VOC emissions of L. cajanderi and L. sibirica

4.1.1 Emission spectra

All of the shoot or branch scale emission studies that have been conducted on the boreal conifer tree species have shown monoterpenes to dominate the mass based emissions spectra (see e.g. Rinne et al., 2009; Figure 8) and such was the case also for bothL. sibirica andL. cajanderi(Papers I and II). Monoterpenes accounted for about 60 to 90% of the total emissions of both trees. Its proportion was lowest in the beginning of the summer and gradually increasing over the course of the summer.

Figure 8. Terpenoid emission spectra of boreal conifersL. cajanderi (Paper II),L. sibirica (Paper I),P. abiesandP. sylvestris in June. Emission ofp. abies have been measured at three different locations: Sodankylä (Sod) in the northern Finland, Järvenpää (Jär) in the southern Finland and at SMEAR II (SME), while P. sylvestris emissions have been measured at SMEAR II.

In case of L. sibirica, substantial and constant throughout the summer sesquiterpene emissions (about 10%) were found as well (Paper I). The rest of the terpenoid emissions were comprised of isoprene, MBO and 1,8-cineol. At the end of the growing season, the sesquiterpene emissions declined notably, and their contribution declined to about 3%.

Based on the results ofPaper I, substantial sesquiterpene emissions were expected from L. cajanderi as well (Paper II). However, this was not the case, as sesquiterpenes only accounted for about 1−2% of the total terpenoid emissions. In addition to monoterpenes, linalool contributed prominently to the emission spectra of both measured trees – varying between 7 and 37% for tree A and 3 and 19% for tree B. Linalool emissions were highest in June and decreased during the summer.

The monoterpene emission spectra of both L. sibirica and L. cajanderi remained fairly constant throughout the growing season. Some differences were found in the emissions of these twoLarix species. Emission spectrum ofL. sibiricawas be dominated by sabinene, which was not detected for L. cajanderi. In general sabinene has not been reported to dominate the monoterpene emissions of any boreal tree species, though it has been identified from the emissions of bothP. abies andP. sylvestris (Figure 9). Other important monoterpenes fromL. sibiricawere ∆³-carene, α- and β-pinene which accounted for about 20%, 10% and 10%, respectively. The rest of the monoterpene emission included limonene, β-phellandrene, terpinolene, tricyclene, α-phellandrene and camphene. In case of both measuredL. cajanderi trees A and B, similar monoterpene emission spectra were observed;

however, the emission total rates of tree B were substantially higher than those of tree A.

About half of the monoterpene emissions were comprised of ∆³-carene, while α- and β-pinene accounted for 20−30% and 10−15% of the total, respectively. Also camphene, limonene, terpinolene and p-cymene were detected. Monoterpene emission spectra of L.

cajanderi were somewhat similar to the emission spectra of P. abies and P. sylvestris (Figure 9).

Figure 9. Monoterpenes emitted by boreal conifersL. cajanderi,L. sibirica,P. abiesand P. sylvestris in June. Emissions ofP. abies have been measured at three different locations:

Sodankylä (Sod) in the northern Finland, Järvenpää (Jär) in the southern Finland and at SMEAR II (SME), whileP. sylvestrisemissions have been measured at SMEAR II.

It has been shown that the emission spectra of different tree individuals of the same tree species can vary considerably. Bäck et al. (2012) reported the VOC emissions of 40 P.

sylvestris trees growing at the SMEAR II site and found large variation in the monoterpene emission spectra of those trees. Also Figure 9 shows large differences in the emission spectra of individualP. abies andP. sylvestris trees.

4.1.2. Emission potentials

BothLarix species had a distinct daily emission pattern which followed both temperature and light (Figure 6 and Figure 1 inPaper I). Though it should be noted, however, that temperature and radiation are strongly coupled in field conditions, especially in the summer in the high latitude boreal forest; hence it can be difficult to differentiate whether the

emissions are driven by temperature, light or both. In case of bothLarixspecies, emission rates decreased towards the end of the summer. Notable seasonal variation has been reported for different plants (Kesselmeier and Staut, 1998). This variation can be associated with plant growth and phenology such as bud burst, flowering or shedding of leafs. For example, Hakola et al. (1998) discovered high monoterpene emissions from Populus tremula andSalix phylicifolia, which are normally substantial isoprene emitters, during the bud burst and leaf, growth and the isoprene emissions started only when the leaves where fully grown. Aalto et al. (2015) reported episodes of monoterpene emission bursts during the early stages of the photosynthetic recovery in the spring.

There was also a substantial difference in the emission rates of the two measured L.

cajanderi trees as the emissions from tree B were up to ten times higher than those of tree A. Some previous studies have shown emission potentials to vary significantly between different tree individuals of, for example,Betula pubenscens (Hakola et al., 2001),Betula pubescens spp. czerepanovii (Haapanala et al., 2009) and Pinus sylvestris (Bäck et al.

2012).

Mono- and sesquiterpene emission potentials of both tree species were determined using both pool (equation 2) and synthesis (equation 3) algorithms, and the same fitting parameters were used for both trees species (Papers I and II). The monoterpene emission potentials calculated using pool algorithm with constant β were 5.2−21 µg gdw-1 h^-1 forL.

sibirica, 1.5−2.2 µg gdw-1

h^-1 forL. sibiricaA and 0.46−19 µg gdw-1

h^-1 for L. cajanderiB.

Respective values for sesquiterpenes were: 0.40−1.8, 0.02−0.03 and 0.04−0.36 µg gdw-1 h

-1. Thus, monoterpene emission potentials of L. cajanderi and L. cajanderi tree B were somewhat comparable, while emission potential ofL. cajanderi tree A was clearly lower.

The high sesquiterpene emissions of L. sibirica are reflected by its high sesquiterpene emission potential, which is actually almost as high as the monoterpene emission potential ofL. cajanderiA. Several studies have been conducted on the monoterpene emissions from other boreal tree species, and the normalized monoterpene emission potentials of between 0.2 and 8.3 µg gdw-1

h^-1 have been reported for different tree species (see table 2 in Rinne et al., 2009). Thus with respect toL. cajanderithe results of our study are more or less in the same range as the previous measurements while the emission potentials ofL. sibirica were higher than those of other boreal trees.

In order to get an idea of how the above canopy monoterpene concentrations due to L.

cajanderi emissions compare to the monoterpene concentrations above the forest dominated by P. sylvestris, the ambient monoterpene concentrations measured at Spasskaya Pad and SMEAR II were plotted against ambient temperature (Figure 10). The relation between the ambient concentration and temperature is similar at both sites, indicating comparable monoterpene emissions from both forests when the temperatures are the same. However one should keep in mind that atmospheric concentrations are a consequence of a large number of different complicated processes including losses/formation due to oxidation and atmospheric dilution and mixing. Thus temperature is not the only driving factor.

Figure 10. Daily mean above canopy monoterpene concentrations (in ppbv) measured with a PTR-MS as a function of temperature (in ºC) at Spasskaya Pad and SMEAR II.