UEF//eRepository
DSpace https://erepo.uef.fi
Rinnakkaistallenteet Luonnontieteiden ja metsätieteiden tiedekunta
2018
Scots pine provenance affects the emission rate and chemical
composition of volatile organic compounds of forest floor
Kivimäenpää, Minna
Canadian Science Publishing
Tieteelliset aikakauslehtiartikkelit
© Authors
All rights reserved
http://dx.doi.org/10.1139/cjfr-2018-0049
https://erepo.uef.fi/handle/123456789/7209
Downloaded from University of Eastern Finland's eRepository
1
Scots pine provenance affects the emission rate and chemical composition of volatile 1
organic compounds of forest floor 2
3
Minna Kivimäenpääa, Juha-Matti Markkanena, Rajendra P. Ghimirea, Toini Holopainena, 4
Martti Vuorinenb, Jarmo K. Holopainena 5
6
aDepartment of Environmental and Biological sciences, University of Eastern Finland, 7
P.O.Box 1627, 70211 Kuopio, Finland 8
bNatural Resources Institute Finland (Luke), Juntintie 154, 77600 Suonenjoki, Finland 9
10
Corresponding author: Minna Kivimäenpää (email: minna.kivimaenpaa@uef.fi) 11
12
2 Abstract
13
Scots pine (Pinus sylvestris L.) is an important source of biogenic volatile organic 14
compounds (BVOCs) in the boreal zone. BVOC emission rate and profile affect air quality, 15
climate forcing, plant stress tolerance, and thus the growing conditions of forests. BVOC 16
emission profile of shoots and forest floor, and emission rates from forest floor, were studied 17
in a latitudinal provenance experiment with 19-year-old Scots pine common garden in 18
Central Finland. The provenances studied were Saaremaa (SAA 58°22´), Korpilahti (KOR 19
62°0´), Suomussalmi (SUO 65°10´) and Muonio (MUO 67°56´). A chemotype with high 20
proportion of Δ-3-carene, terpinolene, sabinene, γ-terpinene and α-terpinene was 21
significantly more common for the southern SAA than the northern SUO and MUO 22
provenances. A chemotype with high proportion of α-pinene, β-pinene, limonene and 23
myrcene was more common in the three northernmost provenances. The main compounds 24
emitted by forest floor were α-pinene, Δ-3-carene and camphene. Similarly to shoot 25
emissions, forest floor emissions from SAA had highest proportion of Δ-3-carene. Average 26
total VOC emission rate from forest floor was 50 µg m-2 h-1 at the end of August. Total 27
emission rates were 65 % higher in KOR than in MUO. High emission rates were explained 28
by the high amount of decomposing needle litter and low moss coverage.
29 30
Keywords: tree provenance, BVOCs, forest floor, needle litter, mosses 31
32
3 1. Introduction
33
Boreal forest or taiga is the world’s largest terrestrial biome dominated by coniferous tree 34
species and it covers one third of the total forest area of the Earth (Gauthier et al. 2015).
35
Carbon sequestration by boreal forests is estimated to represent 20% of the annual C sink of 36
global forests (Gauthier et al. 2015). In Finland, 86 % of the land area, ca. 26 million 37
hectares, is covered by forests, and Scots pine (Pinus sylvestris) occupies 65 % of the forest 38
land (Peltola 2014). Global emissions of biogenic volatile organic compounds (BVOCs), 39
such as methanol, acetone and terpenoids isoprene, monoterpenes and sesquiterpenes, from 40
vegetation were modelled to be 659.5 Tg C y-1 and boreal forests area has a considerable 41
contribution to monoterpene emission during summer (Messina et al. 2016). Terpenoids are 42
components of resin, synthesized, stored and transported in the resin canals of coniferous 43
trees (Trapp and Croteau 2001). Emissions of volatile terpenoids of Scots pine needles 44
originate both from storage and de novo synthetis (Ghirardo et al. 2010). In addition, stems 45
(Heijari et al. 2011) and roots (Lin et al. 2007) emit BVOCs.
46
Monoterpenes 3-carene and α-pinene are the two terpenoids with the highest emission 47
rates from Scots pines. Terpene profile of Scots pine is under genetic control, and the 48
division of Scots pines to chemotypes of ‘low’ or ‘high proportion of 3-carene’ is well 49
known (Baradat and Yazdani 1988). Scots pine trees growing in Northern Finland have 50
lower proportion of high-carene chemotypes than those in Southern Finland (Muona et al.
51
1986) and similar latitudinal difference was observed between Scots pine provenances in a 52
multi-site common garden experiment established in three locations in Finland (Nerg et al.
53
1994). Nerg et al. (1994) also reported that opposite to 3-carene, shoot α-pinene 54
concentration increased from southern to northern provenances. Relatively similar terpene 55
profiles have been reported for pine shoot (needles+stems) (Nerg et al. 1994), needle and 56
wood terpene concentrations (Manninen et al. 2002) and stump BVOC emissions 57
(Kivimäenpää et al. 2012) in the same common garden. Chemotypes characterized by α- 58
4
pinene and 3-carene were also found from Scots pine shoot BVOC emissions in a pine- 59
dominated mixed forest stand in southern Finland (Bäck et al. 2012).
60
Plants use terpenoids for defense against biotic stresses, such as pathogens and herbivores 61
(Trapp and Croteau 2001). They can be toxic or repellent for herbivores, or attract natural 62
enemies of herbivores or their egg parasitoids (Mumm and Hilker 2006). BVOC profile is 63
important as it is used as chemical cue both for location of herbivores and for their enemies 64
(Mumm and Hilker 2006). BVOCs also increase plant resistance against abiotic stress, e.g.
65
heat and ozone, a phytotoxic air pollutant (Loreto and Schnitzler 2010).
66
In the atmosphere, BVOCs affect air quality, temperature and PAR (photosynthetically 67
active radiation), and thus forest productivity (Holopainen 2011). Namely, BVOCs 68
contribute to both formation and breakdown of ozone (Lerdau and Slobodkin 2002) that 69
with prevailing concentrations can have adverse effects on Scots pine (Huttunen and 70
Manninen 2013). In addition, BVOCs reduce oxidation of greenhouse gas methane, and 71
thus, have a warming effect on the climate (Peñuelas and Staudt 2010). Moreover, in 72
reactions with atmospheric oxidants, Scots pine BVOCs contribute to the formation of 73
secondary organic aerosols (SOA) and via increase of cloud condensation nuclei promote 74
cloudiness (Kulmala et al. 2013). SOA and cloud formation can increase tree photosynthesis 75
and primary production in diffuse light (Holopainen 2011) and cool the climate (Scott et al.
76
2018). Reactivity of VOCs with various tropospheric oxidants such as ozone and OH and 77
NO3 radicals (Atkinson and Arey 2003) and consequent SOA mass yields (Lee et al. 2006) 78
differs between individual compounds. Larger BVOC emissions of stressed Scots pine 79
seedlings led to larger SOA formation (Joutsensaari et al. 2015). Therefore, both emission 80
rates and profile, especially from dominant forest species, are important in considering 81
significance of BVOCs on climate.
82
5
Forest floor can be a considerable source of BVOCs, because understory plant species, 83
such as Calluna vulgaris (Tiiva et al. 2017) or Rhododendron tomentosum (Himanen et al.
84
2010) shrubs, conifer needle litter (Isidorov et al. 2010), litter decomposers (Isidorov et al.
85
2016), roots (Lin et al. 2007), rhizosphere (Rasheed et al. 2017), ectomycorrhizal fungi and 86
endophytes (Bäck et al. 2010) emit BVOCs. BVOC emissions from Scots pine-dominated 87
forest floor have been mainly measured in one location in Southern Finland, and estimations 88
of emissions rates have high variation, ranging from 0-373 µg m-2 h-1 (summarized in Mäki 89
et al. 2017). Common garden experiments where Scots pines from different provenances 90
grow in the same location, provide a way to estimate forest floor emission rates from a wider 91
latitudinal range. Provenance experiments also enable to study if shoot and forest floor VOC 92
emission profiles differs between latitudes, similarly as latitude affected terpene 93
concentration profiles of pine shoot (Nerg et al. 1994), needle and wood (Manninen et al.
