Passive adsorption of neighbouring plant volatiles linked to associational 1 susceptibility in a subarctic ecosystem

1 susceptibility in a subarctic ecosystem. 2 3 Adedayo Mofikoya1, Kazumi Miura1,2,, Toini Holopainen1, Jarmo K. Holopainen1 4 5 1 Department of Environmental and Biological Sciences, University of Eastern Finland, P. O. Box 6 1672, 70211, Kuopio, Finland. 7 2 Institute of Biology, Free University of Berlin, Haderlebener Str.9, 12163, Berlin, Germany 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Corresponding author E-mail: adedayo.mofikoya@uef.fi 29


54
Plants growing in the subarctic are adapted to fast growth owing to the relatively short growing season 55 in the region. The vegetation composition of subarctic ecosystems typically consists of sedges, 56 bryophytes and shrubs; the majority of the tree species in these ecosystems are dwarf or miniature 57 versions of the same species found in warmer climates -a coping mechanism for low nutrient   shrubs, we found a number of other species growing in the understorey (Appendix A1).

146
Under 24 mountain birch trees, a 1m 2 quadrant was marked and a MB branch growing directly above 147 each quadrant was selected for VOC sample collection. The density of the Rt shoots growing in the 148 quadrant was used to place the trees into three different categories; low, medium and high density.

149
The low Rt density group had Rt coverage under the tree of 2-17% in the 1m 2 quadrant (n=8). The 150 medium density group was considered to have an average Rt density, which was 20 -25% of the 151 Biogeosciences Discuss., doi:10.5194/bg-2016-464, 2016 Manuscript under review for journal Biogeosciences Published: 22 November 2016 c Author(s) 2016. CC-BY 3.0 License. quadrant (n = 4). The high Rt density group had a coverage of 40-80% (n = 6). We marked 6 trees 152 with no Rt in the understorey as the control group. Due to the small sample size of the medium Rt 153 density sites, we merged low and medium density into a new 12 tree group -Moderate density. The 154 branch used for VOC sampling as well as three other branches at the same heights were used for 155 arthropod analysis. The damage levels on leaves and number of arthropods on leaves were observed 156 visually. We recorded and counted the species of arthropods on the leaves and the number of leaves 157 with gall mite colonies per branch. We also counted the number of leaves with holes and other injuries 158 per branch and visually estimated the size of the damage area. was the carrier gas. Oven temperature was at 40°C for one minute, then raised to 210°C at 5°C min -1 185 and further to 250°C at 20°C min -1 . The compounds (Terpenes and GLVs) were identified by 186 comparing their mass spectra, retention time and peak with those in the Wiley library and pure 187 standards. A palustrol standard was unavailable, so a ledol standard was used to calculate its emission.

188
Other unavailable compounds were quantified by comparing peak areas to corresponding peak area 189 and amount (ng) of -pinene in pure standard. Emission rates were expressed as ng g -1 LA m -2 h -1 for 190 birch leaves and ng g -1 leaf DW h -1 for R. tomentosum shoots.  The volatile compounds emitted by Rt branches included the monoterpene, -myrcene (58% of total 217 VOC emissions), the sesquiterpene, aromadendrene (8%) and the sesquiterpene alcohols palustrol 218 and ledol (15 and 3% respectively) ( Table 1). There was no difference in the means of Rt compounds 219 emitted from sampled branches from high and moderate Rt quadrants. Rt branches from high density 220 quadrants had higher emission rates per emitting unit (ng g -1 h -1 ) of these four compounds than those 221 from moderate Rt density quadrants -60 vs 38% for -myrcene, 8.9 vs 4.4% for aromadendrene, 16 222 vs 10% and 3.9 vs 2.4% for palustrol and ledol respectively.

224
The total monoterpenoid emissions from MB trees was highest in the control group and then 225 decreased with increasing Rt density. The control (Mann-Whitney U = 4.0; P = 0.026) and moderate 226 Rt (Mann-Whitney U = 14, P = 0.04) groups differed significantly from the high Rt group (Table 2).

227
There was a marginally significant decrease in the emission of -pinene from MB branches across 228 the three Rt groups. In between groups, trees growing above high Rt density had significantly lower 229 -pinene emission compared to the control group (Mann-Whitney U = 3.5, P = 0.037) ( Table 2).

230
There was also a non-significant decreasing trend in sesquiterpene emissions from MB branches with 231 increasing Rt density, the lowest total sesquiterpene emission was from trees growing above high Rt 232 density (Table 2).

233
There was a marginally significant reduction in total MB VOC emission across the three groups 234 (Table 2), control group (Mann-Whitney U = 3, P = 0.015) had significantly higher total volatile 235 emissions compared to high Rt group (Table 2). The emission rates of -myrcene, the major monoterpene released by Rt increased across the MB tree 238 groups from control to high Rt quadrants (Table 2). There was higher emission of -myrcene from 239 MB trees in high and moderate Rt quadrants compared to control (Mann -Whitney U = 3.5, P = 240 0.015; and Mann-Whitney U = 18.5, P = 0.05 respectively) ( Table 2).

241
The other terpenoid compounds recovered from sampling the MB branches were sesquiterpenoid 242 compounds emitted characteristically by R. tomentosum (palustrol, ledol and aromadendrene). There 243 was increase in the palustrol recovery from MB trees across the three treatment groups (Table 2).

244
Palustrol emissions from MB branches sampled from both moderate (Mann-Whitney U = 9, P = 245 0.01) and high density (Mann-Whitney U = 3, P = 0.015) Rt quadrants were significantly higher than 246 emissions from control trees (Table 2). tr e e s w a s r e c ov e r e d f r om f ol i a g e i n b oth m od e r a te a n d h i gh d e n si ty q u a d r a n ts a t r a te s b el ow 250 5ng/m 2 /h. (Table 2).

251
There was also an increase in total adhered emissions (TAE) (i.e. -myrcene, aromadendrene, ledol 252 and palustrol) across the three groups. The adhered emissions in MB branches in moderate Rt (Mann-253 Whitney U = 11, P = 0.018) and high Rt (Mann-Whitney U = 3.5, P = 0.05) quadrants were higher 254 than those from the control group (Table 2).  The relationship between the recovery of adhered compounds and temperature was tested only among 263 treatment groups -moderate and high Rt. The recovery of all adhered compounds except -myrcene 264 showed a negative correlation with temperature. The association of temperature with the recovery of 265 adhered compounds from MB leaves was strongest in ledol (n = 18, b = -0.598, P = 0.002) (Fig. 2a).

271
Herbivore density was very low during our sampling. We found a small number of insects (aphids, with -myrcene recovery. (Table 3).

287
The dominating Rt monoterpene, -myrcene was also recovered in significantly higher amounts in 288 MB trees growing above high density Rt shoots compared to the control group in our study. Although,

289
-myrcene is among the monoterpenes synthesized and emitted in minor quantities by Betula spp.    to Rt-exposed birch leaves. Similarly, only one birch feeding herbivore, Phaedinus flavipes preferred 361 non-exposed birch leaves to Rt-exposed leaves when two different herbivores were tested (Himanen