SEARCH

SEARCH BY CITATION

Keywords:

  • Leaf structure;
  • nitrogen content;
  • photosynthetic capacity;
  • mesophyll conductance;
  • CO2 drawdown;
  • growth irradiance;
  • leaf age

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

Mature non-senescent leaves of evergreen species become gradually shaded as new foliage develops and canopy expands, but the interactive effects of integrated light during leaf formation (QintG), current light (QintC) and leaf age on foliage photosynthetic competence are poorly understood. In Quercus ilex L., we measured the responses of leaf structural and physiological variables to QintC and QintG for four leaf age classes. Leaf aging resulted in increases in leaf dry mass per unit area (MA), and leaf dry to fresh mass ratio (DF) and decreases in N content per dry mass (NM). N content per area (NA) was independent of age, indicating that decreases in NM reflected dilution of leaf N because of accumulation of dry mass (NA = NM MA). MA, DF and NA scaled positively with irradiance, whereas these age-specific correlations were stronger with leaf growth light than with current leaf light. Area-based maximum ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylase activity (VcmaxA), capacity for photosynthetic electron transport (JmaxA) and the rate of non-photorespiratory respiration in light (RdA) were also positively associated with irradiance. Differently from leaf structural characteristics, for all data pooled, these relationships were stronger with current light with little differences among leaves of different age. Acclimation to current leaf light environment was achieved by light-dependent partitioning of N in rate-limiting proteins. Mass-based physiological activities decreased with increasing leaf age, reflecting dilution of leaf N and a larger fraction of non-photosynthetic N in older leaves. This resulted in age-dependent modification of leaf photosynthetic potentials versus N relationships. Internal diffusion conductance (gm) per unit area (gmA) increased curvilinearly with increasing irradiance for two youngest leaf age classes and was independent of light for older leaves. In contrast, gm per dry mass (gmM) was negatively associated with light in current-year leaves. Greater photosynthetic potentials and moderate changes in diffusion conductance resulted in greater internal diffusion limitations of photosynthesis in higher light. Both area- and mass-based gm decreased with increasing leaf age. The decrease in diffusion conductance was larger than changes in photosynthetic potentials, leading to larger CO2 drawdown from leaf internal air space to chloroplasts (ΔC) in older leaves. The increases in diffusion limitations in older leaves and at higher light scaled with age- and light-dependent increases in MA and DF. Overall, our study demonstrates a large potential of foliage photosynthetic acclimation to changes in leaf light environment, but also highlights enhanced structural diffusion limitations in older leaves that result from leaf structural acclimation to previous rather than to current light environment and accumulation of structural compounds with leaf age.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

Leaves in different canopy positions are exposed to widely different irradiances. Plant photosynthetic capacities acclimate to the strong canopy light gradient, resulting in positive correlations between long-term average integrated light (Qint) availability in the canopy and photosynthetic capacity (e.g. Meir et al. 2002; Niinemets, Tenhunen & Beyschlag 2004). However, increases in Qint are also associated with higher temperature, greater evaporative demand and overall greater water stress (see Baldocchi, Wilson & Gu 2002 for review; Niinemets & Valladares 2004). Leaf acclimation to water stress brings about a series of modifications in foliage structure such as increased cell wall thickness, greater degree of cell wall lignification and overall greater density of foliage (Salleo & Lo Gullo 1990; Grossoni et al. 1998; Niinemets & Kull 1998). Such chemical and structural alterations collectively improve the leaf tolerance of low water potentials, but they may result in greater limitations of photosynthesis by CO2 diffusion from intercellular air space to carboxylation sites in chloroplasts, thereby reducing the efficiency of nitrogen use in photosynthesis. Particularly, in Mediterranean environments, interacting canopy light and drought gradients are expected to be especially severe (Tenhunen et al. 1985; Sala et al. 1994), and strongly modify foliage photosynthetic functioning in high light. Possibly because of strong water limitations in higher light, leaf photosynthetic capacity versus canopy light relationships were strongly curvilinear with only moderate changes above c. 30% full canopy light in Mediterranean oaks (Rambal et al. 1996; Niinemets et al. 2004).

It has been assumed for a long time that photosynthetic capacity and internal diffusion conductance (gm) are positively correlated such that leaves of different structure operate at the same CO2 drawdown from intercellular air space to chloroplasts (ΔC) (von Caemmerer & Evans 1991; Loreto et al. 1992; Evans & von Caemmerer 1996; Evans & Loreto 2000; Piel et al. 2002; Loreto, Centritto & Chartzoulakis 2003; Flexas et al. 2004; Grassi & Magnani 2005), contrasting the hypothesis of canopy variation in diffusion limitations of photosynthesis. However, the correlations between gm and photosynthetic capacity are often curvilinear (DeLucia, Whitehead & Clearwater 2003; Loreto et al. 2003; Flexas et al. 2004; Niinemets & Sack 2006), and recent evidence demonstrates that lower diffusion conductances are associated with larger reductions of CO2 concentrations between intercellular air space and chloroplasts (Niinemets et al. 2005; Warren & Adams 2006; Niinemets & Sack 2006).

In evergreen species, a further complication is that the canopies consist of various-aged leaves. In Mediterranean evergreens, the fraction of leaf area older than one year is often 0.4–0.6 of total (Sala i Serra 1992). As the older leaves become gradually shaded by growing foliage, foliage physiological characteristics acclimation to their growth irradiance should re-acclimate to changed light environment. Re-acclimation of foliage photosynthetic characteristics has been studied in temperate conifers (Brooks, Hinckley & Sprugel 1994; Brooks, Sprugel & Hinckley 1996; Niinemets 1997), but information is scarce for broad-leaved evergreens. Studies in temperate conifers highlight re-allocation of photosynthetic nitrogen from compounds limiting light-saturated rate of photosynthesis to light harvesting (Warren & Adams 2001), but also conflicting requirements for enhanced N investment in light harvesting and overall decrease of N with age (Brooks et al. 1994, 1996; Niinemets 1997). Previous work in Mediterranean species has further demonstrated an increase of internal diffusion limitations of photosynthesis with increasing age of fully illuminated various-aged leaves (Niinemets et al. 2005). Given that leaf thickness remains essentially constant in various-aged mature leaves (Brooks et al. 1996; Oguchi, Hikosaka & Hirose 2003, 2005), photosynthetic potentials and gm may become increasingly uncoupled with leaf aging and shading in evergreen canopies. The interactions between leaf age and light environment on photosynthetic potentials and gm have major implications for parameterization of whole canopy carbon gain models, but the evidence available so far is inadequate.

We investigated within-canopy variation in structure, N and photosynthetic potentials in various-aged leaves of Mediterranean evergreen species Quercus ilex L. to test the following hypotheses: (1) diffusion limitations of photosynthesis are larger in leaves exposed to higher light; (2) foliage structural characteristics of older leaves are acclimated to growth irradiance, while photosynthetic potentials re-acclimate to current light; and (3) limited structural re-acclimation results in stronger diffusion limitations in older shaded leaves than in the current-year leaves exposed to the same irradiance. The results of this analysis demonstrate complex interactions between growth and current light environments and leaf aging, and provide a valuable extension to understand how do evergreen broad-leaved canopies function.

MATERIALS AND METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

Study site

The study was carried out in Monte Brione, Riva del Garda, Trentino, Italy (45°53′N, 10°52′E, altitude c. 300 m). The site has Mediterranean climate with an average winter temperature of 5.4 °C, average summer temperature of 22.7 °C, annual average temperature of 14 °C and annual precipitation of 930 mm. The measurements were conducted in October 2004 after the onset of autumn rains when the plants were recovered from summer drought. Current-year leaves were 6 months old by the time of the study.

The stand supports a Q. ilex-dominated evergreen forest with an average height of 10–11 m. Deciduous species Ostrya carpinifolia Scop. and Fraxinus ornus L. were minor associates in the mid-canopy. Total stand plant area index (including leaves, branches and stems) determined by hemispheric photography according to the methodology of van Gardingen et al. (1999) was 5.0 ± 0.5 m2 m−2, with an average Markov clumping index of 0.61 typical of a moderate aggregated canopy (Nilson 1971). The soil is a chromic-calcaric luvisol (FAO-UNESCO 1988) formed on a silty calcareous bedrock (pH = 7.75 for AE horizon and pH = 8.35 for Ck horizon) (Sartori, Corradini & Mancabelli 1995).

Foliage sampling and determination of light availability in leaf location

Foliage for gas-exchange measurements was collected along the light gradient from canopy top to bottom in six canopy trees using a mobile lift between 8:00 and 13:00 h. For every branch selected, two hemispherical pictures above the branch tips bearing current-year leaves were taken using a Nikon D70 digital camera (Nikon Corporation, Tokyo, Japan) equipped with a Sigma 8 mm F4 Ex fish-eye lens (Sigma Corporation, Kanagawa, Japan). Matrix metering was used, the aperture was set to a fixed small value (f = 7.4) to minimize lens vignetting, and the images were underexposed by one stop to enhance the contrast of the canopy silhouettes (Hale & Edwards 2002). In addition, two synchronized LI-2000 plant canopy analysers (PCAs, Li-COR Inc. 1992) were used to estimate the fraction of penetrating diffuse light for current-year, and 1- and 2-year-old leaves. One PCA was set in a nearby completely open location, and the other was used in the canopy to measure the transmittances. Five readings were taken for every leaf age class on a given branch. After taking the hemispherical pictures and PCA readings, the branches were harvested and immediately re-cut under water. The cut ends were kept in water while the branches were transported to the laboratory.

The hemispherical photographs were converted to binary images on the basis of the iterative algorithm of Ridler & Calvard (1978), suggested by Jonckheere et al. (2005) as the most appropriate method for automatic image segmentation. Binary images were used to calculate the fraction of diffuse irradiance of open sky (diffuse site factor, ID) and the fraction of potential penetrating direct solar radiation of open sky [direct site factor, IB(i)] at hour i. Relative estimates of light availability were converted to average daily integrated quantum flux density (Qint, mol m−2 d−1) for the period of leaf growth and development (from 1 May 2004 to 31 August 2004) using hourly values of above-canopy global solar radiation, RG(i) (J m−2 s−1) measured at the weather station in Riva del Garda:

  • image(1)

where γ (mol J−1) is a conversion factor for Qint to global solar radiation; n is the time conversion constant (number of d during the study period per number of s in a day); pD(i) is the proportion of diffuse quantum flux density estimated at hourly time-step according to the algorithm of Weiss & Norman (1985). An average seasonal value of pD = 0.47 was obtained for the period of leaf growth, providing an average integrated above-canopy Qint of 36.0 mol m−2 d−1. Estimates of Qint for the same branch from the two hemispherical photographs were averaged. Average standard deviation (SD) of the measurements was 2.1 mol m−2 d−1.

