• Biomass;
  • eddy covariance;
  • Fagus sylvatica;
  • forest ecosystem;
  • respiration


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

1. We present measurements of CO2 fluxes over 2 years above and within a young Beech stand in the east of France. This site is part of the Euroflux network set up to monitor fluxes over representative European forests.

2. The net ecosystem carbon (C) exchange was derived from continuous eddy flux measurements. Major components of the total flux (i.e. soil and above-ground biomass respiration and assimilation of leafy branches) were measured independently using chambers. The main C stocks (i.e. root, stem and branch biomass) were also quantified.

3. Daily minima of CO2 flux were typically around −20 µmol CO2 m−2 s−1 during the period of full leaf expansion, while night-time ecosystem respiration varied between 5 and 15 µmol CO2 m−2 s−1. The seasonal pattern of net ecosystem assimilation was very close to that of net assimilation at the single branch scale. The seasonal variation of net ecosystem exchange was closely related to leaf expansion and soil water content during the dry year of 1996.

4. Measurements of ecosystem respiration (eddy flux) were corrected for CO2 storage within the stand. This C flux showed a seasonal pattern, the maximum rates (4–7 g C m−2 day−1) occurring in spring and summer, and appeared to be correlated with soil temperature. Temporal variation of soil respiration showed the same pattern, and effects of both temperature and soil drying were found. Annual soil respiration was ≈ 70% of ecosystem respiration. Root respiration was 60% of the total below-ground respiration.

5. Annual net C exchange was −218 and −257 g C m−2 in 1996 and 1997, respectively, corresponding to net C uptake by the forest. These values are much lower than the annual biomass increment (stems and large roots) of the stand: 427 and 471 g C m−2 year−1, respectively. The difference may be explained by a release of CO2 from the decomposition of woody debris.

6. Ecosystem C loss by respiration was 800–1000 g C m−2 year−1. Gross C gain was 1000–1300 g C m−2 year−1. Ecosystem respiration therefore played a major role in the annual C balance of this forest.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

To elucidate the influence of forests on the global carbon (C) cycle and their response to the increase in atmospheric CO2 concentration ([CO2]), the main CO2 sinks, sources and stocks of C must be quantified more accurately. CO2 emissions from fossil fuel burning plus the release of C from deforestation represent ≈ 7·1 Pg C year−1 for 1980–89, while the corresponding increase in atmospheric CO2 is only ≈ 3·2 Pg C year−1 (Houghton et al. 1996). This imbalance is attributed to C sequestration by oceans and vegetation, although the relative importance of each C sink is unknown. Large variations in the C sequestration capacities of various forest ecosystems have been reported (Nabuurs & Mohren 1995). These variations depend on climate, species, site productivity and sylvicultural regime. Temperate and boreal forests may sequestrate ≈ 0·7 Pg C year−1 (Dixon et al. 1994; Jarvis 1995), so reducing the rate of CO2 release to the atmosphere. Recent estimates show that forest plantations fix ≈ 0·2 Pg C year−1 (Winjum & Schroeder 1997). However, the estimates of global fluxes are highly uncertain, and there is a need for direct measurements of these fluxes.

In order to obtain more information on C fluxes over and within forests, the European research programme Euroflux (‘long-term carbon dioxide and water vapour fluxes of European forests and interactions with the climate system’) was initiated in which a large-scale network of 15 representative forest sites was set up, from western oceanic to continental zones, and from boreal to Mediterranean countries (Baldocchi et al. 1996; Tenhunen et al. 1998). This programme aims: (1) to characterize and analyse CO2, energy and water fluxes over representative forest sites in relation to climate, species and stand structure, to provide functions and parameters for global models, and (2) to quantify the sink strength of these forests for CO2 and its spatial and temporal variation.

This paper presents the results obtained from 2 years of measurements in a young Beech stand. Different methodologies were used to evaluate independently the major components in the forest C balance. C fluxes estimated at different spatial scales (leaf, branch and stand) are compared. A companion paper (Granier, Biron & Lemoine 2000) deals with the water balance of the same forest stand.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References


The experimental plot (Euroflux site FR02) is in the state forest of Hesse, France (48°40′ N, 7°05′ E), in a stand composed mainly (90%) of naturally established Beech (Fagus sylvatica L.; see Table 1). Other tree species represented are Carpinus betulus L., Betula pendula Roth, Quercus petraea (Matt.) Liebl., Larix decidua Mill., Prunus avium L. and Fraxinus excelsior L. The stand is on a site of good productivity in the first class of Schober's yield table for Beech which predicts a mean increment of 9 m3 ha−1 year−1 at age 100 years (Le Goff 1981). The experimental plot (0·6 ha) was in the central part of a 65-ha zone composed mainly of young Beech. The plot and the surrounding stands were thinned 2 years before C flux measurements began. Most of the root biomass was in the top 40 cm of soil, but roots < 1 mm diameter occurred down to 150 cm.

Table 1.  Stand characteristics of the Hesse forest
Stand density (stems ha−1) 3480
Stand age in 1996 (years)   30
Mean height of Beech stand (m)   13
Mean girth per tree (cm)   24
Dry weight of trunks (kg ha−1)64500
Dry weight of branches (kg ha−1)11400
Dry weight of leaves (kg ha−1) 2400
Dry weight of roots (kg ha−1)15200
Total dry weight (kg ha−1)93500
Understorey vegetationsparse
Ground area (m2 ha−1)   20·7
Slope (%)   <5
Elevation (m above sea level)300
Soil typeluvisol/stagnic luvisol
Soil clay content (%) 0–0·1 m depth   25–35
Soil clay content (%) > 0·1 m depth   40
Mean annual rainfall (mm)  820
Mean annual temperature (°C)    9·2

Three towers were erected. One (18 m high) was used for eddy covariance (EC) and microclimate measurements. The two others (15 m high) were devoted to ecophysiological measurements. Sixty subplots were defined for geostatistical studies of soil and vegetation. Nine of these were used for soil respiration and temperature measurements and 12 others for measurements of tree girth and height.

Climate and microclimate

The following instruments were installed above the stand, at 18 m: a net radiometer (REBS, Seattle, WA, USA), a solar radiometer (Model CE-180, Cimel, Paris, France), a ventilated psychrometer (manufactured in-house and equipped with Pt 100 temperature sensors), a rain-gauge (Model ARG 100, Campbell Scientific, Logan, UT), and a switching anemometer (Vector Instruments, Rhyl, UK).

Within and below the canopy we measured: global radiation at heights of 12, 10, 8 and 1 m using 33-cm-long linear radiometers (INRA, Versailles, France); trunk and branch temperatures of one tree at four heights (1·5, 6, 8 and 10 m); soil heat flux using two heat flux transducers (REBS) at −5 cm; soil temperature at −10 cm (five replicates) and in one vertical profile (−5, −10, −20, −40 and −80 cm) in a central plot using copper/constantan thermocouples. Each of the five trees on which net C assimilation was measured was fitted with two 30-cm-long linear PAR sensors constructed from 20 amorphous silicon cells each 4 × 10 mm (Solems, Palaiseau, France).

Data were acquired every 10 s and 30-min averages were stored (Model CR7 datalogger, Campbell Scientific, Courtaboeuf, France).

