The effect of canopy architecture on the patterning of “ windflecks ” within a wheat canopy

Under field conditions, plants are subject to wind-induced movement which creates fluctuations of light intensity and spectral quality reaching the leaves, defined here as windflecks. Within this study, irradiance within two contrasting wheat ( Triticum aestivum ) canopies during full sun conditions was measured using a spectroradiometer to determine the frequency, duration and magnitude of low- to high-light events plus the spectral composition during wind-induced movement. Similarly, a static canopy was modelled using three-dimensional reconstruction and ray tracing to determine fleck characteristics without the presence of wind. Corresponding architectural traits were measured manually and in silico including plant height, leaf area and angle plus biomechanical properties. Light intensity can differ up to 40% during a windfleck, with changes occurring on a sub-second scale compared to (cid:1) 5 min in canopies not subject to wind. Features such as a shorter height, more erect leaf stature and having an open structure led to an increased frequency and reduced time interval of light flecks in the CMH79A canopy compared to Paragon. This finding illustrates the potential for architectural traits to be selected to improve the canopy light environment and provides the foundation to further explore the links between plant form and function in crop canopies.

and structure of leaf material, coupled with local weather conditions including sun elevation, cloud cover plus wind speed and direction (Burgess et al., 2016;de Langre, 2008;Grace, 1988).
Historically, research in plant and crop photosynthesis has focused on rates of CO 2 uptake under steady state light conditions. However, it is increasingly apparent that photosynthetic productivity is also determined by dynamic changes in environmental variables, and there is often a time lag before photosynthesis can respond . The heterogeneity of the light environment influences how plants respond to and exploit available resources for photosynthesis and crop production. This has been recently demonstrated through changes in biomass production via altering capacity for photoprotection and the speed of recovery (Hubbart et al., 2018;Kromdijk et al., 2016). However, to quantify the impact of a particular photosynthetic process on potential productivity requires knowledge of the precise "signature" of light dynamics and the accompanying changes to spectral composition. For example, rapid fluctuations may lead to an overall increase in productivity due to maintenance of a higher induction state of photosynthesis (Acevedo-Siaca et al., 2020;Burgess et al., 2016;Retkute et al., 2015;Roden & Pearcy, 1993a;Wang, Burgess, de Becker, & Long, 2020). Longer periods of low light cause the de-activation of enzymes and stomatal closure and vice versa. Another benefit of wind-induced changes in the light environment could arise through homogenizing the light available such that photosynthesis will be increased if there is a narrower distribution for a given time-integrated photon flux density due to the non-linear response. Understanding the precise spatiotemporal light dynamics in different canopy structures is thus essential for predicting the impact of these different processes on whole-plant photosynthesis.
The majority of existing studies have characterized the effect of sunflecks within a forestry setting, where periods of high light can persist for minutes or even hours depending on the structure of tree crowns (Chazdon & Pearcy, 1991;Pallardy, 2008;Way & Pearcy, 2012). In such cases, the sunflecks, on a background diffuse irradiance (i.e., shade), can contribute a large percentage of incident irradiance for understorey plants (Barradas, Jones, & Clark, 1998;Pfitsch & Pearcy, 1989;Roden & Pearcy, 1993a;Tang, Washitani, Tsuchiya, & Iwaki, 1988). However, more interest has arisen recently on the effect of fluctuating light in the agricultural setting, where the structure of a crop stand leads to very different patterns of radiation over smaller spatial, and often temporal, scales, with direct consequences in terms of photosynthetic productivity Murchie et al., 2018;Murchie, Pinto, & Horton, 2009;Slattery, Walker, Weber, & Ort, 2018;Wang et al., 2020).
Wind affects both the plant canopy and its interactions with the environment, according to both the wind speed and the duration of gusts. In turn, and as a consequence, the structure of a plant is constrained and shaped by wind such as dwarfing characteristics (Gardiner, Berry, & Moulia, 2016). The resulting in canopy light environment will be altered in terms of frequency, duration and amplitude of high light events. It can be expected that the most drastic effects of wind-induced movement are felt in lower canopy layers, where the movement of overhanging leaf material can lead to increased light penetration. A period of high light intensity becomes more likely as the canopy starts to move, but the average duration of such periods may be lower than during still conditions (Tong & Hipps, 1996). The effect (and possibly biological function) of movement, especially in upper layers, therefore, becomes that of light scattering and distribution, facilitating photosynthesis in lower leaf layers. Previously, this has been compared to a disco "mirrorball" spinning at fast or slow speeds; with spin speed correlating with likelihood of a high light event (Burgess et al., 2016).
