SEARCH

SEARCH BY CITATION

Keywords:

  • isoprene emission;
  • emission modeling;
  • within-canopy gradients;
  • physiological plasticity;
  • big-leaf model

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

[1] Isoprene emission potential (ES) varies in tree canopies, and such variations have potentially major implications for predicting canopy level emissions. So far, quantitative relationships of ES with irradiance are missing, and interspecific variation in ES plasticity and potential effects on canopy level emissions have not been characterized. ES, foliage structural, chemical, and photosynthetic characteristics were studied relative to integrated within-canopy daily quantum flux density (Qint) in temperate deciduous tree species Quercus robur, Populus tremula, Salix alba, and Salix caprea, and canopy isoprene emissions were calculated considering observed variation in ES and under different simplifying assumptions. Strong positive curvilinear relationships between nitrogen and dry mass per unit area, photosynthetic potentials and ES per area with Qint were observed. Structural, chemical, and photosynthetic traits varied 1.5-fold to 4-fold and ES per area 3-fold to 27-fold within the canopy. ES variation reflected accumulation of mesophyll cell layers and greater emission capacity of average cells. Species with largest structural and photosynthetic plasticity had greatest plasticity in ES. Relative to the simulation considering within-canopy variation in ES, the bias from assuming a constant ES varied between −8% and +68%, and it scaled positively with ES plasticity. The bias of big-leaf simulations varied between −22% and −35%, and it scaled negatively with ES plasticity. A generalized canopy response function of ES developed for all species resulted in the lowest bias between −11% and 6% and can be recommended for practical applications. The results highlight huge within-canopy and interspecific variation in ES and demonstrate that ignoring these variations strongly biases canopy emission predictions.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

[2] Isoprene, a volatile compound produced by certain plant species through the deoxyxylulose phosphate pathway [Rohdich et al., 2001, 2003], is involved in the protection of plants from heat stress [Behnke et al., 2007; Sharkey and Singsaas, 1995; Sharkey et al., 2008], possibly reflecting the capacity of lipid-soluble isoprene to detoxify lipophilic free radicals [Copolovici et al., 2005; Laothawornkitkul et al., 2009; Velikova et al., 2004; Velikova and Loreto, 2005; Vickers et al., 2009]. Apart from physiological significance, isoprene released by the emitting species plays a major role in tropospheric ozone formation worldwide [Arneth et al., 2008; Guenther et al., 2006; Laothawornkitkul et al., 2009; Sharkey et al., 2008] and possibly also in the formation of secondary organic aerosols [Claeys et al., 2004; Hallquist et al., 2009], thereby importantly altering climate [e.g., Kulmala et al., 2004].

[3] Because of the importance for large-scale processes, major modeling efforts are directed towards estimating the isoprene emissions from vegetation [Arneth et al., 2008; Grote and Niinemets, 2008; Guenther et al., 2006; Zimmer et al., 2003]. The emission models are based on leaf level algorithms resembling photosynthesis: an exponential increase until a maximum and subsequent decline beyond the maximum to describe the temperature dependence and a hyperbolic equation to describe the light dependence [e.g., Guenther et al., 2006, 1993; Niinemets et al., 2010b]. Alternatively, isoprene emissions have been directly linked to photosynthetic metabolites [Martin et al., 2000; Niinemets et al., 1999d; Zimmer et al., 2003]. Leaf level estimates are then integrated to yield whole canopy, landscape and regional level emissions [Baldocchi, 1991; Fuentes et al., 1995; Guenther et al., 2000; Lamb et al., 1996].

[4] The way the leaf level estimates are scaled up to canopy level can critically alter predicted emission flux of reactive hydrocarbons [Geron et al., 1994, 1997; Grote, 2007; Lamb et al., 1996]. Plant canopies are characterized by extensive light gradients, often 10-fold to 20-fold variation in average daily integrated light between canopy top and bottom [Cescatti and Niinemets, 2004; Niinemets, 2010; Sinoquet et al., 2007]. This vast variation in light availability results in profound modifications in structure, chemistry, and photosynthesis of leaves acclimated to different within-canopy light conditions. As key modifications in foliar traits, nitrogen content and leaf dry mass per unit area, and net assimilation rate per unit area increase 2-fold to 4-fold from canopy bottom to top (see Niinemets [2007] for a review). Such increases in resource investment optimize whole-canopy photosynthesis for given foliage biomass in leaves, and according to simulations, result in 10%–20% larger whole-canopy carbon gain than if all leaves in the canopy had a constant average photosynthetic capacity [Baldocchi and Harley, 1995; Gutschick, 1988; Niinemets and Anten, 2009].

[5] Apart from photosynthesis, isoprene emission potentials have been found to be larger in upper than in lower canopy [Funk et al., 2006; Geron et al., 1997; Harley et al., 1996, 1997; Niinemets et al., 2010a]. However, in parameterizing the emission models, it is common practice to measure foliage isoprenoid emission potentials only for high-light exposed foliage [Buckley, 2001; He et al., 2000; Llusià and Peñuelas, 2000; Pio et al., 2005; Sabillón and Cremades, 2001; Winters et al., 2009]. In simplest integration schemes, whole-canopy isoprene emission is found by just multiplying a leaf level constant emission potential, modified for variations in temperature and incident light, with leaf area index [Fuentes et al., 1995; Owen and Hewitt, 2000; Owen et al., 2003; Stewart et al., 2003]. Because of strong environmental gradients within the canopy, such an approach leads to highly biased canopy emission estimate. In more advanced integration schemes, models with sunlit/shaded leaf area fractions [Steinbrecher et al., 2009], layered canopy models simulating light distribution [Fuentes et al., 1999; Guenther et al., 1995; Lamb et al., 1996], or more detailed canopy models estimating light and temperature distributions and turbulence [Baldocchi et al., 1995; Geron et al., 1997; Huber et al., 1999; Lamb et al., 1996] have been used, but still employing a constant isoprene emission potential. In a few cases, within-canopy variation in isoprene emission potentials has been included in the canopy models [Baldocchi et al., 1999; Funk et al., 2006; Geron et al., 1994, 1997; Guenther et al., 1999; Harley et al., 1996, 1997] (see Grote et al. [2010] for inclusion of within-canopy variation in monoterpene emission potential), but these simulations were based on hypothetical correlations of isoprene emission potential with canopy leaf area index [Baldocchi et al., 1999; Geron et al., 1994; Guenther et al., 1999] or linking the emissions to canopy height or leaf area index rather than to light [Funk et al., 2006; Geron et al., 1997; Grote et al., 2010; Harley et al., 1996, 1997]. Big-leaf models have also been used (e.g., GOTILWA+ [Keenan et al., 2009]) that are based on a premise that isoprene emission is directly proportional to light availability, thereby enabling analytical integration of whole-canopy isoprene emission fluxes based on only upper canopy values. Finally, canopy level isoprene emission factors have been used, and a constant factor of 0.57 has been employed to account for the reduction of emissions due to light and temperature gradients within the canopy [Guenther et al., 2006].

[6] Because of lack of quantitative relationships between isoprene emission potentials and canopy light environment, it is currently impossible to assess the effect of variations in within-canopy isoprene emission potentials on whole-canopy isoprene emission fluxes. Furthermore, important variations in species structural and photosynthetic plasticity to light have been demonstrated [Niinemets and Anten, 2009; Portsmuth and Niinemets, 2007; Sánchez-Gómez et al., 2006; Valladares et al., 2002], but interspecific variation in plastic adjustments of isoprene emission potentials has not been studied.

[7] Here we developed quantitative relationships between isoprene emission potential and canopy light environment in four temperate tree species and tested the key hypotheses that variation in foliage isoprene emission potentials throughout the canopy follows the profiles of photosynthesis and that species differences in plastic adjustment in isoprene emission potentials are correlated with species differences in structural and photosynthetic plasticity. We further tested the capacity of different simplified scaling-up schemes to predict canopy isoprene emissions. We focus here at the long-term variation in within-canopy light conditions and provide a snapshot of within-canopy profiles of isoprene emission to underscore the overall importance of consideration of within-canopy variation in predicting emission fluxes. The list of acronyms used with corresponding units is provided in the Notation.

2. Material and Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

2.1. Study Design

[8] The study was conducted in the vicinity of Tartu, Estonia (58°23′N, 27°05′E, elevation ca. 40 m above the sea level) in the middle of August 2009. The open early successional stand with a height of 15 m was dominated by Populus tremula L. and Salix species. Canopy-reaching individuals of four isoprene-emitting tree species, P. tremula, Quercus robur L., Salix caprea L., and S. alba L. were selected for the measurements. A mobile lift was used to measure foliage isoprene emissions from the top to the bottom of the plant canopies. The measurements were conducted in 16–18 different canopy locations in all species. The underlying hypothesis in this study was that most of the within-canopy variation in isoprene emission potentials at any given moment of time is associated with within-canopy variation in light environment. Thus, we did not try to replicate the emission measurements at specific canopy location but rather tried to best describe the light conditions of each sample location to describe the variation among the emission potentials of adjacent leaves. All sample locations in different positions are true replicates for studying the response of given leaf trait (structure, photosynthesis, isoprene emission) to environmental gradient (light) (see Hurlbert [1984] for a discussion).

2.2. Determination of Long-Term Leaf Light Environment

[9] A hemispheric photograph was taken above each shoot using Nikon Fisheye Converter lens FC-E8 0.21X mounted on a Nikon Coolpix 990 digital camera. Original 24 bit hemispherical photographs were converted to 1 bit black and white images as detailed previously [Niinemets et al., 2004b]. From the thresholded images, the fractions of diffuse light of open sky (diffuse site factor, ID) and of potential penetrating direct solar radiation of open sky (direct site factor, IB) were determined using WinPhot 5.0 software [ter Steege, 1996]. The direct site factor was calculated as an average during the growing season after leaf maturation (20 June to 15 August 2009). Daily average values of above-canopy direct and diffuse global solar radiation and total solar radiation (G) measured in nearby meteorological station in Tõravere (58°16′N, 26°28′E) were used to derive the average daily integrated quantum flux density, Qint, for each canopy location,

  • equation image

where κ (mol MJ−1) converts the values of global solar radiation to photosynthetic quantum flux density (κ = 1.977 mol MJ−1) [Ross and Sulev, 2000] and pD is the fraction of diffuse radiation in total solar radiation [Niinemets et al., 2003b]. An average value of G = 18.89 MJ m−2 d−1 and pD = 0.519 were estimated for the part of the growing season used for determination of IB, giving an average above-canopy Qint of 37.3 mol m−2 d−1.

2.3. Estimation of Effective Leaf Area Index

[10] We further used the gap fraction data estimated from hemispheric photographs by the WinPhot 5.0 software [ter Steege, 1996] to determine the effective leaf area index (Le) above each sample location as detailed by Niinemets et al. [2004a]. Briefly, we employ the inversion procedure developed by Miller [1967] and Campbell and Norman [1989] that is implemented in the Li-Cor Plant Canopy Analyzer [Li-Cor Inc., 1992],

  • equation image

where T(θ) is the gap fraction corresponding to zenith angle θ. For each species, the effective stand leaf area index was taken as the average of the three largest estimates at the bottom of the canopy.

[11] Le is an estimate of leaf area index determined from optical measurements assuming random dispersion of foliage elements. In reality, foliage of temperate trees is often clumped, implying that the canopy transmits more light than a canopy having random foliage dispersion, and thus, Le underestimates the true leaf area index, L [Baldocchi and Collineau, 1994; Cescatti and Niinemets, 2004]. L can only be reliably estimated by direct methods such as destructive harvesting or litter collection [Bréda, 2003]. For modeling purposes (see section 2.9), we estimated a representative value of clumping index λ0 of 0.47 (λ0 = Le/L) based on L estimates from leaf-litter collection [Niinemets and Tamm, 2005] and hemispheric photographs taken below tree canopies for three temperate mixed deciduous forests dominated by P. tremula (O. Kull and Ü Niinemets (unpublished data, 1995) and Niinemets et al. [1999a] for representative light measurements and Niinemets and Tamm [2005] for sites). This value is within the range of values reported for other Populus dominated stands [Niinemets et al., 2004a].

2.4. Gas Exchange Measurements and Volatile Organic Compound Sampling

[12] A portable gas exchange/fluorescence system (GFS-3000, Heinz Walz GmbH, Effeltrich, Germany) was used for measurements of leaf gas exchange and for volatile organic compound (VOC) sampling. The system has an environmental-controlled cuvette with 8 cm2 window area and is equipped with full window leaf chamber fluorimeter for sample illumination and for fluorescence measurements. All measurements were made at the ambient CO2 concentration of 380 μmol mol−1, light intensity of 1000 μmol m−2 s−1 (10% blue and 90% red LED light), while leaf temperature was kept at 25°C, and relative humidity >50% during all measurements. Although VOC measurements are conventionally conducted at 30°C [Guenther et al., 1993], separate laboratory studies suggested that this temperature was already somewhat inhibitory for photosynthesis and isoprene emission in the studied subboreal species (data not shown).

[13] After enclosure of the leaf in the cuvette, light was switched on and the leaf was stabilized until stomata opened and steady state values of net assimilation rate (A) and stomatal conductance (gs) were obtained. Thereafter, steady state fluorescence value F was recorded and a flash of saturating white light of 4500 μmol m−2 s−1 was given and the maximum fluorescence yield Fm was measured.

