Biogeosciences, 6, 1059–1087, 2009 www.biogeosciences.net/6/1059/2009/ Biogeosciences © Author(s) 2009. This work is distributed under the Creative Commons Attribution 3.0 License. A new European plant-specific emission inventory of biogenic volatile organic compounds for use in atmospheric transport models M. Karl1,*, A. Guenther2, R. Koble¨ 1, A. Leip1, and G. Seufert1 1European Commission, Joint Research Centre, Institute for Environment and Sustainability, Ispra, Italy 2National Center for Atmospheric Research, Boulder, CO, USA *now at: NILU, Norwegian Institute for Air Research, Kjeller, Norway Received: 10 September 2008 – Published in Biogeosciences Discuss.: 18 December 2008 Revised: 27 May 2009 – Accepted: 27 May 2009 – Published: 18 June 2009 Abstract. We present a new European plant-specific emis- 1 Introduction sion inventory for isoprene, monoterpenes, sesquiterpenes and oxygenated VOC (OVOC), on a spatial resolution of The amount and composition of plant species that cover the 0.089×0.089 degrees, for implementation in atmospheric land surface is the primary control on the type and magni- transport models. The inventory incorporates more accu- tude of biogenic volatile organic compound (BVOC) flux. rate data on foliar biomass densities from several litterfall On a shorter time scale, variation in atmospheric conditions databases that became available in the last years for the main determines the amount of photosynthetically active photon tree species in Europe. A bioclimatic correction factor was flux density (PPFD) reaching the leaf surface, the leaf tem- introduced to correct the foliar biomass densities of trees perature, and soil moisture. All these factors in turn influ- and crops for the different plant growth conditions that can ence variations in BVOC emissions. Temperature and light be found in Pan-Europe. Long-term seasonal variability of intensity are key driving variables that regulate emissions of agriculture and forest emissions was taken into account by isoprene, monoterpenes and other BVOCs in published emis- implementing a new growing season concept. The 2004– sion algorithms (Tingey et al., 1991; Guenther et al., 1993, 2005 averaged annual total biogenic volatile organic com- 1995, 2006; Schuh et al., 1997; Staudt et al., 2000). pound (BVOC) emissions for the Pan-European domain are Once emitted to the atmosphere, BVOCs react in the pres- estimated to be about 12 Tg with a large contribution from ence of nitrogen oxides to increase the concentration of tro- the OVOC class of about 4.5 Tg and from monoterpenes of pospheric ozone (Atkinson and Arey, 2003), which is a res- about 4 Tg. Annual isoprene emissions are found to be about piratory irritant and major component of smog. Emissions of 3.5 Tg, insensitive to the chosen emission algorithm. Emis- BVOC are significant to ozone production in the European sions of OVOC were found to originate to a large extent from boundary layer (Simpson, 1995; Vogel et al., 1995). The agriculture. Further experiments on crop emissions should be oxidation products of terpenes condense to form secondary carried out to check the validity of the applied standard emis- organic aerosols (Hoffmann et al., 1997; Kavouras et al., sion factors. The new inventory aims at a fully transparent 1998; Griffin et al., 1999) which directly alter Earth’s radia- and verifiable aggregation of detailed land use information tive balance and can serve as cloud-condensation nuclei (An- and at the inclusion of plant-specific emission data. Though dreae and Crutzen, 1997). On the basis of results from smog plant-specific land use data is available with relatively high chamber experiments of Pandis (1991) it has commonly been accuracy, a lack of experimental biomass densities and emis- assumed that the photooxidation of isoprene does not con- sion data on terpenes, sesquiterpenes and oxygenated VOC, tribute to the production of secondary organic aerosol un- in particular for agricultural plants, currently limits the setup der ambient conditions (Seinfeld and Pandis, 1998). How- of a highly accurate plant-specific emission inventory. ever, evidence from both field and laboratory experiments has been obtained in the last four years for isoprene con- tributing to SOA formation (Claeys et al., 2004; Kourtchev et al., 2005; Kroll et al., 2005). Sesquiterpenes are a highly re- Correspondence to: G. Seufert active BVOC class with atmospheric lifetimes of only a few ([email protected]) minutes. These compounds have a high potential to form Published by Copernicus Publications on behalf of the European Geosciences Union. 