Natural ecosystems such as forests are important emission sources of Biogenic Volatile Organic Compounds (BVOC). On a global scale BVOC emissions (~ 1150 Tg C per year) were estimated to be an order of magnitude larger than their anthropogenic counterparts (Guenther et al. 1995). The main emitted biogenic species are isoprene (C5H8) and monoterpenes (C10H16), which account for almost half of the BVOC emissions worldwide. The other half mainly consists of oxygenated VOCs (OVOCs) (alcohols, ketones, aldehydes, acetates, …) and higher terpenoid compounds such as the highly reactive sesquiterpenes.
Due to their large emissions and their high reactivity with the main oxidants (OH, O3, NO3) in the atmosphere (Atkinson and Arey 2003), BVOCs are expected to contribute significantly to atmospheric chemistry. In the presence of nitrogen oxides (important air pollutants mainly resulting from fossil fuel combustion), the atmospheric oxidation of BVOCs (which is a very complex process) may for instance result in net O3 formation and have an important impact on air quality. Less volatile oxidation products can result in formation and/or growth of aerosol particles and as such have an impact on health, visibility, and climate (through cloud formation). In order to be able to quantify net formation of oxidants and aerosols from BVOCs, the physicochemical oxidation and aerosol formation/growth processes have to be well understood. Of equal importance, however, is that also the BVOC emissions need to be well characterised and quantified.
During the last decade, numerous emission studies have been performed and species-specific BVOC standard emission inventories have been started (e.g. the Lancaster University database (UK) or the ACD/NCAR database). Photosynthetic active radiation (PAR) and leaf temperature have long been recognized as important factors controlling emissions from tree foliage, and currently used BVOC emission algorithms are mainly a function of these two parameters. Recent studies have shown that other parameters such as leaf age, leaf growth environment, nitrogen content of the plant, water availability, leaf photosynthesis, relative humidity and CO2 concentration in the air can also have a significant influence on the emissions (e.g. Pétron et al. 2001; Guenter et al. 2006). However, experimental data on how forest ecosystem functioning is contributing to these BVOC emissions, and how this contribution responds to varying environmental conditions during the year are still scarce. Therefore more measurements of BVOC emission dynamics, together with plant physiological activity, are clearly needed (Ruuskanen et al. 2005).
This lack of data on the impact of phenology and environmental parameters on BVOC emissions (which requires long term measurements) is partly due to the fact that, in the past, BVOC emissions were mainly characterised by means of time-consuming sampling techniques, followed by off-line analysis.
Recently, rapid and sensitive technologies (such as the Proton Transfer Reaction Mass Spectrometer (PTR-MS, Lindinger et al. 1998) and the Fast Isoprene Sensor (FIS, Guenther and Hills 1998)), which allow in situ high frequency measurements of isoprene (FIS) and other BVOCs and small oxygenated molecules (PTR-MS), have become available. In combination with high frequency vertical air velocity measurements, these new techniques can even be used to perform direct above-canopy eddy covariance flux measurements and thus to estimate BVOC exchange at stand level. At present, eddy covariance BVOC flux measurements based on PTR-MS have been carried out successfully by a limited number of research groups worldwide (e.g. Rinne et al. 2001, Karl et al. 2002, Grabmer et al. 2004, Spirig et al. 2005) but the number of measurement locations in Europe is still very small and measurements were mainly limited to a small fraction of the growing season.
Moreover, as far as we know from literature, only one group has ever performed PTR-MS based branch enclosure BVOC measurements at forest sites (Ruuskanen et al. 2005). However, this kind of experiments should result in more detailed information on the dynamics of BVOC emissions in real environmental conditions.
The subject of the present proposal is the detailed analysis of the emissions of BVOCs occurring from deciduous (European beech) and coniferous (Douglas fir and Norway spruce) tree species growing in Belgian forest ecosystems. European beech has always been considered as a low isoprene and monoterpene emitter. Recently, however, branch enclosure measurements indicated that this species could be classified as a high monoterpene (mainly sabinene) emitter (T. Dindorf et al. 2005). Norway spruce is reported to emit also monoterpenes (mainly α-pinene and limonene), but also isoprene and acetone (Janson and de Serves 2001). Douglas fir is known to be a monoterpene emitter of which the dominant species are α-pinene, β-pinene, Δ3-carene and limonene (Pressley et al. 2004).
The driving variables behind BVOC emissions of the tree species need to be unravelled by means of well-conceived studies. In order to better understand the effects of environmental and ecophysiological parameters on the emissions, detailed studies at a low organisational level (e.g. leaf level) will be conducted and compared to BVOC emission dynamics at canopy and stand level. These studies will result in new parameterizations of BVOC emissions involving multiple parameters. In combination with a well-validated canopy model and with detailed data for land use, species composition and biomass density, they will result in more accurate BVOC emissions inventories, in particular for Belgium.