94
2002) and stump BVOC emission profiles (Kivimäenpää et al. 2012). We can hypothesize 95
that the influence of tree chemotype is observable in the forest floor emissions via emissions 96
from roots and needle litter.
97
The aims of this study were to examine if tree provenance affects 1) BVOC emission 98
profile from the shoots, 2) BVOC emission profile from forest floor, 3) BVOC emission 99
rates from forest floor and if 4) forest floor characteristics (needle litter, understory 100
vegetation, temperature, moisture) influence on the emission rates.
101 102
2. Material and methods 103
104
2.1. Experimental site and tree provenances 105
This study was conducted in an experimental Scots pine (Pinus sylvestris L.) stand with nine 106
provenances sown in a common garden at Suonenjoki Research Unit (latitude 62º37’) of the 107
Finnish Forest Research Institute (currently Natural Resources Institute Finland) in 1991.
108
6
Trees were grown from seeds originating from a 1200-km South-North transect from 109
Estonia to Northern Finland. Trees of each provenance grew in five replicated 1000 m2 110
blocks in fully replicated rows, and the area of the whole research field was 0.5 ha. More 111
details of the site and growing conditions are described in Manninen et al. (2002). The stand 112
has been thinned several times since its establishment, with 30-40% of the original trees left 113
at the time of this study. The provenances selected for the present study were Saaremaa 114
(SAA, latitude 58º22’), Korpilahti (KOR, latitude 62º0’), Suomussalmi (SUO, latitude 115
65º10’) and Muonio (MUO, latitude 67º56’). The Saaremaa provenance originated in 116
Estonia, the others were from Finland. Trees were 19 years old when the sampling was 117
conducted in 2010.
118 119
2.2. Tree selection, shoot BVOC collection and needle length measurement 120
Ten trees of each of the four provenances in the 0.5 ha common garden stand (Kivimäenpää 121
et. al. 2012) were randomly selected for the study on 2 June 2010. Trees were on average 4.5 122
m tall. One branch per tree from the lower parts of the canopy at a height ca. 3 m was 123
selected for BVOC collection that took place on 21 - 22 June 2010. The youngest needles 124
were still elongating at that time. The number of needle generations was four in all SAA and 125
KOR provenances, but there were less trees with fourth generation needles left among SUO 126
and MUO provenances (Table 1). One shoot per tree with all existing needle generations 127
were enclosed into a pre-cleaned (+120 °C) polyethylene terephthalate (PET) bag (Look 128
45x55 cm) that was tightened with a shutter around a bare stem. A hole was cut to the 129
other corner of the bag, and ozone-free (Ozone Scrubber Cartridge, Environnement S.A., 130
Poissy, France, to avoid degradation of VOCs in the adsorbent), charcoal-filtered air 131
(Wilkerson F03-C2-100, Monterrey, Mexico, to remove VOCs from background air entering 132
the collection bags) was led into the bags via Teflon-tubing at a rate of 0.6 l/min. When the 133
7
bags had expanded and the air had been replaced, the flow rate was reduced to 300 ml min-1. 134
A purified stainless steel tube (ATD sample tubes, Perkin Elmer, Norwalk, CT, USA) filled 135
with approximately 150 mg of Tenax TA adsorbent (mesh 60/80, Supelco, Bellefonte, PA, 136
USA) was inserted into a small hole cut in the other corner of the collection bag and 137
fastened with a shutter. The sample was pulled through the sample tube with a vacuum 138
pump (Thomas 5002 12 V DC, Puchheim, Germany) at a rate of 200 ml min-1 for 15 139
minutes. The tubes were sealed with Teflon-coated brass caps and stored at +4 °C until 140
analysis. Temperatures inside the collection bags were recorded by wireless data loggers 141
(Hygrochron DS1923-f5 iButton, Maxim Integrated products, San Jose, CA, USA). Air 142
temperature (sensor S-THA-M006) outside the collection bags and photosynthetically active 143
radiation (PAR, sensor PAR S-LIA-M003) inside empty collection bag were measured and 144
recorded (datalogger, Hobo Micro Station, Onset Computer Corporation, Bourne, MA, 145
USA) next to shoots used for VOC-collection. Air samples from empty collection bags 146
(blank samples) in the forest site were also collected to confirm the purity of the background 147
air entering the collection bags. PAR in the collection bags varied between 140 and 580 148
µmol m-2 s-1 and temperature between 14 and 26 °C. Bag enclosure increased the 149
temperature on average by 1.4 °C.
150
Needle lengths from all existing needle generations were measured first on 29 June 151
(Table 1). Elongating current year needles from SUO were longer than from SAA, and 152
oldest generation needles were shortest in MUO provenance (Table 1). On 21 July, current 153
year needle lengths were measured again, and no provenance differences were observed any 154
longer when their growth had ended (Table 1). Oldest needle generation had dropped from 155
all studied shoots by 21 July.
156 157
2.3. Forest floor characteristics and BVOC collection 158
8
Soil cylinders (height 12 cm) made of polyethene and covering 68 cm2 of forest floor were 159
installed on 2 June 2010 under the foliage of the same trees as used for shoot BVOC 160
collection (n=10 for KOR, n=9 for other provenances) at the distance of 0.5m (between 161
trunk base and cylinder margin). One end of the cylinder was made narrow and sharp and 162
was pressed to the soil. Cylinders were left open so that needle litter accumulated on soil 163
surface inside the cylinder. BVOC emissions from forest floor were collected between 10 164
am – 3 pm on 23 August 2010. Pre-cleaned PET bag (25 x 55 cm) was tightly tied around 165
the soil cylinder with rubber bands. BVOC collection time was 30 min. Otherwise, BVOC 166
collection was done as described in section 2.2. In addition, blank samples using empty bags 167
or soil cylinder in the bag were separately collected to take into account compounds 168
originating from the collection system. Sensors (S-THA-M006 for temperature; S-LIA- 169
M003 for PAR) of data loggers (Hobo Micro Station, Onset Computer Corporation, Bourne, 170
MA, USA) measured and recorded air temperature on the soil surface and PAR-level during 171
the collection. Temperature varied between 15-19 °C and PAR-level was < 100 µmol m-2 s-1. 172
Soil moisture at the depth of 5 cm was measured by soil moisture sensor (Theta Probe, type 173
ML2, Delta-T devices, Cambridge, UK) after BVOC collection. Temperature and soil 174
moisture was not statistically different between provenances (data not shown.) The coverage 175
(%) of mosses (mainly Polytrichum commune and Pleurozium schreberi), graminoids 176
(mainly dried), lichens (mainly Cladonia rangiferina) and Scots pine bark or cones in the 177
cylinders (Table 2) were estimated visually. The needle litter from the cylinders were 178
collected and its dry weight was determined after drying to constant weight at +60 °C.
179
Forest floor under trees from KOR provenance had the significantly higher amount of needle 180
litter than trees from SUO and MUO provenances (Table 2). Coverage of mosses was 181
significantly higher under trees from MUO than SAA and KOR (Table 2).
182 183
9 2.4. BVOC analysis
184
BVOC samples were analyzed by gas chromatography-mass spectrometry (GC-MS, Hewlett 185
Packard type 6890, Waldbronn, Germany; MSD 5973, Beaconsfield, UK). Compounds 186
trapped in the adsorbent were desorbed (Perkin Elmer ATD400 Automatic Thermal 187
Desorption System, Wellesley, MA, USA) at 250 °C for 10 min, cryofocused in a cold trap 188
at -30°C and subsequently injected onto am HP-5 capillary column (50 m x 0.2 mm id. x 189
0.33 µm film thickness, J&W Scientific, Folsom, CA, USA). The temperature program was 190
40 °C for 1 min, followed by increases of 5 °C min-1 to 250 °C. The carrier gas was helium.
