Tree-ring widths and isotopes of artificially defoliated balsam firs: A simulation of spruce budworm outbreaks in Eastern Canada

Environmental and Experimental Botany 81 (2012) 44–54

Contents lists available at SciVerse ScienceDirect

Environmental and Experimental Botany journal homepage: www.elsevier.com/locate/envexpbot

Tree-ring widths and isotopes of artificially defoliated balsam firs: A simulation of spruce budworm outbreaks in Eastern Canada Sonia Simard a,∗ , Hubert Morin a , Cornelia Krause a , William M. Buhay b , Kerstin Treydte c a b c

Département des Sciences Fondamentales, Université du Québec à Chicoutimi, 555 Boulevard de l’Université, Chicoutimi, Québec, G7H 2B1, Canada Department of Geography, University of Winnipeg, 515 Portage Avenue, Winnipeg, Manitoba, R3B 2E9, Canada Swiss Federal Research Institute WSL, Zürcherstrasse 111, 8903, Birmensdorf, Switzerland

a r t i c l e

i n f o

Article history: Received 10 November 2011 Received in revised form 1 February 2012 Accepted 24 February 2012 Keywords: Choristoneura (Archips) fumiferana Artificial defoliation Carbon isotopes Oxygen isotopes Tree rings Gas exchange

a b s t r a c t Defoliation by insects is a major disturbance influencing the forest dynamics in many ecosystems and can affect forest productivity worldwide. The main objective of this research was to further investigate the potential use of tree-ring widths and isotopic compositions to identify different degrees of past spruce budworm defoliation episodes. A secondary objective was to understand the responses of trees to defoliation episodes using carbon isotopes as a proxy to provide insights into subsequent physiological changes. Tree-ring widths, carbon and oxygen isotopic compositions in wood cellulose and gas exchange measurements were compared among 288 balsam fir (Abies balsamea Mill.) seedlings grown in a controlled experiment that involved different intensities of defoliation. Observations were performed over four growing periods. Moderate to heavy-defoliated seedlings showed reduced radial growth and enriched their cellulose carbon isotopic composition probably as a result of mobilized stored carbohydrates enriched in 13 C. Less severely defoliated seedlings did not show significant reductions in growth and 13 C enrichments. The gas exchange observations and wood cellulose oxygen isotope compositions do not suggest photosynthetic compensation in the remaining needles although a positive trend in the response of both assimilation rate (A) and stomatal conductance (gs ) to defoliation was observed in the first growing period. Thus it remains open as to which mechanisms were employed to compensate for the reduced carbon source in the mildly defoliated seedlings. While further investigations are advised, the results of this study still help promote the utilization of tree-ring widths in combination with carbon isotopic compositions for reconstructing severe past defoliation events. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Defoliation affects forest ecosystems worldwide through a variety of causes. Occasional defoliation episodes related to spruce budworm (SBW, Choristoneura (Archips) fumiferana Clem.) outbreaks are one of the main disturbances influencing the forest dynamics of the boreal forest of eastern North America. By feeding on both developing and old needles (during times of high insect population density) of balsam fir [Abies balsamea (L.) Mill.], white spruce [Picea glauca (Moench) Voss] and black spruce [Picea mariana (Mill.) B.S.P] (Morin, 1994) SBW can severely reduce forest productivity thereby decreasing growth and even causing significant tree mortality (Bouchard et al., 2007; MacLean et al., 1996).

∗ Corresponding author. Present address: Swiss Federal Research Institute WSL, Zürcherstrasse 111, 8903, Birmensdorf, Switzerland. Tel.: +41 44 7392 840; fax: +41 44 7392 215. E-mail addresses: sonia [email protected], [email protected] (S. Simard), hubert [email protected] (H. Morin), cornelia [email protected] (C. Krause), [email protected] (W.M. Buhay), [email protected] (K. Treydte). 0098-8472/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.envexpbot.2012.02.012

Identifying and reconstructing past defoliation episodes on both short and long-term scales is necessary for a better understanding of the fundamentals of SBW outbreaks and related forest dynamics as a whole. Moreover, in a context where ecosystem based management is generally considered the most promising approach to the maintenance of healthy and resilient forest ecosystems, it is essential to better assess the natural long-term variability of insect related disturbances (Gauthier et al., 2008). Long-term annually resolved SBW outbreak records can be derived through a dendrochronological approach using tree rings. Since these insects feed on developing and old needles of boreal forest tree species they can and do invoke both physiological and chemical responses that are recorded in annually produced rings (Blais, 1983; Leavitt and Long, 1986a; Simard et al., 2008). Tree-ring records from living trees, sometimes in combination with historical timbers (Boulanger and Arseneault, 2004; Krause, 1997) and/or sub-fossil trees (Simard et al., 2011) therefore offer the potential for extended records of SBW outbreaks. Specifically by reducing photosynthetic biomass, defoliation episodes impact the source/sink balance in trees which can trigger internal reactions to maintain an equilibrium between sources

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

(photosynthetic tissues or tissues from where carbohydrate remobilization can occur) and sinks for the carbon required for normal growth and maintenance (Pinkard et al., 2011). Following defoliation, trees can respond in several ways. One way is to mobilize reserve carbohydrates stored as starch (Eyles et al., 2009) to supply the carbon required. Another strategy to compensate for source limitation is to increase the rate of photosynthesis in the remaining needles until a source/sink balance is re-established (Chen et al., 2001; Lavigne et al., 2001; Little et al., 2003; Reich et al., 1993; Vanderklein and Reich, 1999; Welter, 1989). Increased photosynthetic rates in response to defoliation may be achieved through different ways. Increased stomatal conductance following enhanced water availability, increased photosynthetic carbon fixation following increased flow of nitrogen to remaining leaves (a large proportion of N is in chlorophyll and Rubisco), or increased Rubisco activity (rather than a change in quantity) in response to greater light reaching the remaining needles, have been observed (Turnbull et al., 2007). All these responses are known to result in chemical changes (enrichment in the carbon isotopic composition) (Farquhar et al., 1982; Francey and Farquhar, 1982). In the case of increased photosynthetic activity in response to defoliation, assimilates enriched in 13 C as a consequence of reduced isotope discrimination against the heavier carbon (13 CO2 vs 12 CO2 ), would also isotopically enrich the molecular components (hemi-cellulose, cellulose, lignin) of the tree-ring wood. Similarly, remobilization and use of stored carbohydrates enriched in 13 C could also lead to 13 C enriched tree-ring components formed during and after defoliation events (Brugnoli et al., 1988; Damesin and Lelarge, 2003; Helle and Schleser, 2004; Le Roux et al., 2001). A recent dendro-isotopic survey of SBW outbreaks (1950–1960; 1970–1980), from mature host and non-host conifers from the boreal forest, shows carbon isotope enrichment signatures (relatively high isotope values) in both severely and more lightly defoliated trees (Simard et al., 2008). Higher carbon isotope values of host-trees and reduced ring widths directly corresponded with the two last SBW outbreaks in the area. Altogether, very few studies have investigated the effects of defoliation on the ␦13 C composition in tree ring cellulose (Kress et al., 2009; Leavitt and Long, 1986a; Weidner et al., 2010) and the results varied among the studies. Investigations on European larch (Larix decidua Mill.) in relation to larch budmoth (LBM, Zeiraphera diniana Gn.) defoliation showed no relationship between tree-ring ␦13 C variations and outbreak periods (Kress et al., 2009; Weidner et al., 2010). However, high ␦13 C values and reduced ring widths in tree rings of evergreen conifers formed during a western SBW outbreak were observed by Leavitt and Long (1986a). Kress et al. (2009) proposed the difference in life cycles between SBW and LBM as a possible explanation for the different signals in tree-ring ␦13 C in LBM-European larch and SBW-black spruce/balsam fir systems (Simard et al., 2008). LBM defoliation events last for one vegetation period and reoccur with a periodicity of 8–10-year intervals while SBW outbreaks appear in decadal scale frequencies (30–40-year intervals) which can last for more than five years. The impact of a one-year-long defoliation outbreak compared to a five-year repeated defoliation event is likely much lower and might explain the discrepancies observed between the two studies. The fact remains that understanding causes vs consequences in dendrochronological studies is challenging. For example, it is not clear from the studies of Simard et al. (2008) and Leavitt and Long (1986a) if the wood cellulose ␦13 C values increased in response to defoliation, or if the trees were experiencing some situation (e.g. localized drought) that increased their ␦13 C values and also lead to greater susceptibility to insect attack. The present study helps address this issue by inducing a “cause” (defoliation) and measuring the “consequence” in a controlled manner.

