Interactive effects of elevated CO2 and soil fertility on isoprene emissions from Quercus robur

The effects of global change on the emission rates of isoprene from plants are not clear. A factor that can influence the response of isoprene emission to elevated CO₂ concentrations is the availability of nutrients. Isoprene emission rate under standard conditions (leaf temperature: 30°C, photosynthetically active radiation (PAR): 1000 μmol photons m^?2 s^?1), photosynthesis, photosynthetic capacity, and leaf nitrogen (N) content were measured in Quercus robur grown in well‐ventilated greenhouses at ambient and elevated CO₂ (ambient plus 300 ppm) and two different soil fertilities. The results show that elevated CO₂ enhanced photosynthesis but leaf respiration rates were not affected by either the CO₂ or nutrient treatments. Isoprene emission rates and photosynthetic capacity were found to decrease with elevated CO₂, but an increase in nutrient availability had the converse effect. Leaf N content was significantly greater with increased nutrient availability, but unaffected by CO₂. Isoprene emission rates measured under these conditions were strongly correlated with photosynthetic capacity across the range of different treatments. This suggests that the effects of CO₂ and nutrient levels on allocation of carbon to isoprene production and emission under near‐saturating light largely depend on the effects on photosynthetic electron transport capacity.     

Studies of the relationship between isoprene emission rate and CO2 or photon-flux density using a real-time isoprene analyser

Abstract. Studies of the isoprene emission rate in response to changes in photon-flux density and CO₂ partial pressure were conducted using a recently developed on-line isoprene analyser combined with a gas exchange system and chlorophyll fluorometer. Upon darkening, the isoprene emission rate from leaves of aspen ( Populus tremuloides Michaux.) began to decline immediately, demonstrating that the internal pool of isoprene, or its precursors, is small and that the instantaneous emission rate is tightly coupled to the rate of synthesis. A post-illumination burst of isoprene was observed within 5 min after darkening and lasted for 15–20 min in four isoprene-emitting species that were examined. In leaves of eucalyptus ( Eucalyptus globulus Labill.), the magnitude of the post-illumination burst was dependent on the photon-flux density that existed before darkening, but not on ambient CO₂ partial pressure. The dependence of the post-illumination burst on photon-flux density paralleled that for the steady-state rate of isoprene emission. A step-wise increase in intercellular CO₂ partial pressure from 24.5 to 60 Pa resulted in an immediate decrease in isoprene emission rate and non-photochemical fluorescence quenching, but an increase in CO₂ assimilation rate. Given the several recent studies that link isoprene emission to chloroplastic processes, the results of this study indicate that the linkage is not dependent on the rate of CO₂ flux through the reductive pentose phosphate pathway, but rather on more complex relationships involving metabolites not appreciably influenced by CO₂ partial pressure.     

An apparatus for pulse and pulse-chase experiments with 14CO2 on attached leaves with known, steady-state rates of photosynthesis

Abstract. An apparatus is described to carry out pulse and pulse-chase experiments with ¹⁴CO₂ on intact, attached leaves with known, steady-state rates of photosynthesis under defined conditions of temperature, vapour pressure deficit and photon flux density. Data are presented which show that the pattern of distribution of ¹⁴C between compounds in extracts of such leaves is a true reflection of the pathways of photosynthetic carbon metabolism in the leaf during steady-state photosynthesis.     

Carbon economy in walnut seedlings during the acquisition of autotrophy studied by long-term labelling with 13CO2

A closed growth chamber was designed to study the acquisition of autotrophy by seedlings of walnut ( Juglans regia L. cv. Lara) in controlled conditions (22°C, 12-h photoperiod) during the first two months of growth. The chamber consisted of two airtight compartments, in which continuous gas exchange was measured on the aerial and subterranean parts of several batches of tree seedlings. Long-term labelling with ¹³CO₂ was used in the chamber to study the import, distribution, and respiratory losses of photoassimilates (autotrophic carbon) in relation to the partitioning and use of reserves of the maternal seed (heterotrophic carbon). The carbon economy of walnut seedlings was estimated by measurements of gas exchange, carbon content, and ¹³C/¹²C isotopic ratio in dry matter and respiratory CO₂. The seedlings were entirely heterotrophic for energy and structural growth during the first 21 days after sowing. From day 22, photosynthesis appeared. At day 29, autotrophic carbon accounted for 25% and 30% of respiration in the root and shoot respectively; these proportions increased to 45% and 65% at day 54. The autotrophic carbon was incorporated into the dry matter of the shoot from day 32 but only after day 40 into the dry matter of the taproot. From day 32, the total contribution of heterotrophic carbon decreased regularly, and until day 43, it was essentially used for root growth. Thereafter, the contribution of heterotrophic carbon was negligible, and at day 54 the walnut seedlings were entirely autotrophic.     

