|
| 
Hohenpeissenberg
-VOCs-
| Observation |
| Category |
: |
Air sampling observation |
| Situation |
: |
ongoing |
| Time zone |
: |
Local time +1 |
| Sampling |
| Sampling height |
: |
18 |
| Description |
: |
continuous |
| Sampling and analysis frequency |
: |
Once per day at noontime (11:00-15:00 CET) |
| Sampling environment |
: |
The measurement site is situated on top of a solitary mountain about 300 m above the surrounding countryside at 985 m altitude. At night measurements generally sample air from the residual layer since the nocturnal inversion layer resides below the site. After full development of the mixed layer in the morning, measurements are generally performed in the well mixed boundary layer except for few days in winter when a constant, shallow inversion may prevail for several days. No major anthropogenic emission sources are in the vicinity of the Observatory, and it resembles typical rural Central European impact from traffic exhaust, solvent evaporation, domestic heating and minor industrial plants. Land use in the surrounding is about 70% pasture and 30% mixed forests with the latter emitting biogenic compounds, mainly isoprene and monoterpenes. |
| Description for sampling analysis |
: |
Custom-built sampling and gas flow system with samples split from a permanently flushed glass manifold. System components: Nafion dryer (Permapure)used until Jan 31, 2002, thereafter replaced by moisture trap at 233 K; ozone removed on heated stainless steel (423 K) until April 2006, thereafter additionally by a PTFE coated glass fibre filter impregnated with Na2O3S2 solution; cryo-adsorption on glass beads (87 K adsorption, 503 K desorption); separation on Al2O3/KCl PLOT column (Varian Chrompack, 50 m x 0.53 mm i.d.) temperature programmed from 313 K to 473 K; materials in contact with sample are silcosteel (Restek), stainless steel (at 423 K) and valve sealing (Valcon E rotor material in Valco GC valves by VICI, 423 K). Methods successfully tested in several international intercomparison experiments (NOMHICE, AMOHA, GAW).
Sampling interval 15 min, typically 1 sample per day between 10 and 15 (CET)
-Duelmer et al., 2002, J. Chromatogr. A, 953, 175-197. |
| Instrument and Analysis |
| Measurement method |
: |
Gas Chromatography (FID) |
| Current status and history of instrument |
: |
GC-FID based on Varian 3600CX (Varian GmbH) with self built on-line sampling system
In use since 1995, data available since Sep 1998
|
| Description of instrument |
: |
About 40 C2-C8 hydrocarbons are analyzed on a routine base. 28 of them are delivered to WDCGG. Detection limit is between 1 and 10 ppt, depending on the individual hydrocarbon. Most of the hydrocarbons are base-line separated, if not, the uncertainty of peak area determination is estimated and considered in the error. Each concentration stated has an uncertainty estimate (accuracy combines systematic and random errors and is given in column “rem) which corresponds to combined 2 sigma errors of calibration, reproducibility of the measurements, peak integration errors due to peak overlap and baseline noise. Linearity of the system was checked and proved to be linear over the range of encountered concentrations. For more details, see Plass-Duelmer et al., 2002, J. Chromatogr. A, 953, 175-197. |
| Calibration |
| Current scale employed in the measurement |
: |
Scale is based on gravimetrically prepared, certified mixtures of 27-30 hydrocarbons in N2 at mixing ratios of 1-100 ppb (by mole fraction) by NPL (National Physical Laboratory, U.K.). Stated uncertainties (95% level of confidence) are mostly better than 3% (see below). Following standards have been used:
1. 27-component mixture C2-C9, 7-100 ppb, Cylinder No. 5601326, (28th July 1995), uncertainty +/- 5% (used until Oct 1998)
2. 30-component mixture C2-C9, 1-12 ppb, Cylinder No. 2851B (Aug 1998), uncertainty < +/- 1% (used from Nov 1998 to Sep 2005)
3. 30-component mixture C2-C9, 1-11 ppb, Cylinder No. APE 358162, (Aug 2005), uncertainty +/-2.5-3.5% (used from Oct 2005-Jan 2010)
4. 30-component mixture C2-C9, 2-4 ppb, Cylinder No. D641624, (Dec 2009),
uncertainty +/-2.5-3.5% (used since Jan 2010-Jan2013)
5. 30-component mixture C2-C9, 1-11 ppb, Cylinder No. D910379, (Nov 2012), uncertainty +/-2.5-3.5% (used since Jan 2013)
|
| Measurement calibration |
: |
In addition to daily measurements of ambient air, at least once every 2 weeks measurements of helium (zero gas), calibration gas, and two different reference gases are carried out. These measurements characterize the performance of the system and are used for blank peak area determination, calibration, and compound identification. Furthermore, standard additions are made at least every 2 months to check for artifacts due to sample line or chemical reactions. Each measurement takes about 1 hour. Peak areas are determined by use of chromatography software and peak baselines are carefully checked and adjusted by trained operators.