45

The objective of this research is to further investigate the potential use of tree ring widths and isotopic compositions for identifying different degrees of past SBW defoliation episodes. A controlled experiment involving different intensities of defoliation on balsam fir seedlings was conducted. This study was designed to determine the relation between defoliation intensities and annual tree-ring widths, cellulose carbon and oxygen isotopic compositions and plant physiology (CO2 assimilation rate and stomatal conductance) over four growing periods. Oxygen isotopic compositions were also measured because they are linked to water source, evaporative demand and stomatal conductance (Barbour and Farquhar, 2000; Saurer et al., 1997). Linking both isotopes could therefore help to further understand the mechanisms responsible for the carbon isotopic compositions of the defoliated seedlings (Scheidegger et al., 2000). The research also aims at understanding what tree responses are to outbreaks by using ␦13 C to determine if they rely on starch reserves to continue growth and metabolic activities or if they undergo other physiological changes. Based on the evidence from Leavitt and Long (1986a) and Simard et al. (2008), carbon isotopic enrichment of the balsam fir treering cellulose is expected with increased levels of defoliation due to the mobilization of 13 C enriched stored carbohydrates for xylem cell production and/or increased carbon fixation as a compensatory mechanism to defoliation. Accompanying tree-ring width reductions are also expected in the ring series of the most severely defoliated seedlings. The oxygen isotope composition of the treering cellulose is expected to respond independently to artificial defoliation (Simard et al., 2008). 2. Material and methods 2.1. Plant material To test the effects of different degrees of defoliation on treering widths, and tree-ring cellulose ␦13 C and ␦18 O, a growth experiment was conducted in a greenhouse setting. Two hundred eighty-eight (288) container-grown five-year-old balsam fir (A. balsamea) seedlings of uniform height (26.7 ± 4.1 cm) and diameter (7.8 ± 1.2 mm) were obtained from a nursery in the Lac Saint-Jean (Quebec, Canada) area. The seedlings were planted in plastic pots containing peat moss (15.2 cm diameter, ∼2.44 l of peat moss/pot) in springtime and allowed to adapt to the new rooting environment during that initial summer before starting the defoliation experiment. 2.2. Experimental design Seedlings organized in a randomized block split-plot design were subjected to one to four growing periods or “pulses” of growth (i.e. time as the main plot), four different intensities of current-year needle defoliation (CYD; 0, 33%, 66%, 99%) and two intensities of old-needles defoliation (i.e. all other age-class needles except the currently developing ones; OND; 0% and 50%) (i.e. defoliation as the subplot). The experiment was replicated within three blocks. Pulse of growth refers to a growing episode between two dormancy periods. Dormancy between pulses of growth was applied by exposing the seedlings to a cold environment protected from snow at or below 0 ◦ C for a period of four to six weeks (Dubuc, Canadian Forest Service, Pers. Comm.). Growing period 1 (end of December 2004 to mid of March 2005) was simulated by increasing the greenhouse temperature and irrigating the seedlings (details provided below). Dormancy was then again initiated by a cold treatment between mid March and end of May (2005). Growing period 2 was initiated between June and September (2005) followed by another cold treatment for dormancy between October and December (2005).

46

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

Growing periods 3 and 4 were repeated in the same manner during the following year (2006). Note that during growing periods 2 and 4 the seedlings were transferred to an open field adjacent to the greenhouse after six to eight weeks of initial growth, when more than 70% of the ring width was already developed (Deslauriers et al., 2003). To foster synchronized bud burst, seedlings were fertilized with 2 g of NPK (20% nitrogen, 8% phosphorous, 20% potassium) dissolved in 500 ml of water at the beginning of the growing periods. Conditions inside the greenhouse were set to 22 ◦ C during the day (18 h) and 17 ◦ C during the night (6 h). Wide-spectrum highpressure sodium bulbs (400-W; Lucalox LU400, General Electric Co., Cleveland, OH) provided a photosynthetic photon flux density (PPFD) of 115 ␮mol m−2 s−1 when natural day-light was low, representing approximately 40% of the light saturated photosynthesis (around 500 ␮mol m−2 s−1 for a light saturated photosynthetic rate of about 7 ␮mol m−2 s−1 ) (Landhäusser and Lieffers, 2001). A drip irrigation system supplied the seedlings daily water to maintain an 80% field capacity hydration level (determined by weighing the seedling pots at full and no hydration). 2.3. Defoliation A factorial design experiment (Quinn and Keough, 2002) was used to investigate the physical (ring width, height, diameter), isotopic (␦13 C and ␦18 O in tree-ring cellulose) and physiological (light-saturated CO2 assimilation rate (A) and stomatal conductance (gs )) effects of both current-year defoliation (CYD) and old-needles defoliation (OND). Eight different combinations of defoliation treatments were randomly assigned to the seedlings. The 0% CYD and OND (CYD0, OND0) treatment served as the experimental control. Other seedling groups were subjected to a 33% current-year defoliation and a 0% old-needles defoliation (CYD33OND0) treatment in addition to CYD66-OND0, CYD99-OND0, CYD0-OND50, CYD33-OND50, CYD66-OND50 and CYD99-OND50 treatments (Fig. 1c and d). The current-year needle defoliation levels were defined to simulate light to severe SBW outbreaks. Oldneedles defoliation simulates SBW feeding on old foliage occurring at very high insect population density. All defoliation treatments were carried out three to four weeks after bud burst to allow for sufficient shoot elongation for needle manipulation. For all treatments, the needles were clipped at their base with scissors. For CYD, newly developed needles on all elongating shoots were removed according to the level of defoliation, as shown in Fig. 1a. For OND, none or 50% of remaining needles of all other age classes except the developing needles were removed (Fig. 1a). The defoliation treatments assigned to each seedling were repeated at each subsequent growing period to measure the cumulative effect of defoliation.

