Herbivore and rust damage of Silver birch (Betula pendula) leaves under climate warming

(1)

Faculty of Science and Forestry

HERBIVORE AND RUST DAMAGE OF SILVER BIRCH (Betula pendula) LEAVES UNDER CLIMATE WARMING

Kishor Kumar Roy

MASTER‘S THESIS WOOD MATERIALS SCIENCE JOENSUU 2020

(2)

Roy, K. K. 2020. Herbivore and rust damage of Silver birch (Betula pendula) leaves under climate warming. University of Eastern Finland, Faculty of Science and Forestry, Master‘s thesis in Department of Wood Materials Science. 41p

ABSTRACT

The alarming climate change is expected to increase herbivore and disease outbreaks in boreal forest. Silver birch (Betula pendula Roth) were studied to investigate whether herbivore and rust damage on leaves will increase because of warming climate, and whether this damage can be linked to warming induced changes in quality of leaves (nitrogen and tannin content).

Seedlings were grown in field condition and elevated temperature (ambient +1.5 °C) and examined for three growing seasons (May–August, 2016‒2018). Leaf samples were collected from ten randomly selected individuals in 2016 and later on from five individuals from each control and treated plots. Leaf features such as leaf area, number of trichomes, area of herbivore damage, rusts spots, leaf nitrogen concentration, water percentage and condensed tannin concentrations were examined to estimate their changes under elevated temperature and to observe their correlation with herbivore and rust damage.

Elevated temperature induced increase in leaf areas and in leaf nitrogen concentration and decrease in condensed tannin concentration including both soluble and insoluble tannins. No difference was noted for leaf eaten area but there was a tendency of more herbivore damage under elevated temperature. Results revealed that, leaves with larger area and low number of rust spots were more likely to be affected by herbivores. The intensity of herbivore damage had positive correlation with leaf nitrogen and leaf area and negative correlation with number of rust spots, trichomes andconcentration of tannins. Less rust infestation was noted in treated leaves which may be drying leaf of surface by heaters. However, a negative correlation between soluble tannin concentration and rust was observed in the leaves under ambient conditions. My results indicate that rise in temperature, affects different leaf features which may increase the susceptibility of silver birch leaves to herbivore damage and rust infestation on future climate.

Keywords: Silver birch,herbivore, rust, climate warming

(3)

FOREWORD

This study was conducted under supervision of Virpi Virjamo, PhD, postdoctoral researcher at University of Eastern Finland. I would like to convey earnest gratitude to my supervisor for providing me the opportunity to perform my practical training, laboratory work, statistical analysis and manuscript writing and the entire related master‘s thesis work under her supervision. Her valuable instructions, inspiration, and untiring replies to my all quires throughout the experiment were immense help for successful completion of this thesis work. I would also like to acknowledge her for providing me all the collected data from her ongoing research project ―Nordic forest in changing climate: ecosystem level approach‖. Moreover, I express heartfelt thanks to my parents and wife who always supported and encouraged me to complete the work successfully.

(4)

Among the effects of climate change, elevation of global temperature has been an imperative issue of discussion by environmental researchers in recent decades because of its several adverse consequences. Anthropogenic global warming is currently increasing on an average 0.2°C per decade due to past and ongoing emissions (IPCC 2018). But the climate related risks depend on several aspects such as geographic location, magnitude, adaptation and mitigation etc. Temperature extremes have been observed more in high latitude than mid. It is projected that due to the alarming climate change at high latitudes temperature may increases up to 4.5°C in the cold season if there is 1.5°C rise of global average temperature by 2052 (IPCC 2018). In Finland mean daily temperature after 1960 has already increased by 0.3°C per decades (Aalto et al. 2016). During the period of 1847-2013, mean annual temperature had already increased by 2.3°C throughout the whole Finland (Mikkonen et al. 2015). At the regional scale (arctic and boreal areas), this global warming effects will include drastic changes, e.g., in precipitation, drought, and cloudiness, affecting forest nutrient availability, spread of insect-pest and diseases (IPCC 2014). In these particular geographic zones the growth and physiology of most plants is naturally temperature limited hence the effects of increased temperature would be especially important. Moreover, increased temperature plays various effects on plant growth, due to its widespread role in the regulation of biochemical reactions, phenological development rates and energy exchange within environment (Amthor 1991; Morison and Lawlor 1999). These may cause severe impacts on biodiversity and ecosystem including species loss and extinction of these particular regions.

The boreal forest ecosystem has already exposed several responses to climate change in the past decades (Soja et al. 2007). Interruption in the favorable condition of plants may increase their stress levels which will likely enhance the sensitivity of trees to insect-pests (Schlyter et al. 2006; Jönsson et al. 2009) and diseases (Sturrock et al. 2011). It is also predicted by several earlier studies that a number of insect- pest species will be capable to expand their geographical range because of climate change (Hughes 2002; Jönsson et al. 2009). Therefore with the predictable warming, outbreaks of insect ‐pests may be more frequent and predicted to expand their range extensively towards high latitudes (Hof and Svahlin 2016). Moreover, climate change may also increase the forest disease outbreaks by affecting lifecycle of the pathogen, the host plant and interaction between them; because most plant diseases are strongly influenced by environmental conditions (Sturrock et al. 2011). Forest diseases are complex and involve several interacting factors such as soil moisture regime, stand density, precipitation, defoliating insects or drought (Manion and Lachance 1992). The effects of

(6)

forest diseases on forest ecosystems will change with the changes in climate because host plant susceptibility to pathogens and pathogen growth, reproduction and infection may affect by several environmental abiotic factors such as temperature, precipitation, air moisture etc.

(see Sturrock et al. 2011). Rise in temperature and change in precipitation may allow the ranges of some species to expand hence the distribution of hosts and diseases will change.

Therefore roles of pathogens will probably increase and most of them will be able to migrate themselves to the locations where climate is more suitable for their reproduction and survival (Sturrock et al. 2011). These can have large negative consequences on forest sector.

Climate change comprises affiliated changes in many components of the abiotic flux which are essential for plant life. Trees shows immediate response to climate warming within a short time by changing patterns of their growth and development (Kirschbaum 1999), growth and phenology (Hänninen and Tanino 2011) or by modifying their quality through changing plant secondary compounds (PSCs) (Holopainen et al. 2018). A considerable increase in atmospheric CO2 and temperature can influence interactions between plants and herbivores by altering the quality of plants as food by making them more stoichiometrically unstable for their herbivores (Sterner and Elser 2002; Loladze 2002). It has been assumed that elevated CO₂generally stimulate carbon partitioning to carbohydrates and various classes of plant phenolic compounds (Bezemer and Jones 1998) but increase in temperature alone has opposite effects, whereas CO₂ and temperature collectively has negligible effects (Zvereva and Kozlov 2006). Thus the countering effects of elevated temperature are expected to play a significant role in alpine, boreal and arctic zones in determining the balance between plants their herbivores populations (Veteli et al. 2007). However, these predictions are mostly based on closed chamber or greenhouse conditions and one year or single growing season experiments whereas open-air or field conditions and multiple year or growing season study may reveal diverse conclusions due to the interaction of different environmental factors within or among different year or growing seasons. Therefore, actual impacts of elevated temperature on trees are not yet fully assumed.

