Master Thesis

Fluxes of monoterpenes from a spruce forest: establishing sampling and analytical procedures

Sebastian Schmitt (07/2013-12/2013)

Support: Andreas Held

MSc-Arbeit, Universität Bayreuth, 48 S.

Oxidation products of biogenic volatile organic compounds (BVOC) are thought to be responsible for the growth of newly formed secondary organic aerosol (Birmili et al., 2003; Kulmala et al., 2000). New particle formation (NPF) events with subsequent particle growth have frequently been observed at many places all over the world including the Waldstein site (spruce forest) in NE Bavaria, Germany. For a better understanding of the observed particle growth, measurements of organic precursor gases are crucial. Therefore, a relaxed eddy accumulation sampler developed by the National Center for Atmospheric Research (NCAR) was installed at the Waldstein main tower for measuring uxes and ambient concentrations of BVOCs. The conditional sampler is based on sorbent-cartridges specific for collecting monoterpenes (MTs). Analysis of the samples was carried out at NCAR using thermodesorption followed by gaschromatographic-mass spectrometry coupled to a ame ionization detector (TD-GC-MS-FID). In parallel, a TD-GDFID system for the detection of MTs was successfully built up at the Atmospheric Chemistry laboratory in Bayreuth. Ambient MT concentrations ranging from 0.1 µg m-3 to 10 µg m-3 were detected during the measurement campaign from May to October. alpha-Pinene showed the highest contribution (about 45 %) to the sum of MT concentrations. In addition, beta-pinene (about 10 %), camphene (about 15 %), limonene and eucalyptol (about 20 %) and delta3-carene (about 10 %) were traced. Both MS and FID data were in good agreement despite for limonene and eucalyptol which were overestimated by FID. Fluxes of BVOCs were determined on selected days showing distinct diurnal cycles with the maximum MT fluxes of about 1 mg m-2 hr-1 at noon. MT fluxes followed the diurnal pattern of the buoyancy flux. In order to asses the uncertainties connected to the flux calculation, three different approaches (Monte Carlo simulations, Gauss propagation of uncertainty and a conservative approach) were tested. Total emissions estimated using the MEGAN model (Guenther et al., 2006) showed good agreement with the measurements. Using a simplified approach it was estimated that the measured concentrations of MTs can explain between 10 % and 100 % of observed particle growth rates. This underlines their importance for atmospheric NPF.