[14] After gas exchange and fluorescence measurements, 2.5 L of the air exiting from the cuvette was sampled in a multibed stainless steel cartridge (10.5 cm length, 3 cm inner diameter, Supelco, Bellefonte, USA) filled with Carbotrap C 20/40 mesh (0.2 g), Carbopack C 40/60 mesh (0.1 g), and Carbotrap X 20/40 mesh (0.1 g) adsorbents (Supelco, Bellefonte, USA) optimized for quantitative analysis of isoprene, lipoxygenase (LOX) pathway products (volatile C6 aldehydes), and mono- and sesquiterpenes. The adsorption was done at a flow rate of 250 mL min−1 for 10 min using a 1003-SKC constant flow sampling pump (SKC Inc., Houston, TX, USA). In addition, a sample was taken from an empty leaf cuvette before and after each measurement. Before the collection of volatiles, the traps were cleaned by the passage of a stream of ultra pure helium at a flow rate of 200 mL min−1 and at temperature of 250°C for 2 h.

2.5. Foliage Structural and Chemical Analyses

[15] The leaf used for physiological measurements and four neighboring leaves in large-leaved species (Q. robur, P. tremula, and S. caprea) and 5–15 neighboring leaves in the smaller-leaved species S. alba were harvested. Leaf area was estimated from digital images by UTHSCSA Imagetool 2.00alpha (University of Texas Health Science Center, San Antonio, TX, USA). Fresh mass immediately after sampling and dry mass after oven-drying at 70°C for 48 h were also determined for individual leaves. From these measurements, average values of dry mass per unit area (MA) and fresh to dry mass ratio were determined. Foliage C, N, and S contents were determined by a Vario MAX CNS analyzer (Elementar Analysensysteme GmbH, Hanau, Germany).

2.6. VOC Analyses

[16] Adsorbent cartridges were analyzed with a combined Shimadzu TD20 automated cartridge desorber and Shimadzu QP2010 Plus GC MS instrument (Shimadzu Corporation, Kyoto, Japan). During primary desorption, He (99.9999%, Elmer Messer Gaas AS, Tallinn, Estonia) purge flow was set at 40 mL min−1, desorption temperature at 250°C, and desorption time was 6 min. Second stage trap temperature during primary desorption was −20°C and hold time was 6 min, while the second stage trap desorption temperature was 280°C.

[17] A Zebron ZB-624 fused silica capillary column (0.32 mm inner diameter, 60 m length, 1.8 μm film thickness, Phenomenex, USA) was used for separation of plant-emitted volatile compounds using the following GC oven program: 40°C for 1 min, 9°C min−1 to 120°C, 2°C min−1 to 190°C, 20°C min−1 to 250°C, 250°C for 5 min. The GC carrier gas was He with a flow rate of 1.48 mL min−1. The Shimadzu QP 2010 Plus mass spectrometer was operated in the electron impact mode. The transfer line temperature was set at 240°C and ion source temperature at 150°C. Although we were primarily interested in isoprene, from all chromatograms, we also estimated the volatile lipoxygenase pathway products (C6 aldehydes) to diagnose for possible environmental and/or biotic stresses [Beauchamp et al., 2005; Loreto et al., 2006]. As there is evidence of emissions of mono- and sesquiterpenes in some Salix and Populus species, especially under stress conditions [Brilli et al., 2009; Hakola et al., 1998], mono- and sesquiterpenes were also analyzed. The compounds were identified by comparing the mass spectrum of an individual compound with the spectra of external standards (GC purity, Sigma-Aldrich, St. Louis, MO, USA) and with the spectra in NIST Library. The GC-MS system was calibrated with GC grade external isoprene, C6 aldehyde, mono- and sesquiterpene standards, and absolute amounts of compounds present in cartridges were determined based on compound-specific target ions (for terpenes in most cases ions with m/z 93 or m/z 69) [Copolovici et al., 2009]. The emission rate (E, mol m−2 s−1) for each compound sampled was calculated as

  • equation image

where Xp (mol) is the amount of the compound in the cartridge with the leaf present in the chamber and Xe with the empty cuvette, P (m2) is the projected leaf area enclosed in the cuvette, V (m3) is the volume of air sampled, and υ (m3 s−1) is the air flow rate through the system.

[18] Isoprene was always the primary compound emitted, and in all cases, the possible induced emissions of mono- and sesquiterpenes were close to detection limit (<0.3 nmol m−2 s−1). No significant emissions of LOX products were detected, indicating that the leaves sampled did not experience any severe environmental or biotic stress. Thus, only the data for isoprene are reported.

2.7. Gas Exchange and Isoprene Emission Rate Calculations

[19] Foliage net assimilation rate (A), stomatal conductance to water vapor (gs), and CO2 concentration in substomatal cavities (ci) were calculated according to von Caemmerer and Farquhar [1981]. Electron transport rate from gas exchange, JG, was calculated as [Brooks and Farquhar, 1985]

  • equation image

where Rd (μmol m−2 s−1) is the rate of nonphotorespiratory respiration continuing in the light and Γ* (μmol mol−1) is the hypothetical CO2 compensation point in the absence of Rd. As direct Rd measurements are time-consuming [e.g., Villar et al., 1995], Rd was taken as 0.025Vcmax [Niinemets et al., 1998b], where Vcmax is the maximum carboxylase activity of Rubisco calculated from ci, A, and Rd as in the work of Niinemets et al. [1999c]. In this calculation, Rd depends on Vcmax, and Vcmax on Rd, and therefore, the calculations were carried out in iterative mode.

[20] The electron transport rate was also calculated based on chlorophyll fluorescence measurements, JF, as [Genty et al., 1989]

  • equation image

where Q is the quantum flux density, ξ is the leaf absorptance (taken as 0.85 in this study), and ρPSII is the fraction of light absorbed by PSII (assumed to be 0.5 [see e.g., Edwards and Baker, 1993]). Because of finite diffusion conductance from substomatal cavities (CO2 concentration ci) to chloroplasts (CO2 concentration cC) [e.g., Flexas et al., 2009], equation (4) may underestimate the true linear electron transport rate required to assimilate carbon with the rate A. On the other hand, JF, (equation (5)) depends on assumptions about ξ and ρPSII. Thus, although both the equations (4) and (5) provide a measure for photosynthetic electron transport, both can somewhat deviate from the “true” value.

[21] As six molecules of carbon need to be assimilated to produce one molecule isoprene [Niinemets et al., 1999d], the percentage of carbon used for isoprene emission, γA, was calculated from isoprene emission rate under study conditions of leaf temperature of 25°C and incident light of 1000 μmol m−2 s−1 (ES, nmol m−2 s−1) and A (μmol m−2 s−1) as

  • equation image

The percentage of electrons used for isoprene emission, γJ, is given analogously

  • equation image

where the first part of the numerator provides the amount of electrons needed to bring the reductions state of carbon in isoprene from the level of that in CO2 to that in sugars, while the second part gives the additional electron requirement needed to increase the carbon reduction state from that in sugars (40% carbon) to isoprene (88% carbon) [Niinemets et al., 2002; Niinemets, 2004]. A second estimate of γJ can be calculated replacing JG by JF.

2.8. Statistical Analyses

[22] Linear and nonlinear regressions were used to test for the effects of long-term variation in light environment (Qint) on foliage chemistry and structure and on photosynthesis and isoprene emission rates. Whenever Qint did not significantly affect a given variable, ANOVA followed by Tukey post hoc test was used to compare the means among species. When Qint was significantly correlated with a given trait, the means were compared by common slope co-variation analyses (ANCOVA). The species differences in the slopes of given trait versus Qint relationships were identified by separate slope ANCOVA analyses. Whenever necessary, Qint was log-transformed in ANCOVA analyses to linearize the relationships. All relationships were considered significant at P < 0.05 [Sokal and Rohlf, 1995].

2.9. Simulation of Whole-Canopy Isoprene Emissions Based on Observed Variations in ES

[23] A simple layered canopy model was used to simulate the whole-canopy isoprene emission. The canopy was divided into 50 layers of equal leaf area, and quantum flux density incident to each layer was simulated using Beer's law modified to consider foliage deviation from random dispersion, i.e., foliage clumping (equation (A11)). First, within-canopy variation in isoprene emission potentials was calculated. For this, Beer's law (equation (A11)) was used to calculate within-canopy variation in Qint at each canopy layer. Then, isoprene emission potentials, ES, for each canopy layer were calculated based on species-specific best-fit nonlinear regressions of ES versus Qint (see section 3.3 for the regressions). Further, the instantaneous quantum flux density Q for each canopy layer was calculated from Beer's law (equation (A11)) for an above-canopy quantum flux density of 1000 μmol m−2 s−1. On the basis of these Q values, isoprene emission rate for each canopy layer was further simulated on the basis of isoprene light response function (equation (A2)), using the values of quantum yield of isoprene emission (α in equation (A2)) of 0.0027 mol mol−1 and scaling coefficient (ɛL in equation (A2)) of 1.066 as originally parameterized [Guenther et al., 1991, 1993]. After integration of the emissions from each layer, whole-canopy isoprene emission rate, EC, was obtained. This EC value most closely corresponds to the empirical data obtained in this study and is therefore labeled as baseline simulation EC,base.

2.10. Alternative Ways of Predicting Whole-Canopy Emissions

[24] In addition to EC,base, we calculated EC values based on linear refitting of ES versus Qint relationships (EC,linear), based on a constant ES value for all leaves in the canopy equal to average ES (EC,average) and based on a constant ES value equal to above-canopy ES (EC,max). For EC,average scenario, the average ES used was taken as leaf-area weighted average, i.e., average of 50 layer-specific ES values determined from nonlinear regressions of ES with Qint.

[25] We further used the method of Poorter et al. [2010] to determine a generalized ES versus Qint response curve. In brief, this method is based on normalization of species-specific responses with respect to a certain value of Qint and developing common regressions across the normalized data. First, we determined for each species the above-canopy ES values (ES,0) using the nonlinear empirical regressions between ES and Qint and substituting Qint by the above-canopy value of 37.3 mol m−2 d−1. Thereafter, we normalized all data for given species with respect to ES,0. The pooled normalized data were further fitted by a common regression to determine the canopy response function f(Qint) that was forced to equal 1.0 at above-canopy Qint. Thus, the within-canopy variation in isoprene emission potential for each species is given as the product of ES,0 conventionally assessed in the field studies on isoprene emission and f(Qint). The simulations of EC (EC,general) were further conducted based on species-specific above-canopy isoprene emission potentials, ES,0, and the generalized within-canopy gradient f(Qint).

[26] Finally, a simplified big-leaf model of canopy isoprene emission (EC,big-leaf) was constructed that provides a means to analytically calculate canopy isoprene emission rate based on above-canopy Q and ES,0 (Appendix A). Implicit in big-leaf models, previously employed to simulate leaf photosynthesis [Amthor, 1994; Lloyd et al., 1995; Niinemets and Anten, 2009], is the assumption that leaf physiological potentials are directly proportional to Qint. Thus, all these additional simulations result in hypothetical situations with either smaller (simulations with constant average or above-canopy ES) or larger (big-leaf model) gradient in ES than in those with the observed within-canopy ES gradients.

2.11. Consideration of Potential Variations in Temperature and Initial Quantum Yield

[27] All the simulations outlined were conducted at a constant temperature of 25°C used during the measurements and also keeping the quantum yield for isoprene emission (α) constant. However, in real plant canopies, there is a certain covariation of temperature with light, on the order of 5°C [Baldocchi et al., 2002; Niinemets et al., 1999b]. In addition, α can vary throughout the canopy [Harley et al., 1996, 1997]. To determine whether the variation in temperature and α can modify the conclusions with respect to different strategies in considering within-canopy variations in isoprene emission potentials, we also conducted simulation runs with realistic variations in temperature and α. We used whole-season average gradient in temperature in relation to Qint (Tav = aLnQint + b, where a and b are empirical coefficients) measured in broad-leaved mixed temperate deciduous canopy having a leaf area index of 6 m2 m−2 from Niinemets et al. [1999b]. The within-canopy gradient in temperature predicted by these data was 4°C [Niinemets et al., 1999b]. We used the shape of this response of Tav to Qint and refitted it to the canopy such that leaf-area weighted average temperature was 25°C. According to the final equation, leaf temperature was 27.1°C at the top and 22.9°C at the bottom of the canopy. Isoprene emission rate at different temperatures was simulated using the temperature parameterization of Guenther [1997] model adjusted to equal to 1.0 at 25°C.

[28] In simulations exploring the importance of variations in α, we assumed a value of 0.00174 mol mol−1 for the top leaves and 0.0040 mol mol−1 for the bottom leaves [Harley et al., 1996] and scaled α linearly with Qint.