1.1 Impacts of climate change in Finnish boreal forest

Finland is a country of forest where about 75% (22.8 million hectares) of her land area is covered by forests representing almost 10% of the total forest area in Europe (Lier et al.

2019). In 2018, this forest sector sheared 4.5% of the Finnish GDP and one fifth of the exports goods came from the industrial products of this sector (Lier et al. 2019). In Nordic countries boreal forest trees are used systematically for timber and biomass production for periods

(7)

(Kellomäki 2017) as well as for heating, electricity production, housing and structural materials. Moreover, forests have role in balancing the effect of climate change through sinking a considerable amount of carbon. Lier et al. (2019) stated recently that 50% of the Finland‘s total emissions (excluding the emissions and removals of land use and forestry) have covered by forest sink. Due to the effects of climate change the area of boreal forests in south is probably to be replaced by temperate forests and in north area of tundra forests is likely to be replaced by boreal forests (Kellomäki 2017). Such changes in structure of boreal forest may effect on the ecosystem and Finnish economy as well.

According to report of IPCC (2018), high-latitude tundra and boreal forests are predominantly at risk of climate change-induced degradation and habitat loss and this will proceed with further warming. Annual mean temperature of the northern hemisphere region (where basically boreal forests are located) varies from -5 to +5 (Kellomäki 2017). It has been widely known as boreal forests are dominated with coniferous trees as well as some deciduous species and especially boreal forests in Finland characterized by small number of tree species.

Regarding to Natural Resources Institute Finland (Luke) 2016; although approximately 30 species of trees are naturally present in Finland but in Finnish boreal forests spruce, pine, downy birch and silver birch are known as Finland‘s national tree species. Among those tree species, 80% of Finnish boreal forests are covered by coniferous species; spruce and pine and 17% are covered by deciduous species birches (Lier et al. 2019). But this bio-geographical distribution of plant species are forecasted potential shifts changing as a response to current global warming for the period 2041–2070 under the IPCC A1B emission scenario using temperature-only models (Villén‐Peréz et al. 2020). Effects of warming on vegetation have already been described in terms of upslope shifts of species ranges in mountainous region (Felde et al. 2012) or increase in plant biomass as well as shifts in relative species abundances at high latitudes (Sistla et al. 2013; Vuorinen et al. 2017). Earlier studies also predicted that, southern areas of northern Europe deciduous species would show dominance over conifers (Falk and Hempelmann 2013) and such trend of changing distribution of plant species is expected to continue over the next decades (Lenoir and Svenning 2015). Also in Finland there is a possibility that the distribution and percentage of birches would increase in future and thereby more study is required to understand and overcome the risks which may hamper the distribution of this most economically viable plant species of Finnish boreal forest.

1.2 Importance of silver birch in boreal forest and its impacts upon climate warming Considering the broadleaved deciduous tree species in northern temperate region birches are

(8)

essential ecological components for the biodiversity of coniferous dominating boreal forests.

Birches are considered as light demanding early successional species which can rapidly occupy large areas after clear-cutting and forest fires due to their abundant seed production and fast juvenile growth (Fischer et al. 2002). Two commercially important birch species occur naturally throughout Europe are silver birch (Betula pendula Roth) and downy birch (Betula pubescens Ehrh.). In general morphological appearance both birches resembles each other and in some cases it is hard to differentiate between silver birch and downy birch in natural field condition (Hynynen et al. 2010). However, both birches vary regarding some of the morphology oftwigs, leaves, branches, bark, seeds as well as cell size and wood anatomy (Johnsson 1974; Bhat and Kärkkäinen 1980; Jonsell 2000). Although both silver birch and downy birch have a widespread natural distribution area on the Eurasian continent, extending from the Atlantic to eastern Siberia up to China and Japan; and found almost of the whole Europe but the temperate and boreal forests of Northern Europe are most abundant in birch resources whereas silver birch is absent in Iceland, Greece and the Iberian Peninsula (Hynynen et al. 2010). Silver birch grows most habitually on fertile forest site and on afforested unrestrained fields wheresandy and silty till soils and finesandy soils are available (Raulo 1977; Gustavsen and Mielikäinen 1984; Niemistö 1995; Hynynen et al. 2010).

Similarly in Finland Betula species are commonly distributed on southern fertile sites with coniferous species.

Birches are commercially the most important and productive broadleaved tree species in north European and Nordic countries. Economically they are also very important tree species in Russia and Belarus. The main reasons why the birch species are most important broadleaved tree species in forestry are their high productivity, fast-growing, combined with straight and slender stems, porous wood and light in weight (Hynynen et al. 2010). Large-sized logs are produced within comparatively short period of time with suitable silvicultural treatment and unique physical and mechanical wood properties allow multipurpose and valuable uses of birch in Northern Europe (Dubois et al. 2020). Birches are valuable source for the pulp for paper, plywood, veneer, timber, furniture, firewood and various board products such as medium-density fiberboards (MDF), high-density fiberboards (HDF) or oriented strand boards (OSB). In general, the wood performance of silver birch is slightly better than that the wood of white birch (Betula papyrifera Marshall) due to its higher wood density, though their industrial uses are not differentiated (Luostarinen and Verkasalo 2000; Heräjärvi 2002). In Finland, more than 80% of all commercial birch wood (excluding firewood) were used by pulp industries and more than 90% of the harvested birch logs were used in plywood

(9)

industries during the period of 2014-2016 (Verkasalo et al. 2017). Thus the commercial importance of birch trees is increasing gradually.

In terms of biodiversity in coniferous forests birch is associated with higher number of specialized flora and fauna species than for other tree species in Europe (Dubois et al. 2020).