Quantification of NMHC measurements in ambient air is based on the assumption of an equal response per carbon number (C-response) of the FID for the alkanes, alkenes, cyclic and aromatic hydrocarbons. We calculate the mixing ratio of a compound according to:
Mixing ratio = (peak area sample - peak area blank) * calibration factor / Volume of sample
The compounds calibration factor is determined from the average carbon response factor by taking into account the carbon number of the corresponding molecule. The average carbon response factor is determined on a monthly base from at least 6 calibration gas measurements (3 after each other every 2 weeks), and considers all baseline resolved C2-C7 hydrocarbons. Until 2010, however, acetylene was calibrated using the response factor for acetylene alone.
Values below detection limit: either the automatically generated value or 0.0 pptv, with uncertainty values corresponding to the detection limit (3 x standard deviation of baseline noise integrated over typical peak widths).
For more details, see Plass-Duelmer et al., 2002, J. Chromatogr. A, 953, 175-197.
|
| Scale and calibration(treasability) |
: |
The scale is based on the certified, gravimetrically prepared gas standards supplied by the National Physical Laboratory (NPL, Teddington, U.K.) (see above) and the assumption of an equal response per carbon number except for acetylene (see above). The standards are contained in passivated aluminum cylinders. The average carbon response factor for the different NPL standard mixtures was intercompared when the standards were changed. No significant change (range of 1%) was observed when changing from Cylinder No. 5601326 to 2851B (in 1998), to APE 358162 (in 2005), to D64 1624 (in 2010) and to D910379 (January 2013) indicating stability of the hydrocarbon mixture in the standard over time.
Measurements were successfully tested in several international intercomparison experiments (NOMHICE, AMOHA, GAW, ACTRIS).
|
| Data Processing |
| Measurement unit |
: |
ppt |
| Data processing |
: |
Data are quality controlled. Hydrocarbons are identified by retention time. 28 hydrocarbons corresponding to the recommended VOC calibration mixture (WMO GAW report No. 111) are listed. Accuracy values comprise the total expanded uncertainty with coverage factor k=2 of calibration, reproducibility of the measurements, and peak integration errors due to peak overlap and baseline noise (values are given in the “Rem-column). No averaging is applied for daily data. |
| Processing for averaging |
: |
Processing for Daily Data:
Results of one measurement per day (between 11:00 and 15:00, mostly at about 13:00, CET) are reported, if available.
Processing for Monthly Data:
Arithmetic means are calculated from daily data. |
| Data flag |
: |
|
| Data remarks |
: |
In the Remarks column total expanded uncerainty with coverage factor k=2 are provided (in ppt). They comprise the 2 sigma errors of calibration, reproducibility of the measurements, and peak integration errors due to peak overlap and baseline noise. |
| Other Information |
| Scientific aim |
: |
C2-C8 hydrocarbons: Seasonal cycles and trends, relation between hydrocarbon patterns and transport, anthropogenic versus biogenic origin, impact on local photochemistry and ozone formation. |
| Reference |
: |
Plass-Dlmer, C.; Michl, K.; Ruf, R.; Berresheim, H. C2 - C8 hydrocarbon measurement and quality control procedures at the Global Atmosphere Watch Observatory Hohenpeissenberg. J. Chromatogr. 953, 175-197; 2002.
C. Plass-Dlmer, N. Schmidbauer, J. Slemr, F. Slemr, and H.D’Souza, European hydrocarbon intercomparison experiment AMOHA part 4: Canister sampling of ambient air, J. Geophys. Res., 111, D04306, doi:10.1029/2005JD006351, 2006.
B. Rappenglck, E. Apel, M. Bauerfeind, J. Bottenheim, P. Brickell, P. Cavolka, J. Cech, L. Gatti, H. Hakola, J. Honzak, R. Junek, D. Martin, C. Noone, C. Plass-Dlmer, D. Travers, and D. Wang, The first VOC intercomparison exercise within the Global Atmosphere Watch (GAW), Atmos. Environ. 40, 7508-7527, 2006.
C. Plass-Dlmer, and H. Berresheim, Volatile Organic Compound Measurements at Hohenpeissenberg as part of GAW, in “The German Contribution to the WMO/GAW Program: Upon the 225th anniversary of GAW Hohenpeissenebrg Observatory, ed. W. Fricke, GAW Report 167, 2006.
D. Helmig, J. Bottenheim, I. E. Galbally, A. Lewis, M. J. T. Milton, S. Penkett, C. Plass-Duelmer, S. Reimann, P. Tans, and S. Thiel, Volatile organic compounds in the global atmosphere, Eos Trans. AGU, 90(52), 513–514, 2009.
E. von Schneidemesser, P.S. Monks, and C. Plass-Duelmer, Global Comparison of VOC and CO Observations in Urban Areas, Atmosph. Environ. 44, 5053-5064, 2010.
Hoerger, C.C. et al., ACTRIS non-methane hydrocarbon intercomparison experiment in Europe to support WMO GAW and EMEP observation networks, Atmos. Meas. Tech., 8, 2715-2736, 2015. |
|
| submitted by Meteorological Observatory Hohenpeissenberg, German Meteorological Service |
|
|