Soil water content

Soil water content was measured weekly. During 1996 and 1997, volumetric water content (θ) was measured with a neutron probe (NEA, Ballerup, Denmark) in eight aluminium access tubes, six of which were 160 cm long, and two of which were 240 cm long. In 1997, a time-domain reflectometer (TDR; Trase system, Soil Moisture, Santa Barbara, CA) was also used to measure the soil water content in the upper soil layers. Twelve 40-cm-long stainless steel wave guides were used, two of which were placed in the middle of each of the six plots used for soil respiration and temperature measurements. Relative extractable water in the soil (REW; m3 m−3) was calculated using the average θ in the 0–160 cm soil layer as: (θ − θm)/(θF − θm), where θm and θF are the volumetric minimum water content and the water content at field capacity, respectively. θF was taken as the average observed θ during the winter when soil was refilled, and θm was estimated from retention curves. REW varies between 0 (soil dry) and 1 (field capacity).

Leaf area index

The leaf area index (LAI) was estimated for each of the two years by litter collection in autumn, using 42 square litter traps each covering 0·25 m2. Litter dry mass was measured weekly during the period of collection (from the beginning of October to the end of December). A subsample of leaves was collected every 2 weeks to estimate leaf area (Delta-T Area meter, Delta–T, Cambridge, UK), in order to convert dry mass into leaf area. Mean LAI was ≈ 5·6 ± 0·5 m2 m−2 averaged over both years. Seasonal variations in LAI were derived from the absorption of global radiation by the canopy (see below) using the Beer–Lambert law and an extinction coefficient of 0·4. LAI was also calculated from direct sampling on 23 representative trees (see below); this method gave a mean LAI of 4·9 ± 0·4 m2 m−2 averaged over both years.

Biomass and biomass increment

Biomass equations were developed from a sample of 23 representative trees near the experimental stand. Fresh weights of stem, branch and foliage were measured in the field, and dry weights obtained from oven-dried samples. Relative biomass increments of stem and branch sections were obtained from the last 5-year wood increments measured on the samples.

The root systems of a subsample of 16 trees were excavated mechanically. A mechanical digger excavated around the root systems, with trunks being cut at the base of the tree, and then the stump was pulled up carefully and the root system cleaned. Fresh weights of coarse and small roots (diameter 2 mm) were obtained after washing soil from the roots. Biomass and biomass increment of the root systems were obtained from the measurements of the last 5-year wood increments and the dry weights of root samples (Le Goff & Ottorini 2000).

Equations relating tree girth at breast height to biomass data were obtained. Total above-ground and below-ground biomass and the annual biomass increment of the woody parts of the experimental stand were derived from stand girth distributions and from these equations, for 1996 and 1997. Total stand biomass increments (±95% confidence limits) were 10 400 ± 1530 kg ha−1 and 11 300 ± 1580 kg ha−1 in 1996 and 1997, respectively.

Biomass data were then converted into C mass, using the following equivalents (Santa Regina et al. 1997): 1 kg dry matter = 0·44 kg C in stems, roots and branches and 0·457 kg C in leaves.

Fine root biomass (< 2 mm diameter) was estimated at the stand level from eight soil cores (i.d. 8 cm, depth 12 cm) collected monthly from March to July 1997. Cores were stored in plastic bags at 4 °C until fine roots were washed free of soil, sorted into apparently living and dead fractions and dried at 60 °C for 48 h. Fine root growth into 14 root-free cores was used to estimate annual fine root production from the amount of root that had grown into each core in 1 year (Persson 1983). Soil cores (i.d. 8 cm, depth 12 cm) were taken in March 1997, all roots were carefully removed, and the sifted soil was replaced in the hole. In April 1998, the cores were retrieved and processed as above. These estimations of fine root biomass and annual fine root production were corrected for the spatial and vertical variations of fine root biomass, assuming that fine root biomass in soil cores represented 45% of the total fine root biomass (Epron et al. 1999b).

Tree girths were measured manually at breast height every 2 weeks on all trees in the 12 subplots used for growth measurements (541 trees). Five trees were fitted with automatic dendrometer bands, installed at three heights on the stem: breast height, at the base of the living crown, and at the middle of the stem.

Ecosystem gas exchange

Water vapour and CO2 fluxes were measured by EC (Leuning & Moncrieff 1990) at 18 m, i.e. 3 m above the mean tree height, from a tower in the middle of the plot. Air was drawn from the top of the tower via an inlet, which was equipped with a 0·2-µm PTFE filter (Model Spiralcap, Pall-Gelman, Ann Arbor, MI, USA) at the base of a 3D anemometer (Solent, Model R2, Gill Instruments Ltd, Lymington, UK), through a 30-m long PTFE tube (4 mm i.d.). Air was sucked through the tube by two piston pumps in series (Model Rotronic 406G, Reciprotor, Skara, Sweden). The pumps were 180 V AC powered, instead of 220 V, to reduce warming. The flow rate of 6 dm3 min−1 was monitored by a mass flow controller (Model 5850, Brooks, Veenendaal, Netherlands). Water vapour and CO2 concentrations were measured with a CO2/H2O infra-red gas analyser (IRGA; Model 6262, Li-Cor Inc., Lincoln, NE). Dry, CO2-free reference air was provided for the IRGA using a diaphragm pump (Model TD3LS, Brailsfort, New York, NY) driving air through two columns of CO2 absorbant and desiccant, respectively. Calibration was performed periodically: zero was checked weekly and span every 3 months. Drift was 1 µmol mol−1. Wind speed and gas concentration were scanned at 10 Hz. edisol software (University of Edinburgh, Edinburgh, UK) was used to calculate on-line sensible heat, CO2 and water fluxes. The moving average time constant was 200 s. An average 6% loss of CO2 fluxes was observed between 0·1 and 1 Hz. Raw data were corrected using a relationship between CO2 flux loss and wind speed.

EC measurements started on 20 May 1996 [day of year (DOY) 141]. By 31 December 1996, 160 out of 226 days (70%) had yielded reliable data. In 1997, we obtained reliable data for 318 days (87%). Some data (1·8% of the total) were excluded if: (1) sensible heat fluxes were abnormally large, corresponding to measurements during rainfall or when the 3D anemometer was wet, or (2) filters were partially blocked, restricting the air flow. We checked that eliminated data were distributed randomly so that no bias was introduced to the C flux estimations. Spectral and cospectral analyses were performed periodically to check the quality of flux measurements. Full energy balance was not achieved even if latent heat flux was corrected (Aubinet et al. 1999) for losses at > 0·5 Hz and if heat storage in air and in vegetation was taken into account using the relation ΦL + H + Hs + G= 0·88 Rn (where ΦL, H, Hs, G and Rn are the latent heatflux, sensible heat flux, heat storage in the air and in the vegetation, heat flux in the soil, and net radiation, respectively, in W m−2). Fetch limits were estimated (Gash 1986) as 509 and 77 m from the tower for 90 and 50% of the fluxes, respectively.

Co2 profiles

To quantify the CO2 stored or released by the forest, [CO2] was monitored at four additional heights within and below the canopy. Air was sucked through 2-mm i.d. PTFE tubes with 1-µm PTFE filters (Model ACRO 50, Pall-Gelman, Ann Arbour, MI, USA) installed at 0·2, 0·7, 2 and 8 m above ground level. A piston pump (Reciprotor 406 G, Reciprotor, Skara, Sweden) drove air to an IRGA (see above) at a flow rate of 1 dm3 min−1. Five solenoid valves, driven by a datalogger (see above), allowed sequential measurement of [CO2] at each height, plus one for zero measurement (air passing through soda lime). The duration of a complete cycle was 15 min. Variation in CO2 content per unit of ground area within the 0–18 m air layer was calculated every 30 min.