Hemispherical photographs provided Qint measurements for current-year leaves. These measurements were further combined with canopy transmittances determined by PCA to derive the estimates of Qint for 1- and 2-year-old leaves. Overall, the light available for older leaves significantly decreased with leaf age as a result of canopy growth and enhanced shading (Fig. 1). The time-dependent decline in Qint values was fitted by regressions in the form of Qint(age) = ae(b*age) (r2 > 0.95), and these regressions were used to predict the values of Qint for 3- and 4-year-old leaves.

image

Figure 1. Correlations between the current seasonal average daily integrated quantum flux density (QintC), and the average integrated light during leaf growth (QintG) for various-aged leaves of Mediterranean evergreen species Quercus ilex. As they age, the leaves developed at certain light regime become gradually shaded by growing canopy. Light availability of current-year leaves (age 0) was used as an estimate of QintG for various-aged leaves on the same branch. Data were fitted by linear regressions. We assumed that in this closed canopy forest, age-dependent changes in irradiance are essentially constant in time, and thus, this routine provided a realistic estimate of growth irradiance for older leaves. This was supported by direct measurements of current leaf irradiance of various-aged leaves.

Download figure to PowerPoint

We define the average seasonal daily Qint estimated above the specific leaf age class as the current average light (QintC). The light availability of current-year leaves is further taken as a proxy of the irradiance to which various-aged leaves of the same branch were exposed during their growth (growth light, QintG). Because of overall canopy growth, this procedure may underestimate the irradiance during growth of the oldest leaves, particularly in the upper canopy where age-dependent changes in light availability are the largest. In this closed-canopy stand, such an underestimation is minor in the mid- and lower canopy. Given further that these mature trees have round-topped canopy shape with several growing points, the canopy height increases only by 10–20 cm year−1. Thus, enhanced shading between neighbouring branches also unlikely resulted in significant underestimation of growth irradiance in the upper canopy. We assume that the amount of light incident to current-year foliage provides an overall realistic estimate of the growth irradiance for older leaves.

Leaf photosynthetic measurements

In the laboratory, the sampled branches were pre-conditioned at room temperature and at dim light of 50–100 µmol m−2 s−1 during an 8 h light period for 2–3 d to stabilize leaf gas-exchange characteristics, and achieve high stomatal conductance during gas-exchange measurements. During stabilization, the branches were re-cut under water each day. Stomatal conductances were high and stable in pre-conditioned branches (see Niinemets et al. 2005 for details). Careful pre-conditioning of branches is essential to avoid rapid stomatal closure during gas-exchange measurements that can result in artefacts in gm estimations (Centritto, Loreto & Chartzoulakis 2003).

A Li-Cor LI-6400 portable photosynthesis system and a LI-6400-40 leaf chamber fluorometer (Li-Cor, Inc., Lincoln, NE, USA) were used to measure leaf net assimilation versus internal CO2 (A/Ci) response curves simultaneously with the effective quantum yields of photosystem II (PSII) as in Niinemets et al. (2005). The Qint was set at 1000 µmol m−2 s−1[20% from blue light-emitting diodes (LED), 80% red LED], leaf temperature at 25 °C, and water vapour pressure deficit at 1.1 kPa. The leaves were kept at 50 µmol mol−1 of ambient CO2 before the start of an A/Ci curve to reach maximum stomatal openness at the beginning of experiments (Centritto et al. 2003). At this low CO2 concentration, maximum stomatal conductances to water vapour (gs) were observed 20 min to 1.5 h after switching the leaf to low CO2 atmosphere. After maximum stomatal opening, steady-state net assimilation rates were measured at eight ambient CO2 concentrations between 50 and 2500 µmol mol−1. Generally, several estimates of net assimilation rate were taken at every ambient CO2 concentration (on average 13.2 data points per an A/Ci response curve) to enhance the robustness of the following calculations of biochemical potentials of photosynthesis. For every steady-state photosynthesis measurement, the corresponding effective quantum yield of PSII was also determined by applying a 1 s pulse of white light of 10 000 µmol m−2 s−1 to close all PSII centres (Fm′ determination).

After measurements of an A/Ci curve, the light was switched off, and leaf respiration rate was determined immediately when the chamber temperature and ambient CO2 concentrations stabilized (typically 1–2 min after switching off the light).

A/Ci curves were measured for 86 leaves (27 current-year, 23 1-year-old, 21 2-year-old, 11 3-year-old and 4 4-year-old leaves) from 23 branches. Because not all leaf age classes were present in all branches, the sample number was lower for older age classes, while for current-year leaves, several leaves were occasionally measured from the same branch to estimate the within-branch variability in gas-exchange characteristics. In the following analysis, gas-exchange characteristics of current-year leaves from the same branch were averaged for analysing the relationships with light, while every leaf was considered separately in correlations between leaf gas exchange characteristics and with the leaf structural and chemical variables. Data for 3- and 4-year-old leaves were analysed together (leaf age class ≥ 3-year old).

Determination of the rate of photosynthetic electron transport (JETR) from fluorescence measurements

Leaf reflectance and transmittance were measured with a Taylor-type integrated sphere (Li-Cor 1800, Li-Cor, Inc.) using a Field Spec Pro spectroradiometer (Analytical Spectral Devices, Inc., Boulder, CO, USA) directly after combined gas-exchange and fluorescence estimations. Leaf absorptance for every wavelength in photosynthetically active spectrum, A(λ), was calculated. The amount of absorbed quantum flux density (Qabs) was determined from the values of A(λ), the spectra of blue and red LEDs of LI-6400-40 leaf chamber fluorometer light source (Li-Cor Inc. 2002), and the total quantum flux density of blue and red LEDs. For the specific spectral composition of the Li-Cor light source LEDs and the fractions of red and blue light used in our study (80% red, 20% blue), the average (± SE) values of leaf absorptance were 0.9234 ± 0.0029 for current-year, 0.9310 ± 0.0014 for 1-year-old, 0.9296 ± 0.0033 for 2-year-old and 0.9328 ± 0.0023 for ≥ 3-year-old leaves.

The maximum fluorescence yield in light-adapted state (Fm′) and the steady-state fluorescence yield in light (F) were used to estimate the effective quantum yield of PSII, ΦPSII, as (Fm′ − F)/Fm′. The rate of photosynthetic electron transport, JETR, was calculated as (Schreiber, Bilger & Neubauer 1994):

  • JETR  =  ɛΦPSIIQabs.(2)

The scaling factor (ɛ) characterizes the partitioning of intercepted light between the photosystem I (PSI) and PSII. We assumed that light is equally distributed between both photosystems (ɛ = 0.5) (Schreiber et al. 1994; Bernacchi et al. 2002; DeLucia et al. 2003).

Estimation of gm, maximum ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylase activity (Vcmax) and the capacity for photosynthetic electron transport (Jmax)

Given that only 2 cm2 of leaf area is enclosed in the LI-6400-40 leaf chamber fluorometer cuvette, A and Ci values were corrected for diffusion according to the recommendations of Li-Cor (2002) before further calculations. A diffusion correction term (k) of 0.445 µmol s−1 was used.

We used the variable electron transport rate method of Harley et al. (1992) to calculate the internal diffusion conductance from the substomatal cavities to chloroplasts (gm) as:

  • image(3)

where Rd is the rate of non-photorespiratory respiration continuing in light, and Γ* is the hypothetical CO2 compensation point in the absence of Rd. Standard Rubisco kinetic characteristics determined in vitro (Jordan & Ogren 1984; Niinemets & Tenhunen 1997) were used to derive a value of Γ* at 25 °C (45.3 µmol mol−1). We preferred the in vitro Rubisco characteristics, as the previous in vivo determinations of Γ* are likely underestimates because of assuming an infinite gm in the calculations (see Ethier & Livingston 2004 for a discussion; also see Niinemets et al. 2005). Rd was taken as half of the rate of respiration measured in the dark as reported in previous studies (Villar, Held & Merino 1995; Piel et al. 2002; Niinemets et al. 2005).

The gm values were calculated for A and JETR pairs determined for internal CO2 concentrations of 100–350 µmol mol−1, and the average gm value was determined for each leaf. For this Ci range, 4–8 measurement points (on average 5.3) were available for every A/Ci response curve (see Niinemets et al. 2005 for representative curve fits). Stomatal conductance was generally stable at these relatively low CO2 concentrations. When time-dependent decreases in stomatal conductance were observed during these measurements, the data were rejected. The average coefficient of variation [100*(SD)/(average of all gm values for a given leaf)], of single-leaf gm estimates was 7.6% (range 0.12–36%).

The internal (Ci) and chloroplastic (Cc) CO2 concentration are related as:

  • Cc  =  Ci  −  A/g.(4)

Using Eqn 4, A/Ci curves were converted to A/Cc curves. Values of Vcmax, Jmax and Rd were determined from fitting of A/Cc curves as in Niinemets et al. (1999).

Evaluation of possible errors in gm determination

The assumptions made in calculations of Rd and JETR may bias the gm estimations. In particular, the scaling factor used to convert the measurements of dark respiration to Rd may deviate from the value of 0.5 used in our study (Villar et al. 1995). The errors in JETR may result from the uneven distribution of light between the two photosystems (ɛ in Eqn 2 <> 0.5). Another potential error source in JETR estimations is that the fluorescence signal from lower leaf layers may be relatively lower than the contribution of these layers to total leaf electron transport rate (Vogelmann & Evans 2002), resulting in possible overestimation of actual leaf electron transport rate by fluorescence measurements.

To test the overall effect of possible errors in Rd and JETR on gm estimates, we conducted a sensitivity analysis in Appendix I. In addition, we calculated an alternative estimate of gm based on changes in the curvature of A/Ci response curves when gm is finite (Ethier & Livingston 2004; Niinemets et al. 2005) (Appendix I). The sensitivity analysis demonstrates that for the part of A/Ci curve used for gm determination in our study, even a large underestimation or overestimation in Rd, by ± 50%, and a large overestimation of potential electron transport rate by fluorescence, had a minor effect on gm (see Appendix I for details). In addition, the agreement between the alternative estimates of gm and those derived by Eqn 3 was excellent (Appendix I), overall suggesting that our derivations of gm are robust. This agrees with the observations of Harley et al. (1992) that over the Ci range used for gm determination in our study, the values of gm are stable, and the calculations of gm are relatively insensitive to minor errors in Γ*, Rd, A and JETR.

CO2 drawdown as a result of finite gm

Average values of A, Rd, stomatal conductance to water vapour and Ci measured at ambient CO2 concentrations of 200–400 µmol mol−1 (average ambient CO2 concentration of 330–340 µmol mol−1) were further calculated for measurements with stable stomatal conductance. From these values, average Cc was estimated by Eqn 4, and average Cc/Ci ratio and CO2 drawdown from the internal air space to chloroplasts (ΔC = Ci − Cc) were calculated. Given that photosynthesis responds linearly to CO2 over this concentration range, and Eqn 4 is linear with respect to both Ci and A, average Cc determined from average Ci, and A is equivalent with average Cc derived from individual combinations of A and Ci.

Chemical and structural analyses

Leaf nitrogen content per dry mass (NM) was estimated using a Perkin Elmer series II CHNS/O Analyzer 2400 (Perkin Elmer Life and Analytical Sciences, Inc., Boston, MA, USA). Leaves were scanned with a resolution of 300 dpi, and leaf area was determined with UTHSCSA Imagetool 2.00alpha (C. Donald Wilcox, S. Brent Dove, W. Doss McDavid and David B. Greer, Department of Dental Diagnostic Science, The University of Texas Health Science Center, San Antonio, TX, USA; ddsdx.uthscsa.edu).