191
The standards were self-made mixtures of commercial standards for monoterpenes (18 192
compounds), homoterpenes (1), sesquiterpenes (4), green leaf volatiles (8) and other plant 193
volatile compounds (3) dissolved in methanol. Volumes of 2 µl standard mixtures were 194
injected into the adsorbent tubes. The compounds were identified by comparing their mass 195
spectra to the using commercial standards and the Wiley library. Compound quantification 196
was based on TIC (total ion counts). Compounds for which commercial standard was not 197
available were quantified using α-pinene (for non-oxygenated monoterpenes), 1,8-cineole 198
(oxygenated monoterpenes) and longifolene (sesquiterpenes) as reference compounds.
199
Emission profiles of shoots and forest floor were presented as a proportion of a compound 200
from total emissions. Shoot emissions were measured as a part of continuing experiment, 201
and shoots could not be interfered for needle area or biomass measurements. Therefore, 202
shoot emissions rates (including needle and bark emissions) were calculated per shoot length 203
as done by Ghimire et al. (2013) (TableS11). The shoot BVOC emission rates were not 204
suitable for provenance comparison, because the needle volume per shoot length was visibly 205
different. Emission rates of BVOCs from forest floor were calculated per forest floor area 206
using a unit µg m-2 h-1. Forest floor emission rates were calculated also as carbon (C) 207
1 Supplementary data are available with the article through the journal Web site
10
emissions, as well as standardized to +20 °C, which is a typical summer temperature in 208
boreal forests, using the algorithm by Guenther et al. (1993).
209 210
2.5. Statistics 211
Differences between the provenances were tested by one-way ANOVA with Tukey test for 212
pairwise comparisons. Before that, the data were logarithm transformed to fulfill the 213
assumptions of ANOVA (data normality and homogeneity of variances) when needed. In 214
case the ANOVA assumptions were not met, provenance differences were tested by 215
Kruskal-Wallis test with pairwise multiple comparison test. Differences in the proportion of 216
individual compounds from total emissions between shoots and forest floor were tested by 217
T-test for pairwise comparisons or Wilcoxon Signed Ranks test, when data were not 218
normally distributed. Statistical analyses were performed by IBM SPSS 21.0.
219
Proportions of shoot BVOCs were subjected to a principal component analysis (PCA).
220
The data were mean-centered and standardized to unit-variance. Compounds found in < 10 221
% of the samples were left out from the analysis. Partial least squares regression (PLSR) 222
was used to analyze the influence of the dry weight of needle litter, soil moisture, air 223
temperature and the cover of vegetation groups on individual BVOCs. One component 224
PLSR models, that were cross-validated using seven cross-validation groups, were extracted 225
separately for each BVOC. The PCA and PLSR analyses were conducted using Simca 14 226
(Umetrics, Umeå, Sweden).
227 228
3. Results 229
3.1. Shoot emissions profile 230
As an average over all the provenances, the majority of the shoot BVOC emissions consisted 231
of monoterpenes (97.4 % ± 0.5) and the major compounds were α-pinene, Δ-3-carene, 232
11
limonene, myrcene, β-pinene, camphene (Table 3), β-phellandrene (3 %), terpinolene (2 %), 233
1,8-cineole (2 %), (E)-β-ocimene (1 %) and sabinene (1 %). In total, pines emitted 27 234
different monoterpenoids (Table S12). An average percentage for sesquiterpenes was 2.4 ± 235
0.5 of total emissions and the major compounds of 19 observed sesquiterpenes (Table S12) 236
were (E)-β-farnesene (0.7 %) and longifolene (0.5 %). The proportion of GLVs and 237
methylsalicylate was low (0.1 %).
238
PCA and PC1 revealed BVOC profiles that differed between the provenances (Fig. 1).
239
The trees that had a high proportion of Δ-3-carene in the emission blend also had a high 240
proportion of sabinene, α-terpinene, γ-terpinene and terpinolene, but a low proportion of α- 241
pinene, β-pinene, limonene and myrcene (Fig. 1b). There were more trees with this type 242
(‘high carene but low pinenes’) among the SAA provenance compared to SUO and MUO 243
(Fig. 1a). Four to five of ten trees among KOR, SUO, and MUO provenances had an 244
opposite profile, i.e., a high proportion of pinenes but low Δ-3-carene, and there were also 245
intermediate profile trees. The difference in the profiles characterized by PC1 was 246
significant between SAA and the northernmost provenances SUO and MUO (Fig. 1a). PC2 247
was characterized by trees with high emissions of oxygenated monoterpenes camphor, 1,8- 248
cineole and sesquiterpenes cis-α-bisabolene, δ-cadinene, (E,E)-α-farnesene and an unknown 249
sesquiterpene (Fig. 1b), but this profile was not significantly more common in any of the 250
provenances. Proportions of the two most dominant compounds, α-pinene, and Δ-3-carene, 251
are also shown in Fig. 2. The proportion of Δ-3-carene was highest in SAA provenance and 252
significantly different to MUO provenance.
253 254
3.2. Emission rates and profile from forest floor 255
2 Supplementary data are available with the article through the journal Web site
12
Unstandardized BVOC emission rates from the forest floor were 50 ± 7 µg m-2 h-1 when 256
averaged over all the provenances. Respective emission rate standardized to +20 °C was 67 257
± 8 µg m-2 h-1. Emission rate calculated as C at 20 °C was 59 ± 7 µg m-2 h-1. Temperature- 258
standardized total emission rates were significantly higher in KOR than other provenances 259
(Table 4). The difference between the provenances with the highest and lowest emission 260
rates was 65 %. Compounds α-pinene and Δ-3-carene contributed a major fraction of the 261
emissions from the forest floor (Table 3). Tricyclene, α-pinene, and camphene contributed 262
higher fraction and limonene and myrcene lower fraction in the forest floor emissions than 263
in the shoot emissions (Table 3). The proportion of Δ-3-carene was the highest in SAA 264
provenance and significantly different from KOR and SUO (Fig. 2). Emission rates of Δ-3- 265
carene, on the other hand, were significantly higher in KOR provenance than MUO and 266
SUO (Table 4). Emission rates of α-pinene were significantly higher in KOR than MUO and 267
marginally significantly (p<0.1) between KOR and other provenances (Table 4). Similar 268
differences between KOR and other provenances were also observed for minor compounds, 269
limonene, bornyl acetate and γ-cadinene (Table 4). Unstandardized and temperature- 270
standardized emission rates of individual compounds showed similar differences between 271
the provenances (data not shown).
272
The PLSR analysis showed that emission rates of α-pinene, Δ-3-carene, myrcene, 273
tricyclene, p-cymene, camphor and bornyl acetate were positively related to the dry mass of 274
the needle litter and emission rates of myrcene and bornyl acetate negatively related to the 275
coverage of mosses (Fig. 3, Fig. S13).
276 277
4. Discussion 278
3 Supplementary data are available with the article through the journal Web site
13
This study showed that Scots pine provenance affected BVOC emissions profile of the 279
shoots and forest floor, and high carene and high pinene chemotypes similar to previous 280
studies were found. Highest proportion of Δ-3-carene both in the shoots and forest floor 281
emissions in the southernmost provenance SAA agree with the previously reported 282
differences in Δ-3-carene concentrations in pine shoot (needles+stems) (Nerg et al. 1994), 283
needle and wood concentrations (Manninen et al. 2002) and stump emissions (Kivimäenpää 284
et al. 2012) between the same provenances. Moreover, these studies showed a similar 285
increase in the proportion of α-pinene with latitude, and decreases in sabinene, terpinolene 286
and γ-terpinene with latitude, as reported here for shoot emission profile. Thus, terpene 287
profile is very similar for concentrations and volatile fraction of different plant parts.
288
Positive correlation between Δ-3-carene and terpinolene, as well as between limonene and 289
α-pinene and β-pinene emissions, observed in shoots of this study, are in line with results by 290
Bäck et al. (2012) of Scots pine shoots in a natural forest stand in southern Finland. Positive 291
correlation between 3-carene and terpinolene also in Scots pine oleoresin composition has 292
been reported (Baradat and Yazdani 1988). These observations might be related to the 293
differences in carbon sources in synthesis and emission of these “groups” of monoterpenes.