Throughout the different phases of succession, a huge number of species feed on or live together with birch, including mycorrhiza forming fungi, wood decaying fungi, saproxylic insects and herbivores (Kennedy and Southwood 1984; Perala and Alm 1990; Patterson 1993;

Hynynen et al. 2010). Hence, in Scandinavia, Belgium, Germany, Spain and North America;

birches are used as a target species to improve biodiversity in multiple ecological restoration projects, especially for coniferous forests (De Schrijver et al. 2009; Felton et al. 2011; Burgess et al. 2015; Dubois et al. 2020). Moreover silver birch has an aesthetic appearance due to its unique white bark, light crown and autumn color hence commonly planted in urban areas, roadsides and parks for its pleasant colors (Dubois et al. 2020). Additionally, sap of birch can be tapped and consumed, either as birch beer or as wine by fermentation of sap, fresh sap as a tonic or as concentrated syrup and medicinal use of birch leaf or bark decoctions as diuretic, anti-rheumatic and anti-fever purgative have been well-known since the medieval period.

It is predicted that birch has a great advantage over many other species in contradiction of climate change and the uncertainties of associated biotic risks due to itsadaptation capacity, great tolerance to a large variety of climates and soils and capability to recolonize damaged areas after disease and pest outbreak (Kellomäki and Wang 2001; Lavola et al. 2013; Dubois et al. 2020). The share of birch in boreal forests is predicted to increase in response of climate warming and therefore the percentage of birch in managed forests in Finland is estimated to increase from 10 to 20 % by the year 2100 (Kellomäki et al. 2008). The height, stem diameter and over all biomass production silver birch in the boreal regions has noted to increase in various experiments under elevated temperature until the optimum temperature is reached (Mäenpää et al. 2011; Kasurinen et al. 2012; Chung et al. 2013; Way and Montgomery 2015).

It has been observer that elevated temperature increases average leaf area and rate of photosynthesis which results higher growth response of silver birch though its growth response varies upon developmental stage of trees since young seedlings are tended to show immediate response to climate change (Kellomäki and Wang 2001). Excessive stem height increment compared to stem diameter was noted in B. pendula seedlings as response to elevated temperature which may reduce their mechanical strength (Kellomäki and Wang 2001). This vigorous growth may also induce changes in their defense photochemistry which

(10)

increases susceptibility to herbivore and diseases. It is widely hypothesized that change in climate directly affects the growth, physiology and defense including induced PSC_s production of forest trees. This may alter the biotic stress affliction of trees with the responses of organisms, such as herbivorous insects, microbial plant pathogens, and other exotic invasive pathogens and herbivores (Metlen et al. 2009;Donnelly et al. 2012; Holopainen et al.

2018). Silver birch contains different kinds of secondary phytochemicals which are considered as protectants against various environmental stress factors (Lavola 1998;

Julkunen-Tiitto et al. 2005) hence they are excellent model species for environmental change experimental studies (Lavola et al. 2013).

1.3 Herbivore pressure on silver birch due to climate warming

Global warming is predicted to increase the abundance of herbivorous insects and to lead to shifts in their distribution towards higher latitudes hence boreal forests might be challenged by increased herbivore pressure where they might face more palatable food sources (Heimonen et al. 2015). Temperature has a direct effect on insect developdevelopment and survivalof most insect species and the majority of temperate and boreal species are expected to shift their distribution to higher latitudes because of rising temperature (Parmesan et al. 1999; Bale et al.

2002; Jepsen et al. 2008). The probable expansion of herbivorous insects towards higher latitudes and their increasing abundance would increase the intensity of herbivore facing towards more palatable host plants which emphasize a potential threat to the productivity and regeneration of boreal trees in future (Heimonen et al. 2015). Earlier studies have also proposed latitudinal palatability hypothesis (Heimonen et al. 2015), which means floras from high-latitude populations are more palatable to herbivores than low-latitude plants (Pennings et al. 2001; Moles et al. 2011;Morrison and Hay 2012). Therefore host plants would face new attackers in the future to which they are not well adapted to and could get heavily damaged.

Silver birch is one of the most common deciduous broad leaved tree species in boreal forests which is intended to stimulate its growth in response of climate warming. Previous studies indicates that, larger plants could be more easily detected by herbivores than smaller plants (Haysom and Coulson 1998; Wise and Abrahamson 2008) and vigorously growing trees supposed to be more attractive to insects (Price 1991). Therefore, species like silver birch which increases their growth the because of warming climate probably at higher risk of herbivore damage than slower growing species (Heimonen et al. 2017). The damage by leaf- chewing and leaf-mining insects to northern birch forests is predicted to more than double because of projected 2–3 °C rise in summer temperatures (Kozlov 2008). It has been proven

(11)

by earlier experiments that herbivore can adversely affect the growth and survival of birch.

Increased height and diameter growth of B. pendula saplings throughremoval of insects was experimented by Silfveret al. (2009) and Prittinen et al. (2003) observed that thesmall shaded silver birch seedlings were especially vulnerable and mortality of seedlings was increased by the insect herbivore. Therefore with increased herbivore pressure in future may be a great threat especially for silver birch seedlings.

Both biotic and abiotic factors are known to affect the physical properties and chemical composition in birch (Lavola and Julkunen-Tiitto 1994; Lavola 1998a,b; Keinänen et al.

1999). The elevated atmospheric temperature and CO2 may affect the primary and secondary metabolism of plants. Laitinen et al. (2000) stated that, their results indicate, the quantity if secondary chemicals in individual trees is differently modified by the environmental stimulus.

Therefore the adaptivity, resistance and defensive behavior of an individual tree in relation to other trees of the similar population is dependent on the specific weather conditions of a particular year (Laitinen et al. 2000). In aspects of silver birch earlier studies suggest both, that elevated temperature alone either may increase (Kuokkanen et al. 2001, 2003) or no significant affect (Lavola et al. 2013) in the leaf defense quality i.e. in allocation of phytochemicals such as total phenolics. But in conclusion most of the researchers suggested that the positive effects of increased CO₂(enhanced carbon partitioning to various classes of phenolic compounds or secondary metabolites) in aspects of herbivore defense of silver birch may less or disappear under elevated temperature; hence effects of elevated temperature were found opposite to the effects of elevated CO₂ acting alone (Kuokkanen et al. 2001, 2003;

Zvereva et al. 2006; Veteli et al. 2007, 2013). Thus the actual effects of elevated temperature on birch–herbivore interactions have not been summarized yet.

1.4 Susceptibility of silver birch to diseases due to climate warming

Betula species are supposed to be relatively resistant to diseases, but climate warming and its change could enhance this of problems in birch forests as well (Dubois et al. 2020). Increase in temperature and moisture directly affect the pathogens as many pathogens are sensitive to precipitation and humidity and therefore reproduction, spread and infection rate increases when conditions are moist (Harvell et al. 2002). During recent years some examples have been reported such as dieback of birch saplings in Scotland after 5–10 years of planting on field sites which was probably caused by fungal pathogens (Green and MacAskill 2007;Silva et al. 2008) and crown degeneration and early dieback of roadside birches in Germany due to the ―Birch leaf roll-associated virus‖ (Rumbou et al. 2018). However, most of the diseases

(12)

known to cause problems in forestry species these days are fungal and they gain access to the host plant through localized sites of damaged or diseased tissue or during periods of physiological stress (Lilja et al. 1997).