Fresh mass of leaves was determined directly after gas-exchange measurements. The leaves were then oven-dried at 70 °C for at least 48 h, and dry mass was determined. From these measurements, we calculated leaf dry to fresh mass ratio (DF) and dry mass per unit area (MA).

Apparent nitrogen fractions in photosynthetic machinery

As age-dependent variations in net assimilation can result either from variations in total leaf N content or fractional investment of N in rate-limiting proteins, or from both changes in total content and fractional investments, we further calculated two indices to standardize photosynthetic potentials and estimate the photosynthetic nitrogen use efficiency of various-aged leaves. MA, nitrogen content and values of Jmax and Vcmax were used to calculate the apparent nitrogen fractions in the rate-limiting proteins of photosynthetic electron transport, PB (Niinemets & Tenhunen 1997):

  • image(5)

and in Rubisco (PR):

  • image(6)

where Jmc is the capacity for photosynthetic electron transport per unit cytochrome f (mol e (mol cyt f)−1 s−1 (Jmc = 156 mol e (mol cyt f)−1 s−1 at 25 °C) (Niinemets & Tenhunen 1997), and Vcr is the maximum rate of ribulose-1,5-bisphosphate carboxylation per unit Rubisco protein [Vcr = 20.5 µmol CO2 (g Rubisco)−1 s−1 at 25 °C] (Niinemets & Tenhunen 1997). The scaling coefficients 8.06 and 6.25 are based on the stoichiometry of rate-limiting proteins and nitrogen content of proteins. Although the actual nitrogen fractions of N in photosynthetic apparatus may vary because of species-specific variability in Rubisco specific activity and the rate of electron transport per unit cytochrome f (Niinemets & Tenhunen 1997; Spreitzer 1999), the values of Jmc and Vcr are assumed to be conserved among C3 plants in this analysis. This may lead to species-specific bias in apparent nitrogen investments, but within a species, the values of PB and PR are proportional with the fraction of N in rate-limiting proteins.

Data analyses

Linear and non-linear regressions were used to explore the statistical significance of the effects of current seasonal average daily QintC and QintG (Fig. 1) on foliage structural, chemical and physiological characteristics, and to analyse the correlation among foliage characteristics. Non-linear regressions in the form of y = a + log(Qint) were employed when they yielded a higher degree of explained variance (r2) than the linear regressions.

Structural, chemical and physiological characteristics of leaves with different age were compared by covariance analyses (ancova) when seasonal average daily Qint significantly affected the specific variable according to regression analyses. For non-linear relationships, ancova was conducted with Log(Qint) as the covariate. Separate ancovas were conducted using both estimates of Qint– the QintC and QintG (Fig. 1) as covariates, or for non-linear regressions using the corresponding log-transformed estimates of light availability. Separate-slope ancova analyses were conducted first to test the difference among the slopes. Because the interaction term, age × Log(Qint), was only significant for gm per unit dry mass (gmM; P < 0.03), common-slope analyses were conducted in all cases to test for the difference among the means of characteristics of various-aged leaves at common irradiance.

When the effect of Qint or Log(Qint) was non-significant according to regression analyses (P > 0.05), the means were compared by analysis of variance (anova) [for NM and mass-based Rd values (RdM)]. After conduction of ancova or anova, Bonferroni tests were used to separate the means between the age classes. All statistical effects were considered significant at P < 0.05.

Our study includes a complex array of diverse structural and physiological characteristics of various-aged leaves from different canopy positions. Three criteria were employed to judge whether a specific leaf character of various-aged leaves tended to be more strongly determined by growth or current light environment. The first criterion was the difference in the fraction of variance explained by QintG and QintC when all data for various-aged leaves pooled. The second criterion was the deviation of age-specific regressions from the mean regression line. Thus, when the regression with all data pooled provided a larger r2 with QintG than with QintC, and the age-dependent regressions did not deviate from the mean regression according to ancova analyses, the specific characteristic was concluded to depend more strongly on QintG, otherwise on QintC. To further increase the robustness of these conclusions, we sorted out the pairs of QintG and QintC that deviated most from the regression line of QintG versus QintC for all various-aged leaves pooled (Fig. 1). The studentized residual ≥ 1.0 was used as the threshold for the selection of outlying light estimates. Using this truncated data set with 22 observations for various-aged leaves, we developed again the regressions with both light estimates and leaf photosynthetic characteristics, allowing conclusive testing at P < 0.05 level whether specific variable is correlated with QintG or with QintC without the interfering correlation between the different light estimates.

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

Light- and age-dependent modifications in leaf structure and nitrogen content

Leaf light availability strongly declined as older leaves became shaded by younger leaves (Fig. 1). Because of time-dependent changes in leaf light availability, the seasonal average daily integrated quantum flux densities experienced by leaves during the current season (QintC) and during leaf formation (QintG) were only weakly correlated for older leaves (Fig. 1).

Leaf dry mass per unit area (MA, Fig. 2a & b) and nitrogen content per area (NA; Fig. 2c & d) increased with increasing both the QintC (Fig. 2a & c) and QintG (Fig. 2b & d), but the correlations with current light were generally weaker and not significant for older leaves (cf. Fig. 2a, c & 2b, d). For the data of all age classes pooled, the r2 values were 0.18 for MA and 0.41 for NA versus QintC relations (Fig. 2a & c), while r2 values were 0.67 for MA and 0.70 for NA versus QintG relations (Fig. 2b & d; P < 0.001 for all), further indicating a stronger correlation of MA and NA with leaf growth light. Overall, these relationships were curvilinear and were linearized by log-transformation of Qint (inset in Fig. 2b).

image

Figure 2. Dependencies of leaf dry mass per unit area (MA; a & b) and nitrogen content per unit area (NA; c & d) on current integrated average light (QintC a & c) and the light availability during foliage development (QintG b & d) for various-aged leaves of Quercus ilex. Data were fitted by non-linear regressions in the form of y = a + blog(x). Inset in (b) demonstrates the linearized relationships between Log(QintG) and MA. Non-significant regressions (P > 0.05) are depicted by dashed lines. The correlations between QintC and QintG for various-aged leaves are depicted in Fig. 1. Table 1 provides the comparison of the intercepts of y = a + blog(x) relations between different age classes.

Download figure to PowerPoint

Nitrogen content per dry mass (NM) was independent of light for all age classes (P > 0.1), indicating that light-driven changes in NA were primarily dependent on modifications in MA (NA = MA NM).

Leaf dry to fresh mass ratio (DF) also scaled positively with irradiance, but the correlations for different age classes were stronger with non-transformed than with log-transformed light (data not shown). Linear regressions of DF versus QintC were significant for current (r2 = 0.38, P < 0.001), 1-year-old (r2 = 0.29, P < 0.02), and 2-year-old leaves (r2 = 0.31, P < 0.02), but not for ≥ 3-year-old leaves (r2 = 0.07, P > 0.3). Linear regressions of DF versus QintG were stronger and significant for all age classes (e.g. r2 = 0.41, P < 0.02 for ≥ 3-year-old leaves). For all data pooled, DF versus QintC relationship was not significant (r2 = 0.00, P > 0.6), while the relationship was significant with QintG (r2 = 0.20, P < 0.001).

Nitrogen content per dry mass (NM) declined with increasing leaf age (Table 1). According to the ancovas, at a common irradiance, MA and DF were larger in older leaves, while NA was independent of leaf age (Table 1). Thus, age-dependent increases in MA exactly balanced the age-dependent decrease in NM such that NA remained constant.

Table 1.  Age-dependent variation in leaf structural and physiological characteristics in Quercus ilex leaves (averages ± SE)a
CharacteristicbLeaf age (year)
012≥ 3
  • a

    Means at common irradiance were compared by common slope covariance analyses (ancova) when seasonal average daily integrated quantum flux density (Qint) significantly affected the specific variable according to regression analyses. When Qint was non-significant (P > 0.05), the means were compared by analysis of variance (anova) (for NM and RdM). ancovas were conducted with Log(Qint) (Figs 2, 3, 4 & 6), except for CO2 drawdown from intercellular air space (ΔC) where non-transformed Qint was used as a covariate (Fig. 8). ancovas were conducted with both the leaf current (QintC) and growth light (QintG, Fig. 1) as covariates. After conduction of ancova or anova, the means were separated between the age classes by Bonferroni test. Means with the same letter are not significantly different (P > 0.05). Whenever the ancovas with QintC and QintG yielded different results with respect to the statistical significance, two letters denoting statistical significance are depicted and correspond to: difference according to ancova with QintC/difference according to ancova with QintG.

  • b

    MA, leaf dry mass per unit area; DF, leaf dry to fresh mass ratio; NM, leaf N content per dry mass; NA, leaf N content per area; Vcmax, maximum carboxylase activity of ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) per unit area (VcmaxA), and per unit dry mass (VcmaxM); Jmax, capacity for photosynthetic electron transport per unit area (JmaxA) and per unit dry mass (JmaxM); Rd, rate of non-photorespiratory respiration in light per unit area (RdA) and per unit dry mass (RdM); PB, apparent fraction of leaf N in bioenergetics (Eqn 5); PR, apparent fraction of leaf N in Rubisco (Eqn 6); gS, stomatal conductance to water vapour per unit area; Ci/Ca, ratio of intercellular (Ci) to ambient (Ca) CO2 concentrations for average Ca values of 330–340 µmol mol−1; gm, internal diffusion conductance (Eqn 3) per unit area (gmA) and per unit dry mass (gmM); ΔC, CO2 drawdown from intercellular air space (average ± SE Ci was 215 ± 3 µmol mol−1) to chloroplasts.