294
CO2 labeling experiment by Lüpke et al. (2017) showed that α-pinene, β-pinene, limonene 295
and myrcene in Scots pine monoterpene emissions were rapidly labeled with 13C indicating 296
de novo synthesis while Δ-3-carene was almost non-labeled indicating carbon source from a 297
storage pool.
298
Forest floor emitted mainly monoterpenes, α-pinene, Δ-3-carene and camphene being 299
the main compounds, similarly as reported by Aaltonen et al. (2011) from Scots pine- 300
dominated forest in southern Finland. PLSR analysis showed that emission rates from forest 301
floor were largely explained by the amount of needle litter. Thus, needle litter likely 302
explained some of the differences in the emission profile of the forest floor between the 303
14
provenances. Highest proportion of Δ-3-carene from forest floor under the trees of the 304
southernmost provenance SAA is in accordance with the shoot emission profile and needle 305
concentrations (Manninen et al. 2002), but the similar latitudinal difference in forest floor 306
emission profile between the other provenances was not observed. One reason may be that 307
both Δ-3-carene and α-pinene dominate emissions from fungal species that decompose pine 308
litter (Isidorov et al. 2016), and potentially different amounts of fungal hyphae and activity 309
of these decomposers between the provenances may mask the emissions from decaying 310
litter. On the other hand, emissions of ectomycorrhiza species occurring in Scots pine forest 311
were dominated by linalool and limonene and those of endophyte species by sesquiterpenes 312
(Bäck et al. 2010). In a study by Ditengou et al. (2015) sesquiterpenes were the main 313
compounds emitted by ectomycorrhizal fungi species typical of Scots pine. Thus, the 314
influence of ectomycorrhiza on forest floor emissions and profiles may have been of minor 315
importance in this study. The difference in terpene profile between forest floor and shoot 316
emissions supports the role of litter as a major emission source of forest floor and that the 317
majority of the litter was decomposing, not fresh. Specifically, the higher proportion of 318
tricyclene, camphene and α-pinene but lower proportion of limonene in the forest floor 319
emission compared to the shoot emissions are consistent with results by Kainulainen and 320
Holopainen (2002) who followed monoterpene concentrations in decomposing Scots pine 321
needles for 19 months and compared the concentration to the freshly cut and the living 322
needles in Central Finland. Isidorov et al. (2010) reported similar terpene profile changes in 323
pine litter concentration and BVOC emissions in Poland, Central Europe. Kainulainen and 324
Holopainen (2002) also reported increase in concentrations of oxygenated monoterpenes, 325
such as verbenol and verbenone, in decomposing needle litter, but these compounds were 326
not observed in forest floor BVOC emissions in this study. The reason may be the lower 327
volatility of oxygenated monoterpenes and their hydrophilicity and their consequent 328
15
dissolution to moist forest floor. Litter from forest floor emits also other volatile compounds, 329
such as C1 – C2 compounds methanol, acetone, and acetaldehyde (Greenberg et al. 2012) 330
that could not be measured by the technique used in our study.
331
The emission rates of monoterpenes from forest floor were highest in the KOR 332
provenance and lowest in the two northernmost provenances, SUO and MUO, and the 333
differences were explained by the amount of needle litter. Lowest amount of litter leading to 334
lowest terpenoid emission rates in the northern provenances is in accordance with previously 335
reported shorter shoots with lower biomass in the northern than southern provenances 336
(Manninen et al. 1998), as well as with the shortest needle lengths in some years in the 337
northernmost MUO provenance, observed in this study. Moreover, at the age of seven years, 338
KOR was the best growing provenance (Manninen et al. 2002). Terpene concentration of the 339
needle litter hardly explains differences in the emission rates, because the needle terpene 340
concentrations increased with latitude towards the North (Manninen et al. 1998). Scots pine 341
roots are also a source of BVOC emissions (Lin et al. 2007), but their contribution to 342
emissions could not be measured in this study. However, if the growth responses in the roots 343
are the same as in the shoots (Manninen et al. 1998), smaller root volume and root BVOC 344
emission could be expected from the northern provenances.
345
Our results about the needle litter as a major emission source from forest floor 346
emissions supports the previous study conducted in pine-dominated forests (Mäki et al.
347
2017). Mäki et al. (2017) also showed that litterfall and the fraction of needles in the litter 348
explained the autumnal peak monoterpene emissions from forest floor in Scots pine 349
dominated forests in southern Finland. Autumnal litterfall did not take place before the 350
BVOC sampling from the forest floor in Suonenjoki in the study year. The oldest needle 351
generation (fourth in the study region) senesces, i.e., rapidly turns yellow, at the end of 352
August – beginning of September, and drops later in Suonenjoki region (Kivimäenpää and 353
16
Sutinen 2007). However, some needles from the lower parts of the canopy can drop earlier 354
during the growing season, which was observed in all provenances here. The BVOC 355
composition of the forest floor supports the role of decomposing and not freshly fallen litter 356
as an emission source (Kainulainen and Holopainen 2002), as discussed above. Volatile 357
emissions from forest floor peaks also in spring or beginning of growing season, and are 358
influenced e.g. by growth state of the vegetation and air temperature (Aaltonen et al. 2011).
359
Thus, emission rates of this study should be compared to the other measurements conducted 360
in the middle or end of the growing season before litter fall. The average unstandardized 361
emission rate of BVOCs, 50 µg m-2 h-1, consisting primarily of monoterpenes in our study, is 362
similar to emission rate of total monoterpenes, 49 µg m-2 h-1, reported from the floor of a 363
forest stand dominated by 55-year old Scots pines in southern Finland also at the end of the 364
growing season (Mäki et al. 2017). Forest floor of mixed Scots pine and Norway spruce 365
forest in the southern Sweden also emitted monoterpenes up to 50 µg C m-2 h-1 during the 366
growing season (June-September) (Janson et al. 1999). Mäki et al. (2017) summarized the 367
previous estimations of boreal forest floor emissions, which showed high variation, ranging 368
from 0-373 µg m-2 h-1. Our study showed that tree provenance is an additional source of 369
variation, as it caused 65 % differences to forest floor emissions rates.
370
Negative relationship between myrcene and bornyl acetate emissions with moss 371
coverage in our study may be because mosses have acted as sinks of monoterpenes (Mäki et 372
al. 2017). Adhesion of myrcene on mosses and decomposition by ozone may be another 373
explanation to negative relationship between myrcene emission rates and moss coverage.
374
Myrcene was shown to adhere on plant surfaces from a surrounding source and being a 375
compound sensitive to oxidation by slightly elevated, on average 42 ppb, ozone 376
concentrations (Mofikoya et al. 2017). In Kuopio, 40 km NE from the study area, monthly 377
means for ozone concentration were 29 ppb in June, 30 ppb in July and 26 ppb in August 378
17
(Kivimäenpää et al. 2017), and in Ähtäri EMEP station (62 °33´) 27, 31 and 24 ppb, 379
respectively (Hjellbrekke et al. 2012) during the study year. Higher coverage of soil mosses 380
in the two northernmost provenances is likely a consequence of lower amount of litter, 381
which have allelopathic influence on understory vegetation growth (Reigosa and González 382
2006). For example, monoterpene-rich conifer needle litter may reduce N mineralization of 383
forest soil (Paavolainen et al. 1998) and thus, may have reduced N availability of mosses for 384
growth under trees of provenances KOR and SAA. Higher moss coverages in MUO and 385
SUO may also have physically restricted potential BVOC emissions from roots and soil 386
microbiota.
387
Scots pine is one of the species that will benefit from climate change in northern 388
Europe (Reich and Oleksyn 2008). Shoot biomass of Scots pine will increase in a warmer 389
climate (Rasheed et al. 2017), and thus the amount of litter and forest floor BVOC emissions 390
will increase. Moreover, warming increases BVOC emissions from pine foliage 391
(Kivimäenpää et al. 2016). Therefore, BVOC emission rates are expected to increase at the 392
forest level. Plant VOCs have a cooling net effect on climate (Unger 2014), thus Scots pine 393
and boreal forests can provide valuable ecosystem services against climate change.