Birches are generally attacked by fungal diseases which include birch rust, stem lesions and top dying of birch, leaf lesions of birch etc. Among them rust disease is prone to more destructive and sever for birches. The responsible pathogen for birch rust is Melampsoridium betulinum (Fr.) Kleb. and severity of these disease occur in birch stands especially during summers when the weather is rainy and wet (Hynynen et al. 2010). Rate of precipitation are projected to be greater in northern hemisphere high-latitude and several high-elevation regions due to 1.5°C of global warming (IPCC 2018) which may provide more favorable weather condition for the survival, spread and reproduction of this pathogen. Moreover, the epidemic phase of Melampsoridium betulinum occurs in late summer (Lilja et al. 2010) therefore lengthening the summer season due to global warming may enhance the severity of rust infection in silver birch in upcoming future. It is essential to control birch rust especially in nurseries because this disease may cause high mortality and reduction of plant growth after planting in the natural field condition (Lilja et al. 2010).

Earlier experiments reported silver birch which grown at low nitrogen levels were more resistant to rust than those grown at higher levels (Poteri and Rousi 1996) and birch seedlings grown at high nitrogen supply and with high density create wetter conditions on leaf surfaces and facilitate urediniospore germination (Sharp et al. 1958). On the other hand, there is a possibility that expected rise in temperature due global warming may increase leaf area and photosynthetic capacity of silver birch which may lead to increase the leaf nitrogen concentrations (Kellomäki and Wang 1997, 2001; Kuokkanen et al. 2001; Li et al. 2019).

Therefore there is an undoubted possibility that silver birch would more susceptible rust infestation in near future.

1.5 Importance of this study

The intensity of herbivore damage and rust infection in Betula pendula due to global warming have not been studied in detail especially for natural field condition and for multiple growing seasons. Most of the experiments that have been conducted previously are mainly in green house conditions and concluded the results based on one year or growing season (Kellomäki and Wang 1997, 2001; Veteli et al. 2002; Kuokkanen et al. 2001, 2003). Changes in climatic conditions occur gradually and several climatic factors (temperature, precipitation, drought,

(13)

humidity etc.) may significantly vary among years, therefore plants grown in field condition experience these natural variations in temperature, soil moisture and humidity, light, ultraviolet-B levels, wind etc. On contrary in green houseexperiment plants do not experience these types of natural variations. Therefore, to avoid these kinds of problems multiple year field study setup is compulsory to obtain more reliable results and trustworthy conclusion.

(14)

2 AIM AND HYPOTHESIS OF THE STUDY

The aim of this study was to understand does the warming climate increase birch leaves palatability to herbivores and susceptibility to rust damage. I evaluated the effects of elevated temperature on the herbivore and rust damage on silver birch seedlings gown in outdoor field conditions for continuous three growing seasons. Different leaf features such as leaf area, number of glandular trichomes, eaten area by herbivores, rusts spots and leaf quality such as leaf nitrogen concentration, leaf condensed tannin concentration including its proportion of soluble and insoluble tannins were measured to estimate their changes under elevated temperature and to observe their correlation with herbivore and rust damage.

More precisely, objectives of this study were to assess the following research questions:

a) Will the leaf parameters of silver birch be affected by elevated temperature in the field conditions?

b) Is there more herbivore damage and rust in individuals grown under elevated temperature?

c) Can the herbivore damage and rust abundance be linked to quality of leaves (e.g.

nitrogen and tannin content)?

d) Are the changes in leaf parameters caused by elevated temperature proposing shift in silver birch palatability or susceptibility to diseases in future climate?

Following hypotheses were madeto complete the study:

1. Elevated temperature will increase the leaf area and nitrogen concentration and decrease concentration of condensed tannins of silver birch leaves.

2. There is more herbivore and rust damage in individuals grown under elevated temperature compared to controls.

3. The herbivore damage and rust abundance correlate with quality of leaves.

4. Silver birch becomes more palatable and susceptible to diseases in future climate.

(15)

3 MATERIALS AND METHODS

This experiment has been conducted on the basis of data collected (2016-2018) as a part of an ongoing research carried out by the Department of Environmental and Biological Science, University of Eastern Finland. Practical part of this Master thesis consists of measurements for herbivore damage, leaf area, glandular trichomes, rust and nitrogen content for birch leaf samples from season 2018. For statistical analyses, data was completed with leaf parameters for years 2016-2017 and condensed tannins concentrations, provided by the research project.

3.1 Experimental field

The experimental field was adapted from UVB-temperature field at the botanical garden known as ―Botania‖ of Joensuu, Eastern Finland (62°35‘N, 29°46‘E) in 2016. To protect saplings from mammal herbivores the entire field area is surrounded by 1.5-m high fence and metal shelter was used at the lower part of the fence to prevent the disturbance of voles. This experimental field was made up of total 36 experimental plots, 12 of them assigned for this study where 0.80 m × 2.40 m size of each plot and 3.0 m distance from each other was maintained. In May 2016, plots were prepared by cleaning existing vegetation, removing old top soil of each plot and filling each plot with fresh 20 cm layer of sand-peat mixture soil (0.8% limed). In this study set-up, half of the plots (6) known as treated plots were equipped with adjustable frames holding two infrared heaters (CIR 110, FRICO, Partille, Sweden) in the mid part of plots. As same way, other 6 controlled plots were equipped with adjustable frames having two wooden pieces instead of infrared heaters, which are as like as IR-heaters in size and shape to even out the shadowing effect of the heaters. But in all plots UVB-tubes (see Nybakken et al. 2012) were left at their places without functioning. At the beginning of the experiment adjustable frames were setup at 145 cm above from the ground. Distance between frames and plants were adjusted by lifting them during the all growing seasons (May- August, 2016-2018) so that the heaters always remain 60 cm above than the tip of the highest plants.

Field was visited daily and proper function of heaters and modulated system was checked twice a week during all growing seasons. Seedlings were watered in 2016 first weeks daily, and later in the summer when dry period last for several consequent days but not in 2017- 2018. Weeding was done regularly when necessary. Seedlings were treated twice in June 2016 with herbicide (Baygon aerosol, Johnson) because of unusually large amounts of aphids in Finland during early summer 2016.