The response of an organ to wind will depend upon its length, surface area and mass. The range of motion or potential risk of breakage will also depend upon the strength of the supporting structure as well as the leaf blade, which is in turn related to dry matter accumulation combined with the strength of the vein and thus the water status (Derzaph & Hamilton, 2013;Gonzalez-Rodrigues, Cournède, & de Langre, 2016). In the case of cereal crops, the size, weight and surface area of the ear will also determine movement properties. However, cultivation in dense stands makes movement difficult to characterise due to collisions between neighbouring plants (Doaré, Moulia, & de Langre, 2004). Nevertheless, at low wind speeds, leaf movement is expected to dominate due to low mass and high surface area, whilst higher wind speeds induce greater movement in supporting structures (i.e., stems or branches). Few studies consider wind-induced movement under low wind speeds, with more work aimed at the biomechanical properties required to prevent stem or root failure under damaging wind speeds during lodging events (Berry, Spink, Foulkes, & Wade, 2003;Berry, Sylvester-Bradley, & Berry, 2007). One key structural trait to reduce lodging risk is plant height; with reduced heights leading to a reduction in the leverage that is imposed by the aerial organs on the supporting structure (Berry, Sterling, & Mooney, 2006). Consequently, plant height will be critical in determining overall biomechanics and light patterning within a crop stand as plant height is closely linked to natural frequency with taller plants generally having a higher natural frequency.
The spatial arrangement of leaf material combined with windinduced movement leads to a complex pattern of both light intensity and spectral composition throughout canopies. As light enters the canopy, leaves will preferentially deplete the most useful wavelengths for photosynthesis, predominantly red and blue, and scatter those absorbed less effectively by chlorophyll such as ultraviolet, green and far red (Evans & Anderson, 1987;Smith, McAusland, & Murchie, 2017). This leads to alterations in the proportion of wavelengths reaching lower canopy layers with a steep decrease in the red: far red (R:FR) ratio (R = λ600-700, FR = λ700-800 nm) and a concurrent increase in UV-A:PAR ratio (UV-A = λ315-400, PAR = λ400-700) followed by a more shallow decline in the blue: green (B:G-where B = λ400-500 nm, G = λ500-600 nm) light gradient (Smith et al., 2017). However, as leaf material alters position, we can expect both the quantity and spectral composition of light to change.
Here we will use "windfleck" to refer to the more rapid changes in light intensity brought about by canopy movement, which are our primary focus, compared to high-light events induced by solar movement penetrating [static] canopy gaps (the traditional "sunfleck").
"Fleck" is used as a term to describe an overall change in irradiance (with or without wind). These periods of high light are often defined as irradiances differing by a certain threshold value or from a certain percentage of the baseline irradiance (Barradas et al., 1998;Miyashita et al., 2012;Roden & Pearcy, 1993b). They can be distinguished based on whether their divergence is positive or negative, with a "sunfleck" referring to an increase in irradiance from a baseline, and a "shadefleck" a decrease, in irradiance from a baseline of direct sun.
Until recently, very little research has been aimed at uncovering the optimal protocol to detect and measure the properties of light fluctuations taking place within canopies. This includes among others, analytical approaches to circumvent arbitrary thresholds described above, and quantifying the impact of measurement frequency and integration time of the measurement of light. Previous work has used diverse measurement frequencies ranging from every 50 ms to 5 s (Pfitsch & Pearcy, 1989;Roden, 2003). This can lead to significant loss of information when measuring frequency, peak intensity and duration of any changes in irradiance (Chazdon, 1988). A new method and an associated algorithm is available to characterize key features of light fluctuations including duration, amplitude and average time between highlight events (Durand, Matule, Burgess, & Robson, 2021).
Whilst studies are beginning to arise that assess light interception within canopies of different crops subject to wind-induced movement , there is very little information on how specific architectural traits influence light patterning at the high-resolution required to assess windflecks. Due to the wide range of architectural differences seen between different crops, it is therefore difficult to attribute a specific trait to the characteristics of windflecks that are observed. Thus, there is a need for comparative studies with diverse varieties of the same crop type. Within this study, we aim to determine how wind-induced movement alters the pattern of light quantity and quality reaching different layers within a wheat canopy. Two architecturally contrasting varieties will be used determine how features such as plant height and leaf stature alter both biomechanics and the resulting light environment. By measuring the canopies subject to wind, and modelling the canopies using ray tracing techniques in a static formation, we aim to characterize the key features of light reaching leaves under field conditions. Previous cropping was winter oats (Avena sativa). The field was ploughed, power harrowed and rolled after drilling. The seed rate was adjusted by genotype according to 1,000 grain weight to achieve a target seed rate of 350 seeds m À2 ; rows were 0.125 m apart. A total of 120 kg/ha N fertilizer as ammonium nitrate was applied in a twosplit program. P and K fertilizers were applied to ensure that these nutrients were not limiting. Plant growth regulator was applied at GS31 (stem elongation) to reduce the risk of lodging. Herbicides, fungicides and pesticides were applied as required to minimize the effects of weeds, diseases and pests. One growth stage was analysed: postanthesis ($GS70; Zadoks, Chang, & Konzak, 1974).