394
Warming affects also the BVOC composition. For example, Kivimäenpää et al. (2016) 395
showed that 3-carene emissions from Scots pine shoots were not affected by long-term 396
warming, while α-pinene emissions were increased by a factor of 1.5-2. Such changes, as 397
well as Scots pine chemotypes, can affect oxidative properties of the atmosphere under 398
climate change because α-pinene is more reactive. For example, lifetime of α-pinene with O3
399
is only 4.6 h while that of Δ-3-carene is 11 h in similar atmospheric conditions (Atkinson 400
and Arey 2003). Our sampling method may underestimate the emission rates of highly 401
reactive sesquiterpenes (Atkinson and Arey 2003) and cause variation in BVOC profile 402
because sesquiterpenes can adhere to enclosure surfaces particularly at low temperatures 403
18
(Schaub et al. 2010). Global BVOC emissions are quantified using models such as MEGAN 404
or ORCHIDEE, which take into account emission profiles from different plant functional 405
types (Messina et al. 2016). The proportions of the two major monoterpenes α-pinene and Δ- 406
3-carene in our study (28 and 26 %, respectively ) and in the ORCHIDEE model for the 407
plant functional type ‘boreal needleleaf evergreen tree’ (35 and 18 %, respectively) (Messina 408
et al. 2016) agrees well with that Scots pine is the major evergreen tree species in the boreal 409
region.
410
Estimations of the contribution of forest floor on forest stand emissions are variable.
411
For example, Räisänen et al. (2009) calculated that one-quarter of emissions from the period 412
June-September could originate from forest floor of mature Scots pine forest, whereas 413
according to Janson et al. (1999) forest floor emissions were a few percents of emissions of 414
mixed Norway spruce and Scots pine forest. This study suggests that the chemotype-related 415
effects of Scots pine provenances on the atmosphere may differ over a 1000 km South-North 416
transect in northern Europe. Emissions from forest floor will have an additive effect on 417
forest scale BVOC profile. The influence of tree chemotypes and litter should be considered 418
in modeling BVOC emissions and their impact on atmospheric quality and climate 419
parameters.
420 421
Acknowledgments 422
University of Eastern Finland (spearhead project no. 929714) and Academy of Finland 423
(M.K. and J.K.H, project no. 278424) are acknowledged for the financial support.
424
19 References
Aaltonen, H., Pumpanen, J., Pihlatie, M., Hakola, H., Hellén, H., Kulmala, L., Vesala, T., and Bäck, J. 2011. Boreal pine forest floor biogenic volatile organic compound
emissions peak in early summer and autumn. Agric. For. Meteor. 151(6): 682-691. doi:
https://doi.org/10.1016/j.agrformet.2010.12.010.
Atkinson, R., and Arey, J. 2003. Gas-phase tropospheric chemistry of biogenic volatile organic compounds: a review. Atmos. Environ 37: 197-219. doi:
https://doi.org/10.1016/S1352-2310(03)00391-1.
Bäck, J., Aalto, J., Henriksson, M., Hakola, H., He, Q., and Boy, M. 2012. Chemodiversity of a Scots pine stand and implications for terpene air concentrations. Biogeosciences 9(2): 689-702. doi: 10.5194/bg-9-689-2012.
Bäck, J., Aaltonen, H., Hellén, H., Kajos, M.K., Patokoski, J., Taipale, R., Pumpanen, J., and Heinonsalo, J. 2010. Variable emissions of microbial volatile organic compounds (MVOCs) from root-associated fungi isolated from Scots pine. Atmos. Environ. 44:
3651-3659. doi: https://doi.org/10.1016/j.atmosenv.2010.06.042.
Baradat, P., and Yazdani, R. 1988. Genetic expression for monoterpenes in clones of Pinus sylvestris grown on different sites. Scand. J. For. Res. 3(1-4): 25-36.
Ditengou, F.A., Müller, A., Rosenkranz, M., Felten, J., Lasok, H., van Doorn, M.M., Legué, V., Palme, K., Schnitzler, J., and Polle, A. 2015. Volatile signalling by sesquiterpenes from ectomycorrhizal fungi reprogrammes root architecture. Nat. Commun. 6: 6279.
doi: 10.1038/ncomms7279.
Gauthier, S., Bernier, P., Kuuluvainen, T., Shvidenko, A.Z., and Schepaschenko, D.G. 2015.
Boreal forest health and global change. Science 349(6250): 819-822. doi:
10.1126/science.aaa9092.
20
Ghimire, R., Markkanen, J.-M., Kivimäenpää, M., Lyytikäinen-Saarenmaa, P., and
Holopainen, J. 2013. Needle removal by pine sawfly larvae increases branch-level VOC emissions and reduces below-ground emissions of Scots pine. Environ. Sci. Technol.
47(9): 4325-32. doi: 10.1021/es4006064.
Ghirardo, A., Koch, K., Taipale, R., Zimmer, I., Schnitzler, J.-P., and Rinne, J. 2010.
Determination of de novo and pool emissions of terpenes from four common
boreal/alpine trees by 13CO2 labelling and PTR-MS analysis. Plant Cell Environ. 33(5):
781-792. doi: 10.1111/j.1365-3040.2009.02104.x.
Greenberg, J.P., Asensio, D., Turnipseed, A., Guenther, A.B., Karl, T., and Gochis, D. 2012.
Contribution of leaf and needle litter to whole ecosystem BVOC fluxes. Atmos.
Environ. 59: 302-311. doi: 10.1016/j.atmosenv.2012.04.038.
Guenther, A.B., Zimmerman, P.R., Harley, P.C., Monson, R.K., and Fall, R. 1993. Isoprene and monoterpene emission rate variability: model evaluations and sensitivity analyses. J.
Geophys. Res. Atmos. 98(D7): 12609-12617. doi: 10.1029/93JD00527.
Heijari, J., Blande, J.D., and Holopainen, J.K. 2011. Feeding of large pine weevil on Scots pine stem triggers localised bark and systemic shoot emission of volatile organic compounds. Environ. Exp. Bot. 71: 390-398. doi:
https://doi.org/10.1016/j.envexpbot.2011.02.008
Himanen, S.J., Blande, J.D., Klemola, T., Pulkkinen, J., Heijari, J., and Holopainen, J.K.
2010. Birch (Betula spp.) leaves adsorb and re-release volatiles specific to neighbouring plants – a mechanism for associational herbivore resistance? New Phytol. 186(3): 722- 732. doi: 10.1111/j.1469-8137.2010.03220.x.
21
Hjellbrekke, A.G., Solberg, S., and Fjæraa, A.M. 2012. Ozone measurements 2010.
EMEP/CCC-Report 2/2012. Norwegian Institute for Air Research. Available from https://www.nilu.no/projects/ccc/reports.html [Accessed 29 January 2018].
Holopainen, J.K. 2011. Can forest trees compensate for stress-generated growth losses by induced production of volatile compounds? Tree Physiol. 31(12): 1356-1377. doi:
10.1093/treephys/tpr111.
Huttunen, S., and Manninen, S. 2013. A review of ozone responses in Scots pine (Pinus sylvestris). Environ. Exp. Bot. 90: 17-31. doi:
https://doi.org/10.1016/j.envexpbot.2012.07.001
Isidorov, V.A., Smolewska, M., Purzyńska-Pugacewicz, A., and Tyszkiewicz, Z. 2010.
Chemical composition of volatile and extractive compounds of pine and spruce leaf litter in the initial stages of decomposition. Biogeosciences 7(9): 2785-2794. doi:
https://doi.org/10.5194/bg-7-2785-2010.
Isidorov, V., Tyszkiewicz, Z., and Pirożnikow, E. 2016. Fungal succession in relation to volatile organic compounds emissions from Scots pine and Norway spruce leaf litter- decomposing fungi. Atmos. Environ. 131: 301-306. doi:
https://doi.org/10.1016/j.atmosenv.2016.02.015.