2.12. Canopy Characteristics in Stand Level Emission Modeling

[29] The values of the effective stand leaf area index (Le, equation (2)) observed at the bottom of the tree canopy (average ± SE of the three largest estimates) were 2.80 ± 0.07 m2 m−2 for Q. robur, 2.8 ± 0.4 m2 m−2 for P. tremula, 2.77 ± 0.10 m2 m−2 for S. caprea, and 2.6 ± 0.6 m2 m−2 for S. alba (means are not statistically different according to one-way ANOVA, P > 0.7) and the overall average of stand Le was 2.74 ± 0.06 m2 m−2. Given the clumping index of 0.47 (see section 2.3), this value corresponds to a total stand leaf area index of 5.83 m2 m−2. As Le versus Qint relationships differed little among the species (Figure 1b, no statistical difference of Qint versus logLe according to ANCOVA analyses), all data were fitted by a single exponential relationship and an extinction coefficient of 0.86 was derived (Figure 1b). Thus, in all simulations, constant values of stand Le of 2.74 m2 m−2, clumping index of 0.47, and extinction coefficient of 0.86 were used.

image

Figure 1. Variation in (a) average daily integrated quantum flux density (Qint, equation (1)) from the top to the bottom of the canopies and (b) the correlation between the effective leaf area index (Le, equation (2)) and Qint in the four studied species. In Figure 1a, the data were fitted by nonlinear regressions in the form y = aebx. r2 = 0.86 for Quercus robur (Qr, n = 16), r2 = 0.85 for Populus tremula (Pt, n = 18), r2 = 0.95 for Salix caprea (Sc, n = 18), and r2 = 0.92 for Salix alba (Sa, n = 16). P < 0.001 for all relationships. In Figure 1b, the data for all species were fitted by Beer's equation considering that Qint at the top of the canopy was 37.3 mol m−2 d−1. Representative hemispherical photographs within the canopy of P. tremula are also shown. The arrows denote Qint values corresponding to the photographs from the top to the bottom (35.6, 16.9, 6.4, and 2.5 mol m−2 d−1).

Download figure to PowerPoint

[30] We note that all these model simulations contain crude simplification, e.g., we do not consider diurnal variations in above-canopy Q values, and therefore, these simulations provide only a snapshot of canopy isoprene emission. In addition, lack of consideration of sunlit and shaded leaf area fractions can significantly alter the flux integration [Cescatti and Niinemets, 2004; de Pury and Farquhar, 1997; Niinemets and Anten, 2009]. Despite these simplifications, we argue that these simulations capture the principal features among the different model parameterizations (variable versus constant) and different models (layered versus big leaf) and provide insight into the effect of within-canopy variations in isoprene emission potentials in the absence of other confounding sources of variation.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

3.1. Light Gradients Within the Canopies and Foliage Structural and Chemical Responses to Light

[31] There was a large variation in light availability (seasonal average integrated quantum flux density, Qint) within the canopy of each studied species. The light availability between canopy top and bottom varied 12-fold in P. tremula and S. caprea, almost 10-fold in Q. robur and almost 7-fold in S. alba (Figure 1a). The rate of change of Qint with canopy height, i.e., the steepness of the canopy light gradient, was the greatest in Q. robur and the smallest in S. alba (Figure 1a). Qint scaled with the effective leaf area index (equation (2)) similarly in all species and fitted closely the simple exponential relationship, i.e., Beer's law (Figure 1b).

[32] Leaf dry mass per unit area (MA, Figure 2a) increased curvilinearly with increasing Qint in all species. The largest range of variation based on minimum and maximum values across the light gradient of 2.2-fold was observed in Q. robur and smallest range of variation of 1.5-fold was observed in S. caprea. Leaf dry to fresh mass ratio also increased curvilinearly with Qint in P. tremula (r2 = 0.54, P < 0.001), S. caprea (r2 = 0.43, P < 0.005), and S. alba (r2 = 0.83, P < 0.001), but not in Q. robur (r2 = 0.19, P > 0.09). The variation in dry to fresh mass ratio was larger in S. caprea (1.4-fold) and smallest in Q. robur and P. tremula (1.2-fold).

image

Figure 2. (a) Leaf dry mass per unit area (MA) and (b) leaf nitrogen content per unit area (NA) in relation to daily integrated quantum flux density in the four studied species. Symbols, line and species codes and sample sizes as in Figure 1. Data were fitted by nonlinear regressions in the form y = axb and y = a + blog(x), whichever of the two yielded larger fraction of explained variance. For Figure 2a, r2 = 0.86 (Q. robur), 0.81 (P. tremula), 0.78 (S. caprea), and 0.92 (S. alba). For Figure 2b, r2 = 0.81 (Q. robur), 0.67 (P. tremula), 0.62 (S. caprea), and 0.56 (S. alba). All regressions are significant at P < 0.001. Error bars in Figure 2a provide ±SE of all leaves sampled from given canopy location.

Download figure to PowerPoint

[33] Leaf nitrogen content per area (NA) increased with increasing Qint similarly to MA (Figure 2b). The relationships were analogous with sulfur content per area (SA, r2 = 0.51–0.70, P < 0.005). The contents of these elements per unit dry mass were poorly related to Qint, with a significant weak positive correlation observed only for nitrogen in Q. robur (r2 = 0.33, P = 0.02). As the element content per unit area is the product of the content per unit dry mass and MA, weak mass-based correlations with Qint suggest that the area-based strong correlations with NA and SA mainly reflect light effects on MA, i.e., stacking of leaf cells with similar elemental composition per unit area.

3.2. Foliage Photosynthetic Potential in Relation to Canopy Light Environment

[34] Foliage photosynthetic capacity per unit area increased with increasing Qint in all four species (Figure 3a). The range of variation based on observed minimum and maximum values across the light gradient was greatest in S. caprea (3.7-fold) and smallest in P. tremula (1.9-fold). Analogously, area-based rates of the electron transport calculated from gas exchange (equation (4), Figure 3b) and from chlorophyll fluorescence (equation (5), Figure 3c) increased curvilinearly with light availability. Positive, almost linear correlations of area-based photosynthetic characteristics with MA and NA were also observed (P < 0.005 in all cases). CO2 concentration in substomatal cavities (ci) was not correlated with Qint in Q. robur, S. alba, and S. caprea (P > 0.05). However, ci decreased curvilinearly in P. tremula with increasing Qint (r2 = 0.54, P < 0.005), but the range was only ca 30 μmol mol−1. Thus, during the study, the stomatal limitations of photosynthesis were similar throughout the canopy.

image

Figure 3. Correlations of (a) net assimilation rate, photosynthetic electron transport rate (b) from gas exchange (equation (4)) and (c) from chlorophyll fluorescence (equation (5)) with daily average integrated quantum flux density in four temperate deciduous tree species (data presentation as in Figure 1, data fitting as in Figure 2). For Figure 3a, r2 = 0.84 (Q. robur), 0.73 (P. tremula), 0.62 (S. caprea), and 0.74 (S. alba). For Figure 3b, r2 = 0.78 (Q. robur), 0.70 (P. tremula), 0.65 (S. caprea), and 0.69 (S. alba). For Figure 3c, r2 = 0.79 (Q. robur), 0.80 (P. tremula), 0.71 (S. caprea), and 0.65 (S. alba). All relationships are significant at P < 0.001.

Download figure to PowerPoint

[35] Mass-based rates of net assimilation and photosynthetic electron transport rates varied much less than the area-based rates and were generally not correlated with Qint, MA or nitrogen content per dry mass. Only in S. caprea, weak positive curvilinear relationships were observed between Qint and net assimilation rate per unit dry mass (r2 = 0.24, P < 0.05) and Qint and mass-based electron transport rate from gas exchange (r2 = 0.31, P < 0.02). Thus, as with the nutrient contents, significant area-based positive relations mainly reflected the accumulation of photosynthetic biomass per unit area (increased MA).

3.3. Isoprene Emission Potential in Relation to Light Availability and Foliage Photosynthesis

[36] Isoprene emission potentials (ES) per unit area (ES,A, Figure 4a) and per unit dry mass (ES,M, Figure 4b) were positively correlated with Qint in all species. The range of variation in ES,A based on minimum and maximum values observed was 27-fold in Q. robur, 14-fold in S. caprea, 6-fold in S. alba and 3-fold in P. tremula. Relatively low range of variation in P. tremula reflected higher ES,A values in low light in this species compared with the other three species (Figure 4a). The range of variation in ES,M values was ca. 1.5-fold less than in ES,A: 17-fold in Q. robur, 10-fold in S. caprea, 4-fold in S. alba, and 2-fold in P. tremula (Figure 4b).

image

Figure 4. Dependencies of leaf isoprene emission rate (a) per unit area and (b) per unit leaf dry mass on daily average integrated quantum flux density in the four studied temperate deciduous tree species (data presentation as in Figure 1). Data were fitted by nonlinear regressions in the form y = a + blog(x). The rate per unit area is the product of the rate per unit dry mass and MA (Figure 2a). For Figure 4a, r2 = 0.91 (Q. robur), 0.85 (P. tremula), 0.93 (S. caprea), and 0.89 (S. alba). For Figure 4b, r2 = 0.76 (Q. robur), 0.63 (P. tremula), 0.85 (S. caprea), and 0.66 (S. alba). All relationships are significant at P < 0.001.

Download figure to PowerPoint

[37] In all species, ES,A was positively correlated with both net assimilation rate per area (r2 values for linear correlations between 0.56 and 0.77), and electron transport rates from gas exchange (JG, equation (4), r2 = 0.51–0.77) and chlorophyll fluorescence (JF, equation (5), r2 = 0.55–0.71, P < 0.001 for all correlations). For all data pooled across the species, the best correlation was observed with JF (r2 = 0.48, P < 0.001), while r2 was only 0.28 (P < 0.001) for both net assimilation rate and JG.

[38] The fraction of carbon used for isoprene emission (equation (6), γA) increased curvilinearly with Qint in all species (Figure 5a), with the largest within-canopy variation of 9-fold in Q. robur and smallest of 2.2-fold in P. tremula. Analogously, the fraction of electrons used for isoprene emission (equation (7), γJ) increased curvilinearly with Qint (Figure 5b for γJ based on JF; r2 = 0.49–0.68 for γJ based on JG). ES,M was positively correlated with both γA (Figure 6a) and with γJ (Figure 6b). Analogously, ES,A was positively correlated with both γA and γJ (r2 = 0.83–0.97 for linear correlations). ES,M (Figure 6c) and ES,A (inset of Figure 6c) were positively correlated with MA, but the correlation was stronger with ES,A, emphasizing that light-dependent changes in ES,A reflect both modifications in emission capacity of single leaf cells (ES,M) and increases in MA.

image

Figure 5. (a) The percentage of carbon (equation (6)) and (b) the percentage of electrons on the basis of chlorophyll fluorescence (equation (7), JG replaced by JF) used for isoprene emission in relation to daily average integrated quantum flux density in Q. robur (r2 = 0.69 for Figure 5a and r2 = 0.68 for Figure 5b), P. tremula (r2 = 0.53 for Figure 5a and r2 = 0.50 for Figure 5b), S. caprea (r2 = 0.69 for Figure 5a and r2 = 0.88 for Figure 5b), and S. alba (r2 = 0.51 for Figure 5a and r2 = 0.66 for Figure 5b). Data presentation as in Figure 1. Data were fitted by nonlinear regressions in the from y = a + blog(x) and y = axb, whichever yielded the larger fraction of explained variance (r2). All regressions are significant at P < 0.005.

Download figure to PowerPoint

image

Figure 6. Correlations of the isoprene emission rate per unit dry mass (ES,M) with the percentage of (a) carbon and (b) electrons used for isoprene emission and (c) the correlation with leaf dry mass per unit area (MA) in the four studied species (data presentation as in Figure 1). The inset in Figure 6c demonstrates the correlations between the isoprene emission rate per unit area (ES,A) and MA. Photosynthetic electron transport rate used to derive the electron flow to isoprene was based on chlorophyll fluorescence (equation (7), JG replaced by JF as in Figure 5b). The data were fitted by linear regressions. For Figure 6a, r2 = 0.94 (Q. robur), 0.82 (P. tremula), 0.73 (S. caprea), and 0.78 (S. alba). For Figure 6b, r2 = 0.90 (Q. robur), 0.83 (P. tremula), 0.97 (S. caprea), and 0.85 (S. alba). For Figure 6c, r2 = 0.52 (Q. robur), 0.45 (P. tremula), 0.50 (S. caprea), and 0.46 (S. alba). For the inset, r2 = 0.75 (Q. robur), 0.79 (P. tremula), 0.72 (S. caprea), and 0.76 (S. alba). All regressions are significant at P < 0.005.

Download figure to PowerPoint

3.4. Species Differences in Structure, Chemistry, Photosynthesis, and Isoprene Emission

[39] According to ANOVA and common-slope ANCOVA analyses with log(Qint) as covariate, P. tremula had larger MA than Q. robur and S. alba (Table 1a and Figure 2a). Populus tremula also had lower average N content per dry mass (Table 1a) and lower NA at given Qint than Q. robur and S. caprea (P < 0.01). Photosynthetic characteristics (A, JG and JF), were generally larger in S. alba and P. tremula than in S. caprea and Q. robur (Figure 3 and Table 1b). Isoprene emission rate per area and dry mass were lower in S. caprea than in the three other species (Figure 4 and Table 1b), while the percentages of carbon (Figure 5 and Table 1b) and electrons (P < 0.001) used for isoprene emission were larger in Q. robur than in the other species.

Table 1a. Mean (±SE) Values of Foliage Structural and Chemical Traits in the Four Studied Temperate Deciduous Tree Speciesa
SpeciesPL (cm2)DF (g g−1)MA (g m−2)NM (%)CM (%)SM (%)n
  • a

    Leaf size (PL); dry to fresh mass ratio (DF); dry mass per unit area (MA); and nitrogen (NM), carbon (CM), and sulfur (SM) percentages. Means with the same lowercase letter are not significantly different among the species (P > 0.05). When a given trait was independent of Qint, means were compared by ANOVA followed by a Bonferroni test. When Qint affected the given trait, common slope ANCOVA analysis was conducted with log(Qint) as the covariate (see Figures 25 for curvilinear relationships with Qint). In case of significant slope differences (Table 2), mean comparisons should be interpreted with caution.

