Fine-root respiration in a loblolly pine and sweetgum forest growing in elevated CO2

Authors

  • K. George,

    Corresponding author
    1. Department of Plant Biology, University of Illinois at Urbana-Champaign, Urbana, IL 61802, USA;
    2. Current address of author: Department of Forest Science, Oregon State University, Corvallis, OR 97331, USA
      Author for correspondence: K. GeorgeTel: +1 (541) 760 4478Fax: +1 (541) 737 1393Email: drkategeorge@yahoo.com
    Search for more papers by this author
  • R. J. Norby,

    1. Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA;
    Search for more papers by this author
  • J. G. Hamilton,

    1. Department of Biology, Ithaca College, Ithaca NY, 14850, USA;
    Search for more papers by this author
  • E. H. DeLucia

    1. Department of Plant Biology, University of Illinois at Urbana-Champaign, Urbana, IL 61802, USA;
    Search for more papers by this author

Author for correspondence: K. GeorgeTel: +1 (541) 760 4478Fax: +1 (541) 737 1393Email: drkategeorge@yahoo.com

Summary

  • • The loss of carbon below-ground through respiration of fine roots may be modified by global change. Here we tested the hypothesis that a reduction in N concentration of tree fine-roots grown in an elevated atmospheric CO2 concentration would reduce maintenance respiration and that more energy would be used for root growth and N uptake. We partitioned total fine-root respiration (RT) between maintenance (RM), growth (RG), and N uptake respiration (RN) for loblolly pine (Pinus taeda) and sweetgum (Liquidambar styraciflua) forests exposed to elevated CO2.
  • • A substantial increase in fine-root production contributed to a 151% increase in RG for loblolly pine in elevated CO2. Root specific RM for pine was 24% lower under elevated CO2 but when extrapolated to the entire forest, no treatment effect could be detected.
  • • RG (< 10%) and RN (< 3%) were small components of RM in both forests. Maintenance respiration was the vast majority of RT, and contributed 92% and 86% of these totals at the pine and sweetgum forests, respectively.
  • • The hypothesis was rejected because the majority of fine-root respiration was used for maintenance and was not reduced by changes in root N concentration in elevated CO2. Because of its large contribution to RT and total soil CO2 efflux, changes in RM caused by warming may greatly alter carbon losses from forests to the atmosphere.

Introduction

A substantial fraction of the flux of CO2 from the soil is from roots (Rouhier et al., 1996; Thierron & Laudelout, 1996), with the rest coming from soil organisms. Published values for the proportion of total soil CO2 efflux originating from roots vary from < 10% to > 90% (Hanson et al., 2000), thus, the loss of carbon through root respiration can be an important component of forest carbon budgets. More than 50% of total net primary productivity (NPP) in forest ecosystems may be allocated below-ground (Vogt et al., 1982; Fahey & Hughes, 1994), and the extent to which NPP becomes long-term carbon storage greatly affects the capacity of forests to store atmospheric CO2. The increase in atmospheric CO2 may alter the partitioning of respiration among functional processes, as well as its absolute magnitude, thereby affecting the carbon cycling of ecosystems.

Fine root respiration supports three important functions: maintenance, growth and nutrient uptake (Johnson, 1983; Lambers et al., 1983). Maintenance respiration provides the energy to turnover proteins and to maintain ion gradients, growth respiration provides energy for construction of new cells and nutrient uptake respiration provides the energy required by epidermal root cells to actively transport ions against a concentration gradient. The partitioning of energy among these major functions will influence water and nutrient uptake by fine roots, which will affect tree growth, yet relatively few studies have quantified the proportional investment in these processes (Veen, 1980, 1981; de Visser & Lambers, 1983; Johnson, 1983; Van der Werf et al., 1988; Poorter et al., 1991; Bouma et al., 1996; Mata et al., 1996).

Nutrient uptake respiration is as high as 60% of total root respiration for maize (Veen, 1980, 1981). By contrast, Quercus suber used the majority of respiration for maintenance and used only 19–31% of its total respiration for nutrient uptake (Mata et al., 1996), reflecting the lower nutrient demand and greater nutrient use efficiency of this species. Nutrient uptake respiration has not been determined in an intact forest ecosystem, where the percentage of total fine root respiration used for nutrient uptake could be high, particularly for trees growing in nutrient-poor soils.

Growth in elevated atmospheric CO2 may alter the absolute rate, as well as the partitioning of fine root respiration. Several studies have documented a decrease in the specific rate of fine root respiration for trees grown in elevated atmospheric CO2 (Callaway et al., 1994; BassiriRad et al., 1997; Crookshanks et al., 1998). Growth under elevated CO2 causes a decrease in the nitrogen concentration of roots (Cotrufo et al., 1998) suggesting a reduction in protein concentration. Thus, the energy required for protein turnover may decline in elevated CO2 causing a reduction in maintenance respiration. If maintenance respiration of fine roots grown in elevated atmospheric CO2 is reduced, then more energy could potentially be available to support growth and nutrient uptake. By contrast to maintenance respiration, elevated atmospheric CO2 stimulates fine root production (Norby et al., 1986; Pregitzer et al., 1995; Crookshanks et al., 1998; Janssens et al., 1998; DeLucia et al., 1999). The decrease in maintenance respiration with elevated CO2 may contribute to increases in growth respiration.