Acquisition and within-plant allocation of 13C and 15N in CO2-enriched Quercus robur plants

We assessed the effects of doubling atmospheric CO₂ concentration, [CO₂], on C and N allocation within pedunculate oak plants (Quercus robur L.) grown in containers under optimal water supply. A short-term dual ¹³CO₂ and ¹⁵NO₃^? labelling experiment was carried out when the plants had formed their third growing flush. The 22-week exposure to 700 μl l^?1 [CO₂] stimulated plant growth and biomass accumulation (+53% as compared with the 350 μl l^?1 [CO₂] treatment) but decreased the root/shoot biomass ratio (-23%) and specific leaf area (-18%). Moreover, there was an increase in net CO₂ assimilation rate (+37% on a leaf dry weight basis; +71% on a leaf area basis), and a decrease in both above- and below-ground CO₂ respiration rates (-32 and -26%, respectively, on a dry mass basis) under elevated [CO₂]. ¹³C acquisition, expressed on a plant mass basis or on a plant leaf area basis, was also markedly stimulated under elevated [CO₂] both after the 12-h ¹³CO₂ pulse phase and after the 60-h chase phase. Plant N content was increased under elevated CO₂ (+36%), but not enough to compensate for the increase in plant C content (+53%). Thus, the plant C/N ratio was increased (+13%) and plant N concentration was decreased (-11%). There was no effect of elevated [CO₂] on fine root-specific ¹⁵N uptake (amount of recently assimilated ¹⁵N per unit fine root dry mass), suggesting that modifications of plant N pools were merely linked to root size and not to root function. N concentration was decreased in the leaves of the first and second growing flushes and in the coarse roots, whereas it was unaffected by [CO₂] in the stem and in the actively growing organs (fine roots and leaves of the third growth flush). Furthermore, leaf N content per unit area was unaffected by [CO₂]. These results are consistent with the short-term optimization of N distribution within the plants with respect to growth and photosynthesis. Such an optimization might be achieved at the expense of the N pools in storage compartments (coarse roots, leaves of the first and second growth flushes). After the 60-h ¹³C chase phase, leaves of the first and second growth flushes were almost completely depleted in recent ¹³C under ambient [CO₂], whereas these leaves retained important amounts of recently assimilated ¹³C (carbohydrate reserves?) under elevated [CO₂].     

Soil 13C–15N dynamics in an N2-fixing clover system under long-term exposure to elevated atmospheric CO2

Reduced soil N availability under elevated CO₂ may limit the plant's capacity to increase photosynthesis and thus the potential for increased soil C input. Plant productivity and soil C input should be less constrained by available soil N in an N₂‐fixing system. We studied the effects of Trifolium repens (an N₂‐fixing legume) and Lolium perenne on soil N and C sequestration in response to 9 years of elevated CO₂ under FACE conditions. ¹⁵N‐labeled fertilizer was applied at a rate of 140 and 560 kg N ha^?1 yr^?1 and the CO₂ concentration was increased to 60 Pa pCO₂ using ¹³C‐depleted CO₂. The total soil C content was unaffected by elevated CO₂, species and rate of ¹⁵N fertilization. However, under elevated CO₂, the total amount of newly sequestered soil C was significantly higher under T. repens than under L. perenne. The fraction of fertilizer‐N (f_N) of the total soil N pool was significantly lower under T. repens than under L. perenne. The rate of N fertilization, but not elevated CO₂, had a significant effect on f_N values of the total soil N pool. The fractions of newly sequestered C (f_C) differed strongly among intra‐aggregate soil organic matter fractions, but were unaffected by plant species and the rate of N fertilization. Under elevated CO₂, the ratio of fertilizer‐N per unit of new C decreased under T. repens compared with L. perenne. The L. perenne system sequestered more ¹⁵N fertilizer than T. repens: 179 vs. 101 kg N ha^?1 for the low rate of N fertilization and 393 vs. 319 kg N ha^?1 for the high N‐fertilization rate. As the loss of fertilizer‐¹⁵N contributed to the ¹⁵N‐isotope dilution under T. repens, the input of fixed N into the soil could not be estimated. Although N₂ fixation was an important source of N in the T. repens system, there was no significant increase in total soil C compared with a non‐N₂‐fixing L. perenne system. This suggests that N₂ fixation and the availability of N are not the main factors controlling soil C sequestration in a T. repens system.     