MA (g m−2)150 ± 7a/a153 ± 7a/ab171 ± 7b/bc170 ± 12b/c
DF (g g−1)0.545 ± 0.006a0.579 ± 0.006b0.605 ± 0.006c0.614 ± 0.010c
NM (%)1.792 ± 0.043a1.606 ± 0.038b1.501 ± 0.043bc1.40 ± 0.06c
NA (g m−2)2.68 ± 0.13a2.45 ± 0.11a2.57 ± 0.14a2.32 ± 0.13a
VcmaxA (µmol m−2 s−1)53.4 ± 2.5a/a45.5 ± 4.1a/ab41.0 ± 3.3a/b34.9 ± 3.2a/b
VcmaxM (µmol g−1 s−1)0.358 ± 0.010a/a0.295 ± 0.022ac/a0.237 ± 0.016b/b0.207 ± 0.017bc/b
JmaxA (µmol m−2 s−1)129 ± 6a/a103 ± 8ab/b94 ± 6b/bc76 ± 7ab/c
JmaxM (µmol g−1 s−1)0.863 ± 0.028a0.678 ± 0.035b0.553 ± 0.030c0.450 ± 0.036c
RdA (µmol m−2 s−1)1.50 ± 0.07a1.28 ± 0.09a1.35 ± 0.10a1.12 ± 0.10a
RdM (nmol g−1 s−1)10.17 ± 0.46a8.4 ± 0.5ab8.0 ± 0.6b6.61 ± 0.41b
PR (g g−1)0.1569 ± 0.0040a/a0.141 ± 0.009ab/ab0.123 ± 0.007b/b0.115 ± 0.007ab/b
PB (g g−1)0.0387 ± 0.0010a/a0.0333 ± 0.0015ab/b0.0296 ± 0.0015bc/b0.0255 ± 0.0015c/c
gS (mol m−2 s−1)0.0530 ± 0.0039a0.0434 ± 0.0048a0.041 ± 0.005a0.047 ± 0.007a
Ci/Ca0.600 ± 0.010a0.633 ± 0.013a0.653 ± 0.016ab0.692 ± 0.017b
gmA (mol m−2 s−1)0.0559 ± 0.0016a0.0370 ± 0.0021b0.0290 ± 0.0018c0.0248 ± 0.0021c
gmM (mmol m−2 s−1)0.384 ± 0.013a0.250 ± 0.013b0.175 ± 0.013c0.154 ± 0.016c
ΔC (µmol mol−1)82.8 ± 4.1a101.3 ± 3.3b117.4 ± 3.8c122.6 ± 3.8c

Foliage photosynthetic potentials, and respiration rate in relation to current and previous irradiance and age

Maximum Rubisco carboxylase activity per unit area (VcmaxA; Fig. 3a & b) and the capacity for photosynthetic electron transport per unit area (JmaxA; Fig. 3c & d) increased with increasing both the current (Fig. 3a & c) and previous (Fig. 3b & d) integrated light. While these correlations were stronger for older age classes with the growth light (cf. Fig. 3b & d versus Fig. 3a & c), data for all age classes pooled tended to fit single uniform relationships with QintC (r2 = 0.57 for Fig. 3a and r2 = 0.63 for Fig. 3c, P < 0.001 for both). ancovas demonstrated that VcmaxA versus QintC relationships did not differ among age classes, and there was only a single significant difference at a common QintC between the JmaxA values in current-year and 2-year-old leaves (Table 1). The photosynthetic potentials generally decreased with increasing leaf age at common QintG (Table 1), and the relationships of VcmaxA (Fig. 3b, r2 = 0.50, P < 0.001) and JmaxA with QintG (Fig. 3d, r2 = 0.45, P < 0.001) were also more scattered when all data were pooled, overall indicating that the photosynthetic potentials were adjusted to current rather than to previous light regime. This suggestion was further substantiated by strong correlations between VcmaxA (r2 = 0.68) and JmaxA (r2 = 0.88, P < 0.001 for both) with QintC in the truncated data set including 22 various-aged leaves for which QintC and QintG were most poorly related (studentized residual in the QintG versus QintC ≥ 1.0). In this truncated data set, VcmaxA and QintG were not correlated (r2 = 0.13, P > 0.1), and the correlation with JmaxA was weak (r2 = 0.31, P < 0.02).

image

Figure 3.  Maximum ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylase activity per unit area (VcmaxA; a & b) and the capacity for photosynthetic electron transport per unit area (JmaxA; c & d) in relation to current (QintC; a & c) and growth (QintG; b & d) light in various-aged leaves of Quercus ilex. Symbols and data fitting as in Fig. 2. The comparison of intercepts of y = a + blog(x) relations is shown in Table 1.

Download figure to PowerPoint

Given that MA was more strongly associated with previous light regime (Fig. 2a & b), these differences in adjustment of area-based photosynthetic potentials to QintC and QintG mainly reflected the contrasting light response of mass-based photosynthetic potentials. Both VcmaxM and JmaxM decreased with increasing leaf age (Table 1), but VcmaxM (r2 = 0.19, P < 0.001 for all age classes pooled) and JmaxM (r2 = 0.21, P < 0.001) were more strongly associated with QintC than with QintG (r2 = 0.05, P < 0.05 for VcmaxM and r2 = 0.02, P > 0.2 for JmaxM). The increase of mass-based photosynthetic potentials resulted from a greater fraction of N in bioenergetics (PB, Eqn 5; for all data pooled, r2 = 0.23, P < 0.001) and in Rubisco (PR, Eqn 6, r2 = 0.19, P < 0.001) at higher QintC. The fractional investments of foliar N in bioenergetics and Rubisco were weakly associated with QintG (for all data pooled, r2 = 0.04, P > 0.09 for PB and r2 = 0.08, P < 0.01 for PR).

Non-photorespiratory respiration rates in light per unit area (RdA) scaled positively with QintC (Fig. 4a) and QintG (Fig. 4b) for most leaf age classes, except for the oldest leaves. Values of RdA did not differ at common QintC and QintG (Table 1), but as with photosynthetic potentials, the relationship tended to be stronger with QintC (Fig. 4a, r2 = 0.39, P < 0.001) than with QintG (r2 = 0.33, P < 0.001). In the truncated data set with 22 various-aged leaves, RdA was correlated with QintC (r2 = 0.27, P < 0.02), but not with QintG (r2 = 0.14, P > 0.09).

image

Figure 4. Rate of non-photorespiratory respiration continuing in the light (RdA) in relation to current (QintC; a) and growth (QintG; b) light in various-aged leaves of Quercus ilex. Non-significant regressions (P > 0.05) are depicted by dashed lines. Symbols and data fitting as in Fig. 2. Intercept differences of y = a + blog(x) relations are demonstrated in Table 1.

Download figure to PowerPoint

Mass-based Rd values (RdM) were independent of light in all cases, and tended to decrease with increasing leaf age (Table 1). As RdA was independent of age, this RdM decrease primarily resulted from age-dependent dilution of leaf N and physiologically active compounds with accumulation of structural carbon (RdA = RdMMA).

Age-dependent relationships between nitrogen and photosynthetic potentials

For all data pooled, leaf photosynthetic potentials were significantly correlated with foliage nitrogen content (Fig. 5). These correlations were somewhat stronger for area-based relations (for all data pooled, r2 = 0.63 for VcmaxA, and r2 = 0.57 for JmaxM, P < 0.001 for both) than for mass-based relations (for all data pooled, r2 = 0.49 for VcmaxM, and r2 = 0.55 for JmaxM, P < 0.001 for both), especially for within-age-class correlations (cf. Fig. 5a & c versus Fig. 5b & d). The larger r2 values of area-based relations within age classes reflect the circumstance that these relations arise because of concomitant variations in photosynthetic potentials and NA with light. For all data pooled, the correlations were weaker than within-age classes, because age affected only photosynthetic potentials but not NA (Table 1, Figs 2c, d & 3).

image

Figure 5. Correlations of the area- and mass-based maximum ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylase activities (Vmax; a & b) and the capacities for photosynthetic electron transport (Jmax; c & d) with foliage N content per unit area (NA; a & c) and per unit dry mass (NM; b & d) in various-aged leaves of Quercus ilex. Non-significant regression (P > 0.05) in (d) is depicted by a dashed line. Data were fitted by linear regressions. Symbols as in Fig. 2.

Download figure to PowerPoint

Overall, for area and mass-based relations, common slope ancovas demonstrated that photosynthetic potentials tended to be larger for younger leaves at common nitrogen content (P < 0.001 for the age effect in all relations). This age-dependent difference of foliage photosynthetic potentials at common N content reflects decreases in apparent nitrogen investments in bioenergetics and Rubisco with increasing leaf age (Table 1).

Light- and age-driven modifications in internal diffusion conductance and the correlations with photosynthetic potentials

The internal diffusion conductance per unit area (gmA) scaled positively with both QintC (Fig. 6a) and QintG (Fig. 6b) for current-year and 1-year-old leaves, while it was independent of light for older age classes. The internal conductance per unit dry mass (gmM = gmA/MA) was negatively associated with QintC (Fig. 6c) and QintG (Fig. 6d) for current-year leaves and independent of light for older leaves. There was a strong age-dependent decline in both gmA and gmM (Fig. 6, Table 1).

image

Figure 6. Relationships of internal diffusion conductance (gm, Eqn 3) per unit area (gmA; a & b) and per unit dry mass (gmM; c & d) with current (QintC; a & c) and growth (QintG; b & d) light in various-aged leaves of Quercus ilex. Symbols and data fitting as in Fig. 2. Non-significant regressions (P > 0.05) are depicted by dashed lines. Table 1 provides the comparison of intercepts of y = a + blog(x) relations.

Download figure to PowerPoint

Both area- (Fig. 7a) and mass-based (Fig. 7b) gm and photosynthetic potentials were correlated. For JmaxA versus gmA, the r2 = 0.45–0.65 for different age classes (P < 0.001 for all), and for JmaxM versus gmM, r2 = 0.25–0.49 (P < 0.01 for all). The correlations tended to be curvilinear, especially for area-based relations (Fig. 7a). According to common slope ancovas (using linear approximations for the relationships between photosynthetic potentials and diffusion conductance), gm was larger in younger leaves at a common Vcmax (Fig. 7) and Jmax (data now shown) for both area- and mass-based relations (P < 0.001 for the age effect in all cases). At common VcmaxA and JmaxA, the values of gmA, and at common VcmaxM, the values of gmM were significantly different among age classes for most comparisons (Fig. 7, P < 0.01 for all), except for the comparisons of 2-year-old and ≥ 3-year-old leaves (P > 0.9). At a common JmaxM, the values of gmM did not differ among 1-year-old and ≥ 3-year-old leaves and 2-year-old and ≥ 3-year-old leaves (P > 0.9).

image

Figure 7. Internal diffusion conductance per unit area (gmA) in relation to maximum ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco) carboxylase activity per unit area (VcmaxA; a) and internal diffusion conductance per unit dry mass (gmM) in relation to Rubisco carboxylase activity per unit mass (VcmaxM; b) in various-aged leaves of Quercus ilex. Data were fitted by non-linear equations in the form of y = a + blog(x), except for current-year leaves in (b), where a linear regression provided a larger fraction of explained variance (r2). Symbols as in Fig. 2.

Download figure to PowerPoint

Diffusion limitations of photosynthesis in relation to light and age

Average stomatal conductance at an average ambient CO2 concentration of 335 µmol mol−1 and saturating light was weakly associated with irradiance (for linear relationships with all data pooled, r2 = 0.20 for QintC, and r2 = 0.16, for QintG, P < 0.001 for both). Stomatal conductance did not depend on leaf age, but the intercellular to ambient CO2 concentration ratio (Ci/Ca) tended to be larger in older leaves because of age-dependent decreases in leaf photosynthetic potentials and corresponding lower CO2 drawdown from ambient air to internal air space.

The CO2 drawdown from internal air space to chloroplasts (ΔC = Ci − Cc) at saturating light of 1000 µmol m−2 s−1 and internal CO2 concentrations of 180–250 µmol mol−1 scaled positively with QintC for current-year and 1-year-old leaves (Fig. 8a), and with increasing QintG in all leaf age classes, except for the oldest leaves (Fig. 8b).

image

Figure 8. Variation in drawdown of CO2 concentration from leaf internal air space (Ci) to carboxylation sites in chloroplasts (Cc; ΔC =Ci − Cc) with current (QintC; a) and growth (QintG; b) light in various-aged leaves of Quercus ilex. Symbols as in Fig. 2. Non-significant regressions (P > 0.05) are depicted by dashed lines. Data were fitted by linear regressions. Table 1 provides the comparison of intercepts of these regressions.