The objective of this study was to estimate total fine root respiration and the proportions used for maintenance, growth and nitrogen uptake in loblolly pine and sweetgum forests growing under ambient and elevated atmospheric CO2. In addition, a survey of the literature was conducted for values of fine root respiration to provide comparisons for the rates reported in this study. Few studies report nitrogen uptake respiration, and none, to our knowledge, have attempted to estimate this process for an intact forest ecosystem. Nitrogen was investigated in this study as it was assumed that it represents the greatest expenditure of energy for nutrient uptake (Veen, 1980, 1981). We hypothesized that a reduction in nitrogen concentration of fine root tissue grown in elevated CO2 would reduce maintenance respiration, and that more energy would be used for fine root growth and nitrogen uptake. Fine root maintenance respiration was measured from gas-exchange of nongrowing roots in the absence of nutrients, and growth respiration was quantified from construction costs and production rate of fine roots. Nitrogen uptake respiration was estimated from the annual nitrogen uptake of trees at each site and from a literature value representing the respiration rate associated with nitrogen uptake.

Materials and Methods

Experimental sites

Measurements were made in two similar-age forests where experimental plots were fumigated with CO2 using free-air CO2 enrichment (FACE) technology (Hendrey et al., 1999). One experimental site is an even-aged loblolly pine (Pinus taeda L.) plantation (Duke Forest, North Carolina, USA 35°97′ N 79°09′ W) seeded in 1983 and left unmanaged since. More than 90% of the total biomass is pine (Hamilton et al., 2002), however, a diverse mixture of hardwood species has become established in the understory (Hartz-Rubin & DeLucia, 2001). The soil at this experimental site is an Ultic Alfisol and is low in total nitrogen and phosphorus. The prefumigation soil concentrations for nitrogen and phosphorus were 0.08% ± 0.01 (SD) and 1.23 p.p.m. ± 0.35, respectively (W. H. Schlesinger, pers. comm.). The other experimental site is Oak Ridge National Laboratory (ORNL; Tennessee, USA 35°54′ N 84°20′ W) where a plantation was established in 1988 with 1-yr-old-seedlings of sweetgum (Liquidambar styraciflua L.). The soil at ORNL forest is classified as an Aquic Hapludult, with higher total nitrogen and phosphorus concentrations (0.13% ± 0.01 and 8.21 p.p.m. ± 1.49, respectively) than the Duke Forest site.

The Duke site has six 30-m diameter experimental FACE plots. Three treatment plots have been fumigated with elevated CO2 beginning 27 August 1996. At the ORNL site there are four 25-m diameter FACE plots, and the fumigation of elevated CO2 for the two treatment plots began on 11 May 1998 (Norby et al., 2001). The elevated CO2 plots have target concentrations of 200 µl l−1 above ambient (global average 369 µl l−1 in 2000). The average daytime CO2 concentration in 2000 in the elevated CO2 plots was 545 ± 58 at the ORNL site and 534 ± 149 at the Duke site.

Maintenance respiration

At each FACE site the rate of CO2 efflux of fine roots from 10 separate locations within each experimental plot was measured using a portable IR gas analysis system with the conifer needle cuvette (Li-Cor 6400; Lincoln, NE, USA) in June and July 2000. Measurements were made at one time period as fine roots of tree species respond to changes in temperature (Q10) in a similar way through the growing season and no acclimation has been observed from natural diurnal and seasonal changes in temperature (Sowell & Spomer, 1986; Weger & Guy, 1991; Burton et al., 1996; Zogg et al., 1996; Burton et al., 1998). Intact roots were gently excavated from the organic layer and kept attached to the rest of the root system throughout the measurements. The intact mats (average 0.12 g dry mass) of fine roots (= 2-mm in diameter) were rinsed with water and blotted dry before being placed in the gas-exchange cuvette. During measurements the cuvette was darkened and the roots were kept moist by adding 10 ml of water to the soda lime tube attached to the Li-Cor 6400. This maintained a relative humidity in the cuvette of = 80% and ensured constant respiration rates for at least 30 min without reductions from drying (data not shown). All measurements were taken when the rate of CO2 efflux had stabilized, typically within 10 min of enclosing the roots in the cuvette. The air temperature within the cuvette was 25°C. To minimize CO2 diffusion between the air space inside the cuvette and the atmosphere (Burton & Pregitzer, 2002), measurements were made at the atmospheric CO2 concentration in each plot. The CO2 concentration within the cuvette was 360 µl l−1 in the ambient plots and at 560 µl l−1 in the elevated plots. Measurements on excavated roots of sweetgum and potted seedlings of loblolly pine indicated that variation in atmospheric CO2 concentration from 400 to 2000 µl l−1 had no affect on the rate of root respiration (K. George unpublished). After gas exchange measurements the roots within the cuvette were removed and dried at 70°C for 48 h for measurement of dry mass, nitrogen content and construction costs. It was assumed that these measurements represented maintenance respiration (RM), as fine root growth was slow at this time (Matamala & Schlesinger, 2000).