Linking sequestration of 13C and 15N in aggregates in a pasture soil following 8 years of elevated atmospheric CO2

The influence of N availability on C sequestration under prolonged elevated CO₂ in terrestrial ecosystems remains unclear. We studied the relationships between C and N dynamics in a pasture seeded to Lolium perenne after 8 years of elevated atmospheric CO₂ concentration (FACE) conditions. Fertilizer‐¹⁵N was applied at a rate of 140 and 560 kg N ha^2?1 y^2?1 and depleted ¹³C‐CO₂ was used to increase the CO₂ concentration to 60 Pa pCO₂. The ¹³C–¹⁵N dual isotopic tracer enabled us to study the dynamics of newly sequestered C and N in the soil by aggregate size and fractions of particulate organic matter (POM), made up by intra‐aggregate POM (iPOM) and free light fraction (LF). Eight years of elevated CO₂ did not increase total C content in any of the aggregate classes or POM fractions at both rates of N application. The fraction of new C in the POM fractions also remained largely unaffected by N fertilization. Changes in the fractions of new C and new N (fertilizer‐N) under elevated CO₂ were more pronounced between POM classes than between aggregate size classes. Hence, changes in the dynamics of soil C and N cycling are easier to detect in the POM fractions than in the whole aggregates. Within N treatments, fractions of new C and N in POM classes were highly correlated with more new C and N in large POM fractions and less in the smaller POM fractions. Isotopic data show that the microaggregates were derived from the macro‐aggregates and that the C and N associated with the microaggregates turned over slower than the C and N associated with the macroaggregates. There was also isotopic evidence that N immobilized by soil microorganisms was an important source of N in the iPOM fractions. Under low N availability, 3.04 units of new C per unit of fertilizer N were sequestered in the POM fractions. Under high N availability, the ratio of new C sequestered per unit of fertilizer N was reduced to 1.47. Elevated and ambient CO₂ concentrations lead to similar ¹⁵N enrichments in the iPOM fractions under both low and high N additions, clearly showing that the SOM‐N dynamics were unaffected by prolonged elevated CO₂ concentrations.     

^14CO2 Fixation and Translocation of Photoassimilates as Selection Criteria of Egyptian Taro Genotypes

The growth characterlstlcs, different physlological parameters, photosynthetic activity （^14CO2 fixation）, and the translocatlon rate of photoassimllates In different taro （Colocasia esculenta L. Schott） genotypes was studled In order to determlne the posslble use of these parameters as selectlon crlterla for dlfferent wldely used genotypes of taro （Delta No. 9, 15, 20, 21, and balady）. The results obtalned suggest that Delta No. 21 shows the most slgnlflcant increase In all parameters tested compared wlth the control （balady）, followed by Delta No. 9, 15, and 20, respectively. The results show a positive correlation between photosynthetlc actlvlty, translocatlon efflclency, and total yield. The selected clone Delta No. 21 Is recommended for cultlvatlon In the delta reglon of Egypt.     

Incomplete 13C labelling of α-pinene content of Quercus ilex leaves and appearance of unlabelled C in α-pinene emission in the dark

α ‐Pinene is formed in and emitted by Quercus ilex leaves. The carbon emitted as α ‐pinene is rapidly and totally labelled by ¹³C in CO₂ in air, but α ‐pinene contained in the leaf shows a fraction of completely unlabelled carbon even after long exposures to air containing only ¹³CO₂. When the labelled leaf is darkened, α ‐pinene emission drops but is still measurable for about 10 h, and carbon becomes partially unlabelled. After an 11 h darkening the α ‐pinene content is still as high as in the light but the carbon is mostly unlabelled. If the leaf is re‐illuminated but photosynthesis is inhibited by removing CO₂ and lowering O₂, a burst of emission occurs and the content of α ‐pinene is depleted. Our experiments suggest that a pool of α ‐pinene which is not directly generated by photosynthesis intermediates exists. Since this pool does not contribute relevantly to the emission in the light, we hypothesize that it is inhibited in the light and spatially located in a different compartment from chloroplasts. As we discuss, glycolysis in the cytoplasm and leucine catabolism in the mitochondria are both possible extra‐chloroplasts sources of carbon for isoprenoids.     