gas exchange measurements to avoid interference with the defoliation treatments. A cylindrical conifer foliage cuvette was placed over the current- and one-year-old shoots on the lateral branch of the second verticils (Fig. 1a). Measurements were performed before midday under supplemental fluorescent lighting at saturating photosynthetically active radiation (1 000 ␮mol m−2 s−1 ). Saturation conditions were chosen to ensure comparable measurements between seedlings and dates of measurements. Immediately after the gas exchange measurements, subsamples of the needles were removed for estimation of their surface area (Bernier et al., 2001; Hebert et al., 2006). Needles were dried at 65 ◦ C for 48 h to obtain leaf dry mass per unit leaf area (LMA). The LMA was then used to calculate the total foliar area in the cuvette. Measured values of light-saturated CO2 assimilation rate (A), stomatal conductance (gs ) and ratios of intercellular to ambient air CO2 partial pressure (ci /ca ) were then recalculated based on actual total needle surface areas. The specific leaf area (SLA) of each seedling was calculated at the end of growing periods 1–3. Given the positive relationship between leaf nitrogen (N), SLA and A (Reich et al., 1998), SLA was used as an indicator of leaf N to evaluate the possible effects of defoliation on A at the seasonal level. 2.6. Sample preparation and stable isotope analyses One seedling per treatment per block (different than the one used for gas exchange measurements) was harvested after the fourth growing period for isotope analysis. Ring widths at the base of the stem were measured on eight radii to account for incomplete rings using a high-resolution scanner and image analyzer (WinDendro LA1600+; Regent Instruments, Quebec, Canada). Annual rings corresponding to growing periods 1–4 were separated with a scalpel under a binocular microscope and stored individually for isotope analysis. The samples were finely chopped using a scalpel and holo-cellulose was isolated by delignification in an acetic-acid-acidified sodium chlorite solution, after first removing oils and resins with toluene-ethanol and ethanol soxhlet extractions (Leavitt and Danzer, 1993). Isotopic analyses of the cellulose samples were accomplished by continuous flow isotope ratio mass spectrometry (CF-IRMS) using a VG-Instruments® IsoPrime attached to a peripheral temperature controlled EuroVector® elemental analyzer (EA) (University of Winnipeg Isotope Laboratory, UWIL). Holo-cellulose samples (0.2–0.3 mg) were loaded into tin (carbon) and silver (oxygen) capsules and placed in the EA auto-sampler along with internally calibrated standards. The ␦13 C and ␦18 O values have analytical uncertainties of ±0.12 and 0.3‰, respectively. Holo-cellulose ␦18 O measurements were analyzed within a short time window after production to prevent errors associated with possible oxygen isotope exchanges in stored samples (Wright, 2008).

2.4. Height and diameter 2.7. Statistical analyses The height and diameter of all seedlings were measured at the beginning and at the end of the experiment. The height was measured as the distance between the base of the stem (soil/stem interface) and the base of the apical bud. The diameter of the stem was measured at its lower part at the soil/stem interface. 2.5. Gas exchange measurements The seedlings were allowed to adapt to their new conditions for two weeks after every defoliation treatment. Then needle gas exchange measurements were done once per growing period using a LI-6400 portable photosynthesis system (LI-COR Inc., Lincoln, NE) on one seedling per treatment per block. A seedling different then the one chosen for other measurements was selected for

2.7.1. Height, diameter, SLA and gas exchange variables Differences in height and diameter between seedlings at the onset and end of the defoliation experiment, as well as CO2 assimilation and stomatal conductance were tested using an analysis of variance (ANOVA). A priori contrasts (control vs defoliation levels) were conducted when a hypothesis of equal means was rejected. 2.7.2. Tree-ring widths and stable isotope composition Tree-ring widths and stable isotope compositions (␦13 C, ␦18 O) were analyzed according to a repeated-measures analysis of variance (ANOVAr) (Gumpertz and Brownie, 1993; von Ende, 1993). For the within-subject analysis, a Huynh–Feldt corrected

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

a)

47

Current-year shoot

Current-year shoot Current-year defoliation (CYD)

0%

33% 66% 99%

2-year-old shoot

1-year-old shoot

d)

Control CYD0-OND0

b)

CYD66-OND0

c)

GP 1 66-0 0-50

GP 2 33-50

0-0

0-0

33-50

0-50

100-50

100-50

66-50

66-0

33-0

100-0

33-0

66-50

100-0

66-50

33-0

66-0

100-50

66-0

100-0

33-0

33-50

33-50

0-50

66-50

100-0

0-0

100-50

0-50

0-0

GP 4

GP 3

Block

CYD66-OND50

CYD66-OND50

GP 1

GP 2

Fig. 1. (a) A schematic balsam fir seedling diagram showing the locations of the current-, first- and second-year shoots, respectively, and the different levels of current-year defoliation (circle inset – CYD 0, 33%, 66% and 99%) applied to all seedling current-year shoots (on all verticils) during each of the four growing periods, (b) a schematic diagram showing an example of current-year defoliation (CYD66-OND0) applied only one growing period (GP), (c) a schematic diagram showing the cumulative impact of current- and old-needles defoliation (CYD66-OND50) over 2 consecutive growing periods, (d) a representation of one of the three experimental blocks, with the eight possible defoliation treatments randomly assigned within the main plots (growing periods 1–4).

probability was used to overcome the sphericity assumption in the case of univariate repeated-measures analysis (von Ende, 1993). A full model was first fitted to the data: block, time (growing periods), CYD (0, 33, 66, 99) and OND (0, 50), and all two- and threeway interactions between time, CYD and OND. If the interaction terms were not statistically significant, they were dropped and the model was rerun with just the main effects. Homogeneity of variance was verified by visual analysis of residuals (Devore and Peck, 1994), and logarithmic transformations were performed when necessary to homogenize the variance. Differences were considered significant at P < 0.05. All statistical analyses were performed using the JMP® 8.0 software (SAS Institute Inc.).

3. Results 3.1. Height and diameter Current-year defoliation of seedlings, independently of oldneedles defoliation, resulted in decreased height and diameter at the end of the experiment with increasing levels of defoliation (P < 0.0001; Fig. 2a and b; Table 1). Both variables were reduced by up to 40% for the CYD99 defoliation treatments. Interestingly, while partly removing old needles from seedlings (OND treatments) induced a significant decrease of the radial growth, it did not have an impact on the vertical growth (Fig. 2c; Table 1).

48

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

120

25

a) P=<0.0001

25

b) P=<0.0001

c) P=0.0172 GP 1 GP 4

100 20

20

* 60

* 40

15

* 10

0 33

66

0

99

10

0

0 0

15

5

5

20

Diameter (mm)

*

Diameter (mm)

Height (cm)

80

33

66

99

% CYD

% CYD

0

50

% OND

Fig. 2. Seedling growth heights (a) n = 6) and diameters (b) n = 6 and (c) n = 12) (mean ± se) before the onset of the defoliation experiment and after the end of the last experimental growing period (GP). Stars indicate statistically significant difference with control trees (see Table 1 for P-values).

3.2. Radial growth and tree-ring stable isotope composition The effect of defoliation treatments on ring width varied in time, starting at growing period 2 only until growing period 4 (P = 0.0006 and P = 0.0438 for CYD and OND, respectively) (Fig. 3a and b; Table 2). Compared to the control, the ring widths of CYD99 seedlings were 60% smaller on average after the second growing period, and up to 80% smaller after both the third and fourth growing period (Fig. 3a; Table 2). The moderate level of defoliation (CYD66) significantly reduced the radial growth of the seedlings only after the fourth growing period whereas no effect was evident with the CYD33. Old-needles defoliation, independently of currentyear needle loss, resulted in significantly smaller ring widths by an extra 30% on average for growing periods 2–4 (Fig. 3b, Table 2). A few seedlings (n = 5) developed incomplete rings in response to severe defoliation (CYD99-OND0, CYD99-OND50) experienced during the third and fourth growing period.

Table 1 ANOVA summary (F- and P-values) of seedling heights and diameters before the onset of the experiment and at the end of the last experimental season. Results presenting greater than 95% significance (P < 0.05) are indicated. n.s. = non significant, CYD = current-year needle defoliation, OND = old-needles defoliation, GP = growing period. Source

Main plot Block Time Subplot CYN OND Time × CYD Contrasts GP 4 0 vs 66 0 vs 99 Time × OND GP 4 a

Heighta

Diameter

F

P

F

1.73 157.00

0.0063

0.70 179.88

0.0055

9.59 1.19 15.60

<.0001 n.s. <.0001

16.07 7.65 11.99

<0.0001 0.0091 <0.0001

24.57 8.08 65.81

<.0001 0.0074 <.0001

27.80

<0.0001

67.62 6.27 13.89

<0.0001 0.0172 0.0007

Ln transformed data.

P

The carbon isotope composition of the tree-ring cellulose significantly varied in time in response to current- and old-needles defoliation treatments (interaction time × CYD × OND, P = 0.0239) (Fig. 4a and b; Table 2). Significant 13 C enrichment compared to control seedlings was observed in the moderate to severe treatments (CYD66-OND0, CYD99-OND0, CYD99-OND50) at the first and third growing periods. An average increase of 1.8‰ for seedlings exposed to 99% current-year defoliation treatment compared to the control ones was observed. This observed increase exceeded 2‰ on average in specific treatments for a given growing period (e.g. growing period 1 – CYD99-OND0 (P = 0.0156); growing period 3 – CYD66OND0 (P = 0.0129); growing period 3 – CYD99-OND50 (P = 0.0062)). While not all defoliation treatments induced statistically significant tree-ring cellulose ␦13 C enrichments, a trend toward 13 C enrichment with defoliation is observed. The trend however represents a high variability in time and between treatments as shown by the large standard deviations. Oxygen isotope values of defoliated vs control seedlings scatter (22–28‰) and also include high standard deviations (Fig. 4 a); Table 2). No statistically significant variation was found. A statistically significant relationship between ␦13 C and ␦18 O values could indicate a common source of variation (i.e. stomatal conductance) for isotopic imprinting in tree rings. However, the correlation between ␦13 C and ␦18 O was non-significant which suggests other influencing factors are responsible for the ␦13 C compositions of the seedlings exposed to the different defoliation treatments (Fig. 5). The different growing periods had observable effects on treering widths (Fig. 3), and ␦13 C, but not ␦18 O (Fig. 4). ␦13 C values were, in general, more depleted during growing periods occurring at wintertime (growing periods 1 and 3) in comparison to summertime (growing periods 2 and 4). Tree rings produced in wintertime were also narrower compared to those produced during the other growing periods. That seasonal pattern was not observed for the ␦18 O values. 3.3. Gas exchange Only the removal of newly developed needles independently of the loss of old needles significantly altered the light-saturated

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

1.5

a) P=0.0006

1.0

1.0

0.5

0.5

Ring width (mm)

Ring width (mm)

1.5

0.0 -0.5 *

*

-1.0 -1.5 *

49

b) P=0.043

GP 1 GP 2 GP 3 GP 4

0.0 *

-0.5 -1.0

*

*

-1.5

*

-2.0

-2.0 33

66

99

0

% CYD

50

% OND

Fig. 3. Measured effects of current-year (a) n = 6) and old-needles defoliation (b) n = 12) on annual ring widths (mean ± se) of balsam fir seedlings. The results are presented as deviation from control trees (CYD0-OND0). Stars indicate statistically significant difference with control trees (see Table 2 for P-values). GP = growing period.

CO2 assimilation rate (A) of balsam fir seedlings during the experiment (P = 0.0004) (Fig. 6; Table 3). Generally A was stimulated in the first growing period of all defoliation intensities, although only statistically significant for CYD66 with an increase of 30% compared to the control (P = 0.0022). In growing period 2–4, A generally decreased with significant values only observed during the second growing period for moderate and severe defoliation treatments (Fig. 6; Table 3). In particular, the remaining needles of the 99% defoliated seedlings showed systematically lower assimilation rates during growing periods 2–4 compared to the control seedlings. The stomatal conductance of the remaining needles were, in most cases, reduced during the second and third growing periods and reached the statistically significant level in response to the moderate and severe levels of defoliation during growing period 2 (CYD66 P = 0.0366, CYD99 P = 0.0006) and light and severe levels during growing period 3 (CYD33 P = 0.0093, CYD99 P = 0.0377) (Table 3;

Fig. 6). No significant modification of the ci /ca ratio was observed (data not shown). Defoliation of current-year needles significantly decreased SLA but the response varied with time (P = 0.0396, Table 3). SLA was mostly reduced during the first and third growing periods compared to control seedlings although only significantly during the first periods in response to the light and severe current-year defoliation levels (CYD33 and CYD99 P = 0.0123). Removing 33% of the newly grown needles after growing period 2 increased SLA by 37% in comparison with control seedlings (P = 0.0105, Table 3, Fig. 7).

4. Discussion Overall, this defoliation experiment revealed a high resilience of balsam fir seedlings to needle loss. No seedling mortality

Table 2 Summary of repeated-measures ANOVA (F- and P-values) for carbon isotope compositions (␦13 C) and oxygen isotope compositions (␦18 O) of wood holo-cellulose and ring widths; within-season contrasts for ␦13 C and ring widths. P-values for repeated-measures ANOVA are presented with Huynh–Feldt corrected probabilities. Results presenting greater than 95% significance (P < 0.05) are indicated. n.s. = non significant, CYD = current-year needle defoliation, OND = old-needles defoliation, GP = growing period. Source

Between-subjects Block CYD OND CYD × OND Within-subject Time Time × CYD Time × OND Time × CYD × OND Contrasts within-season GP 1 GP 2

GP 3

GP 4

␦13 C

␦18 O

Ring width

F

P

F

P

F

P

3.50 2.50 0.12 0.52

n.s. n.s. n.s.