Annual maintenance respiration (inline image) was estimated by adjusting the instantaneous rates (RM) measured at 25°C to the average temperature experienced by fine roots over the year and multiplying by the standing mass of fine roots at each site. Average annual soil temperatures 10 cm below the soil surface were 14°C and 15°C for 2000 at the Duke and ORNL forests, respectively. Instantaneous rates of respiration was adjusted to these temperatures using the following equation from Ryan (1991):

RM = R25[exp(ln(Q10)(T − 25))/10];( Eqn 1)

(RM, instantaneous fine root maintenance respiration at temperature T; R25, the rate of fine root respiration at 25°C.) A value of 2.075 was used for Q10, which was an average of several values for conifers from the literature (Sowell & Spomer, 1986; Ryan et al., 1996, 1997; Clinton & Vose, 1999; Tjoelker et al., 1999). The same Q10 value was applied to sweetgum as we were unable to find estimates for this species in the literature. The Q10 for soil respiration in the sweetgum plots was 2.1–2.2 (P. J. Hanson, pers. comm.). The values of loblolly pine fine root standing mass for 1998 at the Duke site were from Matamala & Schlesinger (2000). The values of sweetgum fine root standing mass for 2000 at the ORNL site were from Norby et al. (2002).

Construction and growth respiration

Annual growth respiration (RG) for each forest was calculated from the tissue-specific construction respiration (RC) times the rate of production of fine roots for each FACE plot; where RC is the energy required to make new tissue on a mass basis. Construction respiration was calculated from the heat of combustion and the carbon content of roots as in Williams et al. (1987) and Carey et al. (1996).

Construction cost (C) of fine roots (g glucose g−1 dry weight tissue) was quantified from the ash free heat of combustion, ash content and total organic nitrogen using the following equation from Williams et al. (1987):

image(Eqn 2)

HC, the ash-free heat of combustion (kJ g−1); A, the ash content (g g−1); k, the oxygen state of nitrogen substrate; N, the organic nitrogen content (g g−1); and EG, the growth efficiency conversion (assumed to be 0.89).) A value of −3 was used for k; this value is appropriate for forest soils where most N is absorbed as ammonium (Christensen & MacAller, 1985). Ash-free heat of combustion was determined by combusting a 10–20 mg sample in a microbomb calorimeter (Gentry Instruments, Aiken, USA). Ash content was determined by combusting samples in a muffle furnace at 500°C for four hours (Carey et al., 1996). Carbon and nitrogen content of ground root tissue were measured with an elemental analyzer (NA1500, Carlo Erba, Milan, Italy). The heat of combustion and ash content were measured on a bulked sample from each experimental plot and the nitrogen content was measured on the individual samples (10 per plot) and averaged to provide a value for each experimental plot.

Construction costs expressed in units of glucose required for tissue synthesis, include both carbon incorporated into tissue and respired during construction (Nobel et al., 1992). RC was calculated by subtracting the structural carbon incorporated in the root tissue (carbon content) from construction costs. Units of glucose were converted to CO2 by assuming six CO2 mol are evolved per glucose mole (Nobel et al., 1992). Annual RG was calculated as the product of fine root production (kg m−2 year−1) and RC (mol CO2 kg−1). The values of loblolly pine fine root production for 1998 at the Duke site were from Matamala & Schlesinger (2000). The values of sweetgum fine root production for 2000 at the ORNL site were from Norby et al. (2002).

Nitrogen uptake respiration

Annual nitrogen uptake respiration (RN) was calculated as the product of nitrogen uptake by loblolly pine (Finzi et al., 2002) and sweetgum (R. J. Norby & D. W. Johnson, pers. comm.) over a year (mol N m−2 yr−1) for each FACE plot and an estimate of the specific cost of nitrogen uptake (0.99 mol O2 mol N−1; Mata et al., 1996). Nitrogen uptake was calculated as the annual increment of nitrogen in wood plus the amount lost as litterfall minus retranslocation, expressed on ground area basis (Finzi et al., 2002). The specific cost of nitrogen uptake was converted to mol CO2 mol N−1 using a respiratory quotient of 0.8 (Penning de Vries et al., 1974; Poorter et al., 1991; Matamala & Schlesinger, 2000). There are few values in the literature for the specific cost of nitrogen uptake. The value reported by Mata et al. (1996) is for an evergreen woody species, Quercus suber, growing on nitrogen-poor soil.

Total respiration

The temperature-adjusted annual rate of RM and the temperature-independent annual RG and RN were summed to calculate annual total fine root respiration (RT) for each experimental plot.

Literature survey

A search of Biological Abstracts (Ovid, New York, USA) covering year 1980–2000 was conducted using ‘respiration’, ‘fine’ and ‘roots’ as keywords. The bibliographies of the articles were then scanned for further reports of fine root respiration. The survey produced 39 publications with respiration rates of fine roots (= 2 mm in diameter) from tree species dating from 1950. All rates in the literature were converted to nmol CO2 g−1 tissue s−1 and to 15°C using Eqn 1 (Appendix 1). Studies were excluded from the survey if they did not report temperature during the measurement. The data were used to compare the rate of fine root respiration for gymnosperm vs angiosperm species, mature trees vs seedlings, and roots that were attached to the tree vs those detached during measurements.

