Natural Renewable Resources

 

Optical properties of grapevine leaves: reflectance, transmittance, absorptance and chlorophyll concetration

 

Propiedades ópticas de las hojas de vid: reflectancia, transmitancia, absorptancia y concentración de clorofila

 

Alejandro Cabello–Pasini, Víctor Macías–Carranza

 

Instituto de Investigaciones Oceanológicas, Universidad Autónoma de Baja California, A.P. 453. 22800. Ensenada, Baja California. (acabello@uabc.edu.mx).

 

Received: june, 2011.
Approved: november, 2011.

 

Abstract

Leaf absorptance (A), transmittance (T) and reflectance (R) of visible solar radiation strongly correlate with chlorophyll concentration in a number of plant species, however, little is known about the optical properties of grapevine leaves. Consequently, the objective of this study was to evaluate the optical properties of intact leaves for estimating chlorophyll concentration in Cabernet Sauvignon, Merlot and Tempranillo vaieties. Leaves from Cabernet Sauvignon, Merlot and Tempranillo vines were collected at San Antonio de las Minas, Baja California, México, in 2009, and A, T and R determined with a spectroradiometer. While leaf age played a significant role on the leaf chlorophyll concentration, the chlorophyll a+b concentration vs. R, T and A (evaluated from 400 to 700 nm) in the leaves of the three varieties studied followed an exponential relationship. There was a clear difference in the chlorophyll content, and the R, T, and A spectra among leaves of different ages in all three varieties studied. Reflectance values at wavelengths (λ) >750 nm increased as the age of the leave increased while λ values at approximately 550 nm remained relatively constant. The mean A values of the Merlot (0.80±0.07), Cabernet (0.82±0.05) and Tempranillo leaves (0.78±0.08) studied were below the assumed A value of 0.84 for vascular plants. Our observations suggest that while the 0.84 A value is acceptable for estimating absolute electron transport rates (ETR) in mature grapevine leaves, relative ETR values must be reported in young leaves. Band ratios were developed to optimize the evaluation of canopy area, canopy water status, canopy chlorophyll concentration and others using R measurements. Optimum band ratios were generated by dividing R at the best fit λ by R at each λ through the 400 to 850 nm range and regressing total chlorophyll concentration vs. ratio value.

Key words: absorptance, chlorophyll, reflectance, transmittance, optical properties.

 

Resumen

La absorptancia (A), transmitancia (T) y reflectancia (R) de la radiación solar visible se correlacionan fuertemente con la concentración de clorofila en numerosas especies de plantas; sin embargo, se sabe poco sobre las propiedades ópticas de las hojas de la vid. Por tanto, el objetivo de este estudio fue evaluar las propiedades ópticas de las hojas intactas para estimar la concentración de clorofila en las variedades Cabernet Sauvignon, Merlot y Tempranillo. Se recolectaron hojas de Cabernet Sauvignon, Merlot y Tempranillo en San Antonio de las Minas, Baja California, México, en 2009, y se determinó su A, T y R con un espectrorradiómetro. Aunque la edad de la hoja tuvo gran importancia en la concentración de clorofila en las hojas, la concentración de clorofila a + b vs R, T y A (evaluada entre 400 y 700 nm) en las hojas de las tres variedades estudiadas reveló una relación exponencial. Hubo una clara diferencia en el contenido de clorofila, y los espectros de R, T, y A entre las hojas de diferentes edades en las tres variedades. Los valores de reflectancia en longitudes de onda (λ) > 750 nm aumentaron con la edad de las hojas mientras que los valores de λ a aproximadamente 550 nm se mantuvieron relativamente constantes. Los valores medios de A de las hojas de Merlot (0.80±0.07), Cabernet (0.82±0.05) y Tempranillo (0.78±0.08) analizadas fueron menores al valor de A de 0.84 de las plantas vasculares. Estas observaciones sugieren que el valor de A de 0.84 es aceptable para calcular las tasas de transporte de electrones (ETR) absolutas en hojas maduras de vid, pero los valores relativos a ETR se deben reportar en las hojas jóvenes. Se desarrollaron relaciones de banda para optimizar la evaluación del área del dosel de plantas, el estado del agua del dosel, su concentración de clorofila y otros mediante mediciones de reflectancia. Las relaciones de banda óptimas se generaron dividiendo la R del mejor ajuste de λ entre la R en cada λ en el rango de 400 a 850 nm y con una regresión de concentración de clorofila total vs el valor de la relación.

Palabras clave: absorptancia, clorofila, reflectancia, transmitancia, propiedades ópticas.

 

Introduction

Optical properties of plant leaves such as solar radiation absorptance (A), transmittance (T), and reflectance (R) are strongly dependent on chlorophyll concentration in their tissue. There are differences in chlorophyll concentration among plant species (Carter and Spiering, 2002), among grape varieties (Bica et al., 2000), and as a result of seasonal fluctuations of light and temperature or nutrient stress (Corp et al., 2003; Cabello–Pasini et al., 2004; Cabello–Pasini and Figueroa, 2005). Differences in pigment concentration are often used to evaluate growth, biomass, nitrogen status or physiological stresses in crops or natural vegetation (Blackmer et al., 1996; Peñuelas and Filella, 1998). Furthermore, leaf A, T and R of visible solar radiation show a strong correlation with chlorophyll concentration in several plant species, including Vitis rotundifolia (Carter and Spiering, 2002).

There are spectral reflectance indices based on the optical properties to estimate the concentration of chlorophylls and other leaf pigments in crop vegetation. These indices are useful estimates of biomass and physiological status in wheat (Triticum spp.), corn (Zea mays L.) and other crops, as well as in aquatic vegetation (Augenstein et al., 1991; Blackmer et al., 1996; Barbar et al., 2006). Hyperspectral imagery based on reflectance properties of grapevine leaves is used to discriminate between Cabernet Sauvignon and Shiraz varieties in South Australia (Lacar et al., 2001). However, since reflectance varies as a function of leaf development (Poni et al., 1994), it is critical to understand the effect of leaf age on the leaf optical properties in order to generate reliable hyperspectral indexes for varietal recognition.

Chlorophyll fluorescence is also a useful tool for studying the effects of environmental stress on plants since photosynthesis is often reduced in plants experiencing adverse conditions, such as water deficit, extreme temperature events, nutrient deficiency, polluting agents, attack by pathogens, etc. (Lang et al., 1998). Electron transport rate (ETR) is used as an indicator of plant stress and its computation requires estