Seasonal Variation in Respiration of 1‐year‐old Shoots of Scots Pine Exposed to Elevated Carbon Dioxide and Temperature for 4 Years

Abstract

Sixteen 20‐year‐old Scots pine (Pinus sylvestris L.) trees growing in the field were enclosed for 4 years in environment‐controlled chambers that maintained: (1) ambient conditions (CON); (2) elevated atmospheric CO₂ concentration (ambient + 350 µmol mol^–1; EC); (3) elevated temperature (ambient +2–6 °C; ET); or (4) elevated CO₂ and elevated temperature (ECT). The dark respiration rates of 1‐year‐old shoots, from which needles had been partly removed, were measured over the growing season in the fourth year. In all treatments, the temperature coefficient of respiration, Q₁₀, changed with season, being smaller during the growing season than at other times. Respiration rate varied diurnally and seasonally with temperature, being highest around mid‐summer and declining gradually thereafter. When measurements were made at the temperature of the chamber, respiration rates were reduced by the EC treatment relative to CON, but were increased by ET and ECT treatments. However, respiration rates at a reference temperature of 15 °C were reduced by ET and ECT treatments, reflecting a decreased capacity for respiration at warmer temperatures (negative acclimation). The interaction between season and treatment was not significant. Growth respiration did not differ between treatments, but maintenance respiration did, and the differences in mean daily respiration rate between the treatments were attributable to the maintenance component. We conclude that maintenance respiration should be considered when modelling respiratory responses to elevated CO₂ and elevated temperature, and that increased atmospheric temperature is more important than increasing CO₂ when assessing the carbon budget of pine forests under conditions of climate change.

INTRODUCTION

Atmospheric CO₂ concentration is rising and is expected to be approximately double the current value by the end of the century. As a consequence, an increase of 1·5–6 °C in global mean temperature is predicted (Schimel et al., 1996). Such broad‐scale atmospheric and climatic changes will affect the physiology and ecology of trees (Gunderson and Wullschleger, 1994; Saxe et al., 1998). It is expected that rates and amounts of respiration will alter with climatic changes (Mitchell et al., 1990). Quantification of the carbon dynamics of forests exposed to elevated CO₂ concentrations is important, as they are a large terrestrial ecosystem with a potentially substantial effect on the global carbon budget. It has been estimated that woody tissue respiration accounts for at least 25 % of the above‐ground respiratory budget of forests (Carey et al., 1996) and is thus a large component of the annual carbon balance of forest ecosystems (Ryan et al., 1996; Carey et al., 1997; Damesin et al., 2002; Rayment et al., 2002).

The effects of elevated CO₂ on respiration of leaves of forest trees are well documented (Ryan, 1991; Ceulemans et al., 1999; Norby et al., 1999; Saxe et al., 2001). Dark respiration decreases in response to CO₂ enrichment (Amthor et al., 1992; Drake, 1992; Wullschleger et al., 1992; Teskey, 1995; Jach and Ceulemans, 2000; Zha et al., 2001, 2002a; Wang et al., 2002). This also occurs in whole plants (Bunce, 1990; Bunce and Caulfield, 1991; Ziska and Bunce, 1993) and, possibly, in roots (Gifford et al., 1985). Respiratory responses to elevated CO₂ may be tissue‐specific in a given species (Wullschleger et al., 1995). However, knowledge of the respiratory responses of woody tissues to elevated CO₂ is scarce, and few data have been reported on the contribution of particular branches or shoots to the total carbon budget of Scots pine (Pinus sylvestris L.). Estimating respiration of woody tissue is therefore critical for predicting the carbon balance of forest ecosystems in a changing climate (Amthor, 1991).

Respiration is measured throughout the growing season and during dormancy in woody shoots of Scots pine grown for 4 years under natural soil conditions. The objectives were: (1) to determine the diurnal and seasonal changes taking place in shoot respiration; and (2) to examine the responses of shoot respiration to elevated CO₂ and elevated temperature.

MATERIALS AND METHODS

Site and treatments

Measurements were made in a naturally seeded stand of Scots pine (Pinus sylvestris L.) trees located near Mekrijärvi Research Station (62°47′N, 30°58′E, 145 m a.s.l), University of Joensuu, Finland. The soil is a sandy loam with a water retention of 40 mm for the top 30 cm at field capacity and 20 mm at wilting point. Details of the site are given by Kellomäki and Wang (1997). In 1996, 16 trees approx. 20 years old with the same crown size and height were selected and enclosed in closed‐top chambers. The trees had an average height of approx. 6 m in 2000 and a diameter at breast height of 5·3 cm.