5.99 1.24 1.03 0.84

n.s. n.s. n.s.

1.39 24.87 16.06 1.44

<.0001 n.s. n.s.

45.10 0.95 1.44 2.47

<.0001 n.s. n.s. 0.0232

4.64 1.28 1.80 1.61

0.0097 n.s. n.s. n.s.

67.39 4.25 2.94 0.59

<.0001 0.0006 0.0438 n.s.

0 vs 99-0 OND CYD 0 vs 99-0 OND CYD 0 vs 66-0 0 vs 99-50 0 vs 99-0 OND CYD 0 vs 66-0 0 vs 99-0

6.59

0.0129 11.86 15.39 39.88 11.86 6.18

0.0011 <.0001 0.0001 0.0011 0.0159

13.11 11.01 18.19 5.17 44.90

0.0006 0.0016 <.0001 0.0268 <.0001

6.23 8.08

0.0156 0.0062

50

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

a)

-20 Old-needles defoliation - 0%

Old-needles defoliation - 50%

-22

δ13C

-24 -26 -28 -30 32 30

δ18O

28 26 24 22 20 18

CYD 0 CYD 33 CYD 66 CYD 99

1

2

3

4

1

2

3

4

growing period

b) 4

Old-needles defoliation - 0% * P=0.023

Old-needles defoliation - 50% *

*

δ13C

2

0

-2

GP 1 GP 2 GP 3 GP 4

-4 33

66

99

% CYD

0

33

66

99

% CYD

Fig. 4. (a) Tree-ring cellulose carbon and oxygen isotopes compositions (␦13 C and ␦18 O) (mean ± sd; n = 3) and, (b) tree-ring cellulose carbon isotope composition (␦13 C) presented as deviation from the control (CYD0-PYD0) (mean ± se; n = 3), in response to current- and old-needles defoliation over the four growing periods (GP).

was documented even under the most severe level of defoliation (CYD99-OND50) where nearly all needles were removed from the seedlings for four consecutive growing periods. Defoliation up to 66% current-year needles did not significantly alter the radial growth or the height of the seedlings until the fourth growing period. However, removing nearly all needles did result in reduced tree-ring widths and even incomplete tree rings through the fourth growing period. The results suggest that, in general, the seedlings compensated to some extent for the loss of photosynthetic biomass during lighter current-year needle defoliation treatments but were very

sensitive to the loss of old needles. Moderate to heavy-defoliated seedlings showed reduced radial growth and enriched their cellulose carbon isotopic composition. Although less severely defoliated seedlings did not show significant 13 C enrichments, positive trends in response to the defoliation of newly formed needles were observed. Different factors (or a combination of factors) may have played a role in the increase in cellulose ␦13 C values. Carbon isotope enrichment may have resulted due to increased photosynthetic rates in response to increased light availability for the remaining needles, changes in stomatal conductance (enhanced water availability), increased flow of nitrogen (Chen et al., 2001; Lavigne et al.,

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54 Table 3 Summary of ANOVA (F- and P-values) for light-saturated CO2 assimilation rate (A) and stomatal conductance (gs ) and specific leaf area (SLA). Result presenting greater than 95% significance (P < 0.05) are indicated. n.s. = non significant, CYD = currentyear needle defoliation, OND = old-needles defoliation, GP = growing period. Aa

Source

Main plot Block Time Subplot CYD OND Time × CYD Contrasts GP 1 0 vs 33 0 vs 66 0 vs 99 GP 2 0 vs 33 0 vs 66 0 vs 99 GP 3 0 vs 33 0 vs 66 0 vs 99 a

gs a

F

P

0.61 11.80

0.0064

5.33 0.56 3.95

0.0023 n.s. 0.0004

3.41

0.0221

10.16

0.0022

10.96

<.0001

5.15

0.0029

8.31 17.87

0.0053 <.0001

4.56 12.93 2.81 7.19

0.0366 0.0006 0.0461 0.0093

4.50

0.0377

F

P

F

P

0.61 9.92

0.0102

0.44 7.80

0.0416

1.86 1.24 2.78

n.s. n.s. 0.0082

3.28 0.58 2.40

0.0278 n.s. 0.0396

3.43 6.71

0.0234 0.0123

6.72 3.52 7.04

0.0123 0.0210 0.0105

P=n.s.

30.0

δ18O

27.5 25.0 22.5 20.0 17.5 -32

-30

-28

-26

-24

-22

-20

δ13C Fig. 5. Relationship between carbon and oxygen isotope compositions (␦13 C and ␦18 O, respectively) of balsam fir tree-ring cellulose.

3

2001; Leavitt, 2010; Little et al., 2003; Reich et al., 1993; Turnbull et al., 2007; Vanderklein and Reich, 1999; Welter, 1989), or remobilization of reserve carbohydrates stored as starch (Eyles et al., 2009).

SLA

Ln transformed data.

32.5

4.1. Photosynthetic rate A number of studies have shown that some trees can maintain their source/sink carbon balance following partial defoliation through an increase in the photosynthetic rate in the remaining foliage (Chen et al., 2001; Lavigne et al., 2001; Little et al., 2003; Reich et al., 1993; Vanderklein and Reich, 1999; Welter, 1989). Such increases, by reducing ci and subsequent discrimination against 13 C, should also result in 13 C enrichment of recent assimilates (Farquhar et al., 1982). No constant increases in photosynthetic rates were observed in response to the different levels of defoliation (Fig. 6) although a consistent positive trend of A and gs was present during the first growing period (only significant for A CYD66). The lack of repeated measurements during the growing periods lowers the statistical power of the analyses and makes it difficult to find statistically significant responses. However, the very different A and gs responses (positive trend) to defoliation during the first growing period compared to the other growing periods may be important. Increased light availability for the remaining needles following defoliation (Leavitt, 2010; Leavitt and Long, 1986a,b) might be a possible source of variation that influenced A. Higher stomatal conductance in response to increased water availability following defoliation has also been observed to stimulate photosynthesis (Turnbull et al., 2007). Pinkard et al. (2011) showed that the degree of a saplings photosynthetic response to defoliation is influenced by whether or not trees suffered from other limitations (e.g. nitrogen availability) that would limit their carbon sink strength. The strongest photosynthetic responses (increased A) occurred when there were no other limiting constraints on saplings. In this experiment, seedlings were maintained under optimal water and nutrient conditions but did not show a photosynthetic response to defoliation (Fig. 6) as strong as was demonstrated by Pinkard et al. (2011). While needle nitrogen levels were not measured in this experiment, Reich et al. (1998) demonstrated a positive relationship between leaf N and SLA for a wide range of species and functional groups. Both Pinkard et al. (2011) and Quentin et al. (2011) also demonstrated a positive relationship between specific leaf area and fixation capacity of the remaining foliage following partial defoliation. Although many studies highlighted the relationship between

0.10

a) P=0.0004

2 *

gs (molH2O m-2 s-1)

A (μmolCO2 m-2 s-1)

51

1 0 -1 -2

b) P=0.0082

GP 1 GP 2 GP 3 GP 4

0.05 0.00 -0.05 -0.10

*

-3

*

33

66

% CYD

*

99

33

66

99

% CYD

Fig. 6. Light-saturated CO2 assimilation rate (A) (a) n = 6) and stomatal conductance (b) n = 6) (mean ± se) of combined current-year and one-year old foliage of balsam fir seedlings in response to different levels of current-year defoliation measured during the four growing periods (GP). The results are presented as deviation form the control trees (CYD0-PYD0). Stars indicate statistically significant difference with control trees (see Table 3 for P-values).