Data analysis

Each plot was treated as an experimental unit and replicate measurements within each plot were averaged to provide a plot mean. The Duke FACE site was treated as a split plot design (n = 3), and a paired t-test was applied to each variable. There was no blocking at the ORNL FACE site and an independent samples t-test was used for all variables (n = 2). The mean distributions of data derived from the literature survey were not analyzed statistically. A meta-analysis could not be performed on the data set as individual publications did not contain data on respiration rates of roots that were both attached to the tree and detached during measurements, from mature trees and seedlings and from gymnosperm and angiosperm species (Appendix 1). Log transformations were performed where data were not normal. All statistical analyses were conducted with SPSS 10.05 (Chicago, IL, USA). Unless otherwise stated, P < 0.05 was the accepted probability level.

Results

Tissue-specific maintenance respiration

At the Duke FACE site, the instantaneous rate of maintenance respiration (RM) of loblolly pine fine roots, expressed on a dry mass basis, was significantly lower in the elevated CO2 treatment (P < 0.05; Table 1), but was not different between treatments when expressed on a nitrogen basis. There was no significant difference between the CO2 treatments for instantaneous RM of sweetgum fine roots from the ORNL site when expressed on a mass or nitrogen basis (Table 1). RM appeared to be higher for sweetgum than for loblolly pine when expressed on a mass basis but not on a nitrogen basis. No relationship was found between nitrogen concentration of the individual fine root tissues and their respective respiration rates for either species (data not shown).

Table 1.  Instantaneous maintenance respiration (RM) rates on a mass and nitrogen basis at 25°C for loblolly pine (Pinus taeda) and sweetgum (Liquidambar styraciflua) fine roots growing at the Duke and ORNL free-air CO2 enrichment (FACE) sites in ambient (∼360 µl l−1) and elevated (∼560 µl l−1) atmospheric CO2
 Loblolly pine Ambient CO2Elevated CO2% E-ASweetgum Ambient CO2Elevated CO2% E-A
  1. Each value is a mean of three plots for loblolly pine and two plots for sweetgum (± 1 SD). The percentage difference in the rates for trees in ambient and elevated CO2 plots is designated ‘% E-A’. The asterisk represents a significant (P < 0.05) difference between CO2 concentrations.

RM (nmol CO2 g−1 s−1)8.93 (1.3)6.91 (0.6)−22.6*10.16 (0.2)11.75 (1.7)15.6
RM (µmol CO2 g−1 N s−1)0.96 (0.3)0.85 (0.1)−11.5 1.06 (0.1) 1.08 (0.1) 2.0

Tissue-specific construction respiration

The ash-free heat of combustion (ΔHC) for loblolly pine fine roots was higher in the elevated CO2 treatment (P < 0.05; Fig. 1c), resulting in a nonsignificant increase in construction respiration (RC) in the elevated CO2 treatment (P = 0.10; Fig. 1f). There was no difference in the other components of RC for pine across treatments (Fig. 1a,b,d,e). For sweetgum no differences between the CO2 treatments for RC or any of its components could be detected (P > 0.05; Fig. 1a–f).

Figure 1.

Effect of ambient (∼360 µl l−1, closed bars) and elevated (∼550 µl l−1, open bars) atmospheric CO2 on (a) percent ash (b) percent nitrogen (c) ash-free heat of combustion (ΔHC) (d) construction cost in g glucose g−1 of tissue (e) percent carbon and (f) construction respiration (RC) of loblolly pine (Pinus taeda) and sweetgum (Liquidambar styraciflua) fine roots. The asterisk represents significant differences (P = 0.05) between the ambient and elevated CO2 treatments within a species. Each bar is a mean of three plots for loblolly pine and two plots for sweetgum (± 1 SD).

Annual total, maintenance, growth and nutrient uptake respiration

Annual growth respiration (RGP < 0.05) and fine root production (Matamala & Schlesinger, 2000) were significantly greater in the elevated CO2 treatment for loblolly pine (Table 2). Annual nitrogen uptake respiration (RN) was marginally greater in the elevated CO2 treatment (P = 0.06) for loblolly pine as the uptake of nitrogen by these trees was increased in elevated CO2 (Finzi et al., 2002). The uptake of nitrogen by loblolly pine was less than half the uptake of nitrogen by sweetgum (Table 2). There was no significant difference in annual RT and RM between CO2 treatments for loblolly pine (P > 0.05; Table 2). The high variance in RT and its components for sweetgum resulted in no significant differences, although annual RT and RM were consistently higher under elevated CO2. It appeared that annual RT and RM were greater for loblolly pine than for sweetgum, whereas annual RG and RN were lower in loblolly pine than sweetgum (Table 2).