Each chamber had eight sides and an internal volume of approx. 26·5 m³, and covered a ground area of 5·9 m². The computer‐controlled heating and cooling system, together with a set of magnetoelectric valves, automatically adjusted the temperature and CO₂ concentration inside the chamber to mimic ambient conditions, or to increase temperature above that of ambient air by 2 °C from 15 Apr. until 15 Sep. and by 6 °C from 16 Sep. to 14 Apr., or to increase the CO₂ concentration (+350 µmol mol^–1) all day throughout the year, or to increase both temperature and CO₂ concentration. During winter, the warming treatments were designed to correspond with the warmest winter predicted for the site after a doubling of atmospheric CO₂ (Hänninen, 1995). The chamber structure, environment control system and long‐term performance have been documented previously (Kellomäki et al., 2000; Zha et al., 2001; Wang et al., 2002).

In summary, there were four CO₂ concentration and temperature treatments: (1) ambient CO₂ concentration and temperature (CON); (2) elevated CO₂ (EC); (3) elevated temperature (ET); and (4) elevated CO₂ and temperature (ECT). Each treatment had four replicates.

Measurement of shoot respiration

Similar 1‐year‐old shoots (formed in 1999) on the fifth whorl from the top of the crown of each tree were selected for repeated measurements of respiration. Their length ranged from 12 cm in CON to 14 cm in ECT, and their basal diameters from 0·21 cm in CON to 0·25 cm in ECT. Needles 2 cm or more below the shoot tip were removed 40 d before the measurements commenced. Dark respiration rates of the shoots were measured at approx. 1‐month intervals from May to October 2000, using a portable photosynthesis analyser (LCA‐4; Analytic Development Co. Ltd, Hoddesdon, UK). Measurements were made 1 h after sunset at the same CO₂ concentration as that under which the trees were growing, and were completed for all treatments within 2–3 days on each occasion. The cuvette was sealed tightly using a dense foam gasket 0·5–0·6 cm thick; these gaskets were cleaned immediately before each measurement and changed if they had lost elasticity.

Shoot diameter was measured using a digital calliper, and the surface area and volume of the shoot enclosed in the leaf chamber were calculated. Respiration rates were expressed relative to shoot volume. The temperature responses of respiration of shoots in each treatment were measured by changing air temperature in the cuvette with a temperature controller (Analytic Development Co., Ltd) in 5 °C temperature increments from 5 to 30 °C throughout growing season. The temperature response was estimated in terms of the temperature coefficient of respiration (Q₁₀) as follows:

where R_T is the instantaneous respiration at temperature T, and R₁₅ is the respiration rate at T = 15 °C.

Throughout the measurement period, total hourly and daily shoot respiration were calculated using individual monthly Q₁₀ and R₁₅ values and hourly temperature data for the chambers. Shoot respiration was partitioned into growth and maintenance components as described by Sprugel (1990). In view of the pattern of the stem diameter growth in trees (Peltola et al., 2002), June, July and August were regarded as the growing season, whereas May, September and October constituted the non‐growing season. Maintenance respiration was estimated as CO₂ efflux during the non‐growing season (Ryan, 1990), and it was assumed that its response to temperature was the same as that in the growing season (Damesin et al., 2002). Growth respiration was estimated as the difference between total and maintenance respiration. Total respiration in each treatment was calculated from individual Q₁₀ values, R₁₅ and hourly temperatures in the chambers.

Statistical analysis

Differences in respiration between the treatments and seasons were analysed using a GLM repeated measures ANOVA in SPSS 11·0 (Chicago, IL, USA). Season was defined as a within‐subjects factor, and the treatment factor was specified as a between‐subjects factor. If treatment significantly affected the dependent variable, Tukey’s HSD of the post hoc multiple comparison tests was performed for each dependent variable separately to compare the differences between the control and other treatments.

RESULTS

Response of respiration to temperature

Respiration rate increased exponentially with temperature during measurement regardless of treatment [Fig. 1; coefficient of determination (R²) > 0·63; P < 0·05]. There were significant (P < 0·001) seasonal changes in Q₁₀ in all the treatments: it was smaller in the growing season than in the non‐growing season (Fig. 2). The relative differences in Q₁₀ between the highest and the lowest values were 20, 25, 22 and 30 % for CON, EC, ET and ECT, respectively.