Products of photosynthetic 14CO2 fixation and related enzyme activities in fruiting structures of chickpea

Fruiting structures of a number of legumes including chickpea are known to carry out photosynthetic CO₂ assimilation, but the pathway of CO₂ fixation and particularly the role of phosphoenolpyruvate carboxylase (EC 4.1.1.31) in these tissues is not clear. Activities of some key enzymes of the Calvin cycle and C₄ metabolism, rates of ¹⁴CO₂ fixation in light and dark, and initial products of photosynthetic ¹⁴CO₂ fixation were determined in podwall and seedcoat (fruiting structures) and their subtending leaf in chickpea (Cicer arietinum L.). Compared to activities of ribulose-1,5-bisphosphate carboxylase (EC 4.1.1.39) and other Calvin cycle enzyme, viz. NADP⁺-glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.13), NAD⁺-glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12) and ribulose-5-phosphate kinase (EC 2.7.1.19), the levels of phosphoenolpyruvate carboxylase and other enzymes of C₄ metabolism viz. NADP⁺-malate dehydrogenase (EC 1.1.1.82), NAD⁺-malate dehydrogenase (EC 1.1.1.37), NADP⁺ malic enzyme (EC 1.1.1.40), NAD⁺-malic enzyme (EC 1.1.1.39), glutamate oxaloacetate transaminase (EC 2.6.1.1) and glutamate pyruvate transaminase (EC 2.6.1.2), were generally much higher in podwall and seedcoat than in the leaf. Podwall and seedcoat fixed ¹⁴CO₂ in light and dark at much higher rates than the leaf. Short-term assimilation of ¹⁴CO₂ by illuminated fruiting structures produced malate as the major labelled product with less labelling in 3-phosphoglycerate, whereas the leaf showed a major incorporation into 3-phosphoglycerate. It seems likely that the fruiting structures of chickpea utilize phosphoenolpyruvate carboxylase for recapturing the respired carbon dioxide.     

Theoretical reconsiderations when estimating the mesophyll conductance to CO2 diffusion in leaves of C3 plants by analysis of combined gas exchange and chlorophyll fluorescence measurements

Existing methods to estimate the mesophyll conductance to CO₂ diffusion ( g _m) are often based on combined gas exchange and chlorophyll fluorescence measurements. However, estimations of average g _m by these methods are often unreliable either because the range of usable data is too narrow or because the estimations are very sensitive to measurement errors. We describe three method variants to estimate g _m, for which a wider range of data are usable. They use curve-fitting techniques, which minimise the sum of squared model deviations from the data for A (CO₂ assimilation rate) or for J (linear electron transport rate). Like the existing approaches, they are all based on common physiological principles assuming that electron transport limits A . The proposed variants were far less sensitive than the existing approaches to 'measurement noise' either created randomly in the generated data set or inevitably existing in real data sets. Yet, the estimates of g _m from the three variants differed by approximately 15%. Moreover, for each variant, a stoichiometric uncertainty in linear electron transport-limited photosynthesis can cause another 15% difference. Any estimation of g _m using gas exchange and chlorophyll fluorescence measurements should be considered with caution, especially when g _m is high.     

The effects of glacial atmospheric CO2 concentrations and climate on isoprene emissions by vascular plants

Isoprene (C₅H₈) emissions by terrestrial vegetation vary with temperature and light intensity, and play an important role in biosphere–chemistry–climate interactions. Such interactions were probably substantially modified by Pleistocene climate and CO₂ cycles. Central to understanding the nature of these modifications is assessment and analysis of how emissions changed under glacial environmental conditions. Currently, even the net direction of change is difficult to predict because a CO₂‐depleted atmosphere may have stimulated emissions compensating for the negative impacts of a cooler climate. Here, we address this issue and attempt to determine the direction of change from an experimental standpoint by investigating the interaction between isoprene emissions and plant growth of two known isoprene‐emitting herbaceous species (Mucuna pruriens and Arundo donax) grown at glacial (180 ppm) to present (366 ppm) CO₂ levels. We found a significant enhancement of isoprene emissions per unit leaf area in M. pruriens under subambient CO₂ concentrations relative to ambient controls but not for A. donax. In contrast, canopy emissions remained unaltered for both plant species because enhanced leaf emissions were offset by reductions in biomass and leaf area. Separate growth experiments with M. pruriens revealed that lowering day/night temperatures by 5°C decreased canopy isoprene emissions irrespective of the CO₂ level. Incorporation of these results into a simple canopy emissions model highlights their potential to attenuate reductions in the total isoprene flux from forests under glacial conditions predicted by standard models.     