Branch gas exchange

Three branch bags were used to monitor CO2/H2O exchange of branches during the 1997 growing season (25 April to 4 November). Two bags were placed at the top of the canopy (13 m above ground level) and a third was placed at 7 m. Each bag comprised a rigid frame covered with polypropylene film, 0·25 m3 in volume. Between measurements, air was blown through the bags (at 0·045 m3 s−1) to minimize overheating. During measurements, bags were closed for 3 min and used as cuvettes operating as a closed system. Bags were scanned sequentially every 30 min and monitored by a datalogger (see above). The CO2 depletion in the bags was measured using an IRGA (see above) operating in absolute mode, and the H2O increase was measured using capacitive sensors (Model HMD 30YB, Vaisala OY, Helsinki, Finland). PAR and air temperature were also measured inside the bags. A similar system was described previously by Dufrêne, Pontailler & Saugier (1993) and used in the BOREAS project (Saugier et al. 1997).

Leaf net co2 assimilation

Net C assimilation by leaves was measured throughout the growing season using a portable system IRGA (LI-6200, Li-Cor Inc., Lincoln, NE) on five trees surrounding one of the towers, from different crown classes: two dominant (i.e. tall trees in which most of the crown is sun-exposed), one co-dominant (medium-sized), and two intermediate (small) trees. Those measurements were made under ambient light on all five trees at two heights in the crowns (≈ 1/3 and 2/3 of the crown extent) on 2–4 leaves selected randomly from branches fitted with linear PAR sensors.

Scaling of net C assimilation to the tree and to the stand was carried out as follows: (1) response curves to PAR were derived from leaf measurements for each tree and each height; (2) response curves plus continuous PAR records were used to calculate photosynthesis per unit of leaf area during the season; (3) we assumed that maximum C assimilation per unit leaf area increased linearly for 30 days from budburst to full leaf expansion, and that the reverse occurred during leaf senescence for 30 days; (4) scaling net C assimilation to the stand was carried out from the areas of sun and shade leaves of each crown class, the total leaf area being estimated as 2·6, 1·6 and 1·5 m2 m−2 for dominant, co-dominant and intermediate/suppressed trees, respectively.

Gross C assimilation was calculated as the difference between net C assimilation and diurnal leaf respiration. The diurnal leaf respiration rate was assumed to be equal to nocturnal leaf respiration. We also assumed that leaf respiration represented half of the respiration by the aerial biomass, and was calculated as Rag = Reco − Rs (see Lebaube et al. 1999), where Rag is the above-ground respiration (g C m−2 s−1), Reco is the ecosystem respiration (g C m−2 s−1), and Rs is the soil respiration, including roots and heterotrophic organisms (g C m−2 s−1).

Soil and root respiration

Six subplots each of ≈ 100 m2 were chosen randomly within the experimental plot for soil CO2 efflux measurements. Two further 3-m2 subplots (2 × 1·5 m) containing no trees were established in June 1996 by digging a trench (1 m deep) around each subplot, lining the trench with a polyethylene film and back-filling the soil. Soil CO2 efflux was measured with a portable IRGA connected to a chamber (0·854 dm3 volume; 0·72 dm2 ground area; Norman, Garcia & Verma 1992). The chamber edge was inserted into the soil to a depth of 1·5 cm. After measuring [CO2] over the soil surface, [CO2] within the chamber was reduced by 15 µmol mol−1, and the subsequent increase in [CO2] recorded for 60 s. Every 2–4 weeks, 12 such measurements were made on each subplot from 0800 to 1600 h local time. Daily means (n = 72 for the main plot and 24 for the trenched plots) and confidence intervals (P = 0·05) were calculated.

The proportion of soil respiration originating from roots was calculated by subtracting the CO2 efflux in trenched plots and taking account of the decomposition of roots killed by trenching. Fine root decomposition was estimated by coring and sorting the remaining dead roots in the trenched plots 2 years after trenching. The remaining fine root necromass was compared with initial fine root biomass and necromass in soil cores. Coarse roots (2–10 mm diameter) excavated during trenching were washed free of soil, cut into pieces 4–6 cm long and placed into 10 × 15 cm litter bags (1 mm mesh size). These were then buried in soil at 10–15 cm depth. Fourteen litter bags were collected on each of five occasions during a 20-month period. Roots were carefully washed free of soil and dried at 60 °C for 5 days. Exponential decay functions (Mt = M0ekt) were fitted to the data, where Mt and M0 are, respectively, the remaining and the initial root dry mass, t is time and k is the decay constant. C loss as CO2 during root decomposition was calculated as (1 − a)cM0(1 − ekt), where the initial C concentration in roots, c, was assumed to be 44%, a is the fraction of C which is incorporated into soil organic matter (SOM), and 1 − a is the fraction lost as CO2 by microbial respiration during initial decay. a was assumed to be 0·22 (see Epron et al. 1999a).

Above-ground wood respiration

Respiration from above-ground wood was estimated by measuring stem CO2 efflux using IRGAs and two-piece cylindrical chambers (see Table 2). Measurements were made at 1·3 m on 15 representative trees of five different diameter classes and also in the crown of a single tree (at three heights). All chambers were covered with aluminium foil to avoid photosynthetic refixation by bark. This simplified the modelling of respiration for estimation of wood CO2 efflux at the stand scale. The protocols for measurements at 1·3 m and in the crown are given in Table 2. The diurnal temperature variation was related to that of CO2 efflux (R expressed either as wood volume or wood area) by an exponential equation expressed in terms of Q10 (the change in rate for a 10 °C change in temperature):

Table 2.  Characteristics of stem CO2 efflux measurements at 1·3 m and in the crown
 Characteristics of CO2 efflux measurements
 at 1·3 min the crown
Chamberacrylic (5 sizes)glass (3 sizes)
Stem diameter class (cm)12–25, 25–32, 32–36, 36–440·25–1·3, 1·3–2·5
Measurement period2–3 days per monthcontinuous
 (March 1997–February 1998)(May 1997–November 1997)
Time-step (h)1·5–2·01·5
IRGA systemCIRAS-1, PP-Systems, Stotfold, UK6262, Li-Cor, Lincoln, NB, USA
Operating modemanual closed systemautomatic open system
  • image(eqn 1)

where R15 is the respiration at 15 °C (base temperature) and T is the air temperature (°C).

To obtain estimates throughout 1997, we used half-hourly air temperatures and monthly values of Q10 and R15. Because global estimates of respiration are often derived from measurements at only one height, we considered two integration methods: (1) the Q10 and R15 coefficients were taken as the values measured on stems at 1·3 m and applied to both stems and branches, or (2) values measured at 1·3 m were applied to stems and those measured in the crown were applied to branches. Both volume- and area-based estimates were made. To estimate the wood volume and area of the stand, we used tree density, tree diameter at 1·3 m and allometric relationships (Schnock 1983; Bartelink 1997). This gave ratios of branch volume to total volume of 12% and branch area to total area of 70–85% depending on tree diameter.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Net ecosystem co2 exchange

A typical time-course of net ecosystem exchange is shown in Fig. 1. Maximum CO2 uptake over the forest was typically c. −20 µmol CO2 m−2 s−1 during full leaf expansion, but some peaks at −30 µmol CO2 m−2 s−1 occurred. (The convention of a negative flux for uptake and a positive flux for release is used here). We also found a curvilinear relationship between net ecosystem production (NEP, g C m−2 s−1) and incident PAR during the day and faster C assimilation under diffuse radiation (e.g. cloudy days) than in direct sunshine. Increases in vapour pressure deficit decreased NEP by closing stomata. During this period, ecosystem respiration was ≈ 5–15 µmol CO2 m−2 s−1 at night, but some peaks exceeded 20 µmol CO2 m−2 s−1.


Figure 1. (a) Time–course of net CO2 flux above the Beech stand (Euroflux site FR02, Hesse forest, France) from 4 to 28 June 1997. Data were averaged half-hourly and corrected for CO2 storage in the 0–18 m air layer. (b) Relationship between CO2 flux and incident PAR for the same period.

Download figure to PowerPoint

The time-course of daily NEP is shown in Fig. 2. A sharp increase in CO2 fixation (i.e. NEP more negative) occurred in spring 1997, corresponding to rapid leaf expansion and increased photosynthesis. Within ≈ 33 days, NEP changed from 0 (DOY 122) to c. −700 mmol CO2 m−2 day−1 (i.e. −8·4 g C m−2 day−1). During late summer, ecosystem C fixation decreased slowly (NEP less negative) because of a seasonal decrease in incoming PAR and drying of the soil.


Figure 2. Top: seasonal course of stem plus leaf area index (SAI), calculated from global radiation intercepted by the canopy. Middle: daily net CO2 exchange above the Beech stand over 2 years. Gaps caused by missing or unreliable data were filled using functions relating NEP to PAR (for daylight hours during the leafed phase) or to soil temperature (otherwise). Negative values indicate C fixation by the forest ecosystem, and positive values indicate loss of C. Bottom: relative extractable water in the soil (REW) calculated from neutron probe measurements in the 0–60 cm soil layer. REW = 0·4 corresponds to the critical soil water below which stomatal closure occurs.

Download figure to PowerPoint

Net canopy co2 exchange

Because of technical problems, one of the branch bags at the top of the canopy produced faulty measurements in July and August 1997; otherwise, it provided results similar to those from the other upper bag.

Under saturating light, C assimilation (per unit leaf area) for branches at the top of the canopy was close to that of single leaves, c. −9 µmol CO2 m−2 s−1. This rate decreased steadily from the end of May. The maximum C assimilation of the shaded branch was about half that of sunlit branches. Mean daily assimilation of top branches, per unit leaf area, was c. −0·3 mol CO2 m−2 day−1 from May to July and decreased thereafter (Fig. 3). That of the shaded branch was more stable, c. −0·1 mol CO2 m−2 day−1.


Figure 3. Daily C balance of three branches over the 1997 growing season. Measurements were made using branch bags. Bags 1 and 2 were at the top of the canopy. Bag 3 was in a low layer. Rates are expressed per unit leaf area.

Download figure to PowerPoint

Nocturnal respiration of branches decreased from 2·5 to 1 nmol CO2 gDM−1 s−1 between May and September (0·8–0·3 µmol CO2 m−2 s−1, per unit leaf area).

From these data, we can estimate the net C assimilation of the canopy (An) for the whole growing season (considering leaves and fine branches only). This estimate, which depends on the distribution of sunlit and shaded leaves, was performed over 191 day (from DOY 117 to 307 in 1997) using LAI = 5·6. Assuming that An decreased linearly from the top to the bottom of the canopy, we obtain a net C exchange of −1930 g C m−2 (ground area). Assuming also that 2/3 of the leaf area was shaded and the rest sunlit, and weighting data from the upper and lower bags accordingly, a smaller estimate of An is obtained: −1640 g C m−2. These large values are explained by the photosynthetic rate of the branch in the lower bag being relatively fast despite the small amount of radiation incident on it: 5–10% of incident PAR.

Using these data and respiration rates measured on small branches at the same site, we estimated the respiration of the leaves alone. First, leaf respiration was computed as total bag respiration minus wood respiration for periods of no photosynthesis (0000 to 0530 h local time). Then, Q10 and R15 values were computed and applied to branch bag data obtained during the day. Over the entire leafed period (from DOY 120 to 305), total leaf respiration was ≈ 240 g C m−2 (60 g C m−2 at night).

Ecosystem respiration

We assumed that the respiration of the ecosystem, Reco, could be estimated (as RE, ecosystem respiration estimated from eddy covariance; g C m−2 s−1) by EC over the whole day when trees had no leaves, or at night when leaves were present. A problem is that this measurement is affected by CO2 storage in the air layer from the soil surface to the height at which EC is measured (18 m). To determine the effect of this, we plotted RE against friction velocity, u* (Aubinet et al. 2000), for a period when temperature and soil water content were stable, i.e. steady respiration rates were likely. RE increased with u* (Fig. 4), indicating that a fraction of CO2 flux was not measured by EC at slow u*. The CO2 content of the air between 0 and 18 m, and its temporal variation, were calculated. Figure 5 shows that CO2 accumulation occurred frequently at night when wind speed was low, while CO2 release was observed during the day as wind speed increased. When wind speed remained > 0·5 m s−1 during the night (indicated by an asterisk in Fig. 5), there was little day-to-night variation in CO2 content. Diurnal variation in CO2 content was much lower during the unleafed period. This was because the air was cooler than during the spring and summer, leading to slower respiration rates, and there was more ventilation and air mixing within the stand.


Figure 4. Nocturnal CO2 flux above the Beech stand as a function of friction velocity (u*) during a period of stable temperature (14–17 °C) and soil water content (REW > 0·7). The curve is the fitted function: CO2 flux = 5·89log(u*) + 9·12 (r2 = 0·31).

Download figure to PowerPoint


Figure 5. Time–course of CO2 storage in the 0–18 m layer in the Beech stand calculated from [CO2] at five heights (0·2, 0·7, 2·0, 8·0 and 18·0 m) in the stand over 32-day periods in summer and winter. CO2 storage is the difference between instantaneous CO2 content minus minimum CO2 content measured during the period. The enlarged section of the graph shows the simultaneous variation of CO2 storage and of u* over a 3-day period. Asterisks correspond to u* > 0·5 m s−1 at night.

Download figure to PowerPoint

Half-hourly CO2 flux data, for the periods when CO2 profiles were measured, were corrected (RE*) by adding the variation in CO2 content within the 0–18 m air layer to the measured CO2 flux above the stand. RE* was not correlated significantly with u*. For the whole of 1997, failure to account for CO2 storage/release would have underestimated RE by ≈ 12%.

RE* during the leafed period was extrapolated to the whole day by assuming that it depended only on temperature, i.e. we assumed that the relationship between respiration and temperature was the same during the day and the night. This assumption is questionable for leaves because mitochondrial respiration is depressed by 40–90% during the day (Brooks & Farquhar 1985). Multiple sources of C efflux over a stand make it impossible to relate RE* to a single temperature measurement: soil plus root respiration would depend closely on soil temperature, while above-ground respiration would probably be related to air or trunk temperature. We tested the dependence of RE* on soil, trunk and air temperatures: the best fit (see Fig. 6) was obtained with soil temperature (°C) at −10 cm (T−10):


Figure 6. Relationship of ecosystem respiration (RE*), corrected for CO2 storage in the 0–18 m air layer (see text), and soil respiration (Rs) to soil temperature at −10 cm (T−10). RE* data are averaged over a night during the leafed phase or over a whole day during the unleafed phase.

