Seasonal patterns of foliage respiration in dominant and suppressed Eucalyptus globulus canopies

Transcription

1 1 Seasonal patterns of foliage respiration in dominant and suppressed Eucalyptus globulus canopies 1,2,3 O Grady, A.P., 1,3 Eyles, A., 2,3 Worledge, D., 2,3 Battaglia, M. 1 School of Plant Science, University of Tasmania, Private Bag 55, Hobart, Tasmania 7001, Australia 2 CSIRO Sustainable Ecosystems, Private Bag 12, Hobart, Tasmania 7001, Australia 3 CRC Forestry, Private Bag 12, Hobart, Tasmania 7001, Australia.

2 2 Summary We examined spatial and temporal dynamics of foliage respiration in canopies of dominant and suppressed Eucalyptus globulus trees to better understand processes regulating foliage respiration in a young fast growing stand. Temperature response functions and seasonal measures of respiration (measured at a reference temperature 15 C, R 15 ) were studied for approximately 1 year to examine; 1. controls on respiration as a function of canopy position, foliar nitrogen and non-structural carbohydrates concentrations and 2. to assess the capacity for thermal acclimation within E. globulus canopies. The short-term temperature response of respiration varied both with canopy position and seasonally. Area-based R 15 measurements declined with increasing canopy depth and were strongly related to foliar N concentrations, especially in upper canopy positions. R 15 was negatively correlated with the average temperature of the preceding 14 days, a pattern consistent with thermal acclimation. In suppressed canopies, R 15 was higher than that at similar canopy heights in dominant trees. Similarly foliar concentrations of non-structural carbohydrates were also relatively higher in suppressed canopies than dominant canopies providing support for a substrate based model of leaf respiration. Our data highlights the dynamic nature of foliar respiration within E. globulus canopies which contrasts with the generally simplistic representation of respiration within most process-based models. Key words Thermal acclimation, size class distributions, Eucalyptus, non-structural carbohydrates

3 3 Introduction Respiration is the process that oxidises photosynthates to produce the carbon skeletons and intermediates required for biosynthesis and thus provide the energy required for maintenance, growth and other metabolic activities such as ion-uptake. Approximately half of the carbon fixed during photosynthesis is released via respiration. Thus, plant respiration plays a major role in the global carbon cycle, returning approximately 60 Gt C yr -1 to the atmosphere (Schimel 1995). By comparison, releases from anthropogenic sources such as the burning of fossil fuels or changes in land use are considerably smaller (5.5 and 1.6 Gt C yr -1 respectively (Schimel 1995; King et al. 2006). At the scale of the individual plant, up to two thirds of daily photosynthetic carbon gain may be released from the foliage, stem and roots via respiration (Atkin and Tjoelker 2003). Foliage respiration contributes significantly to the overall carbon balance of the plant (Cavaleri et al. 2008) and thus plays a critical role in determining the primary productivity of plant canopies (Griffin et al. 2001; Whitehead et al. 2004a) Despite this, we still have only a limited mechanistic understanding of the processes regulating respiration within plant canopies. In particular, few studies have examined in detail spatial and temporal patterns of respiration within the canopy (Tissue et al. 2002; Turnbull et al. 2003; Whitehead et al. 2004a). An improved understanding of this nature is critical to better understand growth responses under contrasting environmental conditions, and to improve our understanding of the role that forest canopies can play in mitigating the effects of climate change in response to increasing atmospheric CO 2 concentrations in warming climate.

4 4 Respiration is a highly temperature dependent process. Over short time scales, respiration rates increase exponentially with temperature (Ryan et al. 1996; Atkin and Tjoelker 2003) such that it is commonly assumed that respiration rates approximately double for each 10 C increase in ambient temperature (Q 10 ~ 2.0, Ryan 1991; Tjoelker et al. 2001). However, reported values of leaf Q 10 vary considerably, reflecting differences between species, growth rates, measurement temperatures and the physiological conditions at the time of measurement, for example, the degree of water stress (Atkin et al. 2000a; Loveys et al. 2003; Wright et al. 2006). There is also increasing evidence that thermal acclimation, even over relatively short time frames of one to two days (Atkin et al. 2000b; Atkin and Tjoelker 2003; Tjoelker et al. 2009) can modulate respiration rates relative to that expected from short term temperature response functions and result in reversible shifts in the shape or elevation of the temperature response function (Tjoelker et al. 2009). For example, an increase in growth temperature is generally associated with a downward shift in respiration and conversely, an upwards adjustment of respiration at cooler temperatures (Atkin and Tjoelker 2003; Tjoelker et al. 2008). The mechanisms underpinning temperature acclimation are poorly understood, but an increasing body of evidence suggests that supply and demand for substrates may play an important role in thermal acclimation and regulation of respiration (Dewar et al. 1999; Atkin and Tjoelker 2003) as respiration rates often co-vary with the concentrations of foliar non-structural carbohydrates. Thus respiration rates are often higher in leaves with higher concentrations of soluble carbohydrates (Griffin et al. 2002; Xu and Griffin 2006). Furthermore, Tjoelker et al. (2008) observed that

5 5 acclimation in Pinus banksiana Lamb. was strongly correlated to soluble carbohydrate concentrations and in a following study, Tjoelker et al. (2009) further observed that seasonal regulation of respiration was driven by increased base respiration rates at cold temperatures mediated principally by increased soluble sugar concentrations and less so via changes in foliar N. These observations contrast with the current representation of respiratory processes within existing modelling frameworks where it is often assumed that respiration is a constant fraction of photosynthetic gain, i.e. carbon use efficiency is constant and conservative (Dewar et al. 1999). While this approach has proven to be reasonably robust over long time scales, it has failed to capture the spatial and temporal flexibility of respiratory processes (Kruse and Adams 2008). Foliar respiration is a key leaf trait (Wright et al. 2004), which also encompasses, among other traits, photosynthesis and foliar nitrogen concentrations. Several studies have demonstrated strong correlations between foliar respiration rates and leaf nitrogen concentration (Ryan, 1995; Reich et al. 1998; Vose and Ryan 2002; Reich et al. 2006; Reich et al. 2008) and it has been shown that these relationships are broadly consistent across a range of species and ecosystems indicating considerable functional convergence within this key life history trait (Reich and Borchet 1988; Wright et al. 2004). Foliar nitrogen also varies considerably within canopies and this variation is often described in relation to the light-nitrogen hypothesis (Field 1983). Thus, nitrogen is distributed within canopies in a manner that maximises canopy production, and gradients in photosynthesis related to light intensity and foliar nitrogen are commonly observed (Warren and Adams 2001; Frak et al. 2002; Whitehead et al. 2004b; O'Grady et al. 2008b). As photosynthesis, foliar nitrogen and respiration are

6 6 strongly correlated, it is also commonly observed that foliar respiration rates are higher in the upper canopy leaves than in the lower canopy (Griffin et al. 2001; Tissue et al. 2002; Whitehead et al. 2004a). Therefore understanding how nitrogen is distributed within canopies provides an important framework for extrapolating leaf scale measurements of respiration to the canopy scale. A characteristic feature of forest growth in even aged stands is the differentiation of trees into differing size classes, a process thought to be an important driver of stand productivity and the commonly observed later-aged decline in forest productivity (Smith and Long 2001; Binkley et al. 2004). Despite being well recognised, a mechanistic understanding of this process is lacking. A recent study of the distribution of photosynthesis and respiration within E. globulus canopies demonstrated strong gradients in these two parameters that were related to light intensity and foliar N (O'Grady et al. 2008b). Furthermore, O'Grady et al. (2008b) observed that area based respiration rates of fully expanded leaves tended to be higher in canopies of suppressed trees relative to that of dominant trees, despite lower growth rates in suppressed trees (Drew et al. 2008), suggesting that suppressed trees had a lower carbon use efficiency, and highlighting the critical role that respiration plays in determining productivity within plant canopies. However, the underlying mechanism for this result remained unresolved. In the current study, we have examined in detail spatial and temporal patterns of leaf respiration within canopies of even-aged E. globulus trees, from contrasting size classes. We aimed to better understand how foliage respiration varies within canopies and in relation to foliar nitrogen and non-structural carbohydrate concentrations.

7 7 Furthermore, we examined seasonal variation in the temperature response of foliar respiration to assess acclimation capacity in E. globulus canopies. We examined respiration in dominant and suppressed canopies growing under uniform environmental conditions assuming that trees of these two contrasting size classes would represent contrasting demands for respiratory substrates and that this may influence respiration rates. Specifically we aimed to: Quantify spatial and temporal patterns of leaf respiration in dominant and suppressed trees with an even-aged stand to better understand the distribution of respiration within forest canopies. Determine temperature response functions and assess the degree of acclimation in measured reference respiration rates (R 15 ) as a function of canopy position and dominance class. And to examine how R 15 co-varies with foliar N and carbohydrate concentrations in dominant and suppressed canopies of trees within an evenaged stand. Methods Site description Research was conducted at the Pittwater research plantation (S 42º 94, E 147º 30 ), approximately 20 km east of Hobart in south-east Tasmania, Australia. The plantation was established in September 2002 and has a climate classified as cool temperate maritime. Average rainfall at the site is approximately 500 mm year -1 and annual evaporation in excess of 1300 mm year -1. Mean daily maximum and minimum

8 8 temperatures vary between 22.5ºC and 12.5ºC and 12ºC and 4ºC for summer and winter respectively (Bureau of Meteorology Soils at the site are duplex (Podosol, Isbell 1996) with an Aeolian derived sandy A-horizon, 1-2 m deep, overlying a sandy clay to clay B horizon at The site consisted of nine growth plots each containing 25 trees planted at a spacing of 3 x 3 m (density 1111 stems ha -1 ). The plantation was divided into two experimental treatments, six irrigated plots and three rain-fed plots. The current study describes work conducted in three of the irrigated plots only. In March 2006, scaffold towers that were designed to give access to approximately 90% of tree height, were constructed around six trees, a dominant and suppressed tree in each of three plots. Each tower contained three narrow platforms providing access to the northern aspect of each tree (sunlight side of the tree). In towers on suppressed trees, platforms were positioned to provide access to the canopy at 3, 5.5, 8.0 and 11 m above ground level (a.g.l, 11 m accessed using pole pruners). In towers on dominant trees, platforms were positioned to provide access to the canopy at 7, 9.5, 12.0 and 15 m a.g.l (15 m, accessed using pole pruners). The plantation has been fertilized on a regular basis since establishment using a full compliment of macro and micro nutrients. A full description of the fertilization regime used at the site is given in O'Grady et al. (2005). Climate Climate at the site was monitored using an automatic weather station, installed approximately 100 m NE of the plantation in an open field. A detailed description of the instrumentation used is given in O'Grady et al. (2008a).

9 9 Foliar respiration (R 15 ) Measurements of dark respiration (R 15 ) followed the protocols outlined by O'Grady et al. (2008b). Briefly, for the measurement of R 15, we used a custom built Perspex chamber of the dimensions 10cm x 10cm x 4cm. The chamber contained an internal fan to ensure adequate mixing within the chamber. Prior to measurements, the chamber was pressure tested by immersing in water and pressurising the chamber to ensure that there were no leaks associated with chamber construction. A copperconstantan thermocouple was placed inside the chamber for monitoring chamber temperature and leaf temperature was assumed to be equilibrated to chamber temperature. All respiration measurements were made using a CIRAS infrared gas analyzer (PP Systems, Herts, UK) with ambient CO 2 concentrations at 370 ppm. Foliar respiration (R 15 ) was measured every 4 to 8 weeks from August 2007 through to June At each sampling period, three fully expanded leaves from the same age cohort (third or fourth leaf from the growing tip) were collected from the northern (sunlit) side of the canopy at two locations: upper and lower positions in the canopy. In suppressed trees leaves were collected at 3m and 11m a.g.l., and in dominant trees, leaves were collected at 8m and 15m a.g.l. All leaves were collected at dawn before there was any direct sunlight on the canopy. Thus, measurements of R 15 were made on detached leaves that were kept in the dark at 15ºC until measurement. Previous studies have demonstrated no differences in rates of respiration of detached or in situ leaves (Turnbull et al. 2001; Callister and Adams 2006; Xu and Griffin 2006) and we confirmed this for our E. globulus leaves. Furthermore rates of dark respiration were

10 10 stable for many hours after detachment (O'Grady et al. 2008b). All measurements were made at constant CO 2 concentration of 370 ppm in a constant temperature room set at 15ºC. Measurement times usually took approximately 20 min or until we were confident that CO 2 evolution was stable (coefficient of variation <1.0%). After each measurement was completed leaves were stored, at -20 o C before processing leaves for specific leaf area (SLA) and foliar chemistry. Leaf area was measured using the scanning software WinRhizo (Regent Instruments, Quebec) and leaves were dried to constant weight at 65 o C. Temperature response In addition to the measurement of seasonal R 15 described above, temperature response functions were developed at three times during the study period; August 2007, November 2007 and May The measurement protocols were similar to those described above. However, measurements were made at four temperatures in a temperature control room; 5, 10, 15 and 20ºC. Furthermore measurements were made at four heights in the canopy of both dominant and suppressed trees (3, 5.5, 7 and 11 m and 7, 9.5, 12 and 15 m in suppressed and dominant canopies respectively). The temperature response of respiration was evaluated using a modified Arrhenius function (Lloyd and Taylor 1994): Eo 1 1 R Rg To Ta = R e o where R o is the respiration rate at a base temperature (15 C, 288 K), T a is the measurement temperature (K) of R and R g is the ideal gas constant (8.314 J mol -1 K -1 ). The model makes the simplifying assumption that the process of respiration can be represented as a single chemical reaction where E o describes the energy of activation

11 11 associated with this simplified chemical reaction. This has been shown to be a reasonable assumption for temperate species (Lyons and Raison 1970). The temperature response of respiration is described by the intercept (R o base respiration rate) and curvature which is a function of both R o and E o (Xu and Griffin 2006). Foliar nitrogen Leaf samples associated with measurements of R 15 were analysed for foliar nitrogen concentration (foliar [N]). Leaves were dried at 70ºC to constant weight and then ground in a cyclone mill. Nitrogen digests followed the sulphuric acid and hydrogen peroxide method of (Lowther 1980). Digested samples were colorimetrically analysed for N (QuikChem method D, Lachat Instruments, Wisconsin, USA) on a continuous-flow injection analyzer (QuikChem 8000, Lachat instruments). Standard samples of known N concentration and blank samples were included to validate the efficiency of digestion and elemental analysis. Foliar non-structural carbohydrates Leaf material for the analysis of leaf carbohydrate concentrations were ground in liquid nitrogen. Starch (St) and Soluble sugars (SS) (g mg -1 ) were extracted using the method (Palacio et al. 2007). Soluble sugars were extracted from 50 mg of dried tissue with a solution of 10 ml of 80% (v/v) ethanol in a 60 C water bath. Soluble starch and any remaining complex sugars were extracted from the resulting residue with 0.2 M sodium acetate and 0.5% amyloglucosidase (Fluka-10115). The concentrations of SS and St were analysed using the Dubois et al. (1956) method as

12 12 modified by (Buysse and Merckx 1993), using a phenol/sulphuric acid colorimetric assay. Absorbance was read at 490 nm on a spectrophotometer (UV-VIS). A glucose solution was used for the standard. Concentration of St was referred to in glucose units (Palacio et al., 2007). Total non-structural carbohydrate (TNC) values were obtained by addition of SS and St values. Data analysis Arrhenius functions were fitted using the solver routines in Microsoft Excel which uses iterative non linear curve fitting procedures to minimise residuals between observed and predicted models. Differences in the derived R 15 and E 0 parameters as a function of dominance class and canopy position were analysed using a group regressions approach (Zar 1999). Seasonal patterns in measured R 15 and foliar nonstructural carbohydrates were analysed using a two-way repeated measures generalized linear model routine with the factors canopy height and date after checking for data normality and variance homogeneity. Post hoc comparisons of means were made using Ryan s test (Ryan 1960). Statistical analyses were conducted using the statistical software Genstat v. 7.1 (VSN International, UK). Results There was a significant (p<0.01) relationship between canopy height and specific leaf area (SLA). SLA increased with depth in the canopy from 3.6 ± 0.08 m 2 kg -1 at 15 m above ground level (a.g.l.) to 5.1 ± 0.07 m 2 kg -1 at 3 m (a.g.l.). There were no differences in the relationship between height and SLA between dominant and

13 13 suppressed trees. SLA=-0.11 (height) + 5.3, r 2 =0.82. Despite the gradient in SLA within canopies, seasonal and spatial patterns in area-based and mass-based respiration within the canopy were similar, thus in this paper we concentrated on area based respiration rates. Seasonal patterns of measured R 15 Measured R 15 varied both as a function of canopy position and season. At all sampling times R 15 was highest in the upper canopies of the dominant trees (15 m), intermediate in the upper canopies of suppressed trees (11 m) and lowest in the lower canopy positions, 3 (suppressed) and 7m (dominant, Fig. 1). Over the course of the study, there was a significant trend for declining in R 15 in the upper canopy of both dominant and suppressed trees (Fig 1). R 15 in the upper canopy of the dominant trees (15 m) declined from 0.84 ± 0.09 µmol m -2 s -1 in August 2007 to 0.45 ± 0.01 µmol m - 2 s -1 in June Changes in R 15 at 15 m were strongly correlated with changes in foliar [N] (R 15 =0.66[N]-2.00, r 2 =0.79, p=0.01) indicating that changes in foliar [N] contributed significantly to the observed decline in R 15 at 15 m. Foliar [N] at 15 m declined from 4.3 g m -2 in August 2007 to 3.6 g m -2 in June R 15 in the upper canopy of suppressed trees at 11 m declined from 0.45 ± 0.14 µmol m -2 s -1 in August 2007 to 0.32 ± 0.02 µmol m -2 s -1 in June 2008 (p<0.05). In the suppressed trees, foliar [N] at 11 m also declined from 3.7 ± 0.3 g m -2 in August 2007 to 2.9 ± 0.3 g m -2 in June 2008 (Fig. 1), however the correlation between R 15 and foliar N at 11 m was weak and non significant (R 15 =0.23[N]-0.32, r 2 =0.27). In contrast, R 15 and foliar [N] in the lower canopy position (7m dominant, 3m suppressed) remained relatively stable throughout the study period. In general, R 15 and foliar [N] were strongly related such

14 14 that when R 15 and foliar [N] data from all seasons and canopy positions were pooled there was a strong exponential relationship (Fig. 2). Temperature response Short-term respiration responses in August 2007, November 2007 and May 2008 were strongly temperature dependent (generally r 2 > 0.9) and varied significantly as a function of tree dominance class and canopy position (Table 1). There were significant differences in the derived R 15 as a function of canopy position in all three sampling periods. R 15 increased with increasing height in the canopy. Group regression analysis found no differences in the slope of the relationship between R 15 and canopy position between dominant and suppressed trees however, suppressed trees had a significantly higher intercept than dominant trees (Fig. 3a). When R 15 was normalised to account for distance from the growing tip there was convergence in the relationship between R 15 and height for the dominant and suppressed trees (Fig. 3b), suggesting that distance to the growing tip of trees is a useful scalar of respiration in these E. globulus canopies. In contrast, there were no significant differences in E 0 as a function of canopy position or between dominant and suppressed trees (mean ± s.e, 73.7 ± 2.3 kj mol -1 ). A summary of the derived R 15 and E 0 parameters from the temperature response functions are given in Table 1. Foliar non-structural carbohydrates

15 15 Soluble sugars on average comprised 70% of the total non structural carbohydrates pool. Mean SS and St concentrations in suppressed trees were 65.2 ± 4.8 and 68.5 ± 2.7 mg g -1 (SS) and 25.6 and 32.4 mg g -1 (St) at 3 and 11 m respectively. In dominant trees at SS and St were 55.7 ± 2.3 and 64.1 ± 2.8 mg g -1 (SS) and 19.2 ± 2.1 and 36.5 ± 5.8 mg g -1 (St) at 7 and 15 m respectively. Area-based foliar soluble sugar (SS) and starch (St) concentrations were variable and varied significantly as a function of height (SS; p< and St; p<0.001) and date (SS; p=0.048, St; p=0.001). Vertical gradients in SS were largely driven by changes in SLA, and thus when this was taken into account, variation in SS concentrations with height were not significant. In contrast both area-based and mass-based St concentrations varied significantly with canopy position. Both SS and St were positively correlated with R 15 (Fig. 4). There was a positive and significant correlation between foliar starch concentrations and R 15 (p<0.01 for both area based and mass based relationships) however the correlation between R 15 and SS was non-significant. Overall SS, St and thus total nonstructural carbohydrates were significantly higher in the upper canopy of both the dominant and suppressed trees than in the lower canopies, however there was considerable seasonal variability. In general, area-based TNC concentrations were relatively higher for a given canopy position in the suppressed trees than in the dominant trees. Seasonal patterns in carbohydrate pools were variable with no consistent patterns observed at each canopy position. Temporal dynamics of SS and St are shown in figure 5. Thermal acclimation

16 16 Given the strong relationship between foliar [N] and R 15 (Fig. 2) and the observed temporal decline in foliar [N] in the upper canopy it was difficult to assess the extent of thermal acclimation using area-based measurements of respiration. Thermal acclimation within the canopy was therefore assessed on the basis of respiration (measured at the reference temperature, 15 o C) per unit nitrogen (R N, µmol g -1 s -1 ). There were significant (p<0.05) negative correlations between R N and the average daily temperature of the preceding two week period at all canopy positions (Fig 6) although the relationship was more variable at 15 m (p=0.043) and 11m (p=0.011) than at lower canopy positions; 7m (p=0.02) 3 m (p=0.01). Discussion Respiratory response to temperature in dominant and suppressed canopies Our results demonstrate that in the E. globulus trees studied, the response of respiration to temperature varied significantly as a function of canopy position and season. These results suggest that modeled estimates of canopy respiration that rely on a single temperature response function within the canopy could be significantly biased. We observed that R 15 declined with increasing canopy depth and that this response was consistent seasonally, although the magnitude of this decline varied seasonally (Table 1). In contrast, there were no differences in the E 0 parameter, either seasonally or as a function of canopy position, suggesting that the energy of activation of dark respiration is stable and that respiration pathways and substrates sources are similar. These results are consistent with previous observations in Quercus rubra (Xu and Griffin 2006) but contrast with those of Turnbull et al. (2003) who found that E 0

17 17 was lower in lower canopy leaves than upper canopy leaves of trees in a deciduous oak forest near New York and in a temperate rainforest in New Zealand. The value of E 0 reported here, ~73.7 kj mol -1 was similar to that previously reported for E. globulus (74.2 kj mol -1, (O'Grady et al. 2008b) but is higher than that reported for Q. rubra (52 kj mol -1, (Xu and Griffin 2006) or for the six species studied by (Turnbull et al. 2003). Reference respiration rates (R 15 ) varied between dominant and suppressed trees. The intercept of the relationship between canopy height and R 15 (Fig. 3a) was significantly higher in suppressed trees than in dominant trees and this pattern is similar to that reported by (O'Grady et al. 2008b). Despite this, there are important differences between these two studies. In the O Grady et al., (2008b) study we pooled seasonal measures of respiration measured at ambient temperatures to generate generalised temperature response functions, acknowledging that this approach would effectively mask acclimation and could influence the interpretation of our results. In this study we aimed to redress this simplification by measuring full temperature response functions seasonally and at four canopy positions in each dominance class as opposed to three in the previous study. Nevertheless the observation that R 15 was relatively higher in the suppressed trees compared to dominant trees was upheld in this study. Within the canopies of these E. globulus trees R 15 varied principally as a function of the distribution of light and foliar [N]. O Grady et al. (2008b) examined in detail vertical gradients in the light environment of trees in both the dominant and suppressed size classes and found no significant difference between light environments at any given height within dominant and suppressed canopies, nor did

18 18 we observe differences in foliar [N] between branches at the same height in dominant and suppressed trees. However despite these observations, R 15 at any given height was higher in suppressed trees than in dominant trees. Differences in function at similar canopy heights in dominant and suppressed canopies have been observed previously. For example, it has been observed that branches on suppressed trees will continue to grow, while branches in the same light environment on dominant trees die (Takenaka 2000; Henriksson 2001; Sprugel 2002). Thus, despite the fact that dominant trees tend to have more resources to allocate, branches on suppressed trees are able to grow at solar irradiances where branches on dominant trees die - a process that has been termed correlative inhibition (Takenaka 2000; Sprugel 2002). If resources are distributed within canopies in a manner that optimises production (Field 1983), this tends to be towards branches higher in the canopy in more favorable light environments. Thus branches at the same height above ground level are relatively higher within their individual canopies than branches at a similar height on a dominant tree and thus might be expected to function in a manner similar to branches at the same relative height as that of branches on dominant trees (Sprugel 2002). Our data provides some support for this observation. When respiration was expressed as a function of relative height (Fig. 3b) there was considerable convergence between the respiration rates of dominant and suppressed trees. While correlative inhibition may provide a useful framework for explaining the observed differences in R 15 between dominant and suppressed trees, what underlying mechanisms would explain these observations? Gradients in R 15 within the canopies of both dominant and suppressed trees appear to be driven primarily by gradients in foliar [N] and non structural carbohydrates (Fig. 2, 4). In this study we observed that

19 19 the non-structural carbohydrates in the foliage of suppressed trees were relatively higher at a given canopy position than that of dominant trees. This observation may help to explain these results when considered in context of the substrate-based model of leaf respiration formulated by Dewar et al. (1999). Central to the Dewar et al., (1999) model of leaf respiration is the concept that respiration is limited by the supply of carbohydrates fixed through photosynthesis. At any given height above ground level, light intensity and light saturated rates of photosynthesis are similar in dominant and suppressed trees (O'Grady et al. 2008b), suggesting that the capacity for generation of photosynthates is similar, however dominant trees exhibited higher over all growth rates (Drew et al. 2008) and this would increase demand for respiratory substrates relative to the suppressed trees (Lavigne et al. 1996; Lavigne and Ryan 1997). Thus if a proportionally larger percentage of leaf carbohydrates in dominant trees is exported to meet higher growth demands of other tree organs, the concentration of carbohydrates available locally for leaf respiration would be reduced, and could potentially limit leaf respiration. Thus while photosynthetic rate at a given point in a tree canopy seems to be determined by the stand level conditions, principally light environment and foliar [N] at that point, respiration is also dependent on substrate availability which may be regulated by demand for substrates in other plant organs. These observations provide further evidence for a substrate mediated model of respiration within plant canopies. There is considerable potential for further research in this area as there is increasing focus on the processes underlying the differentiation of trees within stands into various dominance classes (Binkley 2004), as it is becoming increasingly recognised that

20 20 stand structure plays a crucial role in determining above ground productivity (Smith and Long 2001). Seasonal patterns of measured R 15 There were distinct differences in the seasonal patterns of area-based R 15 at different canopy positions. In the upper canopy positions of both dominant and suppressed trees (15 m and 11 m respectively) R 15 declined over the course of the study. In contrast, at lower canopy positions (7 m and 3 m in dominant and suppressed tree respectively) area-based respiration rates exhibited a more distinct seasonal pattern and were higher during the cooler winter months than during summer (Fig. 1). For example, R 15 at 3 m in the suppressed trees were 45% lower during summer when daily temperatures averaged 18 C compared to winter when daily temperatures average 8 C. The decline in R 15 in the upper canopy leaves was associated with a decline in foliar [N] in these upper canopy positions. Foliar [N] within canopies is dynamic and strongly related to light intensity. For example, Whitehead et al. (2004a) found that foliar [N] declined significantly in experimentally shaded leaves in the upper canopy of Q. rubra trees after only six days. Our study was conducted in a relatively young (5-6 yr old), actively growing E. globulus stand with high leaf area index (~6). Therefore, although measurements were conducted at a constant height above ground level, the relative position and hence light intensity (O Grady et al 2008b) in these upper canopy positions declined over the course of the study, which may have resulted in an export of N to higher more favored branches. Respiration rates are closely coupled to foliar [N] (Griffin et al. 2001; Tissue et al. 2002) demonstrated in this study by the observed strong exponential relationship between

21 21 R 15 and foliar [N] (Fig. 2) and highlighting the close mechanistic coupling between these two leaf traits. Taking these dynamics into account we observed a consistent pattern of declining reference respiration rates (R N ) with increasing temperature (Fig. 6). The seasonal patterns in R N observed within the canopies of the dominant and suppressed E. globulus trees are consistent with a type II acclimation response (Atkin and Tjoelker 2003; Xu and Griffin 2006). Although these types of acclimation responses have been observed previously, the extent of acclimation varies considerably between tissues and species (Ow et al. 2008). Furthermore, it is generally assumed that fully expanded leaf tissues have less capacity to acclimate than developing leaf tissue (Atkin et al. 2005; Loveys et al. 2003). Our foliar respiration data shows that despite these previous observations, fully expanded E. globulus leaves within the canopies of the dominant and suppressed trees studied here exhibited seasonal patterns of leaf respiration consistent with respiratory acclimation. Conclusion The respiration process is at best, only simplistically represented within current mechanistic models such as 3pg (Landsberg and Waring 1997) or CABALA (Battaglia et al. 2004). Our data highlights the dynamic nature of foliar respiration within E. globulus canopies. Not only do respiration rates vary seasonally and as a function of canopy position, respiration also varies as a function of tree dominance, although many of these differences can be explained via differences in foliar [N] and leaf carbohydrate concentrations. Leaves within the canopies of the E. globulus trees also exhibited considerable potential for thermal acclimation. Despite this complexity, incorporation of a substrate based model of respiratory acclimation within canopies

22 22 (Dewar et al. 1999) could potentially significantly improve our representation of respiration within process based modeling frameworks and provide more confidence in our estimates of canopy productivity. Acknowledgements AP O Grady was supported during this research by an ARC linkage project (LP ). Further funding support was provided by the CRC forestry. We thank Charles and Robin Lewis for their support of the Pittwater research plantation on Milford Farm. Ann Wilkinson assisted with foliar N analysis. David Tissue and Libby Pinkard and the reviewers have provided valuable comments on drafts of this paper References Atkin, O.K., D. Bruhn, V.M. Hurry and M.G. Tjoelker Evans Review No. 2: The hot and the cold: unravelling the variable response of plant respiration to temperature. Funct. Plant Biol. 32: Atkin, O.K., J.R. Evans, M.C. Ball, H. Lambers and P. T.L. 2000a. Leaf respiration of snow gum in the light and dark. Interactions between temperature and irradiance. Plant Physiol. 122: Atkin, O.K., C. Holly and M.C. Ball 2000b. Acclimation of snow gum (Eucalyptus pauciflora) leaf respiration to seasonal and diurnal variations in temperature: the importance of changes in the capacity and temperature sensitivity of respiration. Plant Cell Environ. 23:15-26.

23 23 Atkin, O.K. and M.G. Tjoelker Thermal acclimation and the dynamic response of plant respiration to temperature. Trends Plant Sci. 8: Battaglia, M., P. Sands, D. White and D. Mummery CABALA: a linked carbon, water and nitrogen model of forest growth for silvicultural decision support. For. Ecol. Manag. 193: Binkley, D A hypothesis about the interaction of tree dominance and stand production through stand development. For. Ecol. Manag. 190: Binkley, D., J.L. Stape and M.G. Ryan Thinking about efficiency of resource use in forests. For. Ecol. Manag. 193:5-16. Buysse, J. and R. Merckx An improved colorimetric method to quantify sugar content of plant-tissue. J. Exp. Bot. 44: Callister, A.N. and M.A. Adams Water stress impacts on respiratory rate, efficiency and substrates, in growing and mature foliage of Eucalyptus spp. Planta. 224: Cavaleri, M.A., S.F. Oberbauer and M.G. Ryan Foliar and ecosystem respiration in an old-growth tropical rain forest. Plant Cell Environ. 31: Dewar, R.C., B.E. Medlyn and R.E. McMurtrie Acclimation of the respiration/photosynthesis ratio to temperature: insights from a model. Global Change Biol. 5: Drew, D., A.P. O'Grady, G. Downes, J. Read and D. Worledge Daily stem growth patterns in irrigated and non-irrigated Eucalyptus globulus. Tree Physiol. 28: Field, C.B Allocating leaf nitrogen for the maximisation of carbon gain: leaf age as a control on the allocation program. Oecologia. 56:

24 24 Frak, E., X. Le Roux, P. Millard, B. Adam, E. Dreyer, C. Escuit, H. Sinoquet, M. Vandame and C. Varlet-Grancher Spatial distribution of leaf nitrogen and photosynthetic capacity within the foliage of individual trees: disentangling the effects of local light quality leaf irradiance, and transpiration. J. Exp. Bot. 53: Griffin, K.L., D.T. Tissue, M.H. Turnbull, W. Schuster and D. Whitehead Leaf dark respiration as a function of canopy position in Nothofagus fusca trees grown at ambient and elevatedco 2 partial pressures for 5 years. Funct. Ecol. 15: Griffin, K.L., M. Turnbull and R. Murthy Canopy position affects the temperature response of leaf respiration in Populus deltoides. New Phytol. 154: Henriksson, J Differential shading of branches or whole trees: survival, growth, and reproduction. Oecologia. 126: Isbell, R The Australian Soil Classification. CSIRO Publishing, Melbourne. King, A.W., C.A. Gunderson, W.M. Post, D.J. Weston and S.D. Wullschleger Plant respiration in a warmer world. Science. 312 Kruse, J. and M.A. Adams Sensitivity of respiratory metabolism and efficiency to foliar nitrogen during growth and maintenance. Global Change Biol. 14: Landsberg, J.J. and R.H. Waring A generalised model of forest productivity using simplified concepts of radiation use efficiency, carbon balance and partitioning. For. Ecol. Manag. 95: Lavigne, M.B., S.E. Franklin and E.R. Hunt, Jr Estimating stem maintenance respiration rates of dissimilar balsam fir stands. Tree Physiol. 16:

25 25 Lavigne, M.B. and M.G. Ryan Growth and maintenance respiration rates of aspen, black spruce and jack pine stems at northern and southern BOREAS sites. Tree Physiol. 17: Lloyd, J. and J.A. Taylor On the temperature dependence of soil respiration. Funct Ecol. 8: Loveys, B.R., L.J. Atkinson, D.J. Sherlock, R.L. Roberts, A.H. Fitter and O.K. Atkin Thermal acclimation of leaf and root respiration: an investigation comparing inherently fast- and slow-growing plant species. Global Change Biol. 9: Lowther, J.R Use of a single sulphuric acid-hydrogen peroxide digest for the analysis of Pinus radiata needles. Commun. Soil Sci. Plant Anal. 11: Lyons, J.M. and J.K. Raison Oxidative activity of mitochondria isolated from plant tissues sensitive and resistant to chilling injury. Plant Physiol. 45: O'Grady, A.P., D. Worledge and M. Battaglia Temporal and spatial changes in fine root distributions in a young Eucalyptus globulus stand in southern Tasmania. For. Ecol. Manag. 214: O'Grady, A.P., D. Worledge and M. Battaglia 2008a. Constraints on transpiration in Eucalyptus globulus in southern Tasmania. Agric. For. Meteorol. 148: O'Grady, A.P., D. Worledge, A. Wilkinson and M. Battaglia 2008b. Gradients in photosynthesis and respiration within dominant and suppressed Eucalyptus globulus trees. Funct. Plant Biol. 35:

26 26 Ow, L.F., K.L. Griffin, D. Whitehead, A.S. Walcroft and M.H. Turnbull Thermal acclimation of leaf respiration but not photosynthesis in Populus deltoides nigra. New Phytol. 178: Palacio, S., M. Maestro and G. Montserrat-Marti Seasonal dynamics of nonstructural carbohydrates in two species of Mediterranean sub-shrubs with different leaf phenology. Environ. Exp. Bot. 59: Reich, P.B., M.G. Tjoelker, J.-L. Machado and J. Oleksyn Universal scaling of respiratory metabolism, size and nitrogen in plants. Nature. 439: Reich, P.B., M.G. Tjoelker, K.S. Pregitzer, I.J. Wright, J. Oleksyn and J.-L. Machado Scaling of respiration to nitrogen in leaves, stems and roots of higher land plants. Ecol. Lett. 11: Reich, P.B., M.B. Walters, D.S. Ellsworth, J.M. Vose, J.C. Volin, C. Gresham and W.D. Bowman Relationships of leaf dark respiration to leaf nitrogen, specific leaf area and leaf life span: a test across biomes and functional groups. Oecologia. 114: Ryan, M.G Effects of climate change on plant respiration. Ecol. Appl. 1: Ryan, M.G Foliar maintenance respiration of sub alpine and boreal trees in relation to nitrogen content. Plant Cell and Environ. 18: Ryan, M.G., R.M. Hubbard, S. Pongracic, R.J. Raison and R.E. McMurtrie Foliage, fine-root, woody-tissue and stand respiration in Pinus radiata in relation to nitrogen status. Tree Physiol. 16: Ryan, T.H Significance tests for multiple comparisons of proportions, variances and other statistics. Psychol. Bull. 4:

27 27 Schimel, D.S Terrestrial ecosystems and the carbon cycle. Global Change Biol. 1: Smith, F.W. and J.N. Long Age-related decline in forest growth: an emergent property. For. Ecol. Manag. 144: Sprugel, D.G When branch autonomy fails: Milton's Law of resource availability and allocation. Tree Physiol. 22: Takenaka, A Shoot growth responses to light microenvironment and correlative inhibition in tree seedlings under a forest canopy. Tree Physiol. 20: Tissue, D.T., J.D. Lewis, S.D. Wullschleger, J.S. Amthor, K.L. Griffin and O.R. Anderson Leaf respiration at different canopy positions in sweetgum (Liquidamber styraciflua) grown in ambient and elevated concentrations of carbon dioxide in the field. Tree Physiol. 22: Tjoelker, M.G., J. Oleksyn, G. Lorenc-Plucinska and P.B. Reich Acclimation of respiratory temperature responses in northern and southern populations of Pinus banksiana. New Phytol. 181: Tjoelker, M.G., J. Oleksyn and P.B. Reich Modelling respiration of vegetation: evidence for a general temperature-dependent Q 10. Global Change Biol. 7: Tjoelker, M.G., J. Oleksyn, P.B. Reich and R. Zytkowiak Coupling of respiration, nitrogen, and sugars underlies convergent temperature acclimation in Pinus banksiana across wide-ranging sites and populations. Global Change Biol. 14: Turnbull, M.H., D. Whitehead, D.T. Tissue, W.S.F. Schuster, K.J. Brown and K.L. Griffin Responses of leaf respiration to temperature and leaf

28 28 characteristics in three deciduous tree species vary with site and water availability. Tree Physiol. 21: Turnbull, M.H., D. Whitehead, D.T. Tissue, W.S.F. Schuster, K.J. Brown and K.L. Griffin Scaling foliar respiration in two contrasting forest canopies. Funct. Ecol. 17: Vose, J.M. and M.G. Ryan Seasonal respiration of foliage, fine roots, and woody tissues in relation to growth, tissue N, and photosynthesis. Global Change Biol. 8: Warren, C.R. and M.A. Adams Distribution of N, Rubisco and photosynthesis in Pinus pinaster and acclimation to light. Plant Cell Environ. 24: Whitehead, D., K.L. Griffin, M.H. Turnbull, D.T. Tissue, V.C. Engel, K.J. Brown, W.S.F. Schuster and A.S. Walcroft 2004a. Response of total night-time respiration to differences in total daily photosynthesis for leaves in a Quercus rubra L. canopy: implications for modelling canopy CO 2 exchange. Global Change Biol. 10: Whitehead, D., A.S. Walcroft, N.A. Scott, J.A. Townsend, C.M. Trotter and G.N.D. Rogers 2004b. Characteristics of photosynthesis and stomatal conductance in the shrubland species mânuka (Leptospermum scoparium) and kânuka (Kunzea ericoides) for the estimation of annual canopy carbon uptake. Tree Physiol. 24: Wright, I.J., P.B. Reich, O.K. Atkin, C.H. Lusk, M.G. Tjoelker and M. Westoby Irradiance, temperature and rainfall influence leaf dark respiration in woody plants: evidence from comparisons across 20 sites. New Phytol. 169: Wright, I.J., P.B. Reich, M. Westoby, D.D. Ackerly, Z. Baruch, F. Bongers, J. Cavender-Bares, T. Chapin, J.H.C. Cornelissen, M. Diemer, J. Flexas, E.

29 29 Garnier, P.K. Groom, J. Gulias, K. Hikosaka, B.B. Lamont, T. Lee, W. Lee, C. Lusk, J.J. Midgley, M.-L. Navas, U. Niinemets, J. Oleksyn, N. Osada, H. Poorter, P. Poot, L. Prior, V.I. Pyankov, C. Roumet, S.C. Thomas, M.G. Tjoelker, E.J. Veneklaas and R. Villar The worldwide leaf economics spectrum. Nature. 428: Xu, C.Y. and K.L. Griffin Seasonal variation in the temperature response of leaf respiration in Quercus rubra: foliage respiration and leaf properties. Funct Ecol. 20: Zar, J.H Biostatistical analysis. Prentice-Hall, New Jersey.

30 30 Table 1. Predicted values of R 15 and E 0 in dominant and suppressed E. globulus canopies. Parameters were derived from temperature response curves using an Arrhenius function. Data represent mean ± st.err. (n= 3 trees) August 2007 November 2007 May 2008 Height R 15 E 0 R 15 E 0 R 15 E 0 (m) (µmol m -2 s -1 ) (kj mol -1 ) (µmol m -2 s -1 ) (kj mol -1 ) (µmol m -2 s -1 ) (kj mol -1 ) Dominant ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 8.2 Suppressed ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 14.8

31 31 Figure 1. Seasonal patterns in measured R 15 as a function of canopy position in dominant and suppressed E. globulus trees. Data represent the mean ± s.e of three trees in each dominance class. Figure 2. Relations between measured R 15 and foliar N concentration ([N], g m -2 ) in dominant and suppressed E. globulus canopies. Data from all trees and seasons are pooled and each point represents the mean ± s.e. of measurements made on three trees. Figure 3. a. Variation in R 15 during may 2008 as a function of canopy position in dominant and suppressed E. globulus trees. b. Variation in R 15 normalised to account for distance to the apical tip. Position in the canopy was normalised using the (T h M h ) equation 1, where T h = total tree height and M h is the R measurement T h height above ground level. Data represent the mean ± s.e of three trees in each dominance class. Figure 4. Relations between measured R 15 and a. soluble sugars (g m -2 ) and c. starch (g m -2 ) within the canopies of dominant and suppressed E. globulus trees. Data from all trees and seasons are pooled and each point represents the mean ± s.e. of measurements made on three trees. Figure 5. Seasonal patterns in foliar soluble sugar and starch concentrations in suppressed, a. and b) and dominant (c and d) canopies of E. globulus trees. Data represent the mean ± s.e. of leaves from three trees.

32 32 Fig 6. Relations between measured R N and the mean daily temperature of the 14 day period prior to measurement. Data were measured at 7 and 15 m in the dominant trees (a.) and at 3 and 11 m in the suppressed trees (b). Each point represents the mean ± s.e. of leaves from three trees at each canopy position

33 33

34 34

35 35

36 36

37 37

38 38

39 39