1060 M. Karl et al.: European BVOC emissions ther, 1997) and MEGAN (Guenther et al., 2006), were used 1 2 Spatial to calculate emission fluxes of isoprene over Europe. Pre- Tabular GIS ECMWF vious European emission estimates for isoprene were about Database database Forecast 4 Tg per year, and for total BVOC ranged between 7.5 Tg to 29 Tg per year (Lubkert¨ and Schopp¨ , 1989; Andryukov and Timofeev, 1989; Simpson et al., 1999; Steinbrecher et al., 3 Meteorology 2009), indicating the high uncertainty of BVOC inventories. Data Emission CLC/GLC 2000 Bioclim. The reason for this high variability is not always easy to un- factors ε, ICP Forest correct. Foliar CAPRI Agricult. LAI derstand, because inventories are not always fully transparent factor f dens. d (MODIS) Domain 10 km or cannot be compared for various reasons. Therefore, the new inventory presented here aims at full transparency and avoidance of unnecessary complexity in order to allow cal- 4 Leaf culations in the frame of a global chemical transport model. Temperature Emission Canopy Model Light 2 Inventory description (G97/MEGAN) Extinction Seasonality In this work a new plant-specific BVOC emission inven- Factor tory with high spatial resolution for implementation in at- mospheric transport models is developed for Europe. Data Fig. 1. BVOC emission model, schematic overview. on emission potentials at standard conditions for tree and crop species and for landcover classes was adopted from the work of Steinbrecher et al. (2009), the NatAir (Improv- secondary organic aerosol (Hoffmann et al., 1997; Jaoui et ing and Applying Methods for the Calculation of Natural al., 2003). Bonn and Moortgat (2003) suggest that the re- and Biogenic Emissions and Assessment of Impacts on Air action of sesquiterpenes with atmospheric ozone could be Quality) project inventory of VOC emission from natural responsible for the atmospheric new particle formation ob- and semi-natural vegetation in Pan-Europe. Throughout this served frequently in rural locations. work, emission potentials are given for standard conditions, ◦ −2 −1 Recent observations in boreal forests support the role i.e. 30 C leaf temperature and 1000 µmol m s photo- of aerosol production from biogenic hydrocarbon precur- synthetic photon flux density (PPFD). From several litterfall sors in the activation of cloud droplets (Kerminen et al., databases we derived more accurate data on foliar biomass 2005; Tunved et al., 2006). In addition, BVOCs released densities for the most abundant tree species. A biomass cor- to the atmosphere may represent a relevant source term in rection factor was introduced for forest and agriculture land the overall carbon budget of an ecosystem (Guenther, 2002; use classes to adjust foliar biomass densities for growing con- Kesselmeier et al., 2002). ditions in different climates. To account for the long-term The European landscape is characterised by a great vari- seasonal variability of emissions, a new growing season con- ety of climatic and orographic zones and biomes, ranging cept was introduced based on detailed phenology data for from boreal forests in Scandinavia and Russia to Mediter- major agricultural crops and on a seasonal algorithm for de- ranean shrub vegetation. For millennia, the European land ciduous and evergreen vegetation by Staudt et al. (2000). cover has been intensively modified by man to create an For each BVOC class the emissions can be calculated as extremely patchy landscape. In addition the orography of sum over all grid cells of the European domain: Europe is complex, with changes between mountainous and X E = A · · D · γ (1) plain landscapes on a small scale. Type and amount of BVOC b i b,i i CE,i i emissions is highly variable; even within a plant taxonomic group like the important Quercus (oak) genus one can ob- In Eq. (1), Ai is the vegetated and emitting area in a grid 2 serve several species emitting huge amount of isoprene, other cell i (in m ), b,i is the average emission standard potential −1 −1 species emitting monoterpenes, others are close to zero emit- (in µg gDW h ) of each BVOC class, Di the foliar biomass −2 ters (Csiky and Seufert, 1999). Therefore it is of great impor- density related to dry mass (in g m ) and γCE,i is the dimen- tance to use highly resolved land cover data of forests, agri- sionless canopy emission activity factor, often referred to as culture and other land use for modelling European BVOC the environmental correction factor. emissions. The schematic information flow of our BVOC emission Estimates of European isoprene emission depend on the model is depicted in Fig. 1. The model contains four main choice of emission algorithms describing the temperature parts: (1) retrieving geo-referenced spatial data from sev- and light dependence of emissions. Two different represen- eral Geographic Information System (GIS) databases and tations of the temperature and light dependence, G97 (Guen- property data from tabular databases, (2) meteorological Biogeosciences, 6, 1059–1087, 2009 www.biogeosciences.net/6/1059/2009/ M. Karl et al.: European BVOC emissions 1061 input data from the European Centre for Medium Range part of GLC Northern Eurasia v4.0 and the Europe region Weather Forecast (ECMWF) and leaf area index data from corresponds to the extent of GLC Europe v1.0. MODIS, (3) aggregation of the spatial GIS data for the Pan- All land use data was aggregated and then projected European domain, processing of tabular input and writing on a regular latitude/longitude grid.