Photosynthetic 14CO2 fixation and [15N]-ammonia assimilation during UV-B radiation of Lithodesmium variabile

Uptake of [¹⁵N]-ammonia was more sensitive to UV-B exposure than the total ¹⁴CO₂ fixation rate of Lithodesmium variabile Takano. Short-term UV-B radiation (15 min) had practically no effect on the kinetics of [¹⁵N]-ammonia, whereas there was an effect on [¹⁴C]-bicarbonate uptake rate. A significant reduction was found after 30 and 60 min UV-B stress. The time course of photosynthetic uptake of ¹⁵NH₄Cl at several wavelengths was markedly depressed at shorter wavelengths (irradiation with WG 280). A short-term (11 min) exposure to ultraviolet radiation had no influence on the [¹⁴C]-labeled photosynthetic products. However, the [¹⁵N]-label of several amino acids and the ratio of [¹⁵N]-glutamine to [¹⁵N]-glutamic acid varied after irradiation with different ultraviolet wavebands. The results are discussed with reference to UV damage to the key enzymes of nitrogen metabolism.     

Temporal photosynthetic carbon isotope signatures revealed in a tree ring through 13CO2 pulse-labelling

Using a combined method of pulse-labelling trees and analysing detailed distribution of ¹³C tracer within tree rings, we studied how photo-assimilates incorporated on a given day are then distributed in a tree ring. A branch of a 4-year-old Cryptomeria japonica D.Don tree growing in Tsukuba, Japan was pulse-labelled with non-radioactive ¹³CO₂ on two occasions: 29 May 2001 and 18 September 2001. Two discs were cut from the stem on 4 March 2002, one immediately under and the other 0.5 m below the branch and put through high-resolution δ ¹³C analysis. δ ¹³C peaks were observed in both the earlywood and latewood of the concerned tree ring, corresponding to each pulse-labelling date. The earlywood peaks was broader than the latewood peaks, possibly reflecting seasonal variation of the width of wood developing zone. Half-widths of the peaks were measured and used as indicators for the potential time resolution of tree-ring isotope analysis. The half-widths of the peaks indicated a time resolution no finer than 8.7–28 and 33–42 d in the early and latewood, respectively. Holocellulose extraction yielded only a slight change to the shape of the δ ¹³C peaks. ¹³C tracer pulse-labelled in May and September reached tangentially different locations in the lower disc, suggesting a seasonal change in the pathway of carbohydrates. Local consumption of spring assimilates and long-distance downward transport of autumn assimilates were also suggested.     

Non-stomatal limitation of CO2 assimilation in three tree species during natural drought conditions

The effect of drought on CO₂ assimilation and leaf conductance was studied in three northern hardwood species: Quercus rubra L., Acer rubrum L. and Populus grandidentata Michx. Leaf gas exchange characteristics at two CO₂ levels (320 and 620 μl I⁻¹) and temperatures from 20 to 35°C were measured at the end of a dry period and shortly after 10 cm of rainfall. The effects of drought varied with species, temperature and CO₂ level. Calculated values of internal CO₂ concentration showed little or no decline during drought. Differences in assimilation, before vs after the rains, were most apparent at the higher CO₂ level. These latter two observations indicate nonstomatal disruption of CO₂ assimilation during the dry period. In P. grandidentata there was a substantial interaction between drought and temperature, with a resultant shift in the temperature for maximum assimilation to lower temperatures during drought. During drought, internal CO₂ concentrations increased sharply in all three species under the combined conditions of high temperatures and the higher CO₂ level.     