Download figure to PowerPoint

  • image(eqn 2)

The temporal variation of RE* is shown in Fig. 7. To smooth the large day-to-day scattering, which might be due to measurement errors, to the effect of rain showers that often occurred at night, or to an unknown cause, a moving 10-day average was calculated. After budburst (DOY 120), RE* increased rapidly (Fig. 7) up to a maximum rate of ≈ 7 g C m−2 day−1, at around DOY 170 (19 June 1997), corresponding to the period of maximum tree growth. Thereafter, RE* remained between 4 and 7 g C m−2 day−1 until DOY 250 (7 September 1997). Several periods of slower rates were noted, some linked to a decrease in soil temperature. Soil respiration on different dates and simulated above-ground respiration (Rag) are also presented in Fig. 7. They show roughly the same pattern, and are closely related to soil temperature. There was little respiration during several warm periods, for example on DOY 210–230 and DOY 260–270; this decrease could have been caused by soil drying. Nevertheless, in contrast to the situation for soil respiration (see below), accounting for soil water content did not improve the fit of RE* to soil temperature.


Figure 7. Time–courses of net ecosystem C exchange (NEP), corrected ecosystem respiration (RE*, see text), above-ground respiration (Rag) and soil respiration (Rs) during 1997. RE* is a moving 10-day average. Rag is the simulation of above-ground wood respiration derived using volume-based Q10 and R15 obtained at 1·3 m (for stems) or in the crown (for branches). Also shown is soil temperature at a depth of −10 cm (T−10).

Download figure to PowerPoint

Soil respiration

The frequency of soil respiration measurements allowed the 95% confidence intervals to be < 10% of the daily means, despite large spatial variability. In August 1997, the minimum was 0·9 µmol m−2 s−1 while the maximum was 8·6 µmol m−2 s−1. Daily means of soil CO2 efflux ranged from 0·4 µmol m−2 s−1 in winter to 4·1 µmol m−2 s−1 in August. This large seasonal variation could be explained only partly by seasonal changes in soil temperature (Fig. 6; r2 = 0·79). In summer, strong reductions in soil CO2 efflux were associated with soil drying, as observed in September 1997 (see Fig. 7). Soil respiration was best described with an empirical model including θv, the volumetric soil water content, and T−10, the soil temperature (°C), each measured at −10 cm:

  • image(eqn 3)

Using this equation, the annual soil respiration was calculated as 575 g C m−2 year−1 for 1996 and 663 g C m−2 year−1 for 1997. In 1997, annual soil respiration on the trenched plots was 500 g C m−2 year−1. Including the decomposition of roots killed by trenching and the difference in soil water content between the main and trenched plots (Epron et al. 1999b), the heterotrophic component of soil C efflux was calculated as ≈ 260 g C m−2 year−1, representing 40% of the total soil C efflux, with root respiration accounting for the rest (i.e. 400 g C m−2 year−1).

Above-ground wood respiration

For stem CO2 efflux at 1·3 m, Q10 was always stable (data not shown). The annual mean was 1·7 (n = 105, SD = 0·45). Means were usually between 1·6 and 1·8. In contrast to Q10, R15 showed large monthly variations. The minimum occurred in February (10·4 µmol CO2 m−3 s−1 ≡ 0·2 µmol m−2 s−1) and the maximum in July (132 µmol CO2 m−3 s−1 ≡ 3 µmol CO2 m−2 s−1). For branch CO2 efflux, Q10 was usually 2–3. Similarly to the results for stems, at 1·3 m branch R15 values in summer exceeded those in winter. Maxima (on a volume basis) were measured on the narrowest branches (2400 µmol CO2 m−3 s−1). On an area basis, maximum rates occurred in the thickest branches (3 µmol CO2 m−2 s−1). At the stand level, total annual estimates varied according to the calculation method. The use of volume-based Q10 and R15 calculated from data measured in the crown led to larger annual estimates than those obtained using only volume-based parameters calculated from data collected at 1·3 m. Annual totals estimated from area-based Q10 and R15 were much larger than those estimated using volume-based parameters, except when using parameters calculated from data for the smallest diameter of branch (level 3) (Table 3). A seasonal simulation using volume-based parameters (at 1·3 m in the crown at level 2) is shown in Fig. 7.

Table 3.  Annual totals of respiration by above-ground wood (Rag). Estimates were calculated using monthly volume- or area-based Q10 and R15. The Q10 and R15 used were only those from stems at 1·3 m or from stems in the crown (at three levels: 1, 2, 3, at which stem diameters were 2·5, 1·3 and 0·25 cm, respectively)
  Rag (g C m−2 soil)
Q10 + R15 derived fromQ10 + R15 applied toVolume-based Q10 + R15Area-based Q10 + R15
stems at 1·3 mstems + branches235 810
stems at 1·3 mstems202 202
stems at 1·3 m + level 1stems + branches3781073
stems at 1·3 m + level 2stems + branches311 549
stems at 1·3 m + level 3stems + branches894 554

Synthesis–carbon balance from co2 fluxes vs stand growth

The terms for the annual C balance are shown in Table 4 for the 2 years of observations. In 1996, measurements started at the beginning of May; our estimates are given for the period from 1 May 1996 to 30 April 1997.

Table 4.  Annual C fluxes over and within the Beech stand. Values are in g C m−2 year−1. Abbreviations are given in the text
  Period of measurement
VariableMethodMay 1996–April 1997January 1997–December 1997
  • *NPP = net primary productivity (g C m−2 s−1).

  • †In 1996, ecosystem respiration was not corrected for changes in CO2 content in the 0–18 m air layer.

  • Includes below-ground biomass.

NEPEC −218 −257
NPP*NEP-Rheterotrophic −448 −522
GPPLeaf measurements−1178−1298
GPPSum of C fluxes (see text) −1409
RecoEC  793  988
RsChambers  575  663
RheterotrophicTrenched plots  230  265
RrootsRs − Rheterotrophic  345  398
Rag (living wood)Chambers   311–378
RagReco − Rs   325
Biomass increment   456  498

Gross primary productivity (GPP = NEP + Reco; g C m−2 s−1) calculated from scaled leaf measurements was 1298 g C m−2 year−1 in 1997, i.e. ≈ 4% more than the GPP obtained from EC, which was 1245 g C m−2 year−1. Given the underlying assumptions (especially for respiration estimates) and the different sources of errors (leaf and branch sampling, scaling errors, etc.), these estimates are surprisingly close.

Dendrometric measurements allowed the amount of C stored yearly in aerial and below-ground biomass of trees to be calculated. In 1997, this was 498 g C m2 year−1 for the total stand biomass increment (stems, branches and major roots). From these data, and assuming that the C stock did not vary over the year, we computed another estimate of annual GPP (in 1997) by adding: (1) net annual C gain in wood (= 498 g C m−2 year−1); (2) the amount of C corresponding to the annual leaf (CL) and fine root (CR) production (= 131 and 57 g C m−2 year−1, respectively); (3) respiration losses by aerial plus below-ground biomass. The last term was assumed to equal Reco (= 988 g C m−2 year−1) minus C released from SOM decomposition (40% of 663 g C m−2 year−1 = 265 g C m−2 year−1, i.e. 723 g C m−2 year−1). Heterotrophic respiration exceeds the sum of litter production plus fine roots (turnover) = 188 g C m−2 year−1, which may indicate that inputs to SOM from coarse woody detritus cannot be ignored (Raich & Nadelhoffer 1989). For 1997, this calculation gave:

  • image

This estimate of GPP exceeds the 1245 g C m−2 year−1 derived from EC (Table 4). The former may have been overestimated because we did not take into account C losses from the decomposition of coarse dead wood.

However, there is a large discrepancy between the annual C increment of the stand (498 g C m−2 year−1) and NEP (−257 g C m−2 year−1). In addition to multiple measurement errors, which are difficult to quantify, at least part of this difference (241 g C m−2 year−1) could be attributed to dead wood decomposition. Another possible explanation is litter accumulation in the soil. This was not observed at Hesse: the litter from the previous year had disappeared completely before leaf fall. Nevertheless, some C could progressively accumulate in the upper soil layers, but we did not make any measurements to determine whether this was the case.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Peak NEP over our Beech stand was −20 to −30 µmol CO2 m−2 s−1. This is similar to the fluxes measured by EC in various broad-leaved stands over a wide range of latitudes and tree species: −18 to −25 µmol CO2 m−2 s−1 in a temperate mixed forest at Harvard (Wofsy et al. 1993; Goulden et al. 1996), −20 to −30 µmol CO2 m−2 s−1 in a mountain Beech forest (Valentini et al. 1996), −25 µmol CO2 m−2 s−1 in an Aspen (Populus tremuloides) forest in the boreal zone (Black et al. 1996), −20 µmol CO2 m−2 s−1 in a tropical rainforest (Grace et al. 1995), and −23 µmol CO2 m−2 s−1 in a mixed deciduous forest in Canada (Lee et al. 1996).

Nocturnal CO2 fluxes are more variable among sites. Several authors have reported respiration fluxes similar to those observed at Hesse: 8–10 µmol CO2 m−2 s−1 in the Harvard forest (Goulden et al. 1996), 9–11 µmol CO2 m−2 s−1 in an Aspen forest (Black et al. 1996), and 6–9 µmol CO2 m−2 s−1 in a Mediterranean Quercus ilex forest (Piñol, Alcaniz & Roda 1995). Others found smaller nocturnal CO2 fluxes: 2–6 µmol CO2 m−2 s−1 in a mountain Beech stand (Valentini et al. 1996), 2–7 µmol CO2 m−2 s−1 in a Nothofagus forest (Hollinger et al. 1994), and 4–5 µmol CO2 m−2 s−1 in a Brazilian savannah (Miranda et al. 1997).

Several authors calibrated a function of soil temperature to their measurements. We applied their functions for simulating Reco at three temperatures: 10, 15 and 20 °C (Table 5). Our estimates of Reco are much larger than those obtained for an Italian Beech stand (Valentini et al. 1996), whatever the temperature; above 15 °C, Reco is also larger than that obtained for two North American mixed forests (Goulden et al. 1996; Greco & Baldocchi 1996). However, estimates made by Black et al. (1996) in a boreal Aspen forest were much larger than ours.

Table 5.  Estimated values of Reco (µmol m−2 s−1) at three soil temperatures
  Soil temperature (°C)
ReferenceForest type101520
Greco & Baldocchi (1996)Mixed deciduous2·483·16 4·02
Goulden et al. (1996)Mixed deciduous2·363·40 4·90
Black et al. (1996)Aspen boreal3·417·9118·3
Valentini et al. (1996)Beech1·652·43 3·58
This studyBeech1·973·807·33

In Table 6, our estimates of annual NEP, GPP and Reco are compared with other estimates (obtained using the same methods). The largest NEP and GPP were found for productive coniferous plantations of Douglas Fir (Vermetten et al. 1994) and Pinus radiata (Ryan et al. 1996). Our estimates of GPP and NEP are similar to those reported for other broad-leaved forests: −1000 to −1300 g C m−2 year−1 and −100 to −500 g C m−2 year−1 for GPP and NEP, respectively. Our estimates of Reco are greater than those obtained elsewhere.

Table 6.  Estimates (g C m−2 year−1) of GPP, NEP and Reco in various forest ecosystems
  • *

    Overstorey + understorey.

  • Projected LAI.

Baldocchi et al. (1997)BOREASPinus banksiana1·9–2·2   −47 
Black et al. (1996)BOREASPopulus tremuloides5·1*−1020 −130890
Cannell et al. (1996)IrelandFagus, coniferous   740
Frolking et al. (1996)BOREASPicea mariana    −51 
Goulden et al. (1996)Harvard, USAMixed deciduous   −210 
Grace et al. (1995)AmazoniaTropical rainforest   −102 
Greco & Baldocchi (1996)Oak Ridge, USAQuercus, Acer4·9−1085 −525560
Ryan et al. (1996)Camberra, AustraliaPinus radiata3–4·6−2440 to −3440  
Valentini et al. (1996)ItalyFagus sylvatica4·7−1016 −472544
Vermetten et al. (1994)NetherlandsPseudotsuga menziesii8–10 −1045 to −1543 
Wofsy et al. (1993)Harvard, USAMixed deciduous −1110 −370 
This study (1996)Eastern FranceFagus sylvatica5·7−1011 −218793
This study (1997)Eastern FranceFagus sylvatica5·6−1245 −257988

Annual NEP varies from one site to another, even between those with similar climates. This variation is explained at least partly by large differences in stand growth and soil respiration. For instance, in temperate ecosystems, annual NEP ranges between −210 g C m−2 year−1 (Harvard forest, Goulden et al. 1996) and −1300 g C m−2 year−1 (Douglas Fir forest, Vermetten et al. 1994). Our site had a comparatively small NEP of −257 g C m−2 year−1 (even smaller in 1996, a drier year), probably because of its large annual Reco of c. 1000 g C m−2 year−1. Ecosystem respiration is a crucial term in the C balance, varying between 50 and 90% of gross assimilation (80% in our study). In tropical forests, large values of Reco compared with GPP have also been reported (Grace et al. 1995); those, too, resulted in a small NEP, of −102 g C m−2 year−1.

Total soil respiration is often considered an important component of the C balance of an ecosystem. Our estimates of this flux were 600–700 g C m−2 year−1, similar to other estimates: 800 g C m−2 year−1 (Hanson et al. 1993); 485 g C m−2 year−1 (Simmons et al. 1996); 720 g C m−2 year−1 (Davidson, Beck & Boone 1998). At Hesse, soil respiration accounted for ≈ 70% of Reco and 55% of GPP. This agrees with data from other sites indicating that soil respiration represents 60–80% of total ecosystem respiration (Wofsy et al. 1993; Goulden et al. 1996; Davidson, Beck & Boone 1998). Reco reflects the respiration associated with at least three major components: soil (heterotrophic), root, and above-ground biomass. Root and heterotrophic respiration cannot be separated easily, because they are evaluated from chamber measurements and scaled to the stand. We estimated that root respiration was the major component (60%) of soil C efflux at Hesse. A similar estimation was reported by Ewel, Cropper & Gholz (1987) for a Pinus elliottii plantation.

As shown in Table 3, there are difficulties in scaling Rag to the stand from chamber measurements. These difficulties concern: (1) choosing the unit that best expresses respiration; (2) properly expressing the percentage of branches relative to stems, and (3) integrating the respiration variability along the stem and between stems and branches. Because our measurements of volume- and area-based stem R15 were not constant between breast height and crown level, we estimated Rag using measurements obtained at both levels. One independent estimation of growth respiration was made using stem plus branch growth measured on the site (J.-M. Ottorini & N. Le Goff, data not shown), assuming a growth efficiency [defined as growth/(growth + growth respiration), with terms usually expressed in g C] of 0·65. We obtained Rag = 202 g C m−2 year−1. As maintenance respiration generally represents ≈ 40% of total respiration, we assumed a total respiration of 420 g C m−2 year−1. Another estimate of Rag calculated as the difference between Reco (EC) and Rs (soil chambers) gave Rag = 325 g C m−2 year−1. These two estimates provide a reasonable order of magnitude estimate of Rag. Consequently, our annual total of 235 g C m−2 year−1 obtained using only volume-based parameters from stems at 1·3 m probably underestimated the true value. Area-based parameters obviously overestimate Rag. Our best estimates were probably those obtained using volume-based parameters from stems at 1·3 m and the two largest branch diameters in the crown (levels 1 and 2). These estimates fall within those found in the literature. Edwards & Hanson (1996) estimated Rag in a mixed Oak and Red Maple stand, growing under warmer and wetter conditions than those in the present study, to be 149–204 g C m−2 year−1. For conifers in different climates, Ryan et al. (1995) found annual totals of 52–162 g C m−2 year−1. Larger values (910 or 1314 g C m−2 year−1; Whitmore 1984; Müller & Nielson 1965; cited in Ryan et al. 1994) were found in rain forests. Because we did not take bark assimilation into account, we probably overestimated the wood CO2 efflux. Accurate estimations of this flux are now required.

C sequestration in forest stands can also be estimated from forest inventories. We calculated an annual C immobilization of ≈ 400–500 g C m−2 year−1 in above- and below-ground biomass. In Ireland, Cannell et al. (1996) found a mean immobilization of 190 g C m−2 year−1 in Beech stands of various ages and at several sites. In a mountain Beech forest, Valentini et al. (1996) found that 238 g C m−2 year−1 was stored in the above-ground biomass.