Table 2.  Annual total (RT), maintenance (RM), growth (RG), and nitrogen uptake (RN) respiration and fine root-standing mass, construction respiration (RC), fine-root production and nitrogen uptake for loblolly pine (Pinus taeda) and sweetgum (Liquidambar styraciflua) growing at the Duke and ORNL free-air CO2 enrichment (FACE) sites in ambient (∼360 µl l1) and elevated (∼560 µl l1) atmospheric CO2
 Loblolly pine Ambient CO2Elevated CO2% E-ASweetgum Ambient CO2Elevated CO2% E-A
  1. RM was adjusted to the annual soil temperature at each site. Each value is a mean of three plots for loblolly pine and two plots for sweetgum (± 1 SD). The percentage difference in the rates for trees in ambient and elevated CO2 plots is designated ‘% E-A’. Asterisks represent a significant (P = 0.05) difference between CO2 concentrations. 1Fine root standing mass of loblolly pine in 1998 from Matamala & Schlesinger (2000) and of sweetgum in 2000 from Norby et al. (2002). 2Fine root production of loblolly pine in 1998 from Matamala & Schlesinger (2000) and of sweetgum in 2000 from Norby et al. (2002). 3Nitrogen uptake of loblolly pine from Finzi et al. (2002) and sweetgum from R. J. Norby & D. W. Johnson (pers. comm.).

Annual RT (g C m2 yr1)638.6 (168.4)531.2 (47.5) −17245.1 (84.1)454.9 (182.8)+86
Annual RM (g C m2 yr1)631.0 (168.8)517.7 (48.2) −18208.1 (75.8)400.0 (144.5)+92
RM (g C g1 yr1)  1.7 (0.3)  1.3 (0.1) −24*  1.9 (0.0)  2.2 (0.3)+16
1Standing mass (g m2)363.5 (97.3)385.4 (35.9)  +6112.6 (39.5)194.8 (96.2)+73
Annual RG (g C m2 yr1)  2.6 (1.4)  6.5 (2.1)+151* 24.1 (6.5) 41.4 (38.9)+72
RC (g C kg1) 59.2 (20.7) 81.2 (19.6) +37 69.3 (12.7) 59.2 (29.2)−15
2Production (g m2 yr1) 42.8 (13.0) 80.0 (9.9) +87*345.3 (30.8)612.7 (355.4)+77
Annual RN (g C m2 yr1)  5.1 (1.1)  7.0 (1.6) +39 12.9 (1.7) 13.5 (0.6) +5
RN (g C g N1)  1.8 (0.0)  1.8 (0.0)   0  1.8 (0.0)  1.8 (0.0)  0
3N uptake (g N m2 yr1)  2.8 (0.6)  3.9 (0.9) +39  7.1 (1.0)  7.5 (0.3) +5

Literature survey

To provide a context for our results, a review was conducted of fine root respiration rates in the literature for tree species. The distribution of values reported in the literature was highly skewed to lower rates, with the majority of values at 15°C were = 10 nmol CO2 g−1 s−1 (Fig. 2). The average rate of fine root respiration for attached and severed roots was 9.1 nmol CO2 g−1 s−1 and 6.7 nmol CO2 g−1 s−1, respectively, but a few high values contributed to the higher average for attached roots. Average fine root respiration rates from seedlings (9.1 nmol CO2 g−1 s−1) were substantially greater than for mature trees (3.7 nmol CO2 g−1 s−1). For seedling fine roots, 74% of respiration rates were = 10 nmol CO2 g−1 s−1, compared to 97% of the rates for mature trees. There was no apparent difference between the average fine root respiration rates of angiosperm and gymnosperm tree species. Data from this study are consistent with the literature. At 15°C the RM for fine roots of loblolly pine were 4.47, and 3.46 nmol CO2 g−1 s−1, and 5.08 and 5.88 nmol CO2 g−1 s−1 for sweetgum, in the ambient and elevated CO2 treatments, respectively.

Figure 2.

Relative frequency of fine-root respiration at 15°C (R15) of 328 independent measurements from 39 studies (Appendix 1) divided between fine roots that were attached or detached during respiration measurements, mature trees and seedlings and gymnosperm and angiosperm species.

Discussion

Annual growth (RG) and nitrogen uptake respiration (RN) of fine roots were greater in elevated CO2 for loblolly pine but were unchanged for sweetgum. The duration of exposure to elevated CO2 may have contributed to the difference in response of these forests; at the time of our measurements the pine forest had been exposed to elevated CO2 for 4 yr, whereas the sweetgum forest had only been exposed for 2 yr. The increase in annual RG and RN in response to elevated CO2 for loblolly pine was not apparent in total fine root respiration (RT) because they were such small fractions of the total. Annual maintenance respiration (inline image) accounted for 98% and 86% of RT in the loblolly pine and sweetgum forest, respectively. It was initially predicted that a reduction in the nitrogen concentration of fine roots in elevated CO2 would reduce inline image and increase the energy available for growth and nitrogen uptake. There was a significant reduction in instantaneous RM for pine grown under high CO2 but this difference was no longer statistically significant when extrapolated to inline image, and there was no significant change in the nitrogen concentration of fine roots. It appears that the C : N ratio of fine roots grown in elevated CO2 was not altered and consequently did not explain the trend of reduced inline image and the increase in annual RG for loblolly pine.