The utilization of pre-anthesis reserves in grain filling of wheat. Assessment by steady-state 13CO2/12CO2 labelling

δ, C isotope composition relative to Pee Dee Belemnite
WSC, water-soluble carbohydrates
N, nitrogen
C, carbon
cv, cultivar
ME, efficiency of mobilized pre-anthesis C utilization in grain filling (g C g^–1C)

Significant mobilization of protein and carbohydrates in vegetative plant parts of wheat regularly occurs during grain filling. While this suggests a contribution of reserves to grain filling, the actual efficiency of mobilized assimilate conversion into grain mass (ME) is unknown. In the present study the contribution of pre-anthesis C (C fixed prior to anthesis) to grain filling in main stem ears of two spring wheat (Triticum aestivum L.) cultivars was determined by ¹³C/¹²C steady-state labelling. Mobilization of pre-anthesis C in vegetative plant parts between anthesis and maturity, and the contributions of water-soluble carbohydrates (WSC) and protein to pre-anthesis C mobilization were also assessed. Experiments were performed with two levels of N fertilizer supply in each of 2 years. Pre-anthesis reserves contributed 11–29% to the total mass of C in grains at maturity. Pre-anthesis C accumulation in grains was dependent on both the mass of pre-anthesis C mobilized in above-ground vegetative plant parts (r² = 0·87) and ME (defined as g pre-anthesis C deposited in grains per g pre-anthesis C mobilized in above-ground vegetative plant parts; r² = 0·40). ME varied between 0·48 and 0·75. The effects of years, N fertilizer treatments and cultivars on ME were all related to differences in the fractional contribution of WSC to pre-anthesis C mobilization. Multiple regression analysis indicated that C from mobilized pre-anthesis WSC may be used more efficiently in grain filling than C present in proteins at anthesis and mobilized during grain filling. Possible causes for variability of ME are discussed.     

Translocation of labelled assimilates following photosynthesis of 14CO2 by the field bean, Vicia faba

When whole plants were exposed to ¹⁴CO₂, almost the same amount of radioactivity was taken up initially by each leaf regardless of its position on the stem and of the presence of beans at that node. Thus, although developing beans are a powerful sink for assimilated carbon, they do not increase the CO₂ uptake by adjoining leaves.
The distribution of labelled assimilates 6 hours after feeding ¹⁴CO₂ to a single leaf for 1 hour varied with both the position of the treated leaf and the stage of development of the plant. Before any flowers were set most of the radioactivity from all expanded leaves moved downwards to the roots and the stem below the treated leaf (lower stem). Later, during pod-fill, the upper leaves maintained this supply to the roots and lower stem, whilst most of the carbon translocated from the lower and mid-stem leaves went to the beans. However, we found no exclusive relationship between a leaf and the supply to beans developing on the same node.
The amount of radioactivity moving out of a source leaf at a fruiting node increased over successive samplings up to 48 h; the pattern of distribution of the ¹⁴CO₂ however remained virtually unchanged.     

Modelling the discrimination of 13CO2 above and within a temperate broad-leaved forest canopy on hourly to seasonal time scales

Fluxes and concentrations of carbon dioxide and ¹³CO₂ provide information about ecosystem physiological processes and their response to environmental variation. The biophysical model, CANOAK, was adapted to compute concentration profiles and fluxes of ¹³CO₂ within and above a temperate deciduous forest (Walker Branch Watershed, Tennessee, USA). Modifications to the model are described and the ability of the new model (CANISOTOPE) to simulate concentration profiles of ¹³CO₂, its flux density across the canopy–atmosphere interface and leaf‐level photosynthetic discrimination against ¹³CO₂ is demonstrated by comparison with field measurements. The model was used to investigate several aspects of carbon isotope exchange between a forest ecosystem and the atmosphere. During the 1998 growing season, the mean photosynthetic discrimination against ¹³CO₂, by the deciduous forest canopy (Δ_canopy), was computed to be 22·4‰, but it varied between 18 and 27‰. On a diurnal basis, the greatest discrimination occurred during the early morning and late afternoon. On a seasonal time scale, the greatest diurnal range in Δ_canopy occurred early and late in the growing season. Diurnal and seasonal variations in Δ_canopy resulted from a strong dependence of Δ_canopy on photosynthetically active radiation and vapour pressure deficit of air. Model calculations also revealed that the relationship between canopy‐scale water use efficiency (CO₂ assimilation/transpiration) and Δ_canopy was positive due to complex feedbacks among fluxes, leaf temperature and vapour pressure deficit, a finding that is counter to what is predicted for leaves exposed to well‐mixed environments.