52

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

6

P=0.0396 *

SLA (m2 kg-1)

4

2

0

-2

-4 *

-6

-8

GP 1 GP 2 GP 3

33

*

66

99

% CYD Fig. 7. Specific leaf area (SLA) of balsam fir seedlings in response to current-year needle defoliation (mean ± se; n = 6) measured during the four growing periods (GP). The results are presented as deviation form the control trees (CYD0-PYD0). Stars indicate statistically significant difference with control trees (see Table 3 for P-values).

defoliation, increased A and increased leaf N, less consistent results were also observed (Reich et al., 1993; Vanderklein and Reich, 2000). SLA did not increase but, overall, rather decreased in this experiment following defoliation of current-year needles although it responded positively to the lighter defoliation treatment during the second growing period. The results showed little evidence of the effects of defoliation on A, gs or SLA and, by extrapolation, needle N (Figs. 6 and 7). Overall, no significant correlations between A, gs , or the ci /ca ratio, and ␦13 C were detected here (data not shown) suggesting mechanisms at the leaf level are not primarily responsible for the 13 C enrichments observed. The absence of a significant relationship between ␦13 C and ␦18 O (Fig. 5) also suggests that stomatal conductance did not have much of an influence on the carbon isotope composition of the tree-ring cellulose (Barbour et al., 2002; Saurer et al., 1997). In this particular case, it is suggested that mobilizations of 13 C enriched stored reserves seem to have been the main defoliation compensating mechanism, but only after moderate to severe episodes. 4.2. Mobilization of reserve carbohydrates Woody tree species have a large capacity for storage of carbon and nutrients. Evidence from the literature suggests that defoliated trees are able to maintain growth at or near pre-defoliation rates by using reserves and decoupling their growth from the reduced carbon assimilation (Eyles et al., 2009; Quentin et al., 2011). Significant mobilization of reserve carbohydrates could account for the high resilience of the seedlings to our defoliation treatments. A large decrease in soluble and reserve carbohydrates in the roots, stems and remaining foliage is typically observed following partial defoliation (Eyles et al., 2009; Pinkard et al., 2011; Quentin et al., 2011; Vanderklein and Reich, 1999) to overcome the negative carbon balance imposed and provide carbon to active growth zones. In a related experiment, Rossi et al. (2009) found no significant effect on cambial activity, timing of xylem differentiation, or on xylem cell anatomy (cell lumen area, radial cell diameter and wall

thickness) following a 66% current-year needle defoliation treatment. This suggests that the seedlings had access to the resources required to maintain stem growth during the four cycles of defoliation. Developing needles act as a sink for carbon and nitrogen before they become fully functional whereas old needles tend to function as a source for both elements. It seems therefore that current-year defoliations decrease the sink strength for C and N (leaving more resources available for other sinks) while losing old needles significantly decreases the source (and reserves) for subsequent growth. In our experiment decreased C and N reserves as a direct result of old-needle defoliation, resulted in reduced radial growth by the second growing period independent from the level of current-year needle defoliation. Reserve starch has been shown to have higher carbon isotope values compared to recent assimilates and therefore, when used to maintain radial growth, contributes to cellulose 13 C enrichment in tree rings (Brugnoli et al., 1988; Damesin and Lelarge, 2003; Helle and Schleser, 2004; Le Roux et al., 2001). In this experiment tree-ring cellulose 13 C enrichments occurred in response to the most severe defoliation treatments (CYD66-OND0 growing period 3, CYD99-OND0 growing period 1, CYD99-OND50 growing period 3; Fig. 4) although the response was not consistent over time. Radial growth was reduced by these treatments but perhaps at a lesser rate than would have been expected if old needle removal was greater (Figs. 2 and 3). It is therefore likely that the mobilization of reserve starch is responsible for some or even most of the tree-ring cellulose carbon isotope enrichments in these treatments although old needle defoliation was only significant for one of the three cases (CYD99-OND50). Less severe defoliation treatments (CYD33-OND0, CYD0OND50, CYD33-OND50, CYD66-OND50) did not yield a similarly significant positive response with tree-ring ␦13 C (Fig. 4) although a trend in 13 C enrichment was observed. The carbon isotopic composition of organic matter is influenced by multiple post-carboxylation fractionations (Brüggemann et al., 2011). It is therefore possible that the higher tree-ring cellulose ␦13 C values (although not statistically significant) for the light defoliation treatments are a result of a mixture of a higher portion of carbon from fresh assimilates, than from reserve carbohydrates. The experimental design employed allowed us to assess four growing periods within two calendar years, a notable benefit allowing greater assessment of carbon allocation and storage effects. However, this design is not without associated limitations. Namely, (i) the growing season during the two summer periods were one month longer than during the two winter periods, (ii) light intensities (but not durations) were generally higher during the summer and (iii) the CO2 carbon isotope compositions in the ambient air were more 13 C depleted during the winter. Greater carbon isotope enrichment was observed during the summer growing periods, suggesting the influence of secondary effects related to the variable environmental conditions during the winter vs summer periods. During the experiment, the seedlings were maintained for several hours (early mornings and late afternoons) under a minimum PPFD of 115 ␮mol m−2 s−1 in winter months (representing approximately 25% of the PPFD needed to reach light saturated photosynthetic rates; (Landhäusser and Lieffers, 2001). Possible differences in rates of photosynthesis between the growing periods might be responsible for some of the tree-ring ␦13 C variation observed. Moreover, along with the atmospheric CO2 concentration being generally higher, the atmospheric ␦13 CO2 is also more depleted in the winter (e.g. −8.4‰ in winter vs −7.4‰ in summer, Alaska) (Boutton, 1991; Bowling et al., 2008; Inoue and Sugimura, 1985). This could explain why, on average, the ␦13 C of winter grown seedlings (periods 1–3) were 2.4‰ more depleted than the summer