Janson, R., De Serves, C., and Romero, R. 1999. Emission of isoprene and carbonyl compounds from a boreal forest and wetland in Sweden. Agric. For. Meteor. 98-99:
671-681. doi: https://doi.org/10.1016/S0168-1923(99)00134-3.
Joutsensaari, J., Yli-Pirilä, P., Korhonen, H., Arola, A., Blande, J.D., Heijari, J.,
Kivimäenpää, M., Mikkonen, S., Hao, L., Miettinen, P., Lyytikäinen-Saarenmaa, P., Faiola, C.L., Laaksonen, A., and Holopainen, J.K. 2015. Biotic stress accelerates
22
formation of climate-relevant aerosols in boreal forests. Atmos. Chem. Phys. 15(21):
12139-12157. doi: https://doi.org/10.5194/acp-15-12139-2015.
Kainulainen, P., and Holopainen, J.K. 2002. Concentrations of secondary compounds in Scots pine needles at different stages of decomposition. Soil Biol. Biochem. 34(1): 37- 42. doi: https://doi.org/10.1016/S0038-0717(01)00147-X.
Kivimäenpää, M., Ghimire, R.P., Sutinen, S., Häikiö, E., Kasurinen, A., Holopainen, T., and Holopainen, J.K. 2016. Increases in volatile organic compound emissions of Scots pine in response to elevated ozone and warming are modified by herbivory and soil nitrogen availability. Eur. J. For. Res. 135(2): 343-360. doi: 10.1007/s10342-016-0939-x.
Kivimäenpää, M., Magsarjav, N., Ghimire, R., Markkanen, J., Heijari, J., Vuorinen, M., and Holopainen, J.K. 2012. Influence of tree provenance on biogenic VOC emissions of Scots pine (Pinus sylvestris) stumps. Atmos. Environ. 60: 477-485.
10.1016/j.atmosenv.2012.07.018.
Kivimäenpää, M., and Sutinen, S. 2007. Microscopic structure of Scots pine (Pinus
sylvestris (L.)) needles during ageing and autumnal senescence. Trees 21(6): 645-659.
doi: https://doi.org/10.1007/s00468-007-0157-8.
Kivimäenpää, M., Sutinen, S., Valolahti, H., Häikiö, E., Riikonen, J., Kasurinen, A.,
Ghimire, R.P., Holopainen, J.K., and Holopainen, T. 2017. Warming and elevated ozone differently modify needle anatomy of Norway spruce (Picea abies) and Scots pine (Pinus sylvestris). Can. J. For. Res. 47(4): 488-499. doi: https://doi.org/10.1139/cjfr- 2016-0406.
Kulmala, M., Nieminen, T., Chellapermal, R., Makkonen, R., Bäck, J., and Kerminen, V.
2013. Climate feedbacks linking the increasing atmospheric CO2 concentration, BVOC
23
emissions, aerosols and clouds in forest ecosystems. In Biology, controls and models of tree volatile organic compound emissions. Edited by Ü. Niinemets and R. Monson. Tree Physiology, vol 5. Springer, Dordrecht pp. 489-508.
Lee, A., Goldstein, A.H., Kroll, J.H., Ng, N.L., Varutbangkul, V., Flagan, R.C., and Seinfeld, J.H. 2006. Gas-phase products and secondary aerosol yields from the photooxidation of 16 different terpenes. J. Geophys. Res. Atmos. 111(D7). doi:
1029/2005JD006437.
Lerdau, M., and Slobodkin, L. 2002. Trace gas emissions and species-dependent ecosystem services. Trends Ecol. Evol. 17(7): 309-312. https://doi.org/10.1016/S0169-
5347(02)02535-1.
Lin, C., Owen, S.M., and Peñuelas, J. 2007. Volatile organic compounds in the roots and rhizosphere of Pinus spp. Soil Biol. Biochem. 39(4): 951-960. doi:
https://doi.org/10.1016/j.soilbio.2006.11.007.
Loreto, F., and Schnitzler, J.P. 2010. Abiotic stresses and induced BVOCs. Trends Plant Sci.
15(3):154-166. doi: 10.1016/j.tplants.2008.12.006.
Lüpke, M., Leuchner, M., Steinbrecher, R., & Menzel, A. 2017. Quantification of
monoterpene emission sources of a conifer species in response to experimental drought.
AoB Plants, 9(5): plx045. doi: http://doi.org/10.1093/aobpla/plx045.
Mäki, M., Heinonsalo, J., Hellén, H., and Bäck, J. 2017. Contribution of understorey vegetation and soil processes to boreal forest isoprenoid exchange. Biogeosciences 14(5): 1055-1073. doi: https://doi.org/10.5194/bg-14-1055-2017.
24
Manninen, A.-M., Vuorinen, M., and Holopainen, J.K. 1998. Variation in growth, chemical defense, and herbivore resistance in Scots pine provenances. J. Chem. Ecol. 24(8):
1315-1331. doi: https://doi.org/10.1023/A:1021222731991.
Manninen, A.-M., Tarhanen, S., Vuorinen, M., and Kainulainen, P. 2002. Comparing the variation of needle and wood terpenoids in Scots pine provenances. J. Chem. Ecol.
28(1): 211-228. doi: https://doi.org/10.1023/A:1013579222600.
Messina, P., Lathière, J., Sindelarova, K., Vuichard, N., Granier, C., Ghattas, J., Cozic, A., and Hauglustaine, D.A. 2016. Global biogenic volatile organic compound emissions in the ORCHIDEE and MEGAN models and sensitivity to key parameters. Atmos. Chem.
Phys. 16(22): 14169-14202. doi:10.5194/acp-16-14169-2016.
Mofikoya, A.O., Kim, T.H., Abd El-Raheem, A.M., Blande, J.D., Kivimäenpää, M., and Holopainen, J.K. 2017. Passive adsorption of volatile monoterpene in pest control:
Aided by proximity and disrupted by ozone. J. Agric. Food Chem. 65(44): 9579-9586.
doi: 10.1021/acs.jafc.7b03251.
Mumm, R., and Hilker, M. 2006. Direct and indirect chemical defence of pine against folivorous insects. Trends Plant Sci. 11(7): 351-358.
https://doi.org/10.1016/j.tplants.2006.05.007.
Muona, O., Hiltunen, R., Morén, E., and Shaw, D.V. 1986. Analysis of monoterpene
variation in natural stands and plustrees of Pinus sylvestris in Finland. Silva Fenn. 20(1):
1-8. Available from https://helda.helsinki.fi/handle/10138/15435 [Accessed 29 January 2018].
Nerg, A., Kainulainen, P., Vuorinen, M., Hanso, M., Holopainen, J.K., and Kurkela, T.
1994. Seasonal and geographical variation of terpenes, resin acids and total phenolics in
25
nursery grown seedlings of Scots pine (Pinus sylvestris L.). New Phytol. 128(4): 703- 713. doi: 10.1111/j.1469-8137.1994.tb04034.x.
Paavolainen, L., Kitunen, V., and Smolander, A. 1998. Inhibition of nitrification in forest soil by monoterpenes. Plant Soil 205(2): 147-154. doi:
https://doi.org/10.1023/A:1004335419358.
Peltola, A. 2014. Finnish Statistical Yearbook of Forestry 2014. Finnish Forest Research Institute. pp. 1-426. Available from
http://www.metla.fi/metinfo/tilasto/julkaisut/vsk/2014/ [Accessed 29 January 2018]
Peñuelas, J., and Staudt, M. 2010. BVOCs and global change. Trends Plant Sci. 15(3): 133- 144. doi: https://doi.org/10.1016/j.tplants.2009.12.005.
Räisänen, T., Ryyppö, A., and Kellomäki, S. 2009. Monoterpene emission of a boreal Scots pine (Pinus sylvestris L.) forest. Agric. For. Meteor. 149(5): 808-819. doi:
https://doi.org/10.1016/j.agrformet.2008.11.001.