Instantaneous RM on a mass basis was significantly reduced by the elevated CO2 treatment for loblolly pine but not for sweetgum. It has been suggested that a reduction in tissue nitrogen concentration, possibly caused by an increase in carbon content (Cotrufo et al., 1998), reduced respiration rates of tree roots grown under elevated CO2 (Callaway et al., 1994; BassiriRad et al., 1996; Crookshanks et al., 1998). We were unable to detect an effect of elevated CO2 on the nitrogen concentration of fine roots for either species. Instantaneous RM on a mass basis was higher for sweetgum than loblolly pine and this difference was eliminated when expressed per unit N (Table 1), suggesting that the rate of CO2 flux may have been related to nitrogen concentration. However, no relationship was apparent between individual root respiration measurements and corresponding nitrogen concentrations (n = 53 for loblolly pine and n= 33 for sweetgum; data not shown). It appears that while expressing instantaneous RM on a nitrogen basis reduces some of the variation in respiration rates, the observed differences in fine root respiration are not explained completely by nitrogen concentration.

There was a trend of greater construction respiration (RC) under elevated atmospheric CO2 for loblolly pine fine roots but not for sweetgum. The increase in the ash-free heat of combustion of loblolly pine fine roots in elevated CO2 resulted in a small increase in construction costs and RC. The increase in construction costs from elevated CO2 may be related to increases in the lignin concentration of fine roots (Eissenstat, 1992). In terms of glucose equivalents, lignin is one of the most expensive compounds to produce (Penning de Vries et al., 1974) and elevated CO2 has been found to increase the lignin concentration of roots (Booker et al., 2000). Elevated CO2 also affects RC of other plant tissues. Construction costs and RC were reduced in leaves with increasing atmospheric CO2 (Wullschleger & Norby, 1992; Wullschleger et al., 1992; Griffin et al., 1993; Ziska & Bunce, 1994), which was primarily associated with changes in nonstructural carbohydrates and to a lesser extent by lignin (Griffin et al., 1996). Elevated CO2 may result in the construction of more expensive structural compounds in fine roots.

The stimulation of annual RG for loblolly pine under elevated CO2 was caused primarily by the increase in fine root production (Table 2). Both increased RC and fine root production, when extrapolated to the entire forest, contributed to an increase in RG under elevated CO2. But, the stimulation of fine root production by elevated CO2 (87%) was considerably greater than the stimulation of RC (37%). There was no detectable effect of elevated CO2 on RC in sweetgum and only increased root production contributed to the trend of greater RG under elevated CO2 for this species (Table 2). For these two forests it appears that an increase in fine root production is the primary factor contributing to the increase in annual RG under elevated CO2. Loblolly pine had lower fine root production than sweetgum and consequently also had lower annual RG. Trees commonly exhibit an increase in fine root growth under elevated CO2 (Norby et al., 1986; Pregitzer et al., 1995; Crookshanks et al., 1998; Janssens et al., 1998) and in these cases we would also predict an increase in annual RG.

The vast majority of annual RT was used to support cellular maintenance processes (RM), both for loblolly pine and sweetgum. On average inline image was 98% of RT for loblolly pine and 86% for sweetgum, leaving a small proportion of energy annually for RG and RN. The proportion of RT for fine roots allocated to inline image in this study was comparable to a 20-yr old Pinus radiata stand, where RT was 1940 g m−2 y−1 and inline image was 76% of this total (Ryan et al., 1996).

Our estimates of annual RT for loblolly pine were similar to those for another young loblolly pine stand in the Piedmont of North Carolina (663–1062 g C m−2 y−1; Maier & Kress, 2000), but considerably higher than those reported by Matamala & Schlesinger (2000; 349.4 and 401.0 g C m−2 year−1 in ambient and elevated CO2, respectively) at the same site. Differences in methodology may have contributed to this disparity, though this remains uncertain. Unlike this study, Matamala & Schlesinger (2000) measured fine root respiration on severed roots using an oxygen electrode. Over two thirds of the studies in the survey used severed roots and the average respiration was higher for attached than for detached roots. However, the distribution of rates reported in the literature was highly skewed towards lower values, and the greater average for attached roots was caused by a few studies reporting very high rates (> 40 nmol CO2 g−1 s−1; Fig. 2).

Annual soil CO2 efflux from ambient and elevated plots in the pine forest were 928 g C m−2 and 1176 g C m−2, respectively (Andrews & Schlesinger, 2001; Hamilton et al., 2002), and the values for ambient and elevated plots in the sweetgum forest were 960 g C m−2 and 1271 g C m−2, respectively (Norby et al., 2002). The values of total soil CO2 efflux in these forests are close to the average annual value of 1050 g C m−2 for 34 different forest types and similar to other temperate forests (Davidson et al., 2002). In the pine forest, RT estimated from this study contributed 69% and 45% of the total CO2 efflux from ambient and elevated plots, respectively. Using the unusual C isotopic composition of newly fixed C in the elevated plots, Andrews et al. (1999) estimated that 45% of total soil CO2 efflux originated from roots. The proportion of soil CO2 from fine roots was somewhat lower in the sweetgum forest than in the pine forest (ambient, 25%; elevated, 36%).

To calculate the proportion of respiration to support RN annually, we multiplied estimates of total nitrogen uptake in both forests by a literature value for the specific respiration associated with nitrogen uptake. The rate of respiration per unit nitrogen uptake by the trees was taken from a study of Quercus suber. This slow growing evergreen tree species, from xeric, nitrogen-poor soils (Mata et al., 1996), was the best match to loblolly pine and sweetgum in our study (Table 3). Mata et al. (1996) also found similar rates to ours for tissue specific root maintenance (7.13 nmol CO2 g−1 s−1 at 25°C) and growth respiration (51.8 g C kg−1 compare to RCTable 2). The value of RN was 1.8 g CO2 g−1 N for Quercus suber, which was similar to studies of herbaceous plants (Table 3). For example, five studies of herbaceous plants found a narrow range of RN 1.0–3.2 g CO2 g−1 N (Table 3; Veen, 1980, 1981; Johnson, 1983; Van der Werf et al., 1988; Poorter et al., 1991; Bouma et al., 1996). Because RN and the absolute rates of nitrogen uptake are relatively small, the choice of the instantaneous value of RN is not likely to have a large effect on our annual estimate.

Table 3.  Published estimates of nitrogen uptake respiration (RN) and corresponding values of instantaneous maintenance respiration (RM) at 25°C and construction respiration (RC) for various species. The value of RN from Poorter et al. (1991) was based on total anion uptake and is a median of 24 herbaceous species. The values of RN from the other studies were based on the uptake of nitrate.
SpeciesRM (nmol CO2 g−1 s−1)RC (g C kg−1)RN (g CO2 g−1 N)References
Zea mays 3.7104.63.2Veen (1980, 1981)
Helianthus annuus2.0Johnson (1983)
Carex species 4.9 60.51.8Van der Werf et al. (1988)
24 Herbaceous species 64.81.7Poorter et al. (1991)
Solanum tuberosum13.3 47.01.0Bouma et al. (1996)
Quercus suber 7.1 51.81.8Mata et al. (1996)

In our study annual RN was much greater in sweetgum than loblolly pine. This was because the uptake of nitrogen on a ground area basis in the sweetgum stand was nearly double the uptake of nitrogen in the loblolly pine stand. Annual RN required a small expenditure of energy in relation to annual RT, but this proportion was greater for the sweetgum stand (4.1%) compared to the loblolly pine stand (1.1%). The higher nitrogen availability in the sweetgum stand resulted in lower fine root standing mass and RT and greater RN per unit fine root mass compared to the loblolly stand.

In summary, the majority of fine root respiration was used for maintenance and was not reduced by changes in the nitrogen content of the fine roots grown in elevated atmospheric CO2, as initially hypothesized. The future investment of carbon in RM will depend upon the balance between the C : N ratio of tissues and the size of fine root standing biomass. In the loblolly pine forest annual RT was 26% and 19% of gross primary productivity (GPP) in ambient and elevated atmospheric CO2, respectively (Hamilton et al., 2002). Because of its large contribution to RT and total soil CO2 efflux, changes in RM caused by warming or other factors have the potential to greatly alter carbon losses from forests to the atmosphere.

Acknowledgements

We thank E.G. O’Neill for providing fine root standing mass and production for the sweetgum stand and A.C. Finzi for measurements of carbon and nitrogen content of fine root tissue and nitrogen uptake rates of the loblolly pine stand. This research was supported by a grant (DE-FG02–95ER62127) from US Department of Energy, Office of Science, Biological and Environmental Research Program. Additional support was provided by a Dissertation Travel Grant from the Graduate College at the University of Illinois Urbana-Champaign, and a Summer Research Fellowship from the Program in Ecology and Evolutionary Biology at the University of Illinois Urbana-Champaign to KG. We gratefully acknowledge G.R. Hendrey, K.F. Lewin, J. Nagey and W.H. Schlesinger for operating and providing access to the FACTS-1 experiment, and J. Edeburn for managing the Duke Forest.