| Plant material and growth
A weather station was situated in close proximity (approx. 30 m) to the field site. Photosynthetically active radiation (PAR) above the canopy, wind speed, wind direction, temperature, humidity, soil moisture and rainfall were monitored throughout growth using the following sensors: SQ-110 Quantum sensor (Apogee, Utar, USA), WindSonic1 ultrasonic wind sensor (GillInstruments, Lymington, UK), CS215 temperature and humidity probe (Sensirion, Switzerland), 107 thermistor probe (BetaTherm, Galway, Ireland) and the Kalyx-RG aerodynamic rain gauge. The wind sensor was mounted approximately 1.5 m above the ground. Measurements were recorded at every second using the CRX1000X data logger (Campbell Scientific, Leicestershire, UK) from 45 days after sowing (DAS) until harvest.

| Structural, biomechanical and leaf optical measurements
Ten plants of each variety across the three replicate plots were removed for biomechanical and physical analysis using the lodging protocol described in Berry et al. (2000). This included measurements such as plant height, height at the centre of gravity, root and ear number per plant, ear area, internode length and wall width and breaking strength. A validated model for lodging in winter wheat (Berry et al., 2003) was used to calculate biomechanical features of the two contrasting genotypes including natural frequency.
Absorbance by epidermal flavones, anthocyanins and leaf chlorophyll content was measured non-invasively with optical leaf clip Dualex Scientific + (henceforth Dualex) at two canopy positions (at 25 and 50 cm height) at the leaf tip, centre and base of both adaxial and abaxial sides of leaves. The measurements were performed around solar noon (approximately ±2 hr) to exclude potential diurnal variation in UV absorbance by flavones and chloroplast movement (Barnes et al., 2016;Williams, Gorton, & Witiak, 2003). 2.3 | 3D canopy reconstruction for architectural analysis roots to prevent wilting. At least 40 images per plant were taken and reconstructions made as described in Burgess et al. (2015).
Reconstructed canopies were formed by duplicating and randomly rotating the four best reconstructed plants in a 5 by 3 grid, with 13 cm between rows and 5 cm between plants, in accordance with the planting pattern. Each reconstructed canopy is formed of a set of triangles.
Total light per unit leaf area was predicted using a forward raytracing algorithm implemented in fastTracer (version 3; PICB, Shanghai, China; Song et al. (2013). Latitude was set at 53 (for Sutton Bonington, UK), atmospheric transmittance 0.5, light reflectance 7.5%, light transmittance 7.5%, day 184 (3rd July). FastTracer3 calculates light as direct, diffused and transmitted components separately; these were combined together to give a single irradiance level for all canopy positions. Irradiance values were recorded for the 3 hr around solar noon (1130-1430 hr) at 1 min time intervals; the highest resolution possible via this method. This allowed the analysis of fleck patterns in a static canopy (i.e., they do not move with a simulated wind) during full sun. The ray tracing boundaries were positioned so that they bisect the outer plants (e.g., Retkute, Townsend, Murchie, Jensen, & Preston, 2017) to reduce boundary effects.

