) not included in the ImageMagick package. This means that since the upgrade to openSuSE 10.3 this package is missing, and movies are empty. Will be fixed asap.|
|03 May 08||
1. Most (possibly all?) problems with the new API 360EM CO2
sensor solved! New calibrations this morning.
2. The WBGT sensor has been reactivated. May be this will help with getting fine weather!
|30 Apr 08||Life definitively isn't easy... There are serious teething problems with the new API CO2 sensor. I checked all electrical connections and parameters, made several analog output and gaz ( zero and span) calibrations... The main problem are the spikes appearing about every 3 hours, and an irritating "analog calib." warning I am unable to understand.|
|28 Apr 08||
1. New API 360EM CO2 sensor on-line (16:00 UTC). The API CO2
readings are on channel 11, the same place occupied by the
old MIR9000 CO2 sensor. The MIR readings are now on channel 8
(not displayed in the graphics, but available in the data
series). It will be kept running for about a month, than go to
rest and be used as an emergency replacement. A recalibration
check of the API will be made asap.
2. Due to the needed changes in the logger configuration, all logger data have been downloaded, the logger's memory erased and restarted. The complete file with all raw logger data is all_01Mar08to28Apr08.dat.
|25 Apr 08||
1. Maintenance by Envitec. The CO2 readings from 23 March to
today are probably slightly too high
2. Installation and first run of the new Teledyne-Api 360EM CO2 sensor. Will be included in the near-live data asap (probably next week).
|05 Apr 08||Uploaded mar08.dat|
|23 Mar 08||Problem with the calibration of the CO2 sensor. Rechecked today and adjusted calibration factor. (14:30)|
|22 Mar 08||
1, There was a problem with the live lightning page since
29Feb08: the screen was not correctly updated due to a change in
the ftp daemon. Corrected today.
2. Recalibrated CO2 sensor; maintenance work on CO2 and O3 sensors.
|18 Mar 08||Due to a blocked sampling tube CO2 readings were bad from 13/08 to today 08:00.|
|01 Mar 03||
1. The leap-year corruption is severe. I am struggling to put some
order into the data files... there may be some data
2. February 2008 data uploaded to archive. The infamous 29th Feb. is ok! No data loss.
|29 Feb 08||
1. meteo01 is up again with a new system. Some problems with the
lightning pages remain
2. The datalogger can not handle leap years, so there will be no correct graphs today. I will do some housekeeping tomorrow morning to put things again in order.
|27 Feb 08||
1. NOx sensor ok now.
2. Infection of the meteo01.lcd.lu is so severe to command a new system install. Server will be offline for some time (starting 10:30 UTC today). This has repercussions on the lightning pages (no problems for the moment) and possibly on a couple of other running scripts. Most normal operations will be hopefully unaffected!
|26 Feb 08||
1. recalibration and maintenance. There is a problem with the
AC31M NOx sensor
2. our meteo01.lcd.lu server is being hacked. Fight back with a quick kludge
|22 Feb 08||I have added just for fun to the page with the live data a graph with the monthly mean global (land + ocean) temperature anomalies as given by NOAA. The big smoothing is NOT a moving average, but a Bezier smoothing which will always end at the last point. So do not be fooled! The graph will be updated automatically.|
|21 Feb 08||
A couple of problems during the last 3 days:
1. the logger memory ran full so that it stopped logging. I became awre of this four hours too late, so that 4 hours = 8 lines of data have been lost; they will be interpolated in the monthly file.
2. the main Linux computer making the data handling, transfers and graphs has been replaced; a fully new installation with openSuSE 10.3 has been made, together with a new Gnuplot version 4.2 (the old was antique!). Some minor cosmetic changes have been made to the graphs. There are a couple of problems with the "previous days" data and graphs, which will hopefully self correct during the next days. Please be patient...
|04 Feb 08||Maintenance and recalibration work. The CO2 sensor calibration should be taken with caution. There is probably a problem lurking around (see +50ppm step). I hope the new sensor will arrive very soon.|
|03 Feb 08||
1. Uploaded jan08.dat to data archive
2. There are quite a lot of comments on myself and meteoLCD at the New York Times DOT_EARTH blog of Andrew Revkin. Make a search for "Massen" or "LCD" . Comments come from my (non sollicited) inclusion to a list of 400 "non-consensus" scientists of republican Senator Inhofe. Just to say that I agree with most of these non-consensus scepticals.
|23 Jan 08||I am preparing the cabling for the new API CO2 sensor, to be delivered soon. The API will be on channel #8, the channel occupied by the disused Delta_T UVB sensor. This allows to keep the number of data columns constant and avoids software adaptations. It is planned to keep the 2 CO2 sensors working in parallel. For the moment, channel #8 has no sensor attached.|
|18 Jan 07||Maintenance by Envitec. CO2 sensor repaired and working ok since 11:00. Now it's wait and see...|
|10 Jan 07||The MIR9000 CO2 sensor is definitively dying. I ordered a Teledyne API 360EM today for replacement. This order will take some time... please be patient.|
|05 Jan 07||
1. Uploaded dec07.dat and 2007.zip (file with all
concatenated monthly .dat files) into the data archive
2. Also no available in data archive:
2007_only.xls = file with all data, averages etc. from 2007 (20MB)
meteoLCDstats_1998_2007.xls = file with all summaries from 1998 to 2007
|03 Jan 08||The usual new year problem with the 7 days graphs.. as the logger file does not contain the year, Gnuplot gets confused the 1rst week of every new year. This will self correct the 7th January.|