Appendix

Appendix 1

Table 4. Literature survey of fineroot respiration rates of tree species at 15°C (R15). Respiration rates measured on seedlings are indicated by ‘S’ and rates measured on mature trees are indicated by ‘M’. Respiration rates measured on roots detached from the plant are indicated by ‘D’ and rates measured on roots that remained attached to the plant are indicated by ‘A’
SpeciesCategoryR15 (nmol CO2 g−1 s−1)Author
Abies lasiocarpaS, D25.00–29.37Sowell & Spomer (1986)
Acer rubrumS, D23.79Steinbeck & McAlpine (1966)
Acer rubrumS, D 0.53–0.97Carpenter & Mitchell (1980)
Acer rubrumS,D 0.98–1.47Tripepi & Mitchell (1984)
Acer rubrumM, A 1.90Rakonczay et al. (1997b)
Acer rubrumM, D 2.07–6.40Rakonczay et al. (1997a)
Acer saccharumS, D 0.27–0.98Carpenter & Mitchell (1980)
Acer saccharumS, D 0.82–0.85Tripepi & Mitchell (1984)
Acer saccharumS, A 3.98–35.05Walters et al. (1993)
Acer saccharumM, D 2.91–5.89Burton et al. (1996)
Acer saccharumM, D 2.26–5.31Burton et al. (1997)
Acer saccharumM, D 0.90–6.42Pregitzer et al. (1998)
Betula alleghaniensisS, A 6.37–56.08Walters et al. (1993)
Betula nigraS, D 3.86–8.05Boyer et al. (1971)
Betula nigraS, D 0.74–1.38Tripepi & Mitchell (1984)
Betula papyriferaS, A 5.29–79.66Walters et al. (1993)
Betula pendulaS, D 0.83–1.09Tripepi & Mitchell (1984)
Citrus volkamerianaS, A 2.50–12.50Bouma et al. (1997a)
Citrus volkamerianaS, A 1.06–12.37Bouma et al. (1997b)
Citrus volkamerianaS, A 0.40–5.37Bouma et al. (2000)
Fraxinus excelsoirS, D 6.50–7.30Crookshanks et al. (1998)
Larix laricinaS, A 6.26–9.35Conlin & Lieffers (1993)
Liquidambar styracifluaM, A 5.08–5.88George et al. (this study)
Liriodendron tulipiferaS, D 6.97–13.57Boyer et al. (1971)
Liriodendron tulipiferaS, D17.46Steinbeck & McAlpine (1966)
Liriodendron tulipiferaS, D 1.30–4.35Cox (1975)
Malus spps.M, D 5.42–15.12Bouma et al. (2000)
Ostrya virginianaS, A 6.37–58.63Walters et al. (1993)
Picea abiesS, D 3.93–7.80Lahde (1966)
Picea engelmanniiS, D27.48–35.61Sowell & Spomer (1986)
Picea glaucaS, D 0.39–1.72Johnson-Flanagan & Owens (1986)
Picea glaucaS, A 3.90–8.07Conlin & Lieffers (1993)
Picea marianaS, A 3.50–9.72Conlin & Lieffers (1993)
Pinus banksianaS, D 5.55–6.40Lafond (1950)
Pinus banksianaS, D19.40Voigt (1953)
Pinus banksianaS, A 4.17–11.04Conlin & Lieffers (1993)
Pinus contortaS, A 5.61–12.78Conlin & Lieffers (1993)
Pinus echinataS, D11.26–22.52Allen (1969)
Pinus elliottiiM, D 0.91–1.66Cropper & Gholz (1991)
Pinus ponderosaM, D 0.56–0.92Marshall & Perry (1987)
Pinus ponderosaS, D 4.72–6.05BassiriRad et al. (1997)
Pinus resinosaS, D27.02Voigt (1953)
Pinus radiataM, A 2.10–13.90Ryan et al. (1996)
Pinus strobusM, D 2.17–4.31Rakonczay et al. (1997a)
Pinus strobusM, A 0.09–5.72Clinton & Vose (1999)
Pinus sylvestrisS, D 4.16–8.09Lahde (1966)
Pinus sylvestrisS, D 8.19–8.45Crookshanks et al. (1998)
Pinus sylvestrisS, D 3.42–4.95Janssens et al. (1998)
Pinus taedaS, D 6.29–11.10Boyer et al. (1971)
Pinus taedaS, D 1.55–7.66Barnard & Jorgensen (1977)
Pinus taedaS, D 2.05–29.60Drew & Ledig (1981)
Pinus taedaS, A 0.10–0.20Edwards (1991)
Pinus taedaS, A 7.74–10.28BassiriRad et al. (1997)
Pinus taedaM, A 3.46–4.47George et al. (this study)
Prunus serotinaM, A 1.40Rakonczay et al. (1997b)
Pseudotsuga menziesiiS, D 0.04–0.09McCreary & Zaerr (1987)
Quercus petraeaS, D 7.50–8.50Crookshanks et al. (1998)
Quercus rubraS, A 3.19–25.49Walters et al. (1993)
Quercus rubraS, D 1.83–4.71Kelting et al. (1995)
Quercus rubraM, D 1.37–4.54Rakonczay et al. (1997a)
Quercus rubraM, A 2.00Rakonczay et al. (1997b)
Quercus suberS, A 3.75–5.53Mata et al. (1996)
Robinia pseudoacaciaS, D31.33Voigt (1953)
Salix babylonicaS, D25.83Steinbeck & McAlpine (1966)
Salix babylonicaS, D 2.37–4.54Boyer et al. (1971)
Salix nigraS, D29.75Steinbeck & McAlpine (1966)
Taxodium distichum var. distichumS, D 0.60–1.21Carpenter & Mitchell (1980)
Tsuga canadensisS, A11.05–16.74Szaniawski & Adams (1974)
Tsuga heterophyllaS, A 1.38–29.75McDowell et al. (1999)
Mixed hardwoodsM, D 5.57–9.85Fahey & Hughes (1994)
Mixed hardwoods (75%Acer sacharum)M, D 4.63–6.08Zogg et al. (1996)

Ancillary