All modelling was carried out in Mathematica (Wolfram Research
Inc., IL). Cumulative leaf area index (cLAI; leaf area per unit ground area as a function of depth) and cumulative fractional interception (cF; fractional interception as a function of depth) was calculated from each of the canopy reconstructions as in Burgess et al. (2015) for the solar noon time point.
Fleck analysis was performed as described in Section 2.5 using 1,586 time series of triangles (i.e., canopy locations) at 25 and 50 cm (±2.5 cm) per each variety reconstructed canopy.

| Windfleck analysis
The windfleck analysis was performed on a subset of the dataset for increased repeatability of the results. Only measurements recorded with an integration time of 300 ms or lower were used. Each timeseries was reviewed manually to decide whether to identify brief increases (sunfleck) or decreases (shadefleck) of irradiance depending on which dominated the time-series. Because integration time will influence the description of sun-or shadefleck properties, a simple linear model was created to take the integration time into account , whilst a second model was applied to the residuals of the original model to determine the fleck characteristics.
Briefly, time-series of photosynthetically active radiation (PAR, 400-700 nm) were used to detect windflecks using an algorithm based on the properties of first order derivatives. The start and endpoint of fleck can be identified when the numerical derivative crosses zero (i.e., when PAR switches from increasing to decreasing, and vice versa). The algorithm corrects for cases where flecks have multiple peaks; in order to classify as a single fleck those fluctuations that produced multiple peaks without a significant change of PAR . The duration (i.e., time difference between the start and end of the fleck), the difference in PAR between the peak and baseline irradiance, and frequency (average time between two flecks) were measured for each windfleck.

| Statistical analysis
Analysis of Variance (ANOVA) was carried out using Genstat for windows (19th Edition; VSN International Ltd., UK). Data was checked to see if it met the assumption of constant variance and normal distribution of residuals. For optically measured leaf pigments, modelled and measured fleck characteristics, the post hoc Tukey's test was used at a probability level of p < .05.

| Plant structural and biomechanical features
Two contrasting wheat genotypes were chosen to assess the effect of wind-induced movement on the duration, magnitude and frequency of high light events in a field grown canopy ( Figure S1). These differences were reflected in the key structural and biomechanical traits of each genotype given in Table 1. Paragon was significantly taller than CMH79A ($34%), with a correspondingly higher centre of gravity ($32%). Similarly, Paragon had, on average, more ears per plant ($13%), but a much-reduced ear surface area ($51%). These features did not lead to a significant difference in natural frequency between the two varieties.
To further explore the architectural properties of the two genotypes, plants were reconstructed in silico. Figure 1  Leaf area index (LAI) was calculated as the area of mesh (reconstructed plant material) relative to ground area and was 7.10 and 3.69 in Paragon and CMH79A, respectively (Table 1) Figure S2). This indicates striking differences in light distribution for two canopies of spring wheat cultivars with comparable productivity and yield potential.

| Patterns in optically measured leaf pigments
Non-destructive optical measurements of leaf pigments were made at the top of the canopy and in the lower canopy at equivalent heights to the spectral irradiance measurements (Figure 4). Leaf chlorophyll content was up to 50% higher in CMH79A than in Paragon at the base and mid leaf positions in both the top and bottom halves of T A B L E 1 Select plant structural and biomechanical characteristics for each genotype at growth stage (GS) 71-73 (Zadoks et al., 1974) Genotype Height to ear tip (cm)