Quercus robur39.9 ± 2.7a0.457 ± 0.006a99.0 ± 4.9ac2.47 ± 0.05a48.11 ± 0.24a0.185 ± 0.006a16
Populus tremula39.3 ± 1.9a0.470 ± 0.005a106.0 ± 3.9b1.97 ± 0.04b45.86 ± 0.16b0.181 ± 0.006a18
Salix caprea50.7 ± 3.3b0.452 ± 0.011a103.7 ± 3.0ab2.35 ± 0.06a48.92 ± 0.20c0.218 ± 0.008b18
Salix alba6.0 ± 0.3c0.407 ± 0.011b87.6 ± 3.7c2.43 ± 0.06a48.19 ± 0.16a0.245 ± 0.011c16
Table 1b. Mean (±SE) Values of Foliage Physiological Traits in the Four Studied Temperate Deciduous Tree Speciesa
SpeciesA (μmol m−2 s−1)ci (μmol mol−1)JG (μmol m−2 s−1)JF (μmol m−2 s−1)ES,A (nmol m−2 s−1)ES,M (nmol g−1 s−1)γA (%)
  • a

    Net assimilation rate (A), intercellular CO2 concentration (ci), photosynthetic electron transport rate from gas exchange (JG) and fluorescence (JF), isoprene emission rate under standard conditions per unit area (ES,A) and dry mass (ES,M), and percentage of carbon used for isoprene synthesis (γA). Sample sizes are the same as in Table 1a. Means with the same lowercase letter are not significantly different among the species (P > 0.05). When a given trait was independent of Qint, means were compared by ANOVA followed by a Bonferroni test. When Qint affected the given trait, common slope ANCOVA analysis was conducted with log(Qint) as the covariate (see Figures 25 for curvilinear relationships with Qint). In case of significant slope differences (Table 2), mean comparisons should be interpreted with caution.

Quercus robur8.2 ± 0.5a310.8 ± 2.9a53.5 ± 3.3a121 ± 8a18.0 ± 2.5a0.171 ± 0.019a1.31 ± 0.11a
Populus tremula11.22 ± 0.38b310.8 ± 1.5a71.6 ± 2.7b133 ± 5b16.6 ± 1.3a0.154 ± 0.007a0.88 ± 0.05b
Salix caprea8.1 ± 0.6a303.3 ± 3.9a54.6 ± 4.4a129 ± 5a11.4 ± 1.5b0.105 ± 0.012b0.81 ± 0.09b
Salix alba12.4 ± 0.8c324.8 ± 3.5b78 ± 6b138 ± 6b13.0 ± 1.4a0.143 ± 0.012a0.62 ± 0.05b

[40] Interspecific differences in the plasticity of studied characteristics, i.e., the slope of the trait versus Qint relationship, were evaluated by covariance analyses including the species × Qint interaction term. The plasticity in foliage structural, chemical, photosynthetic and isoprene emission traits was generally the lowest in P. tremula. Structural and chemical plasticity and the plasticity in isoprene emission traits was the largest in Q. robur and the plasticity in photosynthetic traits in S. alba (Table 2).

Table 2. Comparison of Slopes of Log(Qint) Versus Leaf Dry Mass Per Unit Area, Nitrogen Content Per Area, Net Assimilation Rate and Photosynthetic Electron Transport Rate From Gas Exchange and Fluorescence, Isoprene Emission Rate Per Unit Area and Dry Mass, and Percentage of Carbon Used for Isoprene Synthesis Among Four Studied Temperate Deciduous Speciesa
SpeciesMA (g m−2)NA (g m−2)A (μmol m−2 s−1)JG (μmol m−2 s−1)JF (μmol m−2 s−1)ES,A (nmol m−2 s−1)ES,M (nmol g−1 s−1)γA (%)
  • a

    Leaf dry mass per unit area (MA), nitrogen content per area (NA), net assimilation rate (A), photosynthetic electron transport rate from gas exchange (JG) and fluorescence (JF), isoprene emission rate per unit area (ES,A) and dry mass (ES,M), percentage of carbon used for isoprene synthesis (γA). Slopes were compared by separate slope ANCOVA analyses with log(Qint) as covariate. Slopes with the same letter are not significantly different (P > 0.05).

Quercus robur62.7a1.89a7.03a41.9a91.0a32.1a0.231a1.60a
Populus tremula42.0b0.911b3.85b26.8a54.5bc14.0b0.068b0.422b
Salix caprea41.5b0.896b6.51a47.8b52.7b19.4c0.147c0.893ab
Salix alba59.8c1.01b12.0c75.6b85.1c21.8c0.168c0.583b

3.5. Whole-Canopy Isoprene Emission Fluxes

[41] Predicted canopy isoprene emission rates (EC) based on observed within-canopy variations in isoprene emission potentials (Figure 4a, EC,base) matched the overall species differences in isoprene emission capacity (Tables 2 and 3). Refitting the observed ES versus Qint relationships with linear regressions and calculating the canopy emission fluxes (EC,linear) resulted in minor differences between EC,base and EC,linear for Q. robur (−0.3% for EC,linear), P. tremula (+0.6%) and S. caprea (−1.5%). However, EC,linear overestimated canopy isoprene emission by ca. 11% in S. alba.

Table 3. Canopy Isoprene Emission Rate (nmol m−2 s−1) in Four Deciduous Temperate Tree Speciesa
SpeciesEC,baseEC,linearEC,generalEC,averageEC,maxEC,big-leafEC,base,TEC,base,αEC,average,TEC,average,α
  • a

    Estimated at constant leaf temperature of 25°C and constant quantum yield of isoprene emission (α) of 0.0027 mol mol−1 and considering the observed nonlinear variations in isoprene emission potential in the canopy (Figure 4a, EC,base), fitting the same relationships using linear regressions (EC,linear), using a generalized “canopy response function of ES (Figure 7, EC,general), assuming that all leaves in the canopy have the same isoprene emission potential equal to average emission potential (EC,average) and to the emission potential of the topmost leaves (EC,max), and using a big-leaf model (EC,big-leaf, Appendix A). For comparison, simulations with realistic within-canopy gradient in temperature (27.1°C at the top and 22.9°C at the bottom, with average leaf-area weighted temperature being 25°C) and α (0.00174 mol mol−1 at the top and 0.0040 mol mol−1 at the bottom) are depicted for the scenario based on observed variation in isoprene emission potentials (EC,base,T and EC,base,α) and for the scenario based on average whole-canopy isoprene emission potentials (EC,average,T and EC,average,α). All simulations were conducted at incident quantum flux density of 1000 μmol m−2 s−1 and for whole-canopy leaf area index of 5.8 m2 m−2 determined from hemispherical photographs using a clumping index of 0.47.

Quercus robur79.979.084.967.1130.761.890.583.471.873.7
Populus tremula69.169.562.063.795.545.276.274.052.769.9
Salix caprea50.549.452.742.881.238.457.152.945.847.0
Salix alba67.574.666.358.9102.148.375.671.262.964.6

[42] The generalized response of standardized isoprene emission versus Qint (“canopy response curve” generally fitted the data well with greatest discrepancies observed for species with lowest (P. tremula) and largest plasticity (Q. robur) (Figure 7). In line with these observations, emission fluxes calculated on the basis of the generalized response (EC,general) deviated from the isoprene emission fluxes based on species-specific regressions (EC,base) most in P. tremula (−11%) and Q. robur (+6%). However, the average deviation was less than 0.5% for the simulations with generalized canopy function.

image

Figure 7. Standardized isoprene emission factor in relation to Qint (“canopy response function”) in the four studied species. In all species, ES,A versus Qint responses were fitted by empirical nonlinear relationships best describing the data (Figure 4a for individual relationships). ES at above-canopy Qint (37.3 mol m−2 d−1) was further calculated for every species (ES,0), and all data for given species were normalized with respect to ES,0. All standardized data pooled were further fitted by a common nonlinear regression equation forced to equal to 1.0 at above-canopy Qint. The relationship obtained (r2 = 0.85 for all species pooled) describes the canopy response function f(Qint). The actual gradient used to simulate canopy emission fluxes is given for every species as ES,0f(Qint).

Download figure to PowerPoint

[43] Assuming a constant ES equal to leaf-area weighted average ES for all leaves (estimate EC,average) resulted in 9% (P. tremula) to 19% (Q. robur) underestimation of canopy isoprene emissions compared with EC,base (Table 3 and Figure 8). In contrast, assuming that all leaves have an ES equal to the topmost leaves (estimate EC,max) overestimated canopy emissions by 40% (P. tremula) to 68% (Q. robur) relative to EC,base (Table 3 and Figure 8). Big-leaf model (Appendix A) that assumes that ES varies in direct proportion to Qint (estimate EC,big-leaf), underestimated canopy emissions by 35% (P. tremula) to 22% (Q. robur) relative to EC,base (Table 3 and Figure 8). The differences between EC values simulated by simplified scaling approaches and EC,base depended on species differences in ES plasticity (Figures 4a and 8). The underestimation of EC by EC,average and overestimation by EC,max relative to EC,base was larger in species with greater within-canopy variation in ES (Figure 8). In contrast, the underestimation by EC,big-leaf, which is based on particularly large within-canopy gradient, decreased with increasing within-canopy variation in ES (Figure 8).

image

Figure 8. Influence of within-canopy variations in isoprene emission potentials (ES) on estimation of canopy isoprene emission rate (EC) by various approaches in P. tremula (Pt), S. alba (Sa), S. caprea (Sc), and Q. robur (Qr). EC was either predicted by considering the observed within-canopy variation in ES (Figure 4a, EC,base), using a generalized shape of within-canopy variation in isoprene emission potential (“canopy response function”, Figure 7, EC,general), or assuming that ES is constant and equal to average ES (EC,average), or assuming that ES is constant and equal to ES of topmost leaves (EC,max) or using a big-leaf model (EC,big-leaf, Appendix A) that assumes that ES is directly proportional to average light availability. Demonstrated are the values of EC,average, EC,max, and EC,big-leaf relative to EC,base (Table 3 for nonnormalized EC values). The ratio of ES values between canopy top and bottom characterizes the magnitude of the ES gradient in the canopy. ES at the top was calculated for above-canopy Qint of 37.3 mol m−2 d−1 from the regressions depicted in Figure 4a, and ES at the bottom was analogously calculated for Qint of 4 mol m−2 d−1. Simulated values for the four species were fitted by nonlinear regressions in the form of y = axb, and all regressions are significant at P < 0.01 (r2 > 0.97).

Download figure to PowerPoint

3.6. Potential Effects of Within-Canopy Variations in Temperature and Quantum Yield for Isoprene Emission

[44] Compared to EC,base simulated using constant temperature and constant quantum yield for isoprene emission (α), the simulation with realistic within-canopy variation in temperature, but still having the same leaf-area weighted average canopy temperature of 25°C (EC,base,T) resulted on average by 12% higher emissions (Table 3), reflecting the circumstance that higher temperatures in the upper canopy combined with higher emission potentials can magnify the effects of within-canopy variations in ES. The comparisons between various models were nevertheless qualitatively similar with generally smaller differences due to temperature gradient than due to consideration of the gradient in isoprene emissions. On average, EC,average underestimated EC,base by 13%, EC,max overestimated by 53% and EC,big-leaf underestimated by 27%, while EC,average,T underestimated EC,base,T by 20%, EC,max,T overestimated by 46% and EC,big-leaf,T underestimated by 28%.

[45] Consideration of within-canopy variation in α (EC,base,α) resulted on average in 5% overestimation of EC,base, reflecting greater isoprene emissions from leaves in lower canopy than predicted by using a constant α (Table 3). Analogously, EC,average,α  overestimated EC,average by 9% (Table 3). However, differences among simulations with different models were little affected by considering within-canopy variation in α: on average EC,average,α underestimated EC,base,α by 10%, EC,max,α overestimated by 60% and EC,big-leaf,α underestimated by 31%. Although these simulations demonstrate that consideration of both within-canopy variations in temperature and α is relevant, they also indicate that lack of consideration of within-canopy variation in the emission potentials leads to largest source of errors in whole-canopy emission predictions.

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

4.1. Within-Canopy Variations in Light, Leaf Structure, Chemistry, and Photosynthesis

[46] As is typical in temperate tree canopies (for reviews, see Niinemets [2007, 2010]), light availability varied ca. an order of magnitude in the studied tree canopies (Figure 1). This extensive light variation was accompanied by significant increases in leaf dry mass per unit area (MA, Figure 2a), nitrogen and sulfur contents per area (Figure 2b) and area-based photosynthetic potentials (Figure 3). These patterns broadly agree with previous observations on plant structural and photosynthetic variation in plant canopies [Kazda et al., 2000; Le Roux et al., 2001; Meir et al., 2002; Niinemets et al., 2004d; Niinemets, 2007], and collectively result in greater canopy photosynthetic productivity than a constant average rate of photosynthesis for all leaves in the canopy [Baldocchi and Harley, 1995; Gutschick and Wiegel, 1988; Niinemets and Anten, 2009].

[47] As in the previous studies (see Niinemets [2007] for a review), the within-canopy plasticity of key limiting mineral nutrients per unit area and photosynthetic potentials per unit area was mainly driven by structural modifications, i.e., accumulation of mesophyll cell layers per unit leaf area, reflected in enhanced MA at higher light. Similarly to previous studies, variations in mass-based chemical element contents and in photosynthetic potentials were generally not related to Qint, demonstrating that photosynthetic potentials of average leaf cells were essentially constant within the light gradients [Evans and Poorter, 2001; Niinemets et al., 1998b].