(16)

3.2 Plant material

One-year old seedlings of silver birch (Betula pendula Roth.) were introduced in May 2016 in 12 plots of the field where six replicates used for elevated temperature (treated plots) and six replicates for elevated temperature (control plots). In 12 plots, total number of silver birch seedlings was 576 where 48 seedlings were planted in each plot maintaining 20 cm plant spacing. Moreover, 228 side plants i.e. 19 plants around each plot with 30 cm intervals was planted as well. All seedlings were seed originated and collected from Tuusniemi forest plantation nursery, Eastern Finland (Fin Forelia Oy). After the plantation few individuals died during the first weeks of experiment which were immediately replaces by new seedlings.

3.3 Treatments

Two treatment levels ambient and elevated temperature (target temperature set +2°C but managed to get +1.5°C on ambient) were monitored in this study during the growing seasons (2016-2018). Among 12 plots six replicates were used for increased temperature and six replicates for control. Modulated heating system was followed where temperature was measured from two heated and two control plots with PT1000 probe elements. IPC100 configuration program was used to compute set point and e-console program for measuring the data. During the year 2016 and 2017, temperature treatment were continued from late May to end of September but in 2018 heaters were shut downed at the middle of the season (early July) as the plant reaches maximum height.

3.4 Sample collection

First mature birch leaves were collected for chemistry, leave size, herbivore and rust analyses in each year at August. Leaf samples were collected from ten randomly selected individuals in 2016 and later on from five individuals from each control and treated plot (total number of leaf samples was 120 n 2016, and 60 for both 2017 and 2018). All collected samples were air dried in 10% RH at room temperature for two days after that stored in -20 °C until analyses.

3.5 Leaf area (cm²), eaten area (cm²), leaf damage (%) and leaf water (%) measurement

Leaf area (cm²) was measured using a portable leaf area meter LI-3000C (LI-COR, Lincoln, NE, USA). Damage of leaves (eaten area per cm²of leaf) throw herbivore was also measured using this portable leaf area meter using the formula; eaten area covered - not covered. How big part (%) eaten area is from total leaf area i.e. leaf damage (%) is measured with the formula; (eaten area/total leaf area) ×100. Leaf water percentage was measured by (leaf fresh

(17)

weight – leaf dry weight)/ leaf fresh weight ×100; for the years 2017–2018.

3.6 Glandular trichomes and rust measurement

Number of glandular trichomes (specialized hair like structure on leaf surface) was counted visually under microscope (2x0.5 cm², both sides of the main vein) from the upper (adaxial) surface of leaves for the years 2017–2018. The number of rust pustules (powdery masses of yellow-orange spores on the lower surface of the leaves) was also visually counted with microscope (2x1 cm², both sides of the main vein) from the lower (abaxial) leaf surfaces.

3.7 Leaf nitrogen (N) and condensed tannin (mg g^-1dw) measurement

Leaf nitrogen (N) concentration percentage was determined using a nitrogen analyzer LECO^® FP-528 (Leco Corp. Svenska AB, Uplands Väsby, Sweden) on 15 mg leaf discs (without veins) and using corn flour 1.7% N as a reference. Amount of total condensed tannin (mg g^-1 dw) content of leaves including its proportion of methanol soluble and insoluble tannins was measured with acid-butanol assay according to Hagerman (2011) where Betula nana tannin was used as a standard.

3.8 Statistical analyses

In statistical analyses, averages from same plot for each year are treated as replicates (n = 6).

Effects of temperature and year on leaf parameters were tested by general linear model (ANOVA) using IBM SPSS statistics for Windows (Version 25.0.0.2). Temperature and year were used as fixed factor and all studied features such as leaf nitrogen concentration(%), leaf water (%), leaf area (cm²), leaf area (%), eaten area (cm²), leaf damage (%), rusts spots (cm^-2), glandular trichomes (cm^-2), total tannins (mg g^-1 dw), soluble tannins (mg g^-1dw) and insoluble tannins (mg g^-1dw) were used as dependent variable. Least Significant difference (LSD) test was selected as post-hoc test for year and unstandardized residuals was saved for residual investigation.

Levene‘s test for homogeneity of variance and normality of residuals of all the variables was checked. To meet assumption of parametric test, square root-transformation was done for leaf area (cm²), while log-transformation was done for soluble tannins (mg g^-1dw) and insoluble tannins (mg g^-1dw). Non-parametric tests were done for leaf water (%), eaten area (cm²), eaten area (%), rusts spots (cm^-2) and glandular trichomes (cm^-2) where Mann-Whitney U tests were followed for treatment and Kruskal-Wallis tests were followed for year. Original data were used without any transformations for these non-parametric tests.

(18)

Correlation (Spearman correlation for non-parametric data) of rust and herbivore damage was tested with all leaf parameters. IR-heaters are assumed to dry the leaf surfaces, which may have direct effect on rust appearance, independent of leaf parameters. Due to this, correlations for control individuals was tested for rust to identify if abundance of rust is linked to underline leaf features or not. This could help to predict their abundance in the future climate. For herbivore damage correlations was tested twice. Once with all individuals from 2017-2018 to see if there are some leaf features explain weather the herbivore damage takes place and another correlations was tested with only attached individuals of 2017-2018 to see if how much is eaten is guided by different leaf properties than why leaf is attached. In both correlations year 2016 data was not included because during this first growing season herbicide was used to protect seedlings from aphids and bugs which may influence the results.

(19)

4 RESULTS

4.1 Leaf nitrogen concentration and leaf water percentage

Elevated temperature, year and their interaction had significant effect on leaf nitrogen concentration (Table 1). In ambient leaves, highest nitrogen content was detected in 2018 while 10% and 27% smaller amounts were recorded for 2016 and 2017, respectively. During first growing year 2016, leaf nitrogen concentration in heated plots increased by 39%

compared to ambient control plots (Fig. 1 A). In following growing season (2017) nitrogen content increased 13% in elevated temperature while in third growing season (2018) 5%

decrease was detected (significant treatment x year interaction, Table 1).

Table 1. F-values and statistical significance (p-value) of ANOVA of all leaf parameters with year, treatment and their interactions. Significant results (p < 0.05) marked with bold.

Leaf parameter Year Treatment Year x Treatment

Nitrogen (%) F = 25.441

p = 0.000

F = 16. 045 p = 0.000

F = 11.465 p = 0.000

Water content (%) p = 0.178 p = 0.977

Leaf Area F = 13.44

p =0.000

F =36.93, p =0.000

F =8.85, p =0.000

Eaten Leaf Area p = 0.003 p = 0.152

Leaf Damage (%) p = 0.003 p = 0.279

Rust spots p = 0.002 p = 0.003

Trichomes p = 0.000 p = 0.410

Soluble tannins F = 74.418 p = 0.000

F = 37.26 p = 0.000

F = 18.31 p = 0.000 Insoluble tannins F = 16.05

p = 0.000

F = 12.60 p = 0.001

F = 15.68 p = 0.000

Total tannins F = 55.70

p = 0.000

F = 21.08 p = 0.000

F = 7.50 p = 0.002

In terms of leaf water content, no significant response detected due to elevated temperature and among years as well (Table 1, Fig. 1 B).