Download figure to PowerPoint

At a common light, ΔC (Fig. 8, Table 1) increased with increasing leaf age, demonstrating overall greater internal diffusion limitations in older leaves that resulted in a shift in chloroplastic CO2 concentrations from c. 125–140 µmol mol−1 in intermediate-light-exposed current-year leaves to c. 90–100 µmol mol−1 in intermediate-light-exposed ≥ 3-year-old leaves. The patterns with the ratio of chloroplastic (Cc) to intercellular (Ci) CO2 concentrations were analogous to those with ΔC. Cc/Ci ratio increased with light and decreased with age from 0.59 in current-year leaves to 0.46 in oldest leaves.

Leaf structure, internal diffusion conductance and CO2 drawdown

For all data pooled, the internal diffusion conductance per unit area, gmA was not correlated with MA (r2 = 0.00, P > 0.9), while it was negatively related to DF (r2 = 0.22, P < 0.001). The internal diffusion conductance per unit dry mass was negatively associated with both MA (r2 = 0.24, P < 0.001) and DF (Fig. 9a).

image

Figure 9. Correlations of internal diffusion conductance per unit dry mass (gmM; a) and CO2 drawdown from internal air space to chloroplasts (ΔC; b) with leaf dry to fresh mass ratio (DF) and the relationship between ΔC and gmM (c) for various-aged leaves of Quercus ilex. Symbols as in Fig. 2. Linear regressions were fitted to all data pooled.

Download figure to PowerPoint

The CO2 drawdown from internal air space to chloroplasts was positively correlated with both MA (r2 = 0.30, P < 0.001) and DF (Fig. 9b). Analogous but negative correlations were observed between Cc/Ci ratio and MA (r2 = 0.39, P < 0.001) and DF (r2 = 0.37, P < 0.001). ΔC (Fig. 9c) and Cc/Ci ratio (r2 = 0.34, P < 0.001) were more strongly associated with gmM than with gmA (r2 = 0.23, P < 0.001 for the correlation with ΔC and r2 = 0.11, P < 0.002 for correlation with Cc/Ci).

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

Influences of previous and current leaf light environment on foliage structure and chemistry

Leaf light availability strongly decreased with increasing leaf age, and for older leaf age classes, the current and growth irradiances were poorly associated (Fig. 1) as consistently observed in crowns of evergreen trees (Brooks et al. 1996; Niinemets 1997; Warren & Adams 2001). To evaluate whether foliar characteristics scale more strongly with current (QintC) or growth (QintG) irradiance, we used several criteria based on the fraction of explained variance, and deviation of age-specific regressions from the average regression with all data pooled. In addition, a truncated data set was developed with QintGQintC pairs diverging most from the regression line of QintG versus QintC (Fig. 1).

In our study, foliage dry mass per unit area (MA, Fig. 2a, b), dry to fresh mass ratio (DF) and nitrogen content per area (NA, Fig. 2c, d) were more strongly associated with growth than with current long-term irradiance. These results agree with previous data indicating that structural characteristics of fully mature leaves are only weakly modified by changes in irradiance in deciduous species (Naidu & DeLucia 1997; Niinemets et al. 2003a; Oguchi et al. 2003, 2005) and in evergreen conifers (Brooks et al. 1994, 1996; Niinemets 1997; Warren & Adams 2001).

Light-dependent increases in MA in broad-leaved evergreens can be associated with both variations in leaf density (ρ) and thickness (T; MA = ρT) (Niinemets, Valladares & Ceulemans 2003b). Previous studies in the vicinity of our site have demonstrated a significant increase of T with increasing light availability in the canopy of Q. ilex (Wagner, Pelaez Menendez & Larcher 1993). However, increases in DF are compatible with accumulation of cell wall materials (Garnier & Laurent 1994), and light-dependent increases in DF are also strongly correlated with ρ in broad-leaved evergreens (Niinemets et al. 2003b). Thus, this evidence suggests that both increases in the thickness and density were responsible for the scaling of MA with light.

Increases in leaf age resulted in significant increases in MA and DF at a common light, and overall decreases in leaf nitrogen contents per area as is typical in evergreen species (Gratani & Bombelli 2000; Warren & Adams 2001; Niinemets & Lukjanova 2003; Niinemets et al. 2004). Importantly, foliar nitrogen contents per area were independent of age at common irradiance (Fig. 2c & d), demonstrating that changes in NM resulted from dilution of nitrogen concentrations because of the accumulation of cell walls and structural compounds, but not from net resorption of N from aging leaves. This result confirms the relative time independence of N contents per unit area in Mediterranean species with long-living foliage (Niinemets et al. 2004, 2005). In addition to Mediterranean species, similar age independence of NA has been observed in some evergreen conifers (Brooks et al. 1994, 1996); but in others, NA strongly declines with needle age (Warren & Adams 2001).

Adjustment of foliage photosynthetic potentials (Jmax and Vcmax) to light in various-aged leaves

Differently from leaf morphology and NA, area-based foliage photosynthetic potentials (VcmaxA and JmaxA, Fig. 3) and non-photorespiratory respiration rates in light (Fig. 4) were more strongly associated with current leaf light environment (cf. Fig. 3a & c versus Fig. 3b, d and Fig. 4a versus Fig. 4b). Acclimation to current leaf light environment was reflected in scaling of mass-based photosynthetic potentials (VcmaxM and JmaxM) as well as the apparent nitrogen fractions in proteins limiting photosynthetic electron transport (PB, Eqn 5) and in Rubisco (PR, Eqn 6) with QintC but not with QintG.

Overall, foliage acclimation to light gradients in deciduous canopies is mainly controlled by changes in MA and leaf thickness that control the possible amount of photosynthetic machinery per unit leaf area (Gutschick & Wiegel 1988; Niinemets & Tenhunen 1997; Evans & Poorter 2001). Several studies have reported a certain acclimation of foliage photosynthetic capacity of fully mature leaves after increases in leaf light availability (Sims & Pearcy 1989; Naidu & DeLucia 1997; Oguchi et al. 2003, 2005). Such acclimation was primarily achieved by enhanced nitrogen investments in components of photosynthetic machinery and increases in the number of chloroplasts per unit leaf surface area (Sims & Pearcy 1989; Oguchi et al. 2003, 2005).

In evergreens, where the older foliage becomes gradually overtopped by the new foliage, foliage photosynthetic capacities and respiration rates decline with decreasing leaf age (our study, Brooks et al. 1994, 1996). Given that at low light, lower investment of N in photosynthetic machinery is required for light saturation, such a decrease can be considered adaptive if N is partitioned in chlorophyll- and light-harvesting proteins that improve light harvesting at low irradiance. Previous work in conifers indicates that the fractional investment of N in light harvesting does increase in shaded older foliage at the expense of N in capacity-limiting proteins (Brooks et al. 1994, 1996). As our study demonstrates, the decrease in area-based photosynthetic potentials in older leaves of Q. ilex is not only associated with a dilution of leaf N and resulting decreases in mass-based capacities, but also with lower fractions of N in Rubisco and bioenergetics (Eqns 5 & 6 and Table 1), leading to lower values of area- and mass-based photosynthetic potentials at common foliar N content (Fig. 4). Thus, these observations agree with the overall hypothesis of shade-driven adjustments in foliage resource partitioning and modification in photosynthetic potentials. However, it is also important that in Mediterranean species, PR and PB may also decrease with increasing leaf age in fully illuminated branches where age and light are not related (Niinemets et al. 2005), possibly because of greater fractions of N associated with cell walls in older leaves. As light and age inherently covaried in our study, conclusive separation of age- and shading-driven modifications that lead to down-regulation of photosynthetic capacities will require further experimentation.

Scaling of internal diffusion conductance with light: greater internal diffusion limitations in higher irradiance

Few studies have investigated area-based internal diffusion conductance (gmA) in relation to long-term leaf light environment (Hanba, Kogami & Terashima 2002; Piel et al. 2002; Warren et al. 2003). Even in these studies, gmA could be investigated at a few canopy levels or growth chamber irradiances. These studies have demonstrated an increase of gmA with increasing light from canopy bottom to top on the order of 20–100% (Piel et al. 2002; Warren et al. 2003). We have developed the full response curves of gmA versus integrated canopy light for four leaf age classes, and found a limited increase in gmA for only the two youngest leaf age classes (Fig. 6a & b). Given that the liquid-phase conductance for CO2 diffusion scales positively with exposed mesophyll to total surface are ratio and chloroplast to total leaf surface area ratio (Evans et al. 1994; Evans & von Caemmerer 1996; Hanba et al. 2001; Terashima et al. 2005), and that these ratios scale with leaf thickness (Nobel 1977; Hanba, Miyazawa & Terashima 1999), light-dependent modification in leaf thickness and number of chloroplasts per unit area in leaves with greater photosynthetic capacity provide the explanation of positive effects of light on gmA in our study.

However, gmA levelled off at highest irradiance (Fig. 6a & b) and mass-based diffusion conductance (gmM = gmA/MA) of current-year leaves was negatively associated with irradiance (Fig. 6c & d). In fact, the chloroplastic (Cc) to internal (Ci) CO2 concentration (Cc/Ci) ratio scaled negatively (data not shown) and the CO2 drawdown from intercellular air space to chloroplasts (ΔC = Ci − Cc) scaled positively (Fig. 8a & b) with integrated light, indicating greater internal CO2 diffusion limitations at higher light. How to reconcile the apparent discrepancies in light responses in gmA and gmM, and the overall effect on chloroplastic CO2 concentration? Firstly, the increase in gmA with light was only c. 1.4-fold for the entire light gradient (Fig. 6a & b), while VcmaxA and JmaxA more than doubled along the light gradient (Fig. 3), implying a greater photosynthetic capacity relative to gm, and accordingly a greater ΔC. Secondly, chloroplastic CO2 concentration is a leaf volume-based estimate and accordingly, Cc/Ci and ΔC should scale with gmM rather than with gmA (Niinemets et al. 2005; Niinemets & Sack 2006) as was confirmed by our study (Fig. 9c). Although the light-dependent change in gm was moderate and most pronounced for the lower canopy (irradiance range of 0–10 mol m−2 d−1, Fig. 6), this evidence collectively demonstrates that the variation (and constancy) in gm observed in our study may significantly influence the carbon gain of different canopy layers.

This negative scaling of gmM with irradiance likely reflects the increase of DF and tissue density with increasing irradiance. Because of a greater evaporative demand in upper canopy, leaves exposed to higher irradiances need to sustain lower leaf water potentials, and have often thicker and more strongly lignified cell walls (Niinemets & Kull 1998), resulting in increased tissue density and reduced efficiency of internal diffusion. It has been previously shown that in Q. ilex, the overall fraction of cell wall materials in current-year leaves, is larger in leaves with greater MA (Rambal et al. 1996). The role of structural characteristics is further underscored by stronger relationships of ΔC with leaf growth irradiance than with current light in various-aged leaves (cf. Fig. 8a versus Fig. 8b). Without growth- and light-driven structural changes (Fig. 2), adjustment of foliage photosynthetic potentials of various-aged leaves to current light (Fig. 3) would have also led to stronger correlations of internal diffusion limitations with current light.