Fig. 1. Relationship between temperature and shoot respiration of Scots pine grown at ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT). Regression curves are described by the equation where r_o and k are coefficients specific to the treatment. The temperature coefficient (Q₁₀) is exp(10k).

Fig. 2. Seasonal changes in temperature coefficient of respiration (Q₁₀) and respiration rate at 15 °C (R₁₅) of Scots pine grown at ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT). Points are means of four measurements. Bars represent s.e.

Q₁₀ was also significantly (P < 0·05) affected by the environmental treatment. Elevated CO₂ increased Q₁₀ in all months except June, but elevated temperature and the combination of elevated CO₂ and elevated temperature consistently reduced Q₁₀. Overall, Q₁₀ was increased by 7 % in the EC treatment relative to the control treatment (CON; P = 0·069), but was reduced by 9 (P = 0·043) and 8 % (P = 0·022), respectively, in the ET and ECT treatments. Annual mean values of Q₁₀ were 1·88, 2·01, 1·71 and 1·74 for CON, EC, ET and ECT, respectively. There was no significant interaction between season and treatment (P = 0·143).

Dark respiration rate

Respiration rates at the reference temperature of 15 °C (R₁₅) varied significantly, being about two‐fold greater during the growing season than in the non‐growing season, regardless of treatment (Fig. 2; P < 0·001). Average R₁₅ values for CON, EC, ET and ECT during the growing period were, respectively, 103, 88, 92 and 91 µmol m^–3 s^–1, and during the non‐growing period were 74, 52, 55 and 53 µmol m^–3 s^–1. Modelled hourly respiration on two typical clear days representing the growing and non‐growing periods paralleled the fluctuations in hourly mean temperature in all treatments (Fig. 3). The highest values for a given treatment occurred at around 1500 h, when the temperature was greatest. When daily respiration rates were modelled from May to October the trend in respiration followed that of temperature in all treatments (Fig. 4).

Fig. 3. Mean hourly environmental temperature in a closed‐top chamber on 26 Jul. and 28 Sep. 2000 and predicted diurnal respiration rate of Scot pine grown at ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT).

Fig. 4. Mean daily temperature and predicted daily respiration rate of Scots pine trees grown under ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT). Measurements made between May and October 2000.

Under EC, ET and ECT conditions, dark respiration at 15 °C was reduced by 22 (P = 0·015), 18 (P = 0·005) and 20 % (P = 0·004), respectively, relative to that in control conditions (Fig. 2). There was no significant difference in R₁₅ among trees in the ET and ECT treatments (P = 0·886). In contrast to R₁₅, respiration rates measured at prevailing temperatures in the chambers decreased by 11 % in EC relative to CON, but increased by 32 and 30 % in ET and ECT (Table 1; P < 0·05). Throughout the year, estimated daily respiration rates in the ET and ECT treatments were consistently higher than those in CON by an average of 45 and 40 %, respectively, while those in the EC treatment were 11 % smaller (Fig. 4). There was no significant interaction between treatment and season (P = 0·253).

Table 1.

Shoot respiration rates of Scots pine grown at ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT)

	Temperature (°C)				Respiration (µmol m–3 s–1)
Date	CON	EC	ET	ECT	CON	EC	ET	ECT
23 May	10·7 ± 0·1	10·7 ± 0·1	14·5 ± 0·1	15·1 ± 02	60·48 ± 6·43	65·22 ± 4·32	78·39 ± 4·21	74·55 ± 7·16
29 Jun.	22·4 ± 0·2	22·0 ± 0·1	24·1 ± 0·1	24·2 ± 02	139·24 ± 7·82	131·23 ± 6·59	157·66 ± 8·62	148·90 ± 7·08
24 Jul.	16·0 ± 0·1	16·2 ± 0·2	18·0 ± 0·2	18·3 ± 0·1	113·92 ± 7·44	104·79 ± 4·88	129·34 ± 7·54	124·92 ± 7·52
27 Aug.	10·8 ± 0·1	10·9 ± 0·1	12·7 ± 0·1	13·4 ± 0·1	87·22 ± 5·26	77·31 ± 5·28	98·66 ± 6·24	101·69 ± 5·46
27 Sep.	2·8 ± 0·2	2·9 ± 0·1	9·1 ± 0·1	9·7 ± 0·1	34·15 ± 2·18	22·38 ± 1·37	59·89 ± 6·15	55·20 ± 3·47
25 Oct.	5·6 ± 0·4	5·9 ± 0·3	11·5 ± 0·1	12·0 ± 0·1	31·38 ± 1·66	27·28 ± 1·26	46·55 ± 3·72	51·68 ± 4·93

Measurements were made at the temperature at which the trees were growing, Each value is the mean of four measurements.