4.2. Canopy Gradients in Isoprene Emission Potentials

[48] Similarly to MA, nitrogen and photosynthetic potentials per unit area, foliage isoprene emission potentials, ES, strongly increased with increasing light availability in the canopy. Differently from other foliage traits, both area (ES,A) and mass-based (ES,M) isoprene emission potentials were strongly correlated with Qint (Figure 4). The range of variation in ES,M values was ca. 1.5-fold less than in ES,A, demonstrating that both modifications in isoprene emission potentials of single cells and light-driven structural modifications in MA played a role in light-driven variations in ES,A (Figure 6c).

[49] Although quantitative relationships of ES with integrated long-term light environment have not been developed so far, greater ES,A values in upper canopy relative to lower canopy [Funk et al., 2006; Harley et al., 1996; 1997; Lerdau and Throop, 2000] and in plants grown under high relative to low light [Litvak et al., 1996] have been observed, and it has been demonstrated that this variation correlates with MA [Harley et al., 1997; Niinemets et al., 1999d]. However, in previous studies, ES,M values were not always related to light availability and MA [Harley et al., 1997; Niinemets et al., 1999d]. Previously reported within-canopy variations in ES,A were also generally less than observed in our study, e.g., 1.6-fold between canopy top and bottom in Q. alba [Harley et al., 1997], ca. 1.5-fold in the work of Brosimum utile and ca. 2.5-fold in Dussia munda [Lerdau and Throop, 2000], 3-fold in Eucalyptus saligna [Funk et al., 2006], on average 3.8-fold for emitting species in a deciduous temperate canopy [Geron et al., 1997], and 4.2-fold in Liquidambar styraciflua [Harley et al., 1996]. The variation was on average 3.2-fold in Populus tremuloides and 2.2-fold in Q. alba trees grown under high and low light [Litvak et al., 1996]. These modest ranges, compared with up to 27-fold variation observed in our study, suggest that full within-canopy variation in isoprene emission potentials may have not been recovered in the previous studies due to lack of explicit light measurements. In fact, when examined in relation to MA, 5-fold to 6-fold range of ES,A has been observed in deciduous oak canopies [Harley et al., 1997], indirectly supporting this conclusion. Typically, “upper” or “sun” canopy is defined as the first few meters of tree crown. However, a simple Beer's model predicts that light availability decreases already almost 30% within the first 1 m2 m−2 leaf area in a canopy with randomly dispersed foliage and light extinction coefficient of 0.5 (equation (A3)). Thus, accurate light measurements are absolutely necessary to characterize the overall variation in foliage plasticity along light gradients.

[50] In previous studies, ES,A values observed across environmental gradients have been found to correlate with foliage nitrogen content and with foliage photosynthetic potentials, in particular, with foliage photosynthetic electron transport rate [Funk et al., 2006; Litvak et al., 1996; Niinemets et al., 1999d; Possell et al., 2004]. Analogous correlations of ES,A with foliage photosynthetic potentials were observed in our study, whereas the best correlation was observed with the rate of electron transport from fluorescence (JF). In developing leaves, the rate of foliage anatomical development and accumulation of rate-limiting proteins of photosynthetic machinery is proportional to incident light [Niinemets et al., 2004c], explaining the gradients in leaf structure and photosynthesis and also partly explaining the gradients in ES,A found in our study.

[51] However, within-canopy variations in isoprene emission potentials were larger than the variations in photosynthesis and photosynthetic electron transport (cf. Figures 3 and 4a). As a result, the fractions of carbon (Figure 5a) and electrons (Figure 5b) used for isoprene emission scaled positively with light. This result suggests that the isoprene synthesis pathway becomes more competitive for photosynthetic carbon and electrons at higher light (Figures 6a and 6b). Such greater competitiveness of the isoprene synthesis pathway can reflect greater investment in terminal enzymes controlling the pathway flux such as isoprene synthase as well as in enzymes responsible for the synthesis of the substrate, dimethylallyldiphosphate (DMADP), for isoprene synthesis. Previous studies have found important variations in isoprene synthase content [Lehning et al., 1999, 2001; Magel et al., 2006; Mayrhofer et al., 2005; Wiberley et al., 2008, 2009], and DMADP pool size [Wiberley et al., 2008, 2009] in plants grown under different conditions. For example, expression of isoprene synthase is elicited earlier and its activity reaches a higher level at greater temperatures [Wiberley et al., 2008, 2009]. Given the warmer temperatures in the upper canopy, these temperature-driven patterns likely contribute to within-canopy variations in ES,A and ES,M. So far, only a few model studies have attempted to include modifications in synthase activities to explain within-canopy variations in isoprene [Niinemets et al., 1999d] and monoterpene [Grote et al., 2010] emission rates, but clearly future experimental studies on expression of isoprene synthase and enzymes responsible for intermediate synthesis under different light conditions are called for.

4.3. What is the Physiological Significance of Within-Canopy Variation in Isoprene Emission Potentials?

[52] While the variation in photosynthetic potentials optimizes canopy photosynthesis for given foliage biomass investment in leaves, the significance of the steep within-canopy gradients in isoprene emission is currently less understood. Isoprene has been mainly implicated in the tolerance of heat stress [Sharkey and Singsaas, 1995; Singsaas et al., 1997]. Light and temperature are typically correlated in temperate tree canopies, and leaves in the canopy top are typically 4°C–5°C warmer than the leaves in the bottom of the canopy [Baldocchi et al., 2002; Niinemets and Valladares, 2004]. Thus, adjustment to higher risk of heat stress in the upper canopy can explain greater isoprene emission potentials in higher light. Furthermore, leaves can encounter highly variable light environments depending on locations of gaps in the canopy, cloudiness conditions, and wind-driven leaf and branch movements. During short-term periods of high-intensity light, lightflecks, leaf temperatures may transiently even rise up to 10°C higher than the ambient temperatures [Singsaas et al., 1999]. While in sunny days upper canopy leaves are exposed to continuous flux of high light, leaves in any position in the canopy can encounter such “heatflecks.” However, due to greater canopy gap fraction, the probability of such heatflecks is larger in the mid than in lower canopy. Acclimation to overall higher temperatures in the upper canopy and to heatflecks in the midcanopy can explain why the decline in ES from canopy top to bottom is relatively small in midcanopy, leading to curvilinearity in ES versus Qint relationships.

[53] There is also evidence that isoprene emission potential acclimates to enhanced irradiance in the absence of concomitant changes in temperature [Hanson and Sharkey, 2001a, 2001b]. In fact, simultaneously with the increase of the production of the smallest isoprenoid compound, isoprene, the content of higher molecular mass isoprenoids, photoprotective carotenoids, also increases with increasing light availability, thereby enhancing the capacity for nonradiative dissipation of excess light energy [Demmig-Adams and Adams, 2006; Hanson and Sharkey, 2001a; Niinemets et al., 1998a, 2003a]. It has been suggested that the role of isoprene emission is to keep the overall isoprenoid synthesis pathway active, allowing for rapid synthesis of higher molecular mass isoprenoids whenever needed [Owen and Peñuelas, 2005; Peñuelas et al., 2005; Rosenstiel et al., 2004]. Thus, according to this hypothesis, higher isoprene emission potential at greater light enables more rapid synthesis of photoprotective carotenoids.

[54] From a broader perspective, isoprene as a lipid-soluble molecule can play an important role in quenching of free radicals and thereby directly participate in reducing the oxidative stress inherent to photoinhibitory conditions. Thus, isoprene may function similarly to other lipid-soluble antioxidants such as α-tocopherol, quenching free radicals in the leaf lipid phase [Havaux and Niyogi, 1999]. As heat stress can amplify photoinhibitory damage [Mishra and Singhal, 1992], enhanced probability for combined high light and heat stress in higher canopy can provide an explanation for scaling of isoprene emission potentials with light. This evidence collectively suggests that steep gradients in isoprene emission may reflect the multiple roles isoprene plays in acclimation to enhanced stress in higher light.

4.4. Species Differences in Adjustment of Isoprene Emission Potentials Vis-à-Vis Structural and Photosynthetic Plasticity

[55] Significant differences in foliage structural (MA), chemical (NA), and photosynthetic plasticity, defined as the slope of trait versus log(Qint) relationships and plasticity in isoprene emission potentials were observed among the species (Table 2). In general, Q. robur was most and P. tremula the least responsive with respect to most traits (Table 2). Greater structural and photosynthetic plasticity was reflected in greater range in isoprene emission potentials as well (Table 2). In addition, higher light-driven variability in isoprene emission was associated with greater changes in the fraction of electrons and carbon used for isoprene emission (Table 2). Important differences in structural, chemical, and photosynthetic plasticity have been demonstrated among temperate tree species in a number of studies [e.g., Niinemets et al., 2003b; Portsmuth and Niinemets, 2007; Sánchez-Gómez et al., 2006; Valladares et al., 2002]. Although significant interspecific variation in isoprene emission potentials at given, typically high, light availability is well known [e.g., He et al., 2000; Kesselmeier and Staudt, 1999; Owen et al., 1998], to our knowledge, this is the first report showing important variations in the plasticity of isoprene emission potentials among temperate tree species. The physiological significance of such interspecific variations is not currently understood. Low capacity for variation in P. tremula may be associated with cooler leaf temperatures due to leaf fluttering in the upper canopy [Roden and Pearcy, 1993]. However, S. alba has strongly hairy foliage, an adaptation that also reduces leaf temperature [Ehleringer and Mooney, 1978], but still had high range of variation in leaf traits. Alternatively, differences in the light-driven changes in isoprene emission may be associated with overall variation in the adjustment of the capacity of isoprenoid synthesis pathway. Previous studies have demonstrated significant species differences in the variability of photoprotective carotenoid and tocopherol pools along light gradients [e.g., García-Plazaola et al., 2004; Niinemets et al., 1998a; 2003a], but the rates of synthesis of isoprene and nonvolatile isoprenoids have not been simultaneously assessed along light gradients.

4.5. Strategies for Estimation of Whole-Canopy Isoprene Emissions

[56] Large gradients and species variation in the gradients of isoprene emission potential have important implications for simulation of canopy isoprene emissions. As summarized in section 1, a variety of approaches have been used to model whole-canopy isoprene emissions, but to our knowledge no studies have so far used quantitative ES versus Qint relationships to parameterize a canopy model.

[57] Our simulations using the quantitative ES versus Qint relationships demonstrated that it is important to consider the within-canopy variation in isoprene emission potentials. Relative to the baseline scenario that considers within-canopy variation in the emission potentials, the constant above-canopy value overestimated the predicted canopy isoprene emission rate on average by 53%, while the constant leaf-area weighted average emission rate underestimated the canopy emission rate on average by 13% (Table 3 and Figure 8). Analogous biases by constant ES values were observed by a three-layer canopy model, but the overestimation by using constant upper canopy values was less (31%) and underestimation by using average values more (29%) [Harley et al., 1996, 1997] than observed here, likely reflecting the smaller gradient used and arbitrary division of canopy between only three layers in these studies.

[58] These discrepancies resulting from the use of constant isoprene emission potentials are somewhat larger than the errors in predicted canopy photosynthesis. Using constant foliage assimilation potentials, canopy photosynthetic production is biased by 10%–20% [Baldocchi and Harley, 1995; Gutschick and Wiegel, 1988; Niinemets and Anten, 2009]. These differences between the bias in canopy isoprene and photosynthesis estimates reflect the smaller within-canopy variation in photosynthetic potentials than in ES.

[59] Because of strong nonlinearity in the emission potentials versus Qint relationships (Figure 4), the big-leaf model that is based on the premise that the variation in isoprene emission potential is directly proportional to light, resulted on average by 27% lower estimates of EC than EC,base (Figure 8 and Table 3). This large negative discrepancy with the big-leaf model reflects the nonlinearity of ES,A versus Qint relationships, as the result of which ES,A decreases more sharply at lower than at higher light (Figure 4a), and accordingly, upper and midcanopy has higher isoprene emission than predicted by the big-leaf model, where ES,A decreases linearly with Qint. Analogously, big-leaf models have been shown to underestimate canopy photosynthetic production [Anten, 1997; Niinemets and Anten, 2009]. In contrast, assuming a linear variation of ES with Qint only moderately altered the simulated emissions (Table 3). Differently from the big-leaf model, linear fits to the actual data have a positive intercept, and thus, ES decreases less deeply than predicted by a big-leaf model.

[60] The discrepancies in canopy isoprene emission fluxes significantly varied among the species. In particular, for both simulations with a constant emission rate, the under- and overestimation relative to the baseline scenario increased as the within-canopy gradient in isoprene emission potentials increased (Figure 8). For the big-leaf model, which assumes the strongest gradient in the emission potentials, the underestimation relative to the baseline scenario decreased with increasing the actual within-canopy gradient in the emission potentials (Figure 8), further emphasizing that differences in species plasticity significantly alter the magnitude of errors. Use of a generalized shape for the canopy function (Figure 7) also resulted in largest deviations for species with smallest and largest actual gradient (Figure 8). Nevertheless, the average deviations were small, on average less than 0.5%, suggesting that generalized canopy response function along with the above-canopy values of isoprene emission factor may provide a practical way to consider variation in ES in large-scale models.