(20)

Figure 1. A) Leaf nitrogen concentration (mean ± SE) and B) Water content (mean ± SE) under ambient (open bars) and elevated temperature (filled bars) at the end of the growing seasons. Water content was not recorded in 2016.

4.2 Leaf area

Differences in leaf areas were significant with treatments, year and treatment-year interactions as well (Table 1). Prominent increase in leaf area was observed with elevated temperature compare to ambient for all growing years (Fig. 2). Maximum leaf area increase (45%) was observed in heated plots than controls 2^nd growing year 2017 (Fig. 2). In the first growing season (2016) average leaf area was increased almost 20% under elevated temperature where no difference was detected in the year of 2018 (Fig. 3).

Figure 2. Leaf Area (mean ± SE) under ambient (open bars) and elevated temperature (filled bars) at the end of the all growing seasons.

4.3 Eaten leaf area and leaf damage from total leaf area (%)

Effect of herbivore on birch leaves was varied among years but no significant difference was noted among treatments for parameters eaten area and leaf damage (Table 1). In both eaten

0 1 2 3 4 5

2016 2017 2018

Nitrogen (%)

Year

A

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

2016 2017 2018

Leaf water (%)

Year

B

0 5 10 15 20 25 30 35 40 45

2016 2017 2018

Leaf Area (cm2)

Year

(21)

area per cm²of leaf (Fig. 3A) and leaf damage (%) (Fig. 3B), it was notable that less herbivore damage occurred in first growing season 2016 than next growing seasons 2017 and 2018 tended to have more herbivore damage for both treatments. Moreover in 2017 and 2018 there was tendency that in more eaten leaf area under warming (Fig. 3A) but difference in herbivore damage between treatments was much smaller if leaf size in consider (Leaf damage from total leaf area) (Fig. 3B).

Figure 3. A) Eaten leaf area (mean ± SE) and B) Leaf damage (mean ± SE) under ambient (open bars) and elevated temperature (filled bars) at the end of the growing seasons.

4.4 Rust spots and leaf trichomes

Average rust spot count per cm² of leaf varied significantly between treatments and among years as well (Table 1). In view of year, highest rust was counted for ambient leaves during the first growing season (2016) and continuous decrease was recorded by following years i.e.

25 % in year 2017 and 95% in year 2018 (Fig. 4A). Similar reducing trend from beginning to end last growing years were noted for treated leaves as well where it was 69 and 72% less in 2017 and 2018 respectively than the year 2016 (Fig. 4A). Considering treatments, rust spots were significantly reduced under elevated temperature where it was 14, 36 and 1.5% less than ambient for year 2016, 2017 and 2018 respectively (Fig. 4A).

Difference in average trichomes number per cm²of leaf area was recorded for year 2017 and 2018 (Fig. 4B). The year has significant effect on average trichomes but not for treatments (Table 1). Higher number of trichomes per cm² of leaf was noted in the year of 2017 than the following year 2018 (Fig. 4B). Leaves in control plots showed 65% reduced number of trichomes whereas leaves in treated plots showed 61% reduced number of trichomes from the year 2017 to 2018 (Fig. 4B).

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

2016 2017 2018

Eaten Leaf Area (cm2)

Year

A

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

2016 2017 2018

Leaf Damge (%)

Year

B

(22)

Figure 4. A) Rust spot count (mean ± SE) and B) Trichomes count under ambient (open bars) and elevated temperature (filled bars) at the end of the all growing seasons. Trichomes count was not recorded in 2016.

4.5 Total, soluble and insoluble tannin of leaf

The treatment, year and their interactions have significant effect on leaf total, soluble and insoluble tannins (Table 1). Under ambient condition highest amount (149.3 mg g^-1 dw) of total tannin was recoded in year 2017 which was 43% higher than year 2016 and 46% higher than year 2018 (Fig. 5). Smaller amount of total tannin was accumulated under increased temperature compared to ambient conditions during the year 2016 and 2017 where it was 54 and 24% less than ambient but during the year 2018 it amount of total tannin was increased by 5% than ambient leaves (Fig. 5). For soluble tannin highest amount (114.4 mg g^-1 dw) was recorded in 2017 for ambient leaves and lowest amount was recorded in 2016 for treated leaves (Fig. 5). Compare to ambient condition, soluble tannin was slightly (6%) increased with treated condition in 2018 while it was reduced by 66 and 30% during year 2016 and 2017 respectively (Fig. 5). Similarly, 4% increased insoluble tannin was recorded under treated condition than control in 2018 while 36% less amount in 2016 and 2% less amount in 2017 was recorded for treated condition than control (Fig. 5). Moreover similar to soluble tannins, highest amount of insoluble tannins was also recorded in 2017 for ambient leaves and lowest amount was recorded in 2016 for treated leaves (Fig. 5).

0 10 20 30 40 50 60

2016 2017 2018

Rust Spot (cm-2)

Year

A

0 20 40 60 80 100 120

2016 2017 2018

Trichomes (cm-2)

Year

B

(23)

Figure 5. Amount of total tannins (means) including its proportion of soluble and insoluble tannins under ambient and temperature condition at the end of all growing seasons.

4.6 Correlation between herbivory and leaf quality

As leaf herbivore damage was almost identical measured either ‗eaten leaf area‘ or ‗leaf damage from total leaf area (%)‘ (Fig. 3) for correlation study only ‗leaf damage (%)‘ was taken into count to evaluate leaf herbivore damage.

Leaf nitrogen concentration had significant correlation with leaf herbivore damage on the attached leaves but no significant effect was noted for all individuals (Table 2). There was no correlation noted for leaf rust spots with leaf nitrogen concentration (Table 3).

Table 2. Correlation of herbivore damage with other leaf parameters for all individuals and for damaged or attached individuals from 2017 - 2018 (***, P < 0.001; **, P < 0.001; *, P <

0.05).