Growth respiration was not significantly (t‐test, P > 0·05) different among plants in the control and the elevated CO₂ and temperature treatments, but maintenance respiration was (t‐test, P < 0·05; Table 2). Differences in maintenance respiration for EC, ET and ECT relative to CON were –23, 68 and 58 %, while differences in growth respiration were –1, 3 and –8 %. Thus, the effect of long‐term elevation of CO₂ concentration and temperature on total shoot respiration was due to the effects on the maintenance component of respiration rather than on the growth component.

Table 2.

Mean daily total, growth and maintenance respiration rates for Scots pine trees growing at ambient CO₂ concentration and temperature (CON), elevated CO₂ (EC), elevated temperature (ET), and elevated CO₂ and temperature (ECT) in July 2000

	Treatment
Respiration rate (mol m–3 d–1)	CON	EC	ET	ECT
Total	10·46	9·79	12·46	11·31
Maintenance	2·61 (25)	2·02 (21)	4·41 (35)	4·12 (36)
Growth	7·84 (75)	7·77 (79)	8·05 (65)	7·19 (64)

Values in parentheses indicate the percentage contribution of growth and maintenance respiration to total shoot respiration for each treatment.

DISCUSSION

The method employed here to measure the respiration rate of woody shoots requires consideration since removal of needles, 2 cm and more from the tip, over a month before measurements started may have affected results. This would have disturbed transport of photosynthate from needle to shoot and altered (probably decreased) the carbohydrate balance. As a result, shoot respiration might have been underestimated. This was not assessed. This effect, however, is expected to be minor, since needles are sparsely distributed on the lower part of shoots, and they are only 1 year old and thus have slower metabolism. Besides, active needles at the shoot tip were retained. Therefore, the effect of removing needles on seasonal changes in shoot respiration, and its response to the environment, is expected to be minor.

On each measurement occasion, the relationship between respiration rate and temperature fitted an exponential equation well (Fig. 1; R² > 0·63, P < 0·05). The Q₁₀ value was greater in the non‐growing season than in the growing season, i.e. respiration was less sensitive to temperature during the growing season. Similar findings have been reported with regard to seasonal variations in Q₁₀ (Paembonan, 1991; Criddle et al., 1994; Lavigne, 1996; Stockfors and Linder, 1998), although Q₁₀ values have also been found to be constant throughout the year (Ryan et al., 1997; Damesin et al., 2002). In addition, we note here that the Q₁₀ values on different measurement occasions during the growing season were similar for a given treatment (Fig. 2), as were those during the non‐growing season. This suggests that some acclimation of respiration to seasonal temperature occurred, and is a gradual process.

That trees grown at elevated CO₂ had higher Q₁₀ values than those grown in control conditions indicates that respiration was more sensitive to temperature at EC. This has been reported for both leaf (Zha et al., 2001) and stem respiration (Carey et al., 1996). The reduced Q₁₀ under ET and ECT conditions, along with lower R₁₅, implies that some negative downward acclimation of shoot respiration may occur in pine under long‐term exposure to elevated temperature.

In agreement with previous results (Paembonan et al., 1991), changes in respiration rate paralleled those in temperature during the day. A similar seasonal pattern of respiration with temperature, with a maximum in mid‐ to late summer, followed by a rapid decrease during the autumn, has been reported by Lavigne and Ryan (1997) and Vose and Ryan (2002). The seasonal change in respiration noted here matched that of needles in our previous study (Zha et al., 2001).

Shoot respiration rates at 15 °C ranged from 37 to 119 µmol m^–3 s^–1 in EC and ET treatments, respectively, and were comparable with values (18–110 µmol m^–3 s^–1) for woody stems in eight boreal forest stands in Canada (Ryan et al., 1997). Assuming that respiration rates in the non‐growing season are maintenance respiration, shoot maintenance respiration at 15 °C would average 57, 50, 64 and 63 µmol m^–3 s^–1 for trees in CON, EC, ET and ECT, respectively. These values are much higher than those quoted for three pines and western hemlock at 15 °C (6·4–11·5 µmol m^–3 s^–1; Ryan et al., 1995), for lodgepole pine (17·1 µmol m^–3 s^–1; Ryan, 1990), for P. radiata (15–39 µmol m^–3 s^–1; Ryan et al., 1996), for white pine (20 µmol m^–3 s^–1; Vose and Ryan, 2002), and for boreal conifers (13–55 µmol m^–3 s^–1; Ryan et al., 1997). However, it should be emphasized that the results cited above apply to stem tissue, whereas the present results are for 1‐year‐old shoot tissues. The higher respiration rates recorded in the present study are therefore understandable since young shoots are metabolically more active than stem tissue.