[61] Additional simulations conducted to evaluate the effect of within-canopy variations in temperature and initial quantum yield of isoprene emission (α) demonstrated that both factors significantly alter the integrated canopy emission fluxes (Table 3). However, the relative differences between various model scenarios were affected only to a minor degree, indicating that independently of variations in temperature and α, the within-canopy variation in the emission potentials exerts a major control on canopy isoprene emissions.

5. Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

[62] These simulations collectively suggest that crude simplifications in the scaling of isoprene emission measurements can lead to large errors. For accurate scaling of leaf level estimates to whole-canopy, within-canopy gradients in the emission potentials must be included in the scaling algorithms. In complex mixed multiple species stands, even a generalized scaling algorithm (“canopy response function”) is superior to using constant above-canopy or constant leaf area weighted values of the emission potential and such a simplification can provide a cost-effective alternative to analyzing the within-canopy variations in each individual species.

Appendix A:: Simple Big-Leaf Isoprene Emission Model

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

[63] Assuming a constant temperature, isoprene emission rate E at any given quantum flux density Q is given as the product of the light response function of isoprene emission f(Q) and the emission rate under standardized quantum flux density ES [Guenther et al., 1991, 1993],

  • equation image

f(Q) = 1 under standardized quantum flux density (typically taken as 1000 μmol m−2 s−1). An hyperbolic function was used to describe the f(Q) function [Guenther et al., 1991, 1993], and thus, E is given as

  • equation image

where α is the quantum yield of isoprene emission for an incident light and ɛL is the scaling parameter to result f(Q) = 1 under standardized Q.

[64] In most simplified way, Q at any depth in an homogeneous canopy can be expressed on the basis of Beer's law as

  • equation image

where Q0 is the quantum flux density above the canopy, LC is the cumulative leaf area index above a given leaf layer in the plant stand, and K is the extinction coefficient for light that determines how strongly Q decreases with increasing LC. To describe the light gradient through the canopy, we define relative quantum flux density qR as Q/Q0. Thus, Q becomes

  • equation image

As is common in big-leaf models, we assume that the quantum yield α is a constant and also that ES at any height in the canopy scales proportionally with qR [Amthor, 1994; Dai et al., 2004; Niinemets and Anten, 2009; Sellers et al., 1992],

  • equation image

where ES,0 is the value of ES for uppermost leaves (qR = 1). Combining equations (A4) and (A5) and substituting ES into equation (A2) gives

  • equation image

Under these assumptions, only qR varies with the depth in the canopy.

[65] Canopy isoprene emission rate EC is obtained by integrating leaf isoprene emission rate (E) over total canopy leaf area index L,

  • equation image

Replacing E from equation (A6), we obtain

  • equation image

Given that ES,0ɛLαQ0 is independent of LC, equation (A8) becomes

  • equation image

Considering that qR = image and integrating the solution of equation (A9) is given as

  • equation image

For internal consistency, ES,0 should be the value of the emission rate observed at Q0, i.e., correspond to the above-canopy quantum flux density and, f(Q) function (equations (A1) and (A2)) should equal to 1 at Q0. However, the standardized Q at which f(Q) = 1, is typically taken as 1000 μmol m−2 s−1 and ES is also defined at this light intensity [Guenther et al., 1991, 1993]. Thus, application of the big-leaf model for higher midday above-canopy Q values requires reparameterization of ɛL such that f(Q0) = 1, and accordingly ES,0 should be defined at Q0. In our study, effective leaf area index (Le) was estimated by hemispheric photography (equation (2), Figure 1b). Estimation of leaf area index from hemispherical photographs assumes that foliage is randomly dispersed. However, in real plant stands, foliage is often aggregated. Effective leaf area index Le is related to total leaf area index, L = Le/λ0, where λ0 is the coefficient of spatial aggregation that decreases with increasing the degree of aggregation [Cescatti and Niinemets, 2004]. If Le is used instead of L, equation (A3) becomes

  • equation image

where Le,C is the effective cumulative leaf area index. Thus, K should be replaced by the effective extinction coefficient Kλ0 in equation (A10). In simulations with varying temperature and initial quantum yield, we used numeric integration, while ES at any given canopy location was determined under the condition specified by equation (A5).

Notation
Definition and units of used acronyms.

 

A

net assimilation rate. (μmol m−2 s−1)

CM

leaf carbon content per dry mass. (%)

ci

CO2 concentration in substomatal cavities. (μmol mol−1)

DF

leaf dry to fresh mass ratio. (g g−1)

EC

canopy isoprene emission rate. (nmol m−2 s−1)

EC,average

EC simulated assuming that ES of all leaves in the canopy equals to the average ES. (nmol m−2 s−1)

EC,average,T

EC simulated as EC,average, but assuming realistic within-canopy gradient in temperature. (nmol m−2 s−1)

EC,average,α

EC simulated as EC,average, but considering possible within-canopy gradient in α. (nmol m−2 s−1)

EC,base

“baseline” simulated EC considering the observed within-canopy variation in ES. (nmol m−2 s−1)

EC,base,T

EC simulated as EC,base, but assuming realistic within-canopy gradient in temperature. (nmol m−2 s−1)

EC,base,α

EC simulated as EC,base, but considering possible within-canopy gradient in α. (nmol m−2 s−1)

EC,big-leaf

EC simulated by a big-leaf model (Appendix A) that assumes that ES varies in direct proportion to Qint. (nmol m−2 s−1)

EC,big-leaf,T

EC simulated as EC,big-leaf, but assuming realistic within-canopy gradient in temperature. (nmol m−2 s−1)

EC,big-leaf,α

EC simulated as EC,big-leaf, but considering possible within-canopy gradient in α. (nmol m−2 s−1)

EC,general

EC simulated using species-specific ES,0 values and a generalized within-canopy gradient f(Qint). (nmol m−2 s−1)

EC,linear

EC simulated as in EC,base, but refitting the nonlinear species-specific ES versus Qint relationships by linear regressions. (nmol m−2 s−1)

EC,max

EC simulated assuming that ES of all leaves in the canopy equals to the above-canopy ES. (nmol m−2 s−1)

ES

isoprene emission rate under standardized conditions (leaf temperature of 25°C, incident quantum flux density of 1000 μmol m−2 s−1), either expressed per unit area or dry mass.

ES,0

ES,A value corresponding to above-canopy Qint. (nmol m−2 s−1)

ES,A

ES expressed on leaf area basis. (nmol m−2 s−1)

ES,M

ES expressed on leaf dry mass basis. (nmol g−1 s−1)

F

steady state fluorescence yield of light-adapted sample.

Fm

maximum fluorescence yield of light-adapted sample.

G

seasonal average total solar radiation. (MJ m−2 d−1)

gs

stomatal conductance to water vapor. (mol m−2 s−1)

IB

fraction of potential penetrating direct solar radiation of open sky (direct site factor).

ID

fraction of diffuse light of open sky (diffuse site factor).

JF

photosynthetic electron transport from chlorophyll fluorescence. (μmol m−2 s−1)

JG

photosynthetic electron transport from gas exchange. (μmol m−2 s−1)

K

light extinction coefficient.

L

stand leaf area index. (m2 m−2)

LC

cumulative leaf area index. (m2 m−2)

Le

effective stand leaf area index. (m2 m−2)

MA

leaf dry mass per area. (g m−2)

NM

leaf nitrogen content per dry mass. (%)

NA

leaf nitrogen content per area. (g m−2)

pD

fraction of diffuse radiation in total solar radiation.

PL

average projected leaf area (leaf size). (cm2)

Q

instantaneous incident photosynthetic quantum flux density. (μmol m−2 s−1)

Q0

Q above the canopy. (μmol m−2 s−1)

Qint

seasonal average daily integrated quantum flux density (“long-term light”). (mol m−2 d−1)

qR

relative quantum flux density (Q/Q0).

Rd

rate of nonphotorespiratory respiration continuing in the light. (μmol m−2 s−1)

SM

leaf sulfur content per dry mass. (%)

SA

leaf sulfur content per area. (g m−2)

T

canopy gap fraction corresponding to zenith angle θ.

Tav

seasonal average leaf temperature. (°C)

Vcmax

the maximum carboxylase activity of Rubisco. (μmol m−2 s−1)

α

quantum yield of isoprene emission. (mol mol−1)

γA

percentage of carbon used for isoprene emission. (%)

γJ

percentage of electrons used for isoprene emission. (%)

Γ*

hypothetical CO2 compensation point in the absence of Rd. (μmol mol−1)

ɛL

scaling parameter of the light response function (equation (A2)).

θ

zenith angle.

κ

ratio of photosynthetic quantum flux density to global solar radiation. (mol MJ−1)

λ0

clumping index.

ξ

leaf absorptance.

ρPSII

fraction of light absorbed by photosystem II.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information