Age-dependent changes in internal diffusion conductance: implications of non-coordinated changes in gm and photosynthetic capacity

The range of gmA values of 0.04–0.10 mol m−2 s−1 in current-year leaves (Fig. 6, Table 1) agrees with previous observations of 0.05–0.2 mol m−2 s−1 in current-year leaves of Mediterranean species (Loreto et al. 1992; Centritto et al. 2003; Niinemets et al. 2005), but is low compared with gmA values of 0.15–0.8 mol m−2 s−1 in herbs and temperate deciduous trees (Evans & Loreto 2000; Flexas et al. 2004; Terashima et al. 2005). Increases in leaf age significantly reduced the gm to values as low as 0.017–0.02 mol m−2 s−1 in the oldest leaves (Fig. 6, Table 1). These low internal diffusion conductances in leaves of Mediterranean sclerophylls agree with a previous theoretical estimates of maximum gm = 0.14 mol m−2 s−1 in full-sun exposed thickest current-year leaves of Q. ilex (Niinemets & Reichstein 2003). This theoretical estimate was obtained using a linear diffusion model, and calculating the serial diffusion resistances from the thickness of diffusion path-lengths in internal air space, cell wall, cytosol and chloroplast stroma as well as by accounting for the porosity of cell walls, membrane solubility of CO2 and chloroplast surface to total leaf area (Niinemets & Reichstein 2003).

There is some evidence indicating that gm may decrease rather rapidly after imposition of water stress (Centritto et al. 2003), possibly because of decreases in aquaporin conductance in plant membranes (Terashima & Ono 2002). With our advanced branch pre-conditioning procedure and measurement protocol, we tried to avoid any water stress in leaves. Although the possible physiological effects may significantly modify the values of gm in stressed leaves, it is important to recognize that leaf structural characteristics (porosity, diffusion path-lengths in liquid- and gas-phase) and the number of chloroplasts per unit leaf surface area provide important structural limits for the maximum values of gm that can further possibly be fine-tuned by changes in aquaporin conductance.

As the maximum values of gm scale with the mesophyll to total surface area and chloroplast to total surface area ratios, the species and life-form differences in gmA values have been associated with differences in foliage photosynthetic potentials. Several studies show a significant positive relationship between foliage photosynthetic capacity and gmA (von Caemmerer & Evans 1991; Loreto et al. 1992; Evans & von Caemmerer 1996; Evans & Loreto 2000; Piel et al. 2002; Loreto et al. 2003; Flexas et al. 2004; Grassi & Magnani 2005), and it has been suggested that because of a coordination between photosynthetic capacity and gmA, internal diffusion limitations of photosynthesis are similar in leaves with different architecture and photosynthetic capacity (Evans & Loreto 2000). In herbaceous species, age-dependent decreases in senescing leaves were associated with decreases in chloroplast number, resulting in proportional decline in photosynthetic capacity and similar ΔC in leaves of different age (Loreto et al. 1994; Evans & Vellen 1996).

Although photosynthetic potentials also decreased in older Q. ilex leaves in our study (Fig. 3), diffusion conductance was higher at common photosynthetic capacity in younger leaves (Fig. 7), indicating that the age-dependent decrease in photosynthetic potentials was less than in diffusion conductance. This resulted in greater internal diffusion limitations of net assimilation rates in older leaves. The latter argument was further supported by strong age-dependent decline in Cc/Ci ratio and ΔC (Fig. 8, Table 1). Overall, our study (Fig. 9b) and recent evidence demonstrate that the correlation between gm and photosynthetic potentials is variable, and that lower gm is associated with greater diffusion limitations of photosynthesis (Hanba et al. 2001; Niinemets et al. 2005; Warren & Adams 2006; Niinemets & Sack 2006).

Non-compatible adjustments of photosynthetic potentials and gm in our study reflected enhanced fractions of cell walls and increased structural investments with increasing irradiance and light (Fig. 9). General negative correlations between gmA and MA have been observed (Syvertsen et al. 1995; Kogami et al. 2001; Niinemets et al. 2005; Terashima et al. 2005), but these correlations depend on the extent to which increases in MA are associated with modifications in thickness and density (see previous discussion). In Mediterranean oaks, it has been shown that the degree of cell wall lignification and total amount of cell walls increase with increasing leaf age (Damesin, Rambal & Joffre 1998), and similar time-dependent changes in cell wall thickness have been observed in young developing leaves of warm temperate broad-leaved species (Miyazawa & Terashima 2001; Miyazawa, Makino & Terashima 2003). These data collectively demonstrate that independent structural and physiological modifications during light acclimation and with foliage aging lead to a strong variation in the internal diffusion limitations of photosynthesis in evergreen broad-leaved species.

CONCLUSIONS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

The results of this study demonstrate strong interactive effects of previous and current light availability on foliage structure and function. While the structure has been acclimated to growth irradiance, foliage physiological potentials acclimate to current light. However, structural adjustment to previous irradiance leads to enhanced internal diffusion limitations in older leaves. These results reinforce the observations that foliage photosynthetic potentials and diffusion conductance may largely vary independently (Niinemets et al. 2005; Warren & Adams 2006; Niinemets & Sack 2006), and this independent variation results in different degrees of diffusion limitations of photosynthesis.

Although in Mediterranean evergreen species, old leaves comprise a significant fraction of the total canopy leaf area, large-scale carbon gain models for Mediterranean ecosystems are generally parameterized with current-year foliage structural, chemical and physiological data and do not consider acclimation of foliage characteristics to light environment (e.g. Infante, Rambal & Joffre 1997; Rambal et al. 2003). In addition, generality of leaf nitrogen versus photosynthesis relationships (e.g. Wright et al. 2004) has often been implemented in parameterization of large scale models on the basis of foliage nitrogen content (Running & Hunt 1993; Woodward, Lomas & Lee 2001). Our study provides important evidence of a strong variation in foliage physiological characteristics with long-term light and leaf age, and significant diversification of the relationships between photosynthetic characteristics and N in dependence on leaf age. These results have important implications for canopy-level photosynthesis model parameterization and scaling, and future simulation analyses.

ACKNOWLEDGMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

We thank the Estonian Science Foundation (Grant 5702), the Estonian Ministry of Education and Science (Grant 0182468As03) the Estonian Academy of Sciences and the Province of Trento, Italy (Grant DL3402) for financial support.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix
  • Baldocchi D.D., Wilson K.B. & Gu L. (2002) How the environment, canopy structure and canopy physiological functioning influence carbon, water and energy fluxes of a temperate broad-leaved deciduous forest – an assessment with the biophysical model CANOAK. Tree Physiology 22, 10651077.
  • Bernacchi C.J., Portis A.R., Nakano H., Von Caemmerer S. & Long S.P. (2002) Temperature response of mesophyll conductance. Implications for the determination of Rubisco enzyme kinetics and for limitations to photosynthesis in vivo. Plant Physiology 130, 19921998.
  • Brooks J.R., Hinckley T.M. & Sprugel D.G. (1994) Acclimation responses of mature Abies amabilis sun foliage to shading. Oecologia 100, 316324.
  • Brooks J.R., Sprugel D.G. & Hinckley T.M. (1996) The effects of light acclimation during and after foliage expansion on photosynthesis of Abies amabilis foliage within the canopy. Oecologia 107, 2132.
  • Von Caemmerer S. & Evans J.R. (1991) Determination of the average partial pressure of CO2 in chloroplasts from leaves of several C3 plants. Australian Journal of Plant Physiology 18, 287305.
  • Centritto M., Loreto F. & Chartzoulakis K. (2003) The use of low [CO2] to estimate diffusional and non-diffusional limitations of photosynthetic capacity of salt-stressed olive saplings. Plant, Cell and Environment 26, 585594.
  • Damesin C., Rambal S. & Joffre R. (1998) Seasonal and annual changes in leaf δ13C in two co-occurring Mediterranean oaks: relations to leaf growth and drought progression. Functional Ecology 12, 778785.
  • DeLucia E.H., Whitehead D. & Clearwater M.J. (2003) The relative limitation of photosynthesis by mesophyll conductance in co-occurring species in a temperate rainforest dominated by the conifer Dacrydium cupressinum. Functional Plant Biology 30, 11971204.
  • Edwards G.E. & Baker N.R. (1993) Can CO2 assimilation in maize leaves be predicted accurately from chlorophyll fluorescence analysis? Photosynthesis Research 37, 89102.
  • Eichelmann H., Oja V., Rasulov B., Padu E., Bichele I., Pettai H., Möls T., Kasparova I., Vapaavuori E. & Laisk A. (2004) Photosynthetic parameters of birch (Betula pendula Roth.) leaves growing in normal and in CO2- and O3- enriched atmospheres. Plant, Cell and Environment 27, 479495.
  • Ethier G.J. & Livingston N.J. (2004) On the need to incorporate sensitivity to CO2 transfer conductance into Farquhar – von Caemmerer – Berry leaf photosynthesis model. Plant, Cell and Environment 27, 137153.
  • Evans J.R. & Von Caemmerer S. (1996) Carbon dioxide diffusion inside leaves. Plant Physiology 110, 339346.
  • Evans J.R. & Loreto F. (2000) Acquisition and diffusion of CO2 in higher plant leaves. In Photosynthesis: Physiology and Metabolism (eds R.C.Leegood, T.D.Sharkey & S.Von Caemmerer), pp. 321351. Kluwer Academic Publishers, Dordrecht, The Netherlands.
  • Evans J.R. & Poorter H. (2001) Photosynthetic acclimation of plants to growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell and Environment 24, 755767.
  • Evans J.R. & Vellen L. (1996) Wheat cultivars differ in transpiration efficiency and CO2 diffusion inside their leaves. In Crop Research in Asia: Achievements and Perspectives. Proceedings of the 2nd Asian Crop Science Conference ‘Toward the Improvement of Food Production under Steep Population Increase and Global Environment Change’ 21–23 August, 1995, the Fukui Prefectural University, Fukui, Japan (eds R.Ishii & T.Horie), pp. 326329. Crop Science Society of Japan, Asian Crop Science Association (ACSA), Fukui, Japan.
  • Evans J.R. & Vogelmann T.C. (2003) Profiles of 14C fixation through spinach leaves in relation to light absorption and photosynthetic capacity. Plant, Cell and Environment 26, 547560.
  • Evans J.R., Jakobsen I. & Ögren E. (1993) Photosynthetic light-response curves. 2. Gradients of light absorption and photosynthetic capacity. Planta 189, 191200.
  • Evans J.R., Von Caemmerer S., Setchell B.A. & Hudson G.S. (1994) The relationship between CO2 transfer conductance and leaf anatomy in transgenic tobacco with a reduced content of Rubisco. Australian Journal of Plant Physiology 21, 475495.
  • FAO-UNESCO (1988) Soil Map of the World. Revised legend. FAO-UNESCO-ISRIC, Rome, Italy.
  • Flexas J., Bota J., Loreto F., Cornic G. & Sharkey T.D. (2004) Diffusive and metabolic limitations to photosynthesis under drought and salinity in C3 plants. Plant Biology 6, 269279.
  • Van Gardingen P.R., Jackson G.E., Hernandez-Daumas S., Russell G. & Sharp L. (1999) Leaf area index estimates obtained for clumped canopies using hemispherical photography. Agricultural and Forest Meteorology 94, 243257.
  • Garnier E. & Laurent G. (1994) Leaf anatomy, specific mass and water content in congeneric annual and perennial grass species. The New Phytologist 128, 725736.
  • Grassi G. & Magnani F. (2005) Stomatal, mesophyll conductance and biochemical limitations to photosynthesis as affected by drought and leaf ontogeny in ash and oak trees. Plant, Cell and Environment 28, 834849.
  • Gratani L. & Bombelli A. (2000) Correlation between leaf age and other leaf traits in three Mediterranean maquis shrub species: Quercus ilex, Phillyrea latifolia and Cistus incanus. Environmental and Experimental Botany 43, 141153.
  • Grossoni P., Bussotti F., Tani C., Gravano E., Santarelli S. & Bottacci A. (1998) Morpho-anatomical alterations in leaves of Fagus sylvatica L. & Quercus ilex L. in different environmental stress conditions. Chemosphere 36, 919924.
  • Gutschick V.P. & Wiegel F.W. (1988) Optimizing the canopy photosynthetic rate by patterns of investment in specific leaf mass. American Naturalist 132, 6786.
  • Hale S.E. & Edwards C. (2002) Comparison of film and digital hemispherical photography across a wide range of canopy densities. Agricultural and Forest Meteorology 112, 5156.
  • Hanba Y.T., Miyazawa S.I. & Terashima I. (1999) The influence of leaf thickness on the CO2 transfer conductance and leaf stable carbon isotope ratio for some evergreen tree species in Japanese warm-temperate forests. Functional Ecology 13, 632639.
  • Hanba Y.T., Miyazawa S.I., Kogami H. & Terashima I. (2001) Effects of leaf age on internal CO2 transfer conductance and photosynthesis in tree species having different types of shoot phenology. Australian Journal of Plant Physiology 28, 10751084.
  • Hanba Y.T., Kogami H. & Terashima I. (2002) The effect of growth irradiance on leaf anatomy and photosynthesis in Acer species differing in light demand. Plant, Cell and Environment 25, 10211030.
  • Harley P.C., Loreto F., Di Marco G. & Sharkey T.D. (1992) Theoretical considerations when estimating the mesophyll conductance to CO2 flux by analysis of the response of photosynthesis to CO2. Plant Physiology 98, 14291436.
  • Infante J.M., Rambal S. & Joffre R. (1997) Modelling transpiration in holm-oak savannah: scaling up from the leaf to the tree scale. Agricultural and Forest Meteorology 87, 273289.
  • Jonckheere I., Nackaerts K., Muys B. & Coppin P. (2005) Assessment of automatic gap fraction estimation of forests from digital hemispherical photography. Agricultural and Forest Meteorology 132, 96114.
  • Jordan D.B. & Ogren W.L. (1984) The CO2/O2 specificity of ribulose 1,5-bisphosphate carboxylase/oxygenase. Dependence on ribulosebisphosphate concentration, pH and temperature. Planta 161, 308313.
  • Kogami H., Hanba Y.T., Kibe T., Terashima I. & Masuzawa T. (2001) CO2 transfer conductance, leaf structure and carbon isotope composition of Polygonum cuspidatum leaves from low and high altitudes. Plant, Cell and Environment 24, 529538.
  • Laisk A. & Oja V. (1998) Dynamics of Leaf Photosynthesis: Rapid-Response Measurements and Their Interpretations. CSIRO Publishing, Canberra, Australia.
  • Li-Cor Inc (1992) LAI-2000 Plant Canopy Analyzer. Operating manual. Li-Cor, Inc., Lincoln, NE, USA.
  • Li-Cor Inc. (2002) Using the LI-6400 Portable Photosynthesis System. Version 5. Li-Cor, Inc., Lincoln, NE, USA.
  • Loreto F., Harley P.C., Di Marco G. & Sharkey T.D. (1992) Estimation of mesophyll conductance to CO2 flux by three different methods. Plant Physiology 98, 14371443.
  • Loreto F., Di Marco G., Tricoli D. & Sharkey T.D. (1994) Measurements of mesophyll conductance, photosynthetic electron transport and alternative electron sinks of field grown wheat leaves. Photosynthesis Research 41, 397403.
  • Loreto F., Centritto M. & Chartzoulakis K. (2003) Photosynthetic limitations in olive cultivars with different sensitivity to salt stress. Plant, Cell and Environment 26, 595601.
  • Meir P., Kruijt B., Broadmeadow M., Barbosa E., Kull O., Carswell F., Nobre A. & Jarvis P.G. (2002) Acclimation of photosynthetic capacity to irradiance in tree canopies in relation to leaf nitrogen concentration and leaf mass per unit area. Plant, Cell and Environment 25, 343357.
  • Miyazawa S.I. & Terashima I. (2001) Slow development of leaf photosynthesis in an evergreen broad-leaved tree, Castanopsis sieboldii: relationships between leaf anatomical characteristics and photosynthetic rate. Plant, Cell and Environment 24, 279291.
  • Miyazawa S.I., Makino A. & Terashima I. (2003) Changes in mesophyll anatomy and sink-source relationships during leaf development in Quercus glauca, an evergreen tree showing delayed leaf greening. Plant, Cell and Environment 26, 745755.
  • Naidu S.L. & DeLucia E.H. (1997) Acclimation of shade-developed leaves on saplings exposed to late-season canopy gaps. Tree Physiology 17, 367376.
  • Niinemets Ü. (1997) Acclimation to low irradiance in Picea abies: influences of past and present light climate on foliage structure and function. Tree Physiology 17, 723732.
  • Niinemets Ü. & Kull O. (1998) Stoichiometry of foliar carbon constituents varies along light gradients in temperate woody canopies: implications for foliage morphological plasticity. Tree Physiology 18, 467479.
  • Niinemets Ü. & Lukjanova A. (2003) Needle longevity, shoot growth and branching frequency in relation to site fertility and within-canopy light conditions in Pinus sylvestris. Annals of Forest Science 60, 195208.
  • Niinemets Ü. & Reichstein M. (2003) Controls on the emission of plant volatiles through stomata: a sensitivity analysis. Journal of Geophysical Research – Atmospheres 108, 4211. DOI: 10.1029/2002JD002626
  • Niinemets Ü. & Sack L. (2006) Structural determinants of leaf light-harvesting capacity and photosynthetic potentials. In Progress in Botany (eds K.Esser, U.E.Lüttge, W.Beyschlag & J.Murata), pp. 385419. Springer Verlag, Berlin, Germany.
  • Niinemets Ü. & Tenhunen J.D. (1997) A model separating leaf structural and physiological effects on carbon gain along light gradients for the shade-tolerant species Acer saccharum. Plant, Cell and Environment 20, 845866.
  • Niinemets Ü. & Valladares F. (2004) Photosynthetic acclimation to simultaneous and interacting environmental stresses along natural light gradients: optimality and constraints. Plant Biology 6, 254268.
  • Niinemets Ü., Tenhunen J.D., Canta N.R., Chaves M.M., Faria T., Pereira J.S. & Reynolds J.F. (1999) Interactive effects of nitrogen and phosphorus on the acclimation potential of foliage photosynthetic properties of cork oak, Quercus suber, to elevated atmospheric CO2 concentrations. Global Change Biology 5, 455470.
  • Niinemets Ü., Kollist H., García-Plazaola J.I. & Becerril J.M. (2003a) Do the capacity and kinetics for modification of xanthophyll cycle pool size depend on growth irradiance in temperate trees? Plant, Cell and Environment 26, 17871801.
  • Niinemets Ü., Valladares F. & Ceulemans R. (2003b) Leaf-level phenotypic variability and plasticity of invasive Rhododendron ponticum and non-invasive Ilex aquifolium co-occurring at two contrasting European sites. Plant, Cell and Environment 26, 941956.
  • Niinemets Ü., Tenhunen J.D. & Beyschlag W. (2004) Spatial and age-dependent modifications of photosynthetic capacity in four Mediterranean oak species. Functional Plant Biology 31, 11791193.
  • Niinemets Ü., Cescatti A., Rodeghiero M. & Tosens T. (2005) Leaf internal diffusion conductance limits photosynthesis more strongly in older leaves of Mediterranean evergreen broad-leaved species. Plant, Cell and Environment 28, 15521566.
  • Nilson T. (1971) A theoretical analysis of the frequency of gaps in plant stands. Agricultural Meteorology 8, 2538.
  • Nobel P.S. (1977) Internal leaf area and cellular CO2 resistance: photosynthetic implications of variations with growth conditions and plant species. Physiologia Plantarum 40, 137144.
  • Oguchi R., Hikosaka K. & Hirose T. (2003) Does leaf photosynthetic light-acclimation need change in leaf anatomy? Plant, Cell and Environment 26, 505512.
  • Oguchi R., Hikosaka K. & Hirose T. (2005) Leaf anatomy as a constraint for photosynthetic acclimation: differential responses in leaf anatomy to increasing growth irradiance among three deciduous trees. Plant, Cell and Environment 28, 916927.
  • Piel C., Frak E., Le Roux X. & Genty B. (2002) Effect of local irradiance on CO2 transfer conductance of mesophyll in walnut. Journal of Experimental Botany 53, 24232430.
  • Rambal S., Damesin C., Joffre R., Méthy M. & Lo Seen D. (1996) Optimization of carbon gain in canopies of Mediterranean evergreen oaks. Annales des Sciences Forestieres 53, 547560.
  • Rambal S., Ourcival J.M., Joffre R., Mouillot F., Nouvellon Y., Reichstein M. & Rocheteau A. (2003) Drought controls over conductance and assimilation of a Mediterranean evergreen ecosystem: scaling from leaf to a canopy. Global Change Biology 9, 18131824.
  • Ridler W. & Calvard S. (1978) Picture thresholding using an iterative selection method. IEEE Transactions on Systems, Man and Cybernetics, 8, 260263.
  • Running S.W. & Hunt E.R. (1993) Generalization of a forest ecosystem process model for other biomes, BIOME-BGC, and an application for global-scale models. In Scaling Physiological Processes: Leaf to Globe (eds J.R.Ehleringer & C.B.Field), pp. 141158. Academic Press, San Diego, CA, USA.
  • Sala i Serra A. (1992) Water relations, canopy structure, and canopy gas exchange in a Quercus ilex forest: variation in time and space.PhD Thesis, Universitat de Barcelona, Facultat de Biologia, Departament d’ecologia, Barcelona, Spain.
  • Sala A., Sabaté S., Gracia C. & Tenhunen J.D. (1994) Canopy structure within a Quercus ilex forested watershed: variations due to location, phenological development, and water availability. Trees: Structure and Function 8, 254261.
  • Salleo S. & Lo Gullo M.A. (1990) Sclerophylly and plant water relations in three mediterranean Quercus species. Annals of Botany 65, 259270.
  • Sartori G., Corradini F. & Mancabelli A. (1995) Verso un catalogo dei suoli del Trentino: 1. I suoli bruni lisciviati (Luvisols). Studi Trentini di Scienze Naturali. Acta Geologica 72, 5577.
  • Schreiber U., Bilger W. & Neubauer C. (1994) Chlorophyll fluorescence as a nonintrusive indicator for rapid assessment of in vivo photosynthesis. In Ecophysiology of Photosynthesis (eds E.-D.Schulze & M.M.Caldwell), pp. 4970. Springer Verlag, Berlin.
  • Sims D.A. & Pearcy R.W. (1989) Photosynthetic characteristics of a tropical forest understory herb, Alocasia macrorrhiza, and a related crop species, Colocasia esculenta grown in contrasting light environments. Oecologia 79, 5359.
  • Spreitzer R.J. (1999) Questions about the complexity of chloroplast ribulose-1,5-bisphosphate carboxylase/oxygenase. Photosynthesis Research 60, 2942.
  • Syvertsen J.P., Lloyd J., McConchie C., Kriedemann P.E. & Farquhar G.D. (1995) On the relationship between leaf anatomy and CO2 diffusion through the mesophyll of hypostomatous leaves. Plant, Cell and Environment 18, 149157.
  • Tenhunen J.D., Lange O.L., Harley P.C., Beyschlag W. & Meyer A. (1985) Limitations due to water stress on leaf net photosynthesis of Quercus coccifera in the Portugese evergreen shrub. Oecologia 67, 2330.
  • Terashima I. & Hikosaka K. (1995) Comparative ecophysiology of leaf and canopy photosynthesis. Plant, Cell and Environment 18, 11111128.
  • Terashima I. & Ono K. (2002) Effects of HgCl2 on CO2 dependence of leaf photosynthesis: evidence indicating involvement of aquaporins in CO2 diffusion across the plasma membrane. Plant and Cell Physiology 43, 7078.
  • Terashima I., Araya T., Miyazawa S.-I., Sone K. & Yano S. (2005) Construction and maintenance of the optimal photosynthetic systems of the leaf, herbaceous plant and tree: an eco-developmental treatise. Annals of Botany 95, 507519.
  • Villar R., Held A.A. & Merino J. (1995) Dark leaf respiration in light and darkness of an evergreen and a deciduous plant species. Plant Physiology 107, 421427.
  • Vogelmann T.C. & Evans J.R. (2002) Profiles of light absorption and chlorophyll within spinach leaves from chlorophyll fluorescence. Plant, Cell and Environment 25, 13131323.
  • Wagner J., Pelaez Menendez S. & Larcher W. (1993) Bioclima e potenziale di produttività di Quercus ilex L. al limite settentrionale dell’areale di distribuzione. Parte III. Adattamento morfologico e funzionale delle foglie alle radiazioni luminose. (Bioclimate and productive potential of Quercus ilex L. at its northernmost distribution limit. Part III. Morphological and functional adaptations of leaves to their light regime). Studi Trentini di Scienze Naturali. Acta Biologica 68, 3751.
  • Warren C.R. & Adams M.A. (2001) Distribution of N, Rubisco and photosynthesis in Pinus pinaster and acclimation to light. Plant, Cell and Environment 24, 597609.
  • Warren C.R. & Adams M.A. (2006) Internal conductance does not scale with photosynthetic capacity: implications for carbon isotope discrimination and the economics of water and nitrogen use in photosynthesis. Plant, Cell and Environment 29, 192201.
  • Warren C.R., Ethier G.J., Livingston N.J., Grant N.J., Turpin D.H., Harrison D.L. & Black T.A. (2003) Transfer conductance in second growth Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) canopies. Plant, Cell and Environment 26, 12151227.
  • Weiss A. & Norman J.M. (1985) Partitioning solar radiation into direct and diffuse, visible and near-infrared components. Agricultural and Forest Meteorology 34, 205213.
  • Woodward F.I., Lomas M.R. & Lee S.E. (2001) Predicting the future productivity and distribution of global terrestrial vegetation. In Terrestrial Global Productivity (eds H.A.Mooney, B.Saugier & J.Roy), pp. 521541. Academic Press, Inc., San Diego, CA, USA.
  • Wright I.J., Reich P.B., Westoby M., et al. (2004) The world-wide leaf economics spectrum. Nature 428, 821827.