Compared with the numerous measurements of respiratory responses of leaves to elevated CO₂, our knowledge of respiratory responses of woody tissue is limited. What few studies there have been have focused on stem tissue (Wullschleger et al., 1995; Carey et al., 1996; Edwards et al., 2002), where respiration tends to be higher in elevated CO₂ than in ambient CO₂, whereas our investigation into respiration in 1‐year‐old shoots shows this to be reduced by elevated CO₂. One explanation for this could be associated with N concentration, which is positively related to respiration, not only for leaves (Wullschleger et al., 1992; Ryan, 1995; Reich et al., 1996, 1998; Tjoelker et al., 1999b; Zha et al., 2002b), but also for woody tissues (Ryan et al., 1996). Moreover, within a species, elevated CO₂ decreases the nitrogen concentration of all tissues (Cotrufo et al., 1998), with the extent of the reduction depending on the tissue. It is thus reasonable to assume that the reduced shoot respiration in the EC treatment is caused by the decreased N concentration. However, this cannot be substantiated because no material was harvested to determine the nitrogen status of the shoots at the time of measurement as this would have affected other measurements. The role of N concentration in shoot tissue therefore requires further analysis.

The effects of elevated temperature on respiration rates are a logical consequence of the temperature sensitivity of the enzymatically catalysed reactions involved in respiration, and of the increased ATP requirements as metabolic rates increase. The temperature stimulation of respiration also reflects the increased demands for energy to support the increased rates of biosynthesis, transport and protein turnover that occur at high temperature. Nevertheless, the elevation of temperature alone (ET), or combined with elevated CO₂ (ECT), decreased the respiration rate relative to that in CON when compared at the reference temperature of 15 °C. This indicates that negative acclimation of respiration occurs in trees grown at elevated temperature in the long term. A similar result has been obtained for leaves (Tjoelker et al., 1999a; Zha et al., 2001), although no significant difference in respiration was noted between ET and ECT treatments. Moreover, the increase in the calculated daily respiration in ET (45 %) and ECT (40 %) was much larger than the reduction under EC conditions (–11 %). These results suggest that the increase in temperature will have a greater impact than rising CO₂ on the carbon budget of trees under climate change.

The estimated growth respiration did not differ between treatments, indicating that the differences in mean total daily respiration rate between them were attributable to the maintenance component, as shown for conifer leaves (Thomas et al., 1993; Zha et al., 2001), where growth respiration decreased in response to the EC treatment (Griffin et al., 1993; Ziska and Bunce, 1993). The present finding that ET and ECT treatments increased maintenance respiration indicates that maintenance respiration is more sensitive to temperature than is the growth component. This could provide support for the notion that long‐term warming is likely to affect the balance between growth and respiration (Ryan, 1991). One explanation for the increased maintenance respiration noted under ET and ECT conditions is that the elevated temperature increased the activities of enzymes related to respiration, whereas reduced maintenance respiration under EC conditions implies that long‐term elevated CO₂ treatment could reduce the maintenance cost. Reduced respiration (mainly maintenance) combined with increased growth of woody tissue (Peltola et al., 2002) in long‐term EC treatment, indicates that Scots pine trees may grow more efficiently under EC than under ambient conditions. However, this result applies to young shoots of pine only; stem tissues remain to be investigated.

In summary, shoot respiration changed diurnally and seasonally regardless of the treatments. It was less sensitive to temperature during the growing season than in the non‐growing season, but the treatments did not significantly affect the seasonal trend. However, shoot respiration was reduced under EC relative to CON and increased under ET or ECT conditions.

ACKNOWLEDGEMENTS

We thank Matti Lemettinen, Alpo Hassinen and Risto Ikonen for assistance with equipment. This work was part of the Finnish Centre of Excellence Programme for Forest Ecology and Management (Project no. 64308), and of the Finnish and Chinese cooperation project (no. 200013). Funding provided by the Academy of Finland, the National Technology Agency (Tekes) and the University of Joensuu is gratefully acknowledged.

Received: 29 November 2002; Returned for revision: 12 February 2003; Accepted: 8 April 2003    Published electronically: 21 May 2003