[66] This work was supported by Human Frontiers of Science Program (www.hfsp.org), Estonian Ministry of Science and Education (grant SF1090065s07), and Estonian Science Foundation (postdoctoral grant JD101 and grant 7645). We acknowledge the skillful technical assistance by Kaia Kask, Miguel Portillo, Pille Randjärv, and Evi Vaino and thank Anne Jõeveer (Estonian Meteorological and Hydrological Institute, Tõravere, Estonia) for global solar radiation data.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information
  • Amthor, J. S. (1994), Scaling CO2-photosynthesis relationships from the leaf to the canopy, Photosynth. Res., 39, 321350.
  • Anten, N. P. R. (1997), Modelling canopy photosynthesis using parameters determined from simple nondestructive measurements, Ecol. Res., 12, 7788.
  • Arneth, A., R. K. Monson, G. Schurgers, Ü. Niinemets, and P. I. Palmer (2008), Why are estimates of global isoprene emissions so similar (and why is this not so for monoterpenes)? Atmos. Chem. Phys., 8, 46054620.
  • Baldocchi, D. D. (1991), Canopy control of trace gas emissions, in Trace Gas Emissions by Plants, edited by T. D. Sharkey et al., pp. 293333, Academic, San Diego, Calif.,
  • Baldocchi, D., and S. Collineau (1994), The physical nature of solar radiation in heterogeneous canopies: Spatial and temporal attributes, in Exploitation of Environmental Heterogeneity by Plants. Ecophysiological Processes Above- and Belowground, edited by M. M. Caldwell, and R. W. Pearcy, pp. 2171, Academic, San Diego, Calif.,
  • Baldocchi, D. D., and P. C. Harley (1995), Scaling carbon dioxide and water vapour exchange from leaf to canopy in a deciduous forest: II. Model testing and application, Plant Cell Environ., 18, 11571173.
  • Baldocchi, D., A. Guenther, P. Harley, L. Klinger, P. Zimmerman, B. Lamb, and H. Westberg (1995), The fluxes and air chemistry of isoprene above a deciduous hardwood forest, Philos. Trans. R. Soc. London. Ser. A, 351, 279296.
  • Baldocchi, D. D., J. D. Fuentes, D. R. Bowling, A. A. Turnipseed, and R. K. Monson (1999), Scaling isoprene fluxes from leaves to canopies: Test cases over a boreal aspen and a mixed species temperate forest, J. Appl. Meteorol., 38, 885898.
  • Baldocchi, D. D., K. B. Wilson, and L. Gu (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 Physiol., 22, 10651077.
  • Beauchamp, J., A. Wisthaler, A. Hansel, E. Kleist, M. Miebach, Ü. Niinemets, U. Schurr, and J. Wildt (2005), Ozone induced emissions of biogenic VOC from tobacco: Relations between ozone uptake and emission of LOX products, Plant Cell Environ., 28, 13341343.
  • Behnke, K., B. Ehlting, M. Teuber, M. Bauerfeind, S. Louis, R. Hänsch, A. Polle, J. Bohlmann, and J.-P. Schnitzler (2007), Transgenic, nonisoprene emitting poplars don't like it hot, Plant J., 51, 485499.
  • Bréda, N. J. J. (2003), Ground-based measurements of leaf area index: A review of methods, instruments and current controversies, J. Exp. Bot., 54, 24032417.
  • Brilli, F., P. Ciccioli, M. Frattoni, M. Prestininzi, A. F. Spanedda, and F. Loreto (2009), Constitutive and herbivore-induced monoterpenes emitted by Populus x euroamericana leaves are key volatiles that orient Chrysomela populi beetles, Plant Cell Environ., 32, 542552.
  • Brooks, A., and G. D. Farquhar (1985), Effects of temperature on the CO2/O2 specificity of ribulose-1,5-bisphosphate carboxylase/oxygenase and the rate of respiration in the light. Estimates from gas exchange measurements on spinach, Planta, 165, 397406.
  • Buckley, P. T. (2001), Isoprene emissions from a Florida scrub oak species grown in ambient and elevated carbon dioxide, Atmos. Environ., 35, 631634.
  • Campbell, G. S., and J. M. Norman (1989), The description and measurement of plant canopy structure, in Plant Canopies: Their Growth, Form and Function, edited by G. Russell et al., pp. 119, Cambridge Univ. Press, Cambridge, U. K.,
  • Cescatti, A., and Ü. Niinemets (2004), Sunlight capture. Leaf to landscape, in Photosynthetic Adaptation. Chloroplast to Landscape, edited by W. K. Smith, T. C. Vogelmann, and C. Chritchley, pp. 4285, Springer, Berlin.
  • Claeys, M., et al. (2004), Formation of secondary organic aerosols through photooxidation of isoprene, Science, 303, 11731176.
  • Copolovici, L. O., I. Filella, J. Llusià, Ü. Niinemets, and J. Peñuelas (2005), The capacity for thermal protection of photosynthetic electron transport varies for different monoterpenes in Quercus ilex, Plant Physiol., 139, 485496.
  • Copolovici, L., A. Kännaste, and Ü. Niinemets (2009), Gas chromatography-mass spectrometry method for determination of monoterpene and sesquiterpene emissions from stressed plants, Stud. Univ. Babes-Bolyai Chem., 54, 329339.
  • Dai, Y.-J., R. E. Dickinson, and Y. P. Wang (2004), A two-big-leaf model for canopy temperature, photosynthesis, and stomatal conductance, J. Clim., 17, 22812299.
  • Demmig-Adams, B., and W. W. Adams III (2006), Photoprotection in an ecological context: The remarkable complexity of thermal energy dissipation: Tansley review, New Phytol., 172, 1121.
  • de Pury, D. G. G., and G. D. Farquhar (1997), Simple scaling of photosynthesis from leaves to canopies without the errors of big-leaf models, Plant Cell Environ., 20, 537557.
  • Edwards, G. E., and N. R. Baker (1993), Can CO2 assimilation in maize leaves be predicted accurately from chlorophyll fluorescence analysis? Photosynth. Res., 37, 89102.
  • Ehleringer, J. R., and H. A. Mooney (1978), Leaf hairs: Effects on physiological activity and adaptive value to a desert shrub, Oecologia, 37, 183200.
  • Evans, J. R., and H. Poorter (2001), Photosynthetic acclimation of plants to growth irradiance: The relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain, Plant Cell Environ., 24, 755767.
  • Flexas, J., F. Loreto, Ü. Niinemets, and T. Sharkey (2009), Preface, J. Exp. Bot., 60, 22152216.
  • Fuentes, J. D., D. Wang, G. den Hartog, H. H. Neumann, T. F. Dann, and K. J. Puckett (1995), Modelled and field measurements of biogenic hydrocarbon emissions from a Canadian deciduous forest, Atmos. Environ., 29, 30033017.
  • Fuentes, J. D., D. Wang, and L. Gu (1999), Seasonal variations in isoprene emissions from a boreal aspen forest, J. Appl. Meteorol., 38, 855869.
  • Funk, J. L., C. P. Giardina, A. Knohl, and M. T. Lerdau (2006), Influence of nutrient availability, stand age, and canopy structure on isoprene flux in a Eucalyptus saligna experimental forest, J. Geophys. Res., 111, G02012, doi:10.1029/2005JG000085.
  • García-Plazaola, J. I., J. M. Becerril, A. Hernández, Ü. Niinemets, and H. Kollist (2004), Acclimation of antioxidant pools to the light environment in a natural forest canopy, New Phytol., 163, 8797.
  • Genty, B., J.-M. Briantais, and N. R. Baker (1989), The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence, Biochim. Biophys. Acta, 990, 8792.
  • Geron, C. D., A. B. Guenther, and T. E. Pierce (1994), An improved model for estimating emissions of volatile organic compounds from forests in the eastern United States, J. Geophys. Res., 99(D6), 12,77312,791, doi:10.1029/94JD00246.
  • Geron, C. D., D. Nie, R. R. Arnts, T. D. Sharkey, E. L. Singsaas, P. J. Vanderveer, A. Guenther, J. E. Sickles, and T. E. Kleindienst (1997), Biogenic isoprene emission: Model evaluation in a southeastern US bottomland deciduous forest, J. Geophys. Res., 102(D15), 18,88918,901, doi:10.1029/97JD00968.
  • Grote, R. (2007), Sensitivity of volatile monoterpene emission to changes in canopy structure: A model-based exercise with a process-based emission model, New Phytol., 173, 550561, doi:10.1111/j.1469-8137.2006.01946.x.
  • Grote, R., and Ü. Niinemets (2008), Modeling volatile isoprenoid emission—A story with split ends, Plant Biol., 10, 828, doi:10.1055/s-2007-964975.
  • Grote, R., T. Keenan, A.-V. Lavoir, and M. Staudt (2010), Process-based simulation of seasonality and drought stress in monoterpene emission models, Biogeosciences, 7, 257274.
  • Guenther, A. (1997), Seasonal and spatial variations in natural volatile organic compound emissions, Ecol. Appl., 7, 3445.
  • Guenther, A. B., R. K. Monson, and R. Fall (1991), Isoprene and monoterpene emission rate variability: Observations with Eucalyptus and emission rate algorithm development, J. Geophys. Res., 96(D6), 10,79910,808, doi:10.1029/91JD00960.
  • Guenther, A. B., P. R. Zimmerman, P. C. Harley, R. K. Monson, and R. Fall (1993), Isoprene and monoterpene emission rate variability: Model evaluations and sensitivity analyses, J. Geophys. Res., 98(D7), 12,60912,617, doi:10.1029/93JD00527.
  • Guenther, A., et al. (1995), A global model of natural volatile compound emissions, J. Geophys. Res., 100(D5), 88738892, doi:10.1029/94JD02950.
  • Guenther, A., B. Baugh, G. Brasseur, J. Greenberg, P. Harley, L. Klinger, D. Serça, and L. Vierling (1999), Isoprene emission estimates and uncertainties for the Central African EXPRESSO study domain, J. Geophys. Res., 104(D23), 30,62530,639, doi:10.1029/1999JD900391.
  • Guenther, A., C. Geron, T. Pierce, B. Lamb, P. Harley, and R. Fall (2000), Natural emissions of nonmethane volatile organic compounds, carbon monoxide, and oxides of nitrogen from North America, Atmos. Environ., 34, 22052230.
  • Guenther, A., T. Karl, P. Harley, C. Wiedinmyer, P. I. Palmer, and C. Geron (2006), Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature), Atmos. Chem. Phys., 6, 31813210.
  • Gutschick, V. P. (1988), Optimization of specific leaf mass, internal CO2 concentration, and chlorophyll content in crop canopies, Plant Physiol. Biochem., 26, 525537.
  • Gutschick, V. P., and F. W. Wiegel (1988), Optimizing the canopy photosynthetic rate by patterns of investment in specific leaf mass, Am. Nat., 132, 6786.
  • Hakola, H., J. Rinne, and T. Laurila (1998), The hydrocarbon emission rates of tea-leafed willow (Salix phylicifolia), silver birch (Betula pendula) and European aspen (Populus tremula), Atmos. Environ., 32, 18251833.
  • Hallquist, M., et al. (2009), The formation, properties and impact of secondary organic aerosol: Current and emerging issues, Atmos. Chem. Phys., 9, 51555236.
  • Hanson, D. T., and T. D. Sharkey (2001a), Effect of growth conditions on isoprene emission and other thermotolerance-enhancing compounds, Plant Cell Environ., 24, 929936.
  • Hanson, D. T., and T. D. Sharkey (2001b), Rate of acclimation of the capacity for isoprene emission in response to light and temperature, Plant Cell Environ., 24, 937946.
  • Harley, P., A. Guenther, and P. Zimmerman (1996), Effects of light, temperature and canopy position on net photosynthesis and isoprene emission from sweetgum (Liquidambar styraciflua) leaves, Tree Physiol., 16, 2532.
  • Harley, P., A. Guenther, and P. Zimmerman (1997), Environmental controls over isoprene emission in deciduous oak canopies, Tree Physiol., 17, 705714.
  • Havaux, M., and K. K. Niyogi (1999), The violaxanthin cycle protects plants from photooxidative damage by more than one mechanism, Proc. Natl. Acad. Sci. U. S. A., 96, 87628767.
  • He, C., F. Murray, and T. Lyons (2000), Monoterpene and isoprene emissions from 15 Eucalyptus species in Australia, Atmos. Environ., 34, 645655.
  • Huber, L., P. Laville, and J. D. Fuentes (1999), Uncertainties in isoprene emissions from a mixed deciduous forest estimated using a canopy microclimate model, J. Appl. Meteorol., 38, 899912.
  • Hurlbert, S. H. (1984), Pseudoreplication and the design of ecological field experiments, Ecol. Monogr., 54, 187211.
  • Li-Cor Inc. (1992), LAI-2000 Plant Canopy Analyzer, operating manual, Lincoln, Nebr.,
  • Kazda, M., J. Salzer, and I. Reiter (2000), Photosynthetic capacity in relation to nitrogen in the canopy of a Quercus robur, Fraxinus angustifolia and Tilia cordata flood plain forest, Tree Physiol., 20, 10291037.
  • Keenan, T., Ü. Niinemets, S. Sabate, C. Gracia, and J. Peñuelas (2009), Process based inventory of isoprenoid emissions from European forests: Model comparisons, current knowledge and uncertainties, Atmos. Chem. Phys., 9, 40534076.
  • Kesselmeier, J., and M. Staudt (1999), Biogenic volatile organic compounds (VOC): An overview on emission, physiology and ecology, J. Atmos. Chem., 33, 2388.
  • Kulmala, M., et al. (2004), A new feedback mechanism linking forests, aerosols, and climate, Atmos. Chem. Phys., 4, 557562.
  • Lamb, B., et al. (1996), Evaluation of forest canopy models for estimating isoprene emissions, J. Geophys. Res., 101, 22,78722,797.
  • Laothawornkitkul, J., J. E. Taylor, N. D. Paul, and C. N. Hewitt (2009), Biogenic volatile organic compounds in the Earth system, New Phytol., 183, 2751.
  • Lehning, A., I. Zimmer, R. Steinbrecher, N. Brüggemann, and J. P. Schnitzler (1999), Isoprene synthase activity and its relation to isoprene emission in Quercus robur L. leaves, Plant Cell Environ., 22, 495504.
  • Lehning, A., W. Zimmer, I. Zimmer, and J. P. Schnitzler (2001), Modeling of annual variations of oak (Quercus robur L.) isoprene synthase activity to predict isoprene emission rates, J. Geophys. Res., 106(D3), 31573166, doi:10.1029/2000JD900631.
  • Lerdau, M., and H. L. Throop (2000), Sources of variability in isoprene emission and photosynthesis in two species of tropical wet forest trees, Biotropica, 32, 670676.
  • Le Roux, X., T. Bariac, H. Sinoquet, B. Genty, C. Piel, A. Mariotti, C. Girardin, and P. Richard (2001), Spatial distribution of leaf water-use efficiency and carbon isotope discrimination within an isolated tree crown, Plant Cell Environ., 24, 10211032.
  • Litvak, M. E., F. Loreto, P. C. Harley, T. D. Sharkey, and R. K. Monson (1996), The response of isoprene emission rate and photosynthetic rate to photon flux and nitrogen supply in aspen and white oak trees, Plant Cell Environ., 19, 549559.
  • Lloyd, J., S. C. Wong, J. M. Styles, D. Batten, R. Priddle, C. Turnbull, and C. A. McConchie (1995), Measuring and modelling whole-tree gas exchange, Aust. J. Plant Physiol., 22, 9871000.
  • Llusià, J., and J. Peñuelas (2000), Seasonal patterns of terpene content and emission from seven Mediterranean woody species in field conditions, Am. J. Bot., 87, 133140.