Rasheed, M.U., Kasurinen, A., Kivimäenpää, M., Ghimire, R., Häikiö, E., Mpamah, P., Holopainen, J.K., and Holopainen, T. 2017. The responses of shoot-root-rhizosphere continuum to simultaneous fertilizer addition, warming, ozone and herbivory in young Scots pine seedlings in a high latitude field experiment. Soil Biol. Biochem. 114: 279- 294. doi: https://doi.org/10.1016/j.soilbio.2017.07.024.
Reich, P.B., and Oleksyn, J. 2008. Climate warming will reduce growth and survival of Scots pine except in the far north. Ecol. Lett. 11(6): 588-597. doi: 10.1111/j.1461- 0248.2008.01172.x.
Reigosa, M.J., and González, L. 2006. Forest ecosystems and allelopathy. In Allelopathy: A physiological process with ecological implications. Edited by M., Reigosa, N. Pedrol
26
and L. González. Springer, Netherlands pp. 451-563. doi: https://doi.org/10.1007/1- 4020-4280-9_20.
Schaub, A., Blande, J.D., Graus, M., Oksanen, E., Holopainen, J.K., and Hansel, A. 2010.
Real-time monitoring of herbivore induced volatile emissions in the field. Physiol.
Plant. 138: 123-133. doi: 10.1111/j.1399-3054.2009.01322.x.
Scott, C.E., Monks, S.A., Spracklen, D.V., Arnold, S.R., Forster, P.M., Rap, A., Äijälä, M., Artaxo, P., Carslaw, K.S., Chipperfield, M.P., Ehn, M., Gilardoni, S., Heikkinen, L., Kulmala, M., Petäjä, T., Reddington, C.L.S., Rizzo, L.V., Swietlicki, E., Vignati, E., and Wilson, C. 2018. Impact on short-lived climate forcers increases projected warming due to deforestation. Nature Commun. 9: 157. doi:10.1038/s41467-017-02412-4.
Tiiva, P., Tang, J., Michelsen, A., and Rinnan, R. 2017. Monoterpene emissions in response to long-term night-time warming, elevated CO2 and extended summer drought in a temperate heath ecosystem. Sci. Total Environ. 580: 1056-1067. doi:
https://doi.org/10.1016/j.scitotenv.2016.12.060.
Trapp, S., and Croteau, R. 2001. Defensive resin biosynthesis in conifers. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52: 689-724. doi: 10.1146/annurev.arplant.52.1.689.
Unger, N. 2014. On the role of plant volatiles in anthropogenic global climate change.
Geophys. Res. Lett. 41(23): 8563-8569. doi:10.1002/2014GL061616.
27 Figure captions
Fig. 1. a) Score and b) loading plots of Principal Component Analysis (PCA) on the proportions of shoot BVOCs from mature Scots pine trees of four provenances Saaremaa (SAA), Korpilahti (KOR), Suomussalmi (SUO) and Muonio (MUO). The variance explained by two first principal components (PC) are shown in parentheses. In a) average scores with SE of PC1 and PC2 of ten trees are shown. Different letters below the bars indicate statistical difference (P < 0.05, pairwise multiple comparison of Kruskal-Wallis test) in the average PC1 score between the provenances. In b) monoterpenoids are shown with normal font, sesquiterpenes in cursive and GLVs emboldened, st1 and st2 refer to unidentified sesquiterpenes.
Fig. 2. Average proportion (+ SE) of emissions α-pinene and Δ-3-carene from total BVOC emissions in shoots of Scots pine trees of four provenances Saaremaa (SAA), Korpilahti (KOR), Suomussalmi (SUO) and Muonio (MUO) and forest floor under the same trees, n=10 for shoots, n=9 for forest floor (expect n=10 for KOR). Different letters above the bars show significant differences between the provenances (P < 0.05 pairwise multiple
comparisons from Kruskal-Wallis test for shoots, Tukey test from ANOVA for forest floor).
Fig. 3. Regression coefficients of partial least squares regression (PLSR) models for the covariance between coverage of vegetation groups, needle litter, bark and cones, air temperature, soil moisture and the emissions of α-pinene (a), myrcene (b) and Δ-3-carene (c). Positive regression coefficients indicate positive relationship and negative ones negative relationship. Error bars show confidence intervals of the regression coefficients. Significant factors (error bars not crossing zero) are shown in black.
Table 1. Average (SE) length (cm) of current (2010) and previous years (2009, 2008, 2007) needles of Scots pines from four provenancesSAA=Saaremaa, KOR=Korpilahti, SUO=Suomussalmi, MUO=Muonio.
Date Needle
generation SAA
58°22´ KOR
62°0´ SUO
65°10´ MUO
67°56´ P-value 29 June 2010 1.4 (0.1) a 1.7 (0.1) ab 1.8 (0.1) b 1.6 (0.1) ab 0.022* 21 July 2010 2.2 (0.2) 2.2 (0.2) 2.4 (0.2) 2.1 (0.1) 0.747* 29 June 2009 2.5 (0.2) 2.2 (0.1) 2.3 (0.2) 2.3 (0.3) 0.649† 29 June 2008 3.2 (0.2) 3.1 (0.2) 3.2 (0.2) 2.6 (0.2) 0.173* 29 June 2007 3.4 (0.2) a 3.3 (0.2) a ‡ 3.3 (0.3) a ‡ 2.3 (0.2) b 0.011*
Note: Different letters between the provenances show statistical difference at P < 0.05 level, n=10.
*ANOVA
† Kruskal-Wallistest
‡ needles left in six of ten trees
Table 2. Average (SE) needle litter (g DW) and coverage (%) (min, max) of vegetation and bark and cones on VOC collection cylinders under Scots pines from four provenancesSAA=Saaremaa, KOR=Korpilahti, SUO=Suomussalmi, MUO=Muonio.
SAA 58°22´ KOR
62°0´ SUO
65°10´ MUO
67°56´ P-value Needle litter 4.7 (0.7) ab 6.6 (0.6) a 3.9 (0.5) b 2.7 (0.7) b 0.001* Mosses 0 (0, 5) a 5 (0, 50) a 30 (0, 100) ab 40 (0, 100) b 0.001† Lichens 0 (0, 5) 0 (0, <5) 0 (0, <5) 10 (0, 45) 0.520† Graminoids 5 (0, 40) 0 (0, <5) 0 (0, <5) 0 (0, <5) 0.506† Bark and cones 0 (0, 0) 0 (0, <5) 0 (0, <5) 0 (0, <5) 0.507†
Note: Different letters between the provenances show statistical difference at P < 0.05 level, n=10 for KOR, n=9 for SAA, SUO, MUO.
*ANOVA
† Kruskal-Wallistest
Table 3. Proportions (%) of key compounds found from both shoot and forest floor emissions as an average (SE) of all Scots pine provenances, n=35.
Compound Shoot Forest floor P-value Tricyclene <0.1 (<0.1) 0.6 (0.2) 0.011† α-Pinene 27.9 (2.7) 52.8 (2.0) <0.001* Camphene 2.6 (0.5) 10.5 (1.3) <0.001* β-Pinene 5.0 (1.1) 3.7 (0.7) 0.276* Myrcene 8.7 (1.5) 0.7 (0.2) 0.011† Δ-3-Carene 25.8 (4.3) 28.9 (1.2) 0.437* p-Cymene 0.4 (0.1) 0.4 (0.1) 0.185† Limonene 11.3 (2.5) 1.4 (0.4) <0.001* Camphor 0.2 (0.1) 0.6 (0.2) 1.000†
Note: P-values show differences in proportions between shoots and forest floor.
*paired samples T-test
† Wilcoxon signed ranks test
Table 4. Emission rates (µg m-2 h-1) of individual compounds and total terpenoids from forest floor under trees from four different Scots pine provenances (KOR=Korpilahti, SAA=Saaremaa,
SUO=Suomussalmi, MUO=Muonio) expressed as standardized to + 20 °C, total C at +20 °C and total non-standardized emissions.