| Flecks in mechanically moving canopies: Windflecks
In general, the irradiance was lower at equivalent heights in Paragon than in CMH79A, especially in the lower canopy ( Figure 7a).
The duration of sunflecks was longer in Paragon than in CMH79A Changes in the frequency and duration of low to high light transitions has consequences for photosynthetic processes. It has previously been shown that there is an inverse relationship between F I G U R E 7 Measured characteristics of sun-and shadeflecks throughout two architecturally diverse wheat canopies during mechanical movement in the wind captured using a spectroradiometer. (a) Average light intensity during the peak and base of the fleck where the inset indicates the percentage difference in PAR between the baseline and peak of each fleck, (b) average sunfleck duration, (c) mean time between sunflecks. The data is grouped based on the nature of the variation of light: either sunfleck (temporary increase of irradiance) or shadefleck (temporary decrease of irradiance) and the height at which the measurement was taken: in the top or bottom half of the canopy. M ± SEM where different letters indicate significant differences following ANOVA and post hoc Tukey's test at the p < .05 level [Colour figure can be viewed at wileyonlinelibrary.com] F I G U R E 8 Changes in the spectral composition of light during a sun-and shadefleck in two architecturally contrasting varieties of wheat. (a) UVA to PAR ratio (b) blue to red ratio (c) blue to green ratio and (d) red to far red ratio. Each inset indicates the percentage difference in the given ratio between the baseline and peak of each fleck. M ± SEM where different letters indicate significant differences following ANOVA and post hoc Tukey's test at the p < .05 level [Colour figure can be viewed at wileyonlinelibrary.com] optimal maximum photosynthetic capacity and the frequency of transitions Yin & Johnson, 2000). However, the average background light intensities (baseline/shade) and those during the sunfleck (peak) are higher in CMH79A relative to Paragon for both the top and bottom parts of the canopy, with up to a 40% increase in intensity over the background values. However, the overall value of PAR is reduced during the measured data (i.e., during wind) compared to the simulated day (i.e., using ray tracing) indicating that whilst the percentage difference was greater, the actual difference between the two varieties during win-induced movement is reduced. Thus, although the frequency of light intensity changes is increased, more photons are received in total. In other words, the integrated and peak photon dose is higher overall (Niinemets & Anten, 2009)  Upon exposure to a sunfleck following a period in shade, a leaf will undergo photosynthetic induction involving several processes which are not instantaneous including activation of the Rubisco enzyme and stomatal opening. Conversely, once the fleck has passed, an induction phase will occur in reverse with enzyme deactivation and stomatal closure. Typically, the reverse induction phase will be slower than the induction, increasing the probability that photosynthesis will be able to respond rapidly to any subsequent sunflecks (Porcar-Castell & Palmroth, 2012). This will also depend on the characteristics of the fleck that are experienced; with the magnitude of change and time spent under each intensity influencing the rate, and potential, of response. However, a trade-off with this potential gain is the response of photoprotective processes including non-photochemical quenching (NPQ). As light intensity increases, excitation energy in photosystem II (PSII) can be dissipated as heat through NPQ (Li, Wakao, Fischer, & Niyogi, 2009) which helps to prevent over-reduction of PSII and electron transport. Similarly to photosynthetic induction, NPQ induction and relaxation are not instantaneous and there will be a lag period between the change in irradiance and response. The rate of NPQ relaxation is slower than that of NPQ induction, intensified by prolonged exposure to high light intensities (Pérez-Bueno, Johnson, Zia, Ruban, & Horton, 2008). Thus, following a decrease in irradiance, CO 2 fixation will be transiently depressed by the slow rate of recovery of NPQ . The balance between competing photosynthetic processes will therefore be important in determining the exact response to fluctuations in light intensity. This has potential implications for the two opposing canopies studied here.
Windflecks are of a much higher frequency and shorter duration than sunflecks considered previously (e.g., Matthews, Vialet-Chabrand, & Lawson, 2018;Morales et al., 2018;Retkute et al., 2017). It is pertinent therefore to consider time constants for each photosynthesis component and compare this to the frequency of the light fluctuations. Previous papers considering sunflecks have accommodated sufficient time for induction and relaxation of photosynthesis, taking into account the typical rates for the slowest elements (Acevedo-Siaca et al., 2020;Kromdijk et al., 2016;Pearcy, 1990;Taylor & Long, 2017). Maximum photosynthesis rates can take up to 20 min to be reached and has been observed to correlate with induction state (Soleh et al., 2017;Taylor & Long, 2017  . Clearly the frequency of windflecks do not, individually, permit time for induction and relaxation of such processes. Light pulses of 20, 50 and 100 Hz have been shown capable of maintaining high photosynthesis rates (Gaudillere, 1977). Although this is slower than time constants for light harvesting and electron transport this might be considered akin to continuous light presumably because during the dark intervals, metabolite pool sizes and enzyme activation states are retained and sufficient electron transport pathways are present, maintaining homeostasis. Windflecks should be highly efficient at allowing leaves to reach and maintain induction.
The question then arises of the frequency and duration of "bursts" or "trains" of windflecks in canopies, and this is unknown. The current study provides estimates of sunflecks and shadeflecks which can be combined with the mathematical model of non-instantaneous leaf response to light patterns  to investigate how light fluctuations of less than a second could further increase the assimilation lost due to induction limitations.
Within and between species there is variation in the speed of response of the photosynthetic machinery (photoprotective and photochemical) to changes in light intensity. This is partly genetic but there is also a component related to the environmental conditions to which a plant is exposed (Acevedo-Siaca et al., 2020;Hubbart et al., 2018;Kromdijk et al., 2016;McAusland et al., 2020;Roden & Pearcy, 1993a;Salter, Merchant, Richards, Trethowan, & Buckley, 2019;Taniyoshi, Tanaka, & Shiraiwa, 2020). This creates a complex picture whereby the fleck characteristics and speed of response of photosynthesis both varies according to genotype and may be different according to environment. Solving this problem requires both physical and biological experimentation with mathematical modelling. Characteristics of sunflecks and shadeflecks estimated in this study takes the first step and will help to advance mathematical models of light dynamics in canopies , photoinhibition  and photoacclimation Townsend et al., 2018). Such work will allow us to examine the potential of selecting varieties for a more efficient use of sunflecks.