Leaf Parameters Leaf Damage (%)

All individuals Attached individuals

Nitrogen concentration NS 0.331**

Leaf Water NS NS

Leaf Area 0.240** 0.406***

Rust spots -0.180* -0.357***

Trichomes NS -0.361***

Soluble tannins NS -0.288**

Insoluble tannins NS NS

Total tannins NS -0.289**

51.69 17.67

114.35 79.92

51.33 54.35 32.75

20.97

34.93 34.17

29.28 30.58

2016 2017 2018

mg g-1 dw

Year

Soluble Tanins Insoluble Tannins

(24)

No significant correlation of leaf water percentage was found for leaf herbivore damage in parts of all individual correlation and attached individual correlation (Table 2). Similarly, leaf water percentage showed insignificant correlation with rust spots of leaf (Table 3).

It was noted that leaf area was significantly affected by herbivore damage for both, all individual and attached individual correlation whereas a negatively significant correlation was noted among leaf area and rust spots (Table 2).

Number of rust spots was negatively correlated with leaf herbivore damage (Table 2).

Likewise, significant negative correlation of rust spot with leaf area and soluble tannins of leaf was detected (Table 3).

Table 3. Correlation of rust spots with other leaf parameters for controls (***, P < 0.001; **, P < 0.001; *, P < 0.05)

Leaf Parameters Rust

Nitrogen concentration NS

Leaf Water NS

Leaf Area -0.191*

Eaten Leaf Area NS

Leaf Damage (%) NS

Trichomes NS

Soluble tannins -0.224*

Insoluble tannins NS

Total tannins NS

In addition, the number of trichomes on leaf surface had significant negative correlations with leaf damage (%) while correlated with only damaged individuals but no effect was noticed while correlated with all individuals (Table 2). There was no correlation found between leaf trichomes and leaf rust spot counts (Table 3).

In correlation study it was found that leaf total tannins and soluble tannins had negative effect on leaf herbivore damage for only damaged or attached individuals (Table 2) whereas only soluble tannins had negative correlation with rust (Table 3). Moreover it was noted that leaf insoluble tannins had no effect on leaf herbivore damage and rust as well.

(25)

5 DISCUSSION

5.1 Leaf properties under ambient growing conditions

In this multiple year field experiment, year-to-year variation in Finnish weather (discussed later) greatly influenced all parameters studied. This could be a vital reason that, some of the findings of this experiment differ with earlier research findings for ambient conditions.

From different years of this study nitrogen concentration for leaves grown under ambient temperature was recorded 2.4% to 3.3%, where it was around 2.8% and 4% for birch seedlings reported by Kuokkanen et al. (2001) and Lavola et al. (2013), respectively.

Concentrations of leaf nitrogen also is in accordance with 25 species survey by Perry and Hickman (2001) where nitrogen% varied from 1.0 to 3.6 (dry weight basis) between tree species.

Leaf water percentage of silver birch recorded here, however, was different than in earlier studies. In ambient growing conditions water% was about 60% of leaf fresh weight while slightly higher amount (about 80%) was noted by Kuokkanen et al. (2001) and Lavola et al.

(2013) in closed-top chamber condition.

Silver birches studied here have also some differences in their defense compare to earlier studies. Average number of glandular trichomes in ambient was counted about 90 and 32 per cm²of leaf adaxial surface in 2017 and 2018, respectively. If calculation is done with average leaf area of those particular years than number of glandular trichomes per leaf would be around 1800 and 1100 for the year 2017 and 2018, respectively; while Thitz et al. (2017);

reported around 1200 glandular trichomes per leaf in ambient conditions during one growing season.

Amount of total condensed tannins including its proportion of soluble and insoluble tannins was different in this study than greenhouse experiments of Kuokkanen et al. (2001) and Lavola et al. (2013). Under ambient conditions, the proportion of soluble tannins found to be higher than insoluble tannins for all growing seasons while opposite was recorded by Lavola et al. (2013) though this her result was based on only for one growing season. This difference in results may happen for genetical variation of plants because significant variation among different genotypes of silver birch was found by Lavola et al. (2013); in levels of both soluble and insoluble condensed tannins. Moreover another reason could be the difference between field and greenhouse conditions. Because in greenhouse condition plants do not face so many

(26)

stresses as like as field condition; for example light conditions are not natural, comparatively higher nutrient availability of soil etc. (Forero et al. 2019).

5.2 Leaf responds on elevated temperature

Empirical evidence suggests both; elevated temperature may increase (Li et al. 2019) or decrease (Dury et al. 1998) this N allocation on plant leaves. According to the results of this study, leaf nitrogen increases under elevated temperature, which agrees to the results of Kuokkanen et al. (2001) for birch seedlings and with the results of Kellomäki and Wang (1997) in aspects of Scots pine. This increase of leaf nitrogen concentrations may correlates with the increased leaf area and photosynthetic capacity of plants under elevated temperature (Kellomäki and Wang 1997, 2001; Kuokkanen et al. 2001; Li et al. 2019). On contrary, no change in leaf nitrogen concentration was reported by Kuokkanen et al. (2003) for same birch (Betula pendula) species for increased temperature. Moreover, Lavola et al. (2013) noted that, depending on the birch genotype nitrogen content reduction varied 3% to 12% in combination with elevated CO2 and temperature. The magnitude and direction of the response may relate to the actual ambient temperature and light conditions experienced by the plants through the experimental period (Kellomäki and Wang 1997, 2001). Additionally, several field and environmental conditions such as; soil fertilization, soil moisture, air humidity, plant genotype etc. may also affects the leaf characteristics (Prittinen et al. 2003; Lavola et al. 2013;

Lihavainen 2016;Wang et al. 2016).

In this experiment, no significant changes were observed for leaf water percentage in response to treatments. Similarly, no significant response was recorded by earlier studies in leaf water content for increased temperature (Kuokkanen et al. 2001; Lavola et al. 2013). Treatment was also significant for leaf area i.e. elevated temperature increased the leaf area which supports by earlier findings (Kellomäki and Wang 2001; Lavola et al. 2013; Li et al. 2019). According to Kellomäki and Wang (2001); elevated temperature increases leaf area and photosynthetic capacity resulting higher growth response thorough higher net carbon intake by plants.

There was no significant effect of treatment noted for leaf glandular trichomes in this experiment, although reduced number of glandular trichomes production under elevated temperature has been reported by Thitz et al. (2017).

Elevated temperature significantly affects the condensed tannin accumulation including soluble and insoluble tannins in this experiment, meanwhile Lavola et al. (2013) reported no significant effects of elevated temperature in allocation of phytochemicals such as soluble and

(27)

insoluble condensed tannins. Similarly no significant effect in allocation of condensed tannins was noted by Kuokkanen et al. (2001) although a reducing tendency was found. Significant decrease as response to elevation in temperature in deciduous woody plants has frequently been reported for total leaf phenolics and low-molecular weight phenolics but not always for more carbon containing i.e. high-molecular weight tannins (Dury et al. 1998; Veteli et al.