Appendix

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. ACKNOWLEDGMENTS
  9. REFERENCES
  10. Appendix

APPENDIX I

Alternative estimates of internal diffusion conductance (gm) and sensitivity of gm to errors
Estimation of gm by a different method

Given that for a set of fixed biochemical characteristics of Rubisco, the shape of A versus Ci response curves deviates from the rectangular hyperbola model when gm is not infinite, an alternative estimate of gm may be determined from A/Ci curve analysis (Ethier & Livingston 2004; Niinemets et al. 2005). We have previously demonstrated that the estimates of gm by Eqn 3 (gm,C) and by the method of Ethier & Livingston (2004) (gm,EL) are strongly correlated for three Mediterranean evergreen species (Niinemets et al. 2005). The estimates of gm,EL were derived for the Rubisco-limited part of the A/Ci curve using an iterative method as in Niinemets et al. (2005). Overall, 5–10 measurement points (on average 7.3) were available for the Rubisco-limited part of every A/Ci response curve (Ci ≤ 250 µmol mol−1).

In this study, we also observed a strong correlation between these two different estimates of gm and close adherence to 1:1 line of these estimates (Fig. A1a). The average difference between the two estimates of gm was 8.4% (0.09–32%), further demonstrating strong convergence of the two approaches. Because of the strong correlation, all statistical relationships were qualitatively identical with both estimates of gm. The strong correspondence between the alternative estimates of gm strongly suggests that our derivations of gm are robust.

Sensitivity of gm to potential errors in respiration and electron transport rate

As Harley et al. (1992) have shown, the estimates of gm may be sensitive to assumptions on derivation of and measurement errors in the rates of non-photorespiratory respiration rate in light (Rd) and the rates of photosynthetic electron transport derived from fluorescence (Eqn 2). Although the two approaches for gm estimation were strongly convergent, we further conducted a sensitivity analysis to assess the influence of the assumptions used for and possible errors in estimation of Rd and JETR values on the derivation of internal diffusion conductance. In particular, we used a factor of 0.5 to convert the measurements of dark respiration rate to values of Rd, but the ‘true’ value of the latter may differ from the former by a factor of 0.3–0.8 (Villar et al. 1995; Piel et al. 2002). Fluorescence estimates of photosynthetic electron transport rate (JETR, Eqn 2) may somewhat overestimate the true rate because the lower leaf layers contribute less to total fluorescence signal than they contribute to the total leaf electron transport (Evans, Jakobsen & Ögren 1993), the exact effect depending on the optical thickness of the leaf and the distribution of the photosynthetic capacity within the leaf (Vogelmann & Evans 2002; Evans & Vogelmann 2003). At the same time, variation in the fractional partitioning of light between the two photosystems (ɛ, Eqn 2) can result in either slight overestimation or underestimation of JETR.

The effect of ± 50% deviation of Rd and JETR from the ‘true’ values was analysed for representative paired values of Rd and JETR and for different net assimilation rates to cover the observed span in gm in various-aged Q. ilex leaves (Fig. A1b). These simulations demonstrate that the overall effect of large errors in Rd on gm estimation was less than 2% (Fig. A1b). This was because within the range of Ci of 100–350 µmol mol−1 used in our study for gm derivation, the values of A + Rd (Eqn 3) were dominated by the values of A, and the contribution of Rd to leaf gross assimilation rate was moderate.

Even a large overestimation by 50% in electron transport rate by fluorescence measurements resulted at most in 4–8% underestimation of gm, and the degree of underestimation was larger for higher values of internal diffusion conductance, that is, for a situation with a smaller difference in JETR and electron transport rate calculated from A + Rd (higher A values in the simulations, Fig. A1b). Given that the photosynthetic capacity decreases from the top to bottom leaf layers (Terashima & Hikosaka 1995), and lower layers intercept much less light than upper layers, the contribution of lower leaf layers to total leaf photosynthetic electron transport is much less than that of the upper layers (Evans & Vogelmann 2003). This reasoning and the good correspondence between JETR and whole leaf electron transport determined from oxygen evolution (Laisk & Oja 1998) suggest that the overestimation of whole leaf electron transport rate by fluorescence measurements is certainly less than 50%. Thus, even the fluorescence may sample a somewhat different chloroplast population than is participating in whole leaf electron transport, the overall influence of this moderate discrepancy on gm calculations in our study is minor.

A very large (50%) underestimation of JETR relative to the ‘true’ value resulted in up to 40% overestimation of gm (Fig. A1b). An underestimation in JETR may result when ɛ < 0.5 (Eqn 2). In practice, this large underestimation is implausible. Extensive experimental studies have demonstrated only a minor variation in ɛ, at most from 0.48 to 0.52 (Edwards & Baker 1993; Eichelmann et al. 2004). In addition, underestimation of electron transport as a result of a larger fraction of light captured by PSI would counteract the overestimation of electron transport that results from non-sampling of lower leaf layers.

Taken together, we conclude that the determination of gm for the leaves of Q. ilex that have inherently low internal diffusion conductance is robust and not biased by the assumptions made in Rd and JETR estimation. This agrees with previous evidence demonstrating that the sensitivity of gm to minor errors in gas-exchange and fluorescence measurements increases with increasing gm (Harley et al. 1992; Ethier & Livingston 2004). Larger scatter between the alternative estimates of gm at higher gm values in our study (Fig. A1a) further supports this conclusion.

image

Figure A1. Comparison (a) of the estimates of internal diffusion conductance (gm) determined by different methods for 86 various-aged leaves of Mediterranean evergreen species Quercus ilex and sensitivity (b) of gm (Eqn 3) to possible error or bias in non-photorespiratory respiration rate in light (Rd) and in the rate of photosynthetic electron transport rate calculated from fluorescence (JETR, Eqn 2). In (a), the values of gm were calculated by a combined chlorophyll fluorescence/gas exchange method (Eqn 3, Harley et al. 1992) and by fitting net assimilation versus internal CO2 concentration responses (A/Ci curves) using the Ethier and Livingston (Ethier & Livingston 2004) approach as described in Niinemets et al. (2005). Linear regression line fitted to the data and one-to-one line are also shown. In (b), the values of gm, Rd and JETR are expressed relative to representative baseline values of Quercus ilex leaves (Rd = 1.62 µmol m−2 s−1, JETR = 200 µmol m−2 s−1). Internal CO2 concentration was set at a representative value of 220 µmol mol−1 and the baseline value of gm was calculated by Eqn 3 for different values of leaf net assimilation rate (A). The baseline values of gm were 0.025 mol m−2 s−1 for A = 4 µmol m−2 s−1, 0.039 mol m−2 s−1 for A = 6 µmol m−2 s−1 and 0.055 mol m−2 s−1 for A = 8 µmol m−2 s−1. Then, either Rd or JETR was varied and the new estimates of gm were calculated again by Eqn 3.