| Canopy architecture determines the spectral composition of light reaching lower canopy layers
Light within lower canopy layers is often considered to be predominately made up of diffuse light interspersed with sunflecks mostly consisting of direct light. As the different wavelengths of light are absorbed by chlorophylls with varying efficiency, the spectral quality within the canopy also dictates the rate of photosynthesis (Hogewoning et al., 2010;Smith et al., 2017;Zhu, Long, & Ort, 2008).
During a fleck, the change between direct-or diffuse-light leads to corresponding changes in the spectral quality of light available to leaves. Chlorophyll absorbs relatively weakly in the green, absorbing mostly in the red and blue regions of the spectrum (Evans & Anderson, 1987;Terashima, Fujita, Inoue, Chow, & Oguchi, 2009).
Coupled with the highly refractive properties of leaf tissue, this leads to a larger portion of green photons in deep mesophyll layers and in the diffuse radiation reaching deeper canopy layers (Smith et al., 2017;Sun, Nishio, & Vogelmann, 1998). This has previously been postulated to contribute to increasing radiation use efficiency with depth in the canopy (Smith et al., 2017).
The light environment in which a plant develops will also affect the accumulation of pigments, which will in turn effect spectral composition available to the photosynthetic machinery. Here, the two varieties differ in the amount and distribution of epidermal flavones and anthocyanins (Figure 4). This pattern reflects the expected primary function of flavones as UV-screening compounds and antioxidants whose accumulation is induced by shortwave solar radiation (blue light, UV-A and UV-B radiation; Righini et al., 2019). Given this, it is interesting that Paragon, in which the extinction of sunlight through the canopy profile is steeper than in CMH79A, generally retained the greater flavone content throughout the canopy for both the abaxial and adaxial epidermis. One possible explanation would be that CMH79A is adapted to higher irradiances in its original environ- It is expected that the lower the natural frequency, the shorter the expected duration of windflecks. This also correlates to the risk of lodging with a reduction in natural frequency associated with an increase in lodging risk (Baker et al., 1998;Piñera-Chavez, Berry, Foulkes, Jesson, & Reynolds, 2016). Therefore, selecting for an increased natural frequency could confer the potential to both reduce the risk of lodging and alter the duration of time spent under different light intensities.
However, for photosynthesis to be fully maximized, further work is required to determine how the time period spent under different light conditions experienced in the field will influence photosynthesis. Whilst previous modelling studies indicate that optimal maximal photosynthesis is related to the frequency of switches, the duration spent under each intensity is also of importance and thus it must be determined whether there are differential effects of lengthening of the time period spent under high-versus low-intensities .
A second architectural feature which can be manipulated to optimize the light environment is the sparseness of the canopy; determined by leaf stature, LAI and planting density. Within this study, the open canopy of CMH79A resulted in higher light intensities at both canopy positions for both the static [modelled] scenario and measured in the field (Figures 6 and 7). This can be related to light attenuation, with an increased probability of light penetration to lower leaf layers.
Erect leaf stature has often been proposed to improve whole canopy photosynthesis in canopies with a larger LAI such as cereals and, as such, is often targeted during breeding as a trait for representative of the idealized plant type; or ideotype. This feature is associated with uniformity of light across the canopy, enhancement of light interception at low solar angles and a reduction in the susceptibility to photoinhibition and heat load when the sun is directly overhead Falster & Westoby, 2003;King, 1997;Peng, Khush, Virk, Tang, & Zou, 2008;Werner, Ryel, Correia, & Beyschlag, 2001). In comparison, the denser canopy of Paragon resulted in a nonsignificant change in light intensity and spectral quality during windflecks, but not in the static [modelled] configuration, thus suggesting that above a certain density (or LAI, etc.), the potential for a windfleck to allow the light to penetrate to lower canopy levels, thus homogenizing the light distribution is more limited.