GPP is related to the amount of solar radiation intercepted by the canopy. Ruimy, Dedieu & Saugier (1996) showed that, as a first approximation, GPP was equal to the product of intercepted PAR and a biological conversion efficiency of 0·02 mol C mol−1 photons ≡ 1·1 g C MJ−1 PAR. At Hesse, incident global radiation (Rg) from DOY 125 to 300 in 1997 was 2752 MJ m−2 ≡ 1376 MJ PAR m−2 assuming PAR/Rg = 0·5. If 95% of incoming PAR was absorbed, this gives a GPP of 1376 × 1·1 × 0·95 = 1438 g C m−2 year−1, a larger estimate than that derived from EC (1245 g C m−2 year−1) or leaf measurements (1298 g C m−2 year−1; Table 4), but smaller than that calculated from branch bag measurements over the growing season (net assimilation of the canopy over 191 days: 1640–1930 g C m−2). Thus, the estimate of GPP remains relatively uncertain.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported by: (1) the EU programme Euroflux (ENV4-CT95-0078) co-ordinated by R. Valentini; (2) the French National Forest Service (ONF); (3) a multidisciplinary programme of the CNRS: ‘Environnement, Vie et Sociétés’. PhD studentships were granted by the governments of France (Ministry of Research) and Luxembourg. The authors are very grateful for the valuable help provided by R. Canta, B. Clerc, P. Garnier, L. Garros, D. Imbert, Y. Lefrèvre, G. Thiriet, F. Willm, T. Bienaimé, S. Daviller, L. Le Goff and D. Pierre.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aubinet, M., Grelle, A., Ibrom, A., Rannik, Ü., Moncrieff, J., Foken, T., Kowalski, A.S., Martin, P.H., Berbigier, P., Bernhofer, C.H., Clement, R., Elbers, J., Granier, A., Grünwald, T., Morgenstern, K., Pilegaard, K., Rebmann, C., Snijders, W., Valentini, R., Vesala, T. (2000) Estimates of the annual net carbon and water exchange of forests: the EUROFLUX methodology. Advances in Ecological Research 30, 113 175.
  • Baldocchi, D., Valentini, R., Running, S., Oechel, W., Dahlman, R. (1996) Strategies for measuring and modelling carbon dioxide and water vapour fluxes over terrestrial ecosystems. Global Change Biology 2, 159 168.
  • Baldocchi, D.D., Vogel, C.A., Hall, B. (1997) Seasonal variation of carbon dioxide exchange rates above and below a boreal jack pine forest. Agricultural and Forest Meteorology 83, 147 170.
  • Bartelink, H.H. (1997) Allometric relationships for biomass and leaf area of beech (Fagus sylvatica L.). Annales des Sciences Forestières 54, 39 50.
  • Black, T.A., Denhartog, G., Neumann, H.H., Blanken, P.D., Yang, P.C., Russell, C., Nesic, Z., Lee, X., Chen, S.G., Staebler, R., Novak, M.D. (1996) Annual cycles of water vapour and carbon dioxide fluxes in and above a boreal aspen forest. Global Change Biology 2, 219 229.
  • Brooks, A. & Farquhar, G.D. (1985) Effect of temperature on the CO2/O2 specificity of ribulose-1,5-biphosphate carboxylase/oxygenase and the rate of respiration in the light. Planta 165, 397 406.
  • Cannell, M.G.R., Cruickshank, M.M., Mobbs, D.C. (1996) Carbon storage and sequestration in the forests of Northern Ireland. Forestry 69, 155 165.
  • Davidson, E.A., Beck, E., Boone, R.D. (1998) Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biology 4, 217 227.
  • Dixon, R.K., Brown, S., Houghton, R.A., Solomon, A.M., Trexler, M.C., Wisniewski, J. (1994) Carbon pools and flux of global forest ecosystems. Science 263, 185 190.
  • Dufrêne, E., Pontailler, J.Y., Saugier, B. (1993) A branch bag technique for simultaneous CO2 enrichment and assimilation measurement on beech (Fagus sylvatica). Plant, Cell and Environment 16, 1131 1138.
  • Edwards, N.T. & Hanson, P.J. (1996) Stem respiration in a closed-canopy upland oak forest. Tree Physiology 16, 433 439.
  • Epron, D., Farque, L., Lucot, E., Badot, P.M. (1999a) Soil CO2 efflux in a beech forest: dependence on soil temperature and soil water content. Annales des Sciences Forestières 56, 221 226.
  • Epron, D., Farque, L., Lucot, E., Badot, P.M. (1999b) Soil CO2 efflux in a beech forest: the contribution of root respiration. Annales des Sciences Forestières 56, 289 295.
  • Ewel, K.C., Cropper, W.P., Gholz, H.L. (1987) Soil CO2 evolution in Florida slash pine plantations. II. Importance of root respiration. Canadian Journal of Forest Research 17, 330 333.
  • Frolking, S., Goulden, M.L., Wofsy, S.C., Fan, S.M., Sutton, D.J., Munger, J.W., Bazzaz, A.M., Daube, B.C., Crill, P.M., Aber, J.D., Band, L.E., Wang, X., Savage, K., Moore, T., Harriss, R.C. (1996) Modelling temporal variability in the carbon balance of a spruce/moss boreal forest. Global Change Biology 2, 343 366.
  • Gash, J.H.C. (1986) A note on estimating the effect of a limited fetch on micrometeorological evaporation measurements. Boundary Layer Meteorology 35, 409 413.
  • Goulden, M.L., Munger, J.W., Fan, S.M., Daube, B.C., Wofsy, S.C. (1996) Measurements of carbon sequestration by long-term eddy covariance: methods and a critical evaluation of accuracy. Global Change Biology 2, 162 182.
  • Grace, J., Lloyd, J., McIntyre, J., Miranda, A., Meir, P., Miranda, H., Moncrieff, J., Massheder, J., Wright, I., Gash, J. (1995) Fluxes of carbon dioxide and water vapour over an undisturbed tropical forest in south-west Amazonia. Global Change Biology 1, 1 12.
  • Granier, A., Biron, P., Lemoine, D. (2000) Water balance, transpiration and canopy conductance in two beech stands. Agricultural and Forest Meteorology 100, 291 308.
  • Greco, S. & Baldocchi, D.D. (1996) Seasonal variations of CO2 and water vapour exchange rates over a temperate deciduous forest. Global Change Biology 2, 183 197.
  • Hanson, P.J., Wullschleger, S.D., Bohlman, S.A., Todd, D.E. (1993) Seasonal and topographic patterns of forest floor CO2 efflux from an upland oak forest. Tree Physiology 13, 1 15.
  • Hollinger, D.Y., Kelliher, F.M., Byers, J.N., Hunt, J.E., McSeveny, T.M., Weir, P.L. (1994) Carbon dioxide exchange between an undisturbed old-growth temperate forest and the atmosphere. Ecology 75, 134 150.
  • Houghton, J.T., Meira Filho, L.G., Callander, B.A., Harris, N., Kattenberg, A., Maskell, K. (1996) Climate Change 1995: The Science of Climate Change. Cambridge University Press, Cambridge.
  • Jarvis, P.G. (1995) The role of temperate trees and forests in CO2 fixation. Vegetatio 121, 157 174.
  • Lebaube, S., Le Goff, N., Ottorini, J.M., Granier, A. (1999) Carbon balance and tree growth in a Fagus sylvatica strand. Annales des Sciences Forestières 57, 49 61.
  • Lee, X.H., Black, T.A., Denhartog, G., Neumann, H.H., Nesic, Z., Olejnik, J. (1996) Carbon dioxide exchange and nocturnal processes over a mixed deciduous forest. Agricultural and Forest Meteorology 81, 13 29.
  • Le Goff, N. (1981) Le Hêtre. 6.2. Tables de Production. INRA, Paris, pp. 326 338.
  • Le Goff, N. & Ottorini, J.-M. (2000) Root biomass and biomass increment in a beech (Fagus silvatica L.) stand in north-east of France. Annales des Sciences Forestières, in press.
  • Leuning, R. & Moncrieff, J. (1990) Eddy-covariance CO2 flux measurements using open-path and closed-path CO2 analysers. Corrections for analyser water vapour sensitivity and damping of fluctuations in air sampling tubes. Boundary Layer Meteorology 53, 63 76.
  • Miranda, A.C., Miranda, H.S., Lloyd, J., Grace, J., Francey, R.J., McIntyre, J.A., Meir, P., Riggan, P., Lockwood, R., Brass, J. (1997) Fluxes of carbon, water and energy over Brazilian cerrado: an analysis using eddy covariance and stable isotopes. Plant, Cell and Environment 20, 315 328.
  • Müller, D. & Nielson, J. (1965) Production brute, pertes par respiration et production nette dans la forêt ombrophile tropicale. Forstlige Forsøgsvaesen i Danmark 29, 69 160.
  • Nabuurs, G.J. & Mohren, G.M.J. (1995) Modelling analysis of potential carbon sequestration in selected forest types. Canadian Journal of Forest Research 25, 1157 1172.
  • Norman, J.M., Garcia, R., Verma, S.B. (1992) Soil surface CO2 fluxes and the carbon budget of a grassland. Journal of Geophysical Research 97, 18845 18853.
  • Persson, H.A. (1983) The distribution and productivity of fine roots in boreal forests. Plant and Soil 71, 87 101.
  • Piñol, J., Alcaniz, J.M., Roda, F. (1995) Carbon dioxide efflux and pCO2 in soils of three Quercus ilex montane forests. Biogeochemistry 30, 191 215.
  • Raich, J.W. & Nadelhoffer, K.J. (1989) Belowground carbon allocation in forest ecosystems: global trends. Ecology 70, 1346 1354.
  • Ruimy, A., Dedieu, G., Saugier, B. (1996) TURC: a diagnostic model of continental gross primary productivity and net primary productivity. Global Biogeochemical Cycles 10, 269 285.
  • Ryan, M.G., Hubbard, R.M., Clark, D.A., Sanford, R.L. (1994) Woody-tissue respiration for Simarouba amara and Minquartia guianensis, two tropical wet forest trees with different growth habits. Oecologia 100, 213 220.
  • Ryan, M.G., Gower, S.T., Hubbard, R.M., Waring, R.H., Gholz, H.L., Cropper, W.P., Running, S.W. (1995) Woody tissue maintenance respiration of four conifers in contrasting climates. Oecologia 101, 133 140.
  • Ryan, M.G., Hubbard, R.M., Pongracic, S., Raison, R.J., McMurtrie, R.E. (1996) Foliage, fine-root, woody-tissue and stand respiration in Pinus radiata in relation to nitrogen status. Tree Physiology 16, 333 343.
  • Santa Regina, I., Tarazona, T., Calvo, R. (1997) Aboveground biomass in a beech forest and a Scots pine plantation in the Sierra de la Demanda area of northern Spain. Annales des Sciences Forestières 54, 261 269.
  • Saugier, B., Granier, A., Pontailler, J.Y., Dufrêne, E., Baldocchi, D.D. (1997) Transpiration of a boreal pine forest measured by branch bag, sap flow and micrometeorological methods. Tree Physiology 17, 511 519.
  • Schnock, G. (1983) Volume, biomasses, surfaces d’échange et autres caractéristiques dendrométriques des tiges de charme (Carpinus betulus L.). Acta Oecologica Oecologia Applicata 4, 325 342.
  • Simmons, J.A., Fernandez, I.J., Briggs, R.D., Delaney, M.T. (1996) Forest floor carbon pools and fluxes along a regional climate gradient in Maine, USA. Forest Ecology and Management 84, 81 95.
  • Tenhunen, J.D., Valentini, R., Köstner, B., Zimmermann, R., Granier, A. (1998) Variation in forest gas exchange at landscape to continental scales. Annales des Sciences Forestières 55, 1 12.
  • Valentini, R., De Angelis, P., Matteucci, G., Monaco, R., Dore, S., Mugnozza, G.E.S. (1996) Seasonal net carbon dioxide exchange of a beech forest with the atmosphere. Global Change Biology 2, 199 207.
  • Vermetten, A.W.M., Ganzeveld, L., Jeuken, A., Hofschreuder, P., Mohren, G.M.J. (1994) CO2 uptake by a stand of Douglas fir: Flux measurements compared with model calculations. Agricultural and Forest Meteorology 72, 57 80.
  • Whitmore, T.C. (1984) Tropical Rainforests of the Far East, 2nd Edn. Clarendon Press, Oxford.
  • Winjum, J.K. & Schroeder, P.E. (1997) Forest plantations of the world: their extent, ecological attributes, and carbon storage. Agricultural and Forest Meteorology 84, 153 167.
  • Wofsy, S.C., Goulden, M.L., Munger, J.W., Fan, S.M., Bakwin, P.S., Daube, B.C., Bassow, S.L., Bazzaz, F.A. (1993) Net exchange of CO2 in a mid-latitude forest. Science 260, 1314 1317.