SAA 58°22´ KOR 62°0´ SUO 65°10´ MUO 67°56´ P-value
Tricyclene 0.6 (0.5) 1.2 (0.7) 0.4 (0.2) 0.3 (0.3) 0.704†
α-Pinene 31.5 (9.5) ab 69.4 (16.8) a 31.4 (8.2) ab 25.5 (5.0) b 0.024*
Camphene 5.0 (1.8) 9.9 (2.4) 7.7 (3.1) 3.9 (2.1) 0.231†
β-Pinene 2.1 (0.7) 3.6 (1.3) 2.8 (1.1) 1.0 (0.5) 0.474†
Myrcene 0.4 (0.3) 1.6 (0.7) 0.2 (0.2) 0.2 (0.2) 0.229†
Δ-3-Carene 18.8 (4.1) ab 31.4 (4.14) a 15.2 (4.3) b 12.1 (2.3) b 0.009*
p-Cymene 0 (0) 0.7 (0.3) 0.5 (0.3) 0.1 (0.1) 0.134†
Limonene 0.4 (0.4) 2.5 (0.8) 1.2 (0.6) 0.2 (0.2) 0.040†
β-Phellandrene 0.1 (0.1) 0.2 (0.2) 0 (0) 0 (0) 0.322†
(E)-β-Ocimene 0 (0) 0.3 (0.2) 0 (0) 0 (0) 0.136†
Camphor 0.2 (0.2) 1.2 (0.7) 0.2 (0.2) 0.2 (0.2) 0.567†
Bornyl acetate 0 (0) a 0.7 (0.3) b 0 (0) a 0 (0) a 0.008† γ-Cadinene 0 (0) 1.1 (0.6) 0 (0) 0 (0) 0.036† δ-Cadinene 0 (0) 0.1 (0.3) 0 (0) 0 (0) 0.136† Total +20 °C 59.1 (13.6) b 124.3 (18.7) a 59.5 (15.3) b 43.5 (8.7) b 0.007* Total C +20 °C 47.2 (11.6) ab 97.9 (14.9) a 51.2 (14.5) b 34.4 (7.0) b 0.007* Total unstand. 40.5 (11.8) ab 81.5 (13.3) a 46.3 (14.5) ab 28.9 (5.9) b 0.011* Note: Different letters between the provenances show statistical difference at p < 0.05 level. n=10 for KOR, n=9 for SAA, SUO and MUO.
*ANOVA
† Kruskal-Wallistest
Table S1. Temperature-standardized (20 °C) emissions rates of terpenes and actual emissions of green leaf volatiles (ng h-1 cm-1 shoot length) collected from branches of Scots pine trees of four provenances (SAA=Saaremaa, KOR=Korpilahti, SUO=Suomussalmi, MUO=Muonio). Values are averages (SE) of ten trees.
Compound SAA 58°22´ KOR 62°0´ SUO 65°10´ MUO 67°56´
-Pinene 28.2 (12.8) 24.8 (8.4) 99.2 (43.0) 191.6 (83.1)
Camphene 1.1 (0.2) 1.6 (0.4) 2.9 (0.8) 4.8 (1.2)
Sabinene 3.4 (1.7) 2.4 (1.2) 4.5 (2.7) 3.4 (1.3)
-Pinene 1.4 (0.6) 19.3 (14.1) 9.8 (3.8) 24.8 (17.0)
Myrcene 21.7 (18.3) 12.9 (7.2) 27.2 (9.6) 137.4 (70.4)
-Phellandrene 0.7 (0.4) 0.4 (0.2) 0.9 (0.5) 1.0 (0.4)
3-Carene 104.3 (54.8) 62.6 (35.7) 125.0 (77.1) 64.9 (41.2)
p-Cymenea 0.8 (0.4) 0.6 (0.2) 0.9 (0.3) 1.1 (0.3)
Limonene 3.0 (1.6) 14.3 (11.6) 30.4 (17.2) 198.9 (85.7)
-Phellandrenea 1.4 (0.6) 7.5 (4.3) 10.0 (5.9) 36.9 (16.6)
1,8-Cineole 1.0 (0.2) 2.5 (1.9) 0.8 (0.3) 2.5 (0.6)
(E)- -Ocimenea 11.8 (11.4) 0.3 (0.1) 0.5 (0.2) 9.0 (6.2)
-Terpinene 0.9 (0.6) 0.5 (0.3) 0.8 (0.4) 0.8 (0.3)
Terpinolene 7.8 (5.2) 4.1 (2.4) 6.6 (3.4) 6.1 (3.4)
Camphor 0.1 (<0.05) 0.1 (0.1) 0.1 (<0.05) 0.4 (0.1)
Bornyl acetate <0.05 (<0.05) 0.3 (0.1) 0.3 (0.1) 0.4 (0.1)
–Terpinenea 0.7 (0.5) 0.4 (0.2) 0.7 (0.3) 0.9 (0.3)
Total
monoterpenes
188.9 (105.0) 155.3 (61.1) 320.5 (118.2) 685.7 (222.8) -Longipinenec 0 (0) <0.05 (<0.05) 0.1 (0.1) <0.05 (<0.05) -Copaene <0.05 (<0.05) 0 (0) 0.1 (<0.05) 0.2 (0.1)
-Bourbobenec 0.1 (0.1) 0.1 (0.1) 0.2 (0.1) 0.5 (0.2)
Longifolene <0.05 (<0.05) 1.2 (1.0) 0.6 (0.2) 0.7 (0.2)
-Caryophyllene 0.1 (0.1) 0.2 (0.1) 0.3 (0.2) 0.6 (0.3)
(E)- -farnesene 0.4 (0.2) 0.4 (0.1) 0.4 (0.2) 3.0 (1.2)
-Humulene 0.1 (0.1) <0.05 (<0.05) <0.05 (<0.05) 0.2 (0.1) Unknown st 1c 0.1 (<0.05) 0.1 (<0.05) 0.1 (0.1) 0.2 (0.1) (E,E)- -
Farnesene c
0.1 (<0.05) <0.05 (<0.05) 0.2 (0.2) 0.1 (0.1) -Selinenec 0.2 (0.1) 0.1 (0.1) <0.05 (<0.05) 0.5 (0.2) -Muurolenec <0.05 (<0.05) 0.1 (<0.05) 0.1 (0.1) 0.2 (0.1) -Selinenec 0.1 (0.1) 0.1 (0.1) <0.05 (<0.05) 0.2 (0.1) _cadinenec 0.1 (<0.05) 0.2 (0.1) 0.1 (<0.05) 0.3 (0.1)
-Cadinene 0.1 (<0.05) 0.3 (0.1) 0.1 (0.1) 0.4 (0.1)
Cis- -bisabolenec 0.1 (<0.05) <0.05 (<0.05) 0.0 (0.0) 0.3 (0.1) Total
sesquiterpenes
5.3 (4.7) 29.7 (11.8) 12.1 (5.2) 42.7 (22.5)
dTotal GLVs 0.2 (0.1) 0.2 (0.2) 0.2 (0.1) 1.2 (0.7) Total VOCs 190.5 (105.3) 158.5 (61.4) 323.0 (119.1) 694.5 (223.9)
a-cEmission rates calculated usinga -Pineneb1,8-Cineole orcLongifole as a reference.
Emissions rates of monoterpenes tricyclene (max. 1.2), linalool (2.7), borneol (0.4), terpinen-4-ol (<0.05), allo-ocimenea (0.3), pinocarvoneb (0.3), verbenoneb (0.3), eucarvoneb (0.6), ocimenea (0.1), piperitoneb(0.7), sesquiterpenes -cubebenec (0.1), longicyclenec (0.4), sativenec (0.3),
aromadendrene (0.2) emitted by 1-3 tree individuals are not shown.
dGLVs consisted mainly of (Z)-3-hexenyl acetat (max. 6.9) and (Z)-3-hexen-1-ol (0.5) and methyl salicylate (0.7).
Fig. S1. Regression coefficients of partial least squares regression (PLSR) models for the
covariance between coverage of vegetation groups, needle litter, bark and cones, air temperature, soil moisture and the emissions of tricyclene (a), p-cymene (b), camphor (c) and bornyl acetate (d).
Positive regression coefficient indicate a positive relationship and negative ones negative
relationship. Error bars show confidence intervals of the regression coefficients. Significant factors (error bars not crossing zero) are shown in black.