2002; Veteli et al. 2007; Paajanen et al. 2011).

5.3 Leaf responds on different growing seasons and their interaction with treatments Statistical analysis showed that, year was significant for most of the studied factors (except leaf water percentage) in this experiment. According to Finish Meteorological Institute, average summer temperature (May to August) were 15.08°C, 12.18°C, 15.78°C and average summer precipitation were 74.38 mm, 50.50 mm, 51.53 mm during the year 2016, 2017 and 2018, respectively in Linnunlahti, Joensuu. Favorable growth temperature for full development of birch leaves is approximately 22°C (Kellomäki and Wang 2001) which is clearly higher than the average summer temperature for all growing seasons. Therefore, it is natural that around 2°Cincreases in temperature than ambient would provide more favorable growth conditions and leaf physical and chemical properties as well (Kellomäki and Wang 2001; Pumpanen et al. 2012; Lavola et al. 2013).

Haukioja et al. (1978) reported that, the concentrations of nitrogen in birch leaves are stable during July and the beginning of August in Finland though in this multiple year study nitrogen concentration varied between years. This may due to the huge weather variation among the years (2016-2018) where both temperature and precipitation may play an important role in soil nitrogen cycles, plant nitrogen intake or due to the difference developmental stages of seedlings as growth response depends on that (Kellomäki and Wang 2001).

Lower leaf area was recorded in the year 2017 than in the other two growing years. It is likely result of variation in growing season conditions as the ambient temperature was also lowest (12.18°C). Accordingly the maximum leaf area increase (about 45%) was observed in heated plots in that particular year, 2017. Also the maximum number of trichomes per cm² of leaf was noted in the year of 2017 although treatment was not significant in this study. While not only elevated temperature but also other environmental factors such as drought, waterlogging, air humidity etc. may significantly affects the production of glandular trichomes in silver birch (Lihavainen 2016; Wang et al. 2016; Thitz et al. 2017), number of trichomes per cm² might also been increased as response smaller leaves (i.e. concentration effect, see Thitz et al.

(28)

2017).

In this experiment amount of total condensed tannins including its proportion of soluble and insoluble tannins significantly varied among years. Similar quantitative variation in birch leaf phenolics (condensed tannins) was observed for two successive growing seasons by Laitinen et al. (2000).

Treatment × year interaction was significant for leaf area, nitrogen and total tannins (soluble and insoluble tannins as well) content in this study. For all parameters, effect of treatment disappears in 2018. In 2018 treatment was stopped in early July, but the leaf samples were collected in August. This is supporting Lavola et al. (2013) who determined that birch is quite quickly to adapt and acclimatize with moderate environmental change.

5.4 Herbivore damage:

In this study, most leaf damage appears in leaves with large area and low number of rust spots. However, when affected leaves are considered, strong positive correlations of severity of herbivore damage with leaf nitrogen concentration and leaf area and negative correlation with rust spots, number of trichomes and condensed tannins was observed. This suggest that different features define weather leaf is attractive to herbivore and its palatability to herbivores.

Herbivore damage generally increases with an increase in nitrogen concentration (Mattson 1980; Scriber and Slansky 1981). Moreover, nitrogen concentration of host plant intensely controls processes such as growth, survivorship, population levels and outbreak frequency for insect herbivores (Throop and Lerdau 2004). Similarly it is a widespread expectation through several previous researches that, the plant secondary compounds such as phenolics (specially condensed tannins) act as qualitative defenses against leaf eating herbivores (Bennett and Wallsgrove 1994; Barbehenn and Constabel 2011; Pearse 2011). Zvereva and Kozlov (2006) noted in their meta-analysis that, the performance of herbivores was significantly improved hence the leaf phenolics was decreased by temperature elevation. However, some other study had failed to find a general defensive role of condensed tannins against herbivores where they suggests that the condensed tannins are more effective against specialist rather than generalist insects for example Coleopteran insects were found being more infected than Lepidoptera insects (Ayres et al. 1997; Boeckler et al. 2014).

To predict plant herbivore damage and insect performance under changing environmental

(29)

conditions, analysis of plant chemistry is widely practiced (Coley 1998; Keinänen et al. 1999) hence the factors affecting the performance of herbivores is frequently linked to the chemical and mechanical quality of host plants (Peñuelas and Estiarte 1998; Johns and Hughes, 2002).

The quality of leaves (most important plant tissue used as food by herbivorous insects) usually depends on both the concentrations of essential nutrients and defensive secondary compounds (Zvereva and Kozlov 2006). Fungal growth in host plant may adversely affect the nutritional quality of leaves. Rust pustule acts as foci for the accumulation of many metabolites through the uptake of nutrients by rust fungi; hence balanced mixture of amino acids, different sugars, inorganic nutrients, steroids etc. are the basic needs of rust fungi (see Mendgen 1981). These accumulated metabolites may decline the foliage quality for herbivore and affect host- herbivore relationship. That could explain the negative correlation of herbivore damage with rust infection i.e. less herbivore damage in rust infected leaves observed here in this experimental study.

Change in global climate implies increase in temperature and atmospheric CO2

simultaneously but their effects on leaf chemistry are different. Several literatures suggested that, elevated CO₂ have negative effects on herbivore by modifying host plant quality through decreasing nitrogen levels and increasing phenolic levels meanwhile an increase in temperature have opposite effect (Bezemer and Jones 1998; Zvereva and Kozlov 2006; Veteli et al. 2007). Higher temperature may directly compensate for some of negative effects of elevated CO2 (Veteli et al. 2007). They also concluded that, the overall herbivore response to elevated temperature in future will be positive.

In this study, elevated temperature did not significantly increase herbivore damage, although tendency to higher damage appears. In this field experiment, insect appearance, species and distribution was not controlled; that can effect on the results. Changing climate affects both physical and chemical properties of leaf and secondary compounds together with the physical plant structures have been shown to facilitate herbivore damage (Tahvanainen 1991; Veteli et al. 2002). A negative correlation was observed between herbivore damage and leaf glandular trichomes in this study. These propose that, if the glandular trichomes reduce, leaf insect damage will more. Earlier studies also revealed that, in silver birch glandular trichomes are produced less under elevated temperature (Thitz et al. 2017; 2019). So it can be predicted that climate warming i.e. elevated temperature may increase the susceptibility of herbivore damage on birch in Finland (Heimonen 2017).

Herbivore and rust damage of Silver birch (Betula pendula) leaves under climate warming

HERBIVORE AND RUST DAMAGE OF SILVER BIRCH (Betula pendula) LEAVES UNDER CLIMATE WARMING

Kishor Kumar Roy

Contents