 |
| Titel |
Development of a relaxed eddy accumulation system for the determination of biosphere-atmosphere exchange fluxes of peroxyacetyl nitrate |
| VerfasserIn |
Alexander Moravek, Ivonne Trebs, Michael Welling, Thomas Foken |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250053801
|
|
|
|
|
|
| Zusammenfassung |
| We have developed a flux measurement system based on the relaxed eddy accumulation
(REA) method using a commercially available gas chromatograph with electron capture
detection (GC-ECD). We made flux measurements of peroxyacetyl nitrate (PAN) over the
largest continuous nutrient-poor steppe-like grassland ecosystem in Rhine Hessen (region
of Rhineland-Palatinate, Germany) during selected days in 2010 and 2011. The
PAN measurements were conducted on the estate of the Mainz Finthen airport
(Mainz, Germany (49.969Ë Â N, 8.148Ë Â E)) together with measurements of other
atmospheric compounds, which included vertical profiles of NO, NO2, O3, CO2, H2O and
eddy covariance fluxes of O3, CO2 and H2O as well as soil biogenic NO emission
fluxes.
The use of the commonly applied REA sampling bags resulted in an unexplained PAN
loss. Hence, conditional sampling was realized by trapping PAN onto two pre-concentration
columns, one for updraft and one for downdraft events. Subsequent analysis of PAN was done
online by a GC-ECD (Meteorologie Consult GmbH, Germany) located downstream of the
splitter valves. The GC was modified for consecutive analysis of PAN accumulated in both
reservoirs resulting in a total analysing time of 15 min. The GC-ECD was calibrated as a
function of PAN mixing ratio and sample mass trapped onto the pre-concentration
columns.
We will present the setup of the PAN-REA system and the quality assurance related to the
application of the REA method. This mainly involves timing considerations and a quality
assessment of PAN measurements. The design ensures constant flow and pressure conditions
in the intake tubes and reservoirs. Special emphasis was given to the determination of the
lag time between the vertical wind speed signal and the splitter valves. The wind
vector is adjusted online using the double rotation method and a fixed deadband for
vertical wind speeds around zero was tested. Finally, the use of high frequency O3
measurements as a proxy for both deadband and b-value calculations will be discussed. |
|
|
|
|
|
|