  • Loreto, F., C. Barta, F. Brilli, and I. Nogues (2006), On the induction of volatile organic compound emissions by plants as consequence of wounding or fluctuations of light and temperature, Plant Cell Environ., 29, 18201828.
  • Magel, E., S. Mayrhofer, A. Müller, I. Zimmer, R. Hampp, and J.-P. Schnitzler (2006), Photosynthesis and substrate supply for isoprene biosynthesis in poplar leaves, Atmos. Environ., 40, S138S151.
  • Martin, M. J., C. M. Stirling, S. W. Humphries, and S. P. Long (2000), A process-based model to predict the effects of climatic change on leaf isoprene emission rates, Ecol. Modell., 131, 161174.
  • Mayrhofer, S., M. Teuber, I. Zimmer, S. Louis, R. J. Fischbach, and J.-P. Schnitzler (2005), Diurnal and seasonal variation of isoprene biosynthesis-related genes in grey poplar leaves, Plant Physiol., 139, 474484.
  • Meir, P., B. Kruijt, M. Broadmeadow, E. Barbosa, O. Kull, F. Carswell, A. Nobre, and P. G. Jarvis (2002), Acclimation of photosynthetic capacity to irradiance in tree canopies in relation to leaf nitrogen concentration and leaf mass per unit area, Plant Cell Environ., 25, 343357.
  • Miller, J. P. (1967), A formula for average foliage density, Aust. J. Bot., 15, 141143.
  • Mishra, R. K., and G. S. Singhal (1992), Function of photosynthetic apparatus of intact wheat leaves under high light and heat stress and its relationship with peroxidation of thylakoid lipids, Plant Physiol., 98, 16.
  • Niinemets, Ü. (2004), Costs of production and physiology of emission of volatile leaf isoprenoids, in Advances in Plant Physiology, edited by A. Hemantaranjan, pp. 241278, Scientific, Jodhpur, India.
  • Niinemets, Ü. (2007), Photosynthesis and resource distribution through plant canopies, Plant Cell Environ., 30, 10521071.
  • Niinemets, Ü. (2010), A review of light interception in plant stands from leaf to canopy in different plant functional types and in species with varying shade tolerance, Ecol. Res., 25, 693714, doi:10.1007/s11284-010-0712-4.
  • Niinemets, Ü., and N. P. R. Anten (2009), Packing photosynthesis machinery: From leaf to canopy, in Photosynthesis in Silico: Understanding Complexity From Molecules to Ecosystems, edited by A. Laisk, L. Nedbal, and Govindjee, pp. 363399, Springer, Dordrecht, Netherlands.
  • Niinemets, Ü., and Ü. Tamm (2005), Species differences in timing of leaf fall and foliage chemistry modify nutrient resorption efficiency in deciduous temperate forest stands, Tree Physiol., 25, 10011014.
  • Niinemets, Ü., and F. Valladares (2004), Photosynthetic acclimation to simultaneous and interacting environmental stresses along natural light gradients: Optimality and constraints, Plant Biol., 6, 254268.
  • Niinemets, Ü., W. Bilger, O. Kull, and J. D. Tenhunen (1998a), Acclimation to high irradiance in temperate deciduous trees in the field: Changes in xanthophyll cycle pool size and in photosynthetic capacity along a canopy light gradient, Plant Cell Environ., 21, 12051218.
  • Niinemets, Ü., O. Kull, and J. D. Tenhunen (1998b), An analysis of light effects on foliar morphology, physiology, and light interception in temperate deciduous woody species of contrasting shade tolerance, Tree Physiol., 18, 681696.
  • Niinemets, Ü., W. Bilger, O. Kull, and J. D. Tenhunen (1999a), Responses of foliar photosynthetic electron transport, pigment stoichiometry, and stomatal conductance to interacting environmental factors in a mixed species forest canopy, Tree Physiol., 19, 839852.
  • Niinemets, Ü., V. Oja, and O. Kull (1999b), Shape of leaf photosynthetic electron transport versus temperature response curve is not constant along canopy light gradients in temperate deciduous trees, Plant Cell Environ., 22, 14971514.
  • Niinemets, Ü., J. D. Tenhunen, N. R. Canta, M. M. Chaves, T. Faria, J. S. Pereira, and J. F. Reynolds (1999c), 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 Biol., 5, 455470.
  • Niinemets, Ü., J. D. Tenhunen, P. C. Harley, and R. Steinbrecher (1999d), A model of isoprene emission based on energetic requirements for isoprene synthesis and leaf photosynthetic properties for Liquidambar and Quercus, Plant Cell Environ., 22, 13191336.
  • Niinemets, Ü., G. Seufert, R. Steinbrecher, and J. D. Tenhunen (2002), A model coupling foliar monoterpene emissions to leaf photosynthetic characteristics in Mediterranean evergreen Quercus species, New Phytol., 153, 257276.
  • Niinemets, Ü., H. Kollist, J. I. García-Plazaola, A. Hernández, and J. M. Becerril (2003a), Do the capacity and kinetics for modification of xanthophyll cycle pool size depend on growth irradiance in temperate trees? Plant Cell Environ., 26, 17871801.
  • Niinemets, Ü., F. Valladares, and R. Ceulemans (2003b), Leaf level phenotypic variability and plasticity of invasive Rhododendron ponticum and noninvasive Ilex aquifolium co-occurring at two contrasting European sites, Plant Cell Environ., 26, 941956.
  • Niinemets, Ü., N. Al Afas, A. Cescatti, A. Pellis, and R. Ceulemans (2004a), Petiole length and biomass investment in support modify light-interception efficiency in dense poplar plantations, Tree Physiol., 24, 141154.
  • Niinemets, Ü., A. Cescatti, and R. Christian (2004b), Constraints on light interception efficiency due to shoot architecture in broad-leaved Nothofagus species, Tree Physiol., 24, 617630.
  • Niinemets, Ü., O. Kull, and J. D. Tenhunen (2004c), Within canopy variation in the rate of development of photosynthetic capacity is proportional to integrated quantum flux density in temperate deciduous trees, Plant Cell Environ., 27, 293313.
  • Niinemets, Ü., J. D. Tenhunen, and W. Beyschlag (2004d), Spatial and age-dependent modifications of photosynthetic capacity in four Mediterranean oak species, Funct. Plant Biol., 31, 11791193.
  • Niinemets, Ü., A. Arneth, U. Kuhn, R. K. Monson, J. Peñuelas, and M. Staudt (2010a), The emission factor of volatile isoprenoids: Stress, acclimation, and developmental responses, Biogeosciences, 7, 22032223.
  • Niinemets, Ü., R. K. Monson, A. Arneth, P. Ciccioli, J. Kesselmeier, U. Kuhn, S. M. Noe, J. Peñuelas, and M. Staudt (2010b), The leaf level emission factor of volatile isoprenoids: Caveats, model algorithms, response shapes and scaling, Biogeosciences, 7, 18091832.
  • Owen, S. M., and C. N. Hewitt (2000), Extrapolating branch enclosure measurements to estimates of regional scale biogenic VOC fluxes in the northwestern Mediterranean basin, J. Geophys. Res., 105(D9), 11,57311,583, doi:10.1029/1999JD901154.
  • Owen, S. M., and J. Peñuelas (2005), Opportunistic emissions of volatile isoprenoids, Trends Plant Sci., 10, 420426.
  • Owen, S. M., C. Boissard, B. Hagenlochera, and C. N. Hewitt (1998), Field studies of isoprene emissions from vegetation in the Northwest Mediterranean region, J. Geophys. Res., 103(D19), 25,49925,511, doi:10.1029/98JD01817.
  • Owen, S. M., A. R. MacKenzie, H. Stewart, R. Donovan, and C. N. Hewitt (2003), Biogenic volatile organic compound (VOC) emission estimates from an urban tree canopy, Ecol. Appl., 13, 927938.
  • Peñuelas, J., J. Llusià, D. Asensio, and S. Munné-Bosch (2005), Linking isoprene with plant thermotolerance, antioxidants and monoterpene emissions, Plant Cell Environ., 28, 278286.
  • Pio, C. A., P. A. Silva, M. A. Cerqueira, and T. V. Nuñes (2005), Diurnal and seasonal emissions of volatile organic compounds from cork oak (Quercus suber) trees, Atmos. Environ., 39, 18171827.
  • Poorter, H., Ü. Niinemets, A. Walter, F. Fiorani, and U. Schurr (2010), A method to construct dose-response curves for a wide range of environmental factors and plant traits by means of a meta-analysis of phenotypic data, J. Exp. Bot., 61, 20432055.
  • Portsmuth, A., and Ü. Niinemets (2007), Structural and physiological plasticity to light and nutrients in five temperate deciduous woody species of contrasting shade tolerance, Funct. Ecol., 21, 6177.
  • Possell, M., J. Heath, C. N. Hewitt, E. Ayres, and G. Kerstiens (2004), Interactive effects of elevated CO2 and soil fertility on isoprene emissions from Quercus robur, Global Change Biol., 10, 18351843.
  • Roden, J. S., and R. W. Pearcy (1993), The effect of flutter on the temperature of poplar leaves and its implications for carbon gain, Plant Cell Environ., 16, 571577.
  • Rohdich, F., K. Kis, A. Bacher, and W. Eisenreich (2001), The nonmevalonate pathway of isoprenoids: Genes, enzymes and intermediates, Curr. Opin. Chem. Biol., 5, 535540.
  • Rohdich, F., et al. (2003), The deoxyxylulose phosphate pathway of isoprenoid biosynthesis: Studies on the mechanisms of the reactions catalyzed by IspG and IspH protein, Proc. Natl. Acad. Sci. U. S. A., 100, 15861591.
  • Rosenstiel, T. N., A. L. Ebbets, W. C. Khatri, R. Fall, and R. K. Monson (2004), Induction of poplar leaf nitrate reductase: A test of extrachloroplastic control of isoprene emission rate, Plant Biol., 6, 1221.
  • Ross, J., and M. Sulev (2000), Sources of errors in measurements of PAR, Agric. For. Meteorol., 100, 103125.
  • Sabillón, D., and L. V. Cremades (2001), Diurnal and seasonal variation of monoterpene emission rates for typical Mediterranean species (Pinus pinea and Quercus ilex) from field measurements—Relationship with temperature and PAR, Atmos. Environ., 35, 44194431.
  • Sánchez-Gómez, D., F. Valladares, and M. A. Zavala (2006), Functional traits and plasticity underlying shade tolerance in seedlings of four Iberian forest tree species, Tree Physiol., 26, 14251433.
  • Sellers, P. J., J. A. Berry, G. J. Collatz, C. B. Field, and F. G. Hall (1992), Canopy reflectance, photosynthesis, and transpiration: III. A reanalysis using improved leaf models and a new canopy integration scheme, Remote Sens. Environ., 42, 187216.
  • Sharkey, T. D., and E. L. Singsaas (1995), Why plants emit isoprene, Nature, 374, 769.
  • Sharkey, T. D., A. E. Wiberley, and A. R. Donohue (2008), Isoprene emission from plants: Why and how, Ann. Bot., 101, 518.
  • Singsaas, E. L., M. Lerdau, K. Winter, and T. D. Sharkey (1997), Isoprene increases thermotolerance of isoprene-emitting species, Plant Physiol., 115, 14131420.
  • Singsaas, E. L., M. M. Laporte, J.-Z. Shi, R. K. Monson, D. R. Bowling, K. Johnson, M. Lerdau, A. Jasentuliytana, and T. D. Sharkey (1999), Kinetics of leaf temperature fluctuation affect isoprene emission from red oak (Quercus rubra) leaves, Tree Physiol., 19, 917924.
  • Sinoquet, H., J. Stephan, G. Sonohat, P.-E. Lauri, and P. Monney (2007), Simple equations to estimate light interception by isolated trees from canopy structure features: Assessment with three-dimensional digitized apple trees, New Phytol., 175, 94106.
  • Sokal, R. R., and J. F. Rohlf (1995), Biometry: The Principles and Practice of Statistics in Biological Research, pp. 412415, W. H. Freeman, New York.
  • Steinbrecher, R., G. Smiatek, R. Köble, G. Seufert, J. Theloke, K. Hauff, P. Ciccioli, R. Vautard, and G. Curci (2009), Intra- and inter-annual variability of VOC emissions from natural and semi-natural vegetation in Europe and neighbouring countries, Atmos. Environ., 43, 13801391.
  • Stewart, H. E., C. N. Hewitt, R. G. H. Bunce, R. Steinbrecher, G. Smiatek, and T. Schoenemeyer (2003), A highly spatially and temporally resolved inventory for biogenic isoprene and monoterpene emissions: Model description and application to Great Britain, J. Geophys. Res., 108(D20), 4644, doi:10.1029/2002JD002694.
  • ter Steege, H. (1996), Winphot 5: A Programme to Analyze Vegetation Indices, Light and Light Quality From Hemispherical Photographs, Tropenbos Guyana Rep. 95-2, 40 pp., Tropenbos Guayana Programme, Georgetown, Guayana.
  • Valladares, F., J. M. Chico, I. Aranda, L. Balaguer, P. Dizengremel, E. Manrique, and E. Dreyer (2002), The greater seedling high-light tolerance of Quercus robur over Fagus sylvatica is linked to a greater physiological plasticity, Trees, 16, 395403.
  • Velikova, V., and F. Loreto (2005), On the relationship between isoprene emission and thermotolerance in Phragmites australis leaves exposed to high temperatures and during the recovery from a heat stress, Plant Cell Environ., 28, 318327.
  • Velikova, V., A. Edreva, and F. Loreto (2004), Endogenous isoprene protects Phragmites australis leaves against singlet oxygen, Physiol. Plant., 122, 219225.
  • Vickers, C. E., J. Gershenzon, M. T. Lerdau, and F. Loreto (2009), A unified mechanism of action for volatile isoprenoids in plant abiotic stress, Nat. Chem. Biol., 5, 283291.
  • Villar, R., A. A. Held, and J. Merino (1995), Dark leaf respiration in light and darkness of an evergreen and a deciduous plant species, Plant Physiol., 107, 421427.
  • von Caemmerer, S., and G. D. Farquhar (1981), Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves, Planta, 153, 376387.
  • Wiberley, A. E., A. R. Donohue, M. E. Meier, M. M. Westphal, and T. D. Sharkey (2008), Regulation of isoprene emission in Populus trichocarpa leaves subjected to changing growth temperature, Plant Cell Environ., 31, 258267.
  • Wiberley, A. E., A. R. Donohue, M. M. Westphal, and T. D. Sharkey (2009), Regulation of isoprene emission from poplar leaves throughout a day, Plant Cell Environ., 32, 939947.
  • Winters, A. J., M. A. Adams, T. M. Bleby, H. Rennenberg, D. Steigner, R. Steinbrecher, and J. Kreuzwieser (2009), Emissions of isoprene, monoterpene and short-chained carbonyl compounds from Eucalyptus spp. in southern Australia, Atmos. Environ., 43, 30353043.
  • Zimmer, W., R. Steinbrecher, C. Körner, and J.-P. Schnitzler (2003), The process-based SIM-BIM model: Towards more realistic prediction of isoprene emissions from adult Quercus petraea forest trees, Atmos. Environ., 37, 16651671.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusions
  8. Appendix A:: Simple Big-Leaf Isoprene Emission Model
  9. Acknowledgments
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
jgrg721-sup-0001-t01a.txtplain text document1KTab-delimited Table 1a.
jgrg721-sup-0002-t01b.txtplain text document2KTab-delimited Table 1b.
jgrg721-sup-0003-t02.txtplain text document1KTab-delimited Table 2.
jgrg721-sup-0004-t03.txtplain text document2KTab-delimited Table 3.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.