This is the last news for this year.
As a recall: MeteoLCD started in 1994 and became fully operationel in 1998, which is the base year from which we compute several of our trends.
2. Replaced running UVB sensor (#5760) with model #1894 at 09:45 UTC, which is on return from Solar Light Co. where it has been recalibrated. The new sensitivity is down by 4% in respect to the previous calibration of the #1894, (which operated until 10 Sep. 2012 ) what means that the new readings will be slightly lower. As the systematic calibration error is between -6.2 and +4.7%, this will not have a big relevance..
12
2. The heat stress WBGT sensor has been stopped.
3. The ground ozone sensor will probably be back during the next week. The icons and plots of the live data page have been returned to the original to show ozone and air temperature, just to be sure to correctly display everything when the ozone sensor is working again.
So do not be bewildered by the curious -560 ug/m3 ozone readings when there is yet no sensor available..
Attention: The ozone sensor has been removed and will be recalibrated at Brussels. The sensor will be unavailable for quite a long time, probably up to November!
2. Uploaded aug12.dat file into data archive
09:16: Well, the problem was more serious that anticipated. After some time, the computer hangs, and rebooting does not cure the problem. Replaced computer by a clone ( I had a recent image of the original HD). Ok now....
2. There are two important problems:
- the NO/NO2 sensor is out order, and will not being repaired befor mid-September
- there is an intermittent problem with the temperature regulation (UVB-Tin) of the UVB sensor, starting the 21 Jul 2012 at 04:00.UVB readings (mMED/h) after that date should be accepted with caveats.
12
1. I am pretty sure that the high rural concentrations (that are used to launch mandatory traffic slowing to 90 km/h on the autobahn's) are mostly caused by VOC's (terpenes, isoprenes...) naturally emitted by conifers.
Reducing traffic speed on the Luxembourg highways (which are far away from the reference station in Vianden) will have no measurable effect.
2. There is a short interview with me at the journal of Radio RTL (link here, spoken language is Luxembourgish; jump to time stamp10:50)
3. See a poster session "10 years ozone measurements at meteoLCD" discussing these problems
4. I think that our actual ground ozone readings at Diekirch are too low by about 10 -20%. In September we will try to bring the instrument to Brussels for a general recalibration
2. The lightning movies seem to be made correctly again ...
12
Detective work to find some errors:
1. the logger stopped communicating at 03:30 (obscure COM1 error: solved by rebooting...hmm!)
2. the NOx sensor now seems ok
3. the NOx icons were not drawn correctly (error in a an FTP command file)
4. some interruption from 08:15 to 09:15 because of imaging full hard disk of the Linux server making the graphs This really had to be done!
5. The main hard-disk of the meteo01 server (Linux) went bonk (probably after some power outages, despite UPS). This server handles most of the lightning related stuff, including the making of the movies. I moved the main page displaying the "radar-screen" to our meteo.lcd.lu server, and disabled the links to the movies (ok at 13:15 UTC). The repair of the dead server is quite a lot of work, thank you for not expecting wonders!
2. There is a serious problem with the NOx sensor. For the moment, readings should be discarded.
Attention #1: the scale of the plots (today and previous days) will adjust automatically. If you want the UVA data, download the ASCII files.
Attention #2: To revert to the original configuration, I had to stop the datalogger, download all data, make the config. changes and restart. This has been done from 11:32 to 11:40, and the logger will resume at 12:00. The plots and data of channel 10 (UVA) will start with hopefully correct values at that moment!
A quick research shows that a multiplier of 0.8955 should be applied to the #5670 readings to make them comparable to those of instrument #1894.
2. The Davis VantagePro backup station had a loose windvane/anemometer arm. Fixed. Check of windvane shows that true North gave reading of 319° (instead of 360 or 0). I am not sure for haw long this offset has been active. Recalibrated and checked for correct reading of North direction.(09:00)
3. Replaced pump of zone sensor (15:00)
4. Important: We plan to send the UVB sensor (#1894, channel 9) back to Solar Light Co. for recalibration. When off, this sensor will be replaced by #5760 (same type). To make a prior intercomparison between the 2 UVB sensors, the UVA sensor at ch. 10 is temporarily replaced by the UVB sensor #5760. So the UVA plot shows the mMED/h of the sensor #5760 (discard the wrong text saying UVA). This comparison will run for about 4 days.
2. Restarted WBGT measurements at 15:30 UTC
3. Installed new NAS server for storing the images of the many computers (disk images used for cloning).
12
2. The Delta_T logger does not handle leap years; this has been manually corrected in the feb12.dat file. The logger has been halted, emptied, date reset and restarted. No data will be lost.
2. Recalibration of AC31 NOx sensor.
2. It is a normal situation that there are no moist enthalpy data: moist enthalpy can be computed only for positive air temperatures, and we are in deep in a cold snap! It might seem inelegant that GNUPLOT does not create anything picturesque in that case, but I think you should bear this.
2. I made a quick and dirty repair of the zero air LNI generator: as we have not yet a tool small enough to remove the main bearing (tool is ordered), I exchanged the motor against an old one, a leftover from a previous repair, keeping the original compressor head. Seems to be ok (16:30), so I am making the usual calibration of the CO2 sensor.
...well it did not last long: I just was able to finish the CO2 calibration, when the compressor stopped again.!
So definitive repair of the zero air generator will be made during the next visit of Envicontrol (26 Jan 2012).
2. Also uploaded 2011.zip = data file containing all months, including meta data
3. And made the traditional 2011_only.xls file with all the data of the year 2011 and the usual statistics.
Remember that the ultrasonic anemometer is out of action (possibly for a longer time). Take wind velocity from ch. 61 (labeled Airspeed) and add an offset of 0.38 m/s to make readings of this lower mounted cup anemometer comparable to the ultrasonic anemometer at ch. 4 (labeled Air_Velo) Wind direction should be read from ch.13 (labeled Windvane); ch. 15 (labeled WindDir) corresponds to the ultrasonic anemometer, and should also be discarded.
So in short, here are the channels from left to right in the data file that should be discarded:
| channel | Label | Comment | channel to use instead |
| 4 | Air_Velo | ultrasonic anemometer out of action | ch. 61 |
| 7 | UVA | old Delta_T sensor | ch. 10 |
| 8 | CO2 | old MIR9000 sensor | ch.11 |
| 15 | WindDir | ultrasonic anemometer out of action | ch. 13 |
| 23 | Solarrad | old Delta_T solar sensor | ch. 6 |
The main reason for keeping the data files unchanged is that a lot of users pick data from these files and rely on a stable format.