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

grown seedlings (periods 2–4). A comparison between the ␦13 C and ␦18 O values with time (Fig. 4a) suggests that variations in the atmospheric CO2 carbon isotope composition and/or light intensities between winter vs summer conditions may account for these observed differences. The rather invariant ␦18 O response with time as opposed to ␦13 C (Fig. 4a) indeed points toward an impact on the carbon metabolism. Nevertheless, control seedlings as well as experimental ones were exposed to the same environmental conditions, thereby promoting defoliation treatment intensities as the primary varying factor. However, the fact that the artificial conditions induced by the experimental design may have affected the physiology of seedlings and their carbon allocation dynamics cannot be excluded. 5. Conclusion The main objective of this research was to further validate the use of carbon isotope compositions in tree rings as an efficient and sensitive indicator of past spruce budworm outbreak periods in light of the more traditional tree-ring width indicators. Additionally, tree-ring cellulose ␦13 C measurements were used to investigate if trees rely on starch reserves to carry on growth and metabolic activities during defoliation outbreaks or if they undergo some other physiological changes. Evidence has emerged suggesting that carbon isotope ratios in tree rings may be in fact sensitive to foliage losses although a mechanistic explanation for this relationship is only speculative at this time. Trends in ␦13 C enrichment were observed in response to the most severe artificial defoliation levels although a high variability in the response was observed among the treatments and growing periods possibly due to a small sample size and the artificial nature of the experimental design. While improvements to the experimental procedure are necessary in the future, this controlled study was still successful in further promoting a relationship between tree-ring cellulose ␦13 C variations and disturbances such as defoliation. Post-carboxylation mechanisms via the mobilization of stored reserves most likely contributed to the enriched carbon isotope signatures in response to defoliation. However, a positive trend in A and gs response to defoliation was observed in the first growing period suggesting possible physiological mechanisms to compensate the effects of defoliation. Further research is needed to confirm, from a statistical point of view, the link between the carbon isotope response at the tree ring level and the physiological processes involved during defoliation. The results also highlighted the resilience of young balsam firs to partial defoliation treatments. The growth of the seedlings in mild defoliation treatments was not significantly altered, likely as a result of compensatory mechanisms as observed in other tree species. These mechanisms include the use of stored carbohydrates and increased photosynthetic activity via increased stomatal conductance following enhanced water availability, increased flow of nitrogen to remaining leaves, or increased photosynthetic activity (maximum rate of carboxylation and/or maximum rate of RuBP regeneration) (Eyles et al., 2009; Quentin et al., 2011; Turnbull et al., 2007). This study helps to promote the use of tree-ring widths in combination with carbon isotopic compositions for reconstructing long-term natural severe defoliation events. Their application for light-defoliating events needs to be investigated further. Acknowledgments This research project was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), The Consortium de Recherche sur la Forêt Boréale Commerciale and the Fondation de l’Université du Québec à Chicoutimi. Sincere thanks go to

53

D. Gagnon, J. Allaire, J.-P. Lebeuf and all those who helped in the development and implementation of the defoliation experiment. We also thank D. Walsh and J.-F. Boucher for support in statistical analysis and discussion. Anonymous reviewers provided valuable comments on earlier versions of the manuscript.

References Barbour, M.M., Farquhar, G.D., 2000. Relative humidity- and ABA-induced variation in carbon and oxygen isotope ratios of cotton leaves. Plant Cell Environ. 23, 473–485. Barbour, M.M., Walcroft, A.S., Farquhar, G.D., 2002. Seasonal variation in ␦13 C and ␦18 O of cellulose from growth rings o Pinus radiata. Plant Cell Environ. 25, 1483–1499. Bernier, P.Y., Raulier, F., Stenberg, P., Ung, C.H., 2001. Importance of needle age and shoot structure on canopy net photosynthesis of balsam fir (Abies balsamea): a spatially inexplicit modeling analysis. Tree Physiol. 21, 815–830. Blais, J.R., 1983. Trends in the frequency, extent, and severity of spruce budworm outbreaks in eastern Canada. Can. J. For. Res. 13, 539–547. Bouchard, M., Kneeshaw, D., Messier, C., 2007. Forest dynamics following spruce budworm outbreaks in the northern and southern mixedwoods of central Quebec. Can. J. For. Res. 37, 763–772. Boulanger, Y., Arseneault, D., 2004. Spruce budworm outbreaks in eastern Quebec over the last 450 years. Can. J. For. Res. 34, 1035–1043. Boutton, T.W., 1991. Stable carbon isotope ratios of natural materials: II. Atmospheric, terrestrial, marine, and freshwater environments. In: Coleman, D.C., Fry, B. (Eds.), Carbon Isotope Techniques. Academic Press, San Diego. Bowling, D.R., Pataki, D.E., Randerson, J.T., 2008. Carbon isotopes in terrestrial ecosystem pools and CO2 fluxes. New Phytol. 178, 24–40. Brüggemann, N., Gessler, A., Kayler, Z., Keel, S.G., Badeck, F., Barthel, M., Boeckx, P., Buchmann, N., Brugnoli, E., Esperschütz, J., Gavrichkova, O., Ghashghaie, J., Gomez-Casanovas, N., Keitel, C., Knohl, A., Kuptz, D., Palacio, S., Salmon, Y., Uchida, Y., Bahn, M., 2011. Carbon allocation and carbon isotope fluxes in the plant-soil-atmosphere continuum: a review. Biogeosciences 8, 3457–3489. Brugnoli, E., Hubick, K.T., Von Caemmerer, S., Wong, S.C., Farquhar, G.D., 1988. Correlation between the carbon isotope discrimination in leaf starch and sugars of C3 plants and the ratio of intercellular and atmospheric partial pressures of carbon dioxide. Plant Physiol. 88, 1418–1424. Chen, Z., Kolb, T.E., Clancy, K.M., 2001. Mechanisms of Douglas-fir resistance to western spruce budworm defoliation: bud burst phenology, photosynthetic compensation and growth rate. Tree Physiol. 21, 1159–1169. Damesin, C., Lelarge, C., 2003. Carbon isotope composition of current-year shoots from Fagus sylvatica in relation to growth, respiration and use of reserves. Plant Cell Environ. 26, 207–219. Deslauriers, A., Morin, H., Begin, Y., 2003. Cellular phenology of annual ring formation of Abies balsamea in the Quebec boreal forest (Canada). Can. J. For. Res. 33, 190–200. Devore, J., Peck, R., 1994. Introductory Statistics, 2nd ed. West Publishing Company, St-Paul. Eyles, A., Pinkard, E.A., Mohammed, C., 2009. Shifts in biomass and resource allocation patterns following defoliation in Eucalyptus globulus growing with varying water and nutrient supplies. Tree Physiol. 29, 753–764. Farquhar, G.D., Oleary, M.H., Berry, J.A., 1982. On the relationship between carbon isotope discrimination and the inter-cellular carbon-dioxide concentration in leaves. Aust. J. Plant Physiol. 9, 121–137. Francey, R.J., Farquhar, G.D., 1982. An explanation of 13 C/12 C variations in tree rings. Nature 297, 28–31. Gauthier, S., Vaillancourt, M.A., Leduc, A., De Grandpré, L., Kneeshaw, D., Morin, H., Drapeau, P., Bergeron, Y., 2008. Aménagement écosystémique en forêt boréale. Presses de l’Université du Québec, Québec. Gumpertz, M.L., Brownie, C., 1993. Repeated measures in a randomized block and split-plot experiment. Can. J. For. Res. 23, 625–639. Hebert, F., Boucher, J.F., Bernier, P.Y., Lord, D., 2006. Growth response and water relations of 3-year-old planted black spruce and jack pine seedlings in site prepared lichen woodlands. For. Ecol. Manag. 223, 226–236. Helle, G., Schleser, G.H., 2004. Beyond CO(2)-fixation by Rubisco—an interpretation of (13)C/(12)C variations in tree rings from novel intra-seasonal studies on broad-leaf trees. Plant Cell Environ. 27, 367–380. Inoue, H., Sugimura, Y., 1985. The carbon isotopic ratio of atmospheric carbondioxide at Tsukuba, Japan. J. Atmos. Chem. 2, 331–344. Krause, C., 1997. The use of dendrochronological material from buildings to get information about past spruce budworm outbreaks. Can. J. For. Res. 27, 69–75. Kress, A., Saurer, M., Buntgen, U., Treydte, K.S., Bugmann, H., Siegwolf, R.T.W., 2009. Summer temperature dependency of larch budmoth outbreaks revealed by Alpine tree-ring isotope chronologies. Oecologia 160, 353–365. Landhäusser, S.M., Lieffers, V.J., 2001. Photosynthesis and carbon allocation of six boreal tree species grown in understory and open conditions. Tree Physiol. 21, 243–250. Lavigne, M.B., Little, C.H.A., Major, J.E., 2001. Increasing the sink: source balance enhances photosynthetic rate of 1-year-old balsam fir foliage by increasing allocation of mineral nutriments. Tree Physiol. 21, 417–426. Le Roux, X., Bariac, T., Sinoquet, H., Genty, B., Piel, C., Mariotti, A., Girardin, C., Richard, P., 2001. Spatial distribution of leaf water-use efficiency and carbon

54

S. Simard et al. / Environmental and Experimental Botany 81 (2012) 44–54

isotope discrimination within an isolated tree crown. Plant Cell Environ. 24, 1021–1032. Leavitt, S.W., 2010. Tree-ring C H O isotope variability and sampling. Sci. Total Environ. 408, 5244–5253. Leavitt, S.W., Danzer, S.R., 1993. Method for batch processing small wood samples to holocellulose for stable-carbon isotope analysis. Anal. Chem. 65, 87–89. Leavitt, S.W., Long, A., 1986a. Influence of site disturbance on ␦13 C isotopic time series from tree rings. In: Jacoby, G.C., Hornbeck, J.W. (Eds.), Proceedings of the International Symposium on Ecological Aspect of Tree-Ring analysis. Tarrytown, New York. Leavitt, S.W., Long, A., 1986b. Stable-carbon isotope variability in tree foliage and wood. Ecology 67, 1002–1010. Little, C.H.A., Lavigne, M.B., Ostaff, D.P., 2003. Impact of old foliage removal, stimulating defoliation by the balsam fir sawfly, on balsam fir tree growth and photosynthesis of current-year shoots. For. Ecol. Manag. 186, 261–269. MacLean, D.A., Hunt, T.L., Eveleigh, E.S., Morgan, M.G., 1996. The relation of balsam fir volume increment to cumulative spruce budworm defoliation. For. Chron. 72, 533–540. Morin, H., 1994. Dynamics of balsam fir forests in relation to spruce budworm outbreaks in the boreal zone of Quebec. Can. J. For. Res. 24, 730–741. Pinkard, E.A., Eyles, A., O’Grady, A.P., 2011. Are gas exchange responses to resource limitation and defoliation linked to source:sink relationships? Plant Cell Environ. 34, 1652–1665. Quentin, A.G., Beadle, C.L., O’Grady, A.P., Pinkard, E.A., 2011. Effects of partial defoliation on closed canopy Eucalyptus globulus Labilladière: growth, biomass allocation and carbohydrates. For. Ecol. Manag. 261, 695–702. Quinn, G.P., Keough, M.J., 2002. Experimental Design and Data Analysis for Biologists. Cambridge University Press, UK. Reich, P.B., Ellsworth, D.S., Walters, M.B., 1998. Leaf structure (specific leaf area) modulates photosynthesis-nitrogen relations: evidence from within and across species and functional groups. Funct. Ecol. 12, 948–958. Reich, P.B., Walters, M.B., Krause, S.C., Vanderklein, D.W., Raffa, K.F., Tabone, T., 1993. Growth, nutrition and gas-exchange of pinus resinosa following artificial defoliation. Trees Struct. Funct. 7, 67–77.

Rossi, S., Simard, S., Deslauriers, A., Morin, H., 2009. Wood formation in Abies balsamea seedlings subjected to artificial defoliation. Tree Physiol. 29, 551–558. Saurer, M., Aellen, K., Siegwolf, R., 1997. Correlating ␦13 C and ␦18 O in cellulose of trees. Plant Cell Environ. 20, 1543–1550. Scheidegger, Y., Saurer, M., Bahn, M., Siegwolf, R., 2000. Linking stable oxygen and carbon isotopes with stomatal conductance and photosynthetic capacity: a conceptual model. Oecologia 125, 350–357. Simard, S., Elhani, S., Morin, H., Krause, C., Cherubini, P., 2008. Carbon and oxygen stable isotopes from tree-rings to identify spruce budworm outbreaks in the boreal forest of Quebec. Chem. Geol. 252, 80–87. Simard, S., Morin, H., Krause, C., 2011. Long-term spruce budworm outbreak dynamics reconstructed from subfossil trees. J. Quat. Sci. 26, 734–738. Turnbull, T.L., Adams, M.A., Warren, C.R., 2007. Increased photosynthesis following partial defoliation of field-grown Eucalyptus globulus seedlings is not caused by increased leaf nitrogen. Tree Physiol. 27, 1481–1492. Vanderklein, D.W., Reich, P.B., 1999. The effect of defoliation intensity and history on photosynthesis, growth and carbon reserves of two conifers with contrasting leaf lifespans and growth habits. New Phytol. 144, 121–132. Vanderklein, D.W., Reich, P.B., 2000. European larch and eastern white pine respond similarly during three years of partial defoliation. Tree Physiol. 20, 283–287. von Ende, C.N., 1993. Repeated-measures analysis: growth and other timedependant measures. In: Scheiner, S.M., Gurevitch, J. (Eds.), Design and Analysis of Ecological Experiments. Chapman and Hall, Inc., New York. Weidner, K., Heinrich, I., Helle, G., Loffler, J., Neuwirth, B., Schleser, G.H., Vos, H., 2010. Consequences of larch budmoth outbreaks on the climatic significance of ring width and stable isotopes of larch. Trees Struct. Funct. 24, 399–409. Welter, S.C., 1989. Arthropod impact on plant gas exchange. In: Bernays, E.A. (Ed.), Insect–Plant Interactions. CRC Press, Inc., Boca Raton, FL. Wright, W.E., 2008. Statistical evidence for exchange of oxygen isotopes in holocellulose during long-term storage. Chem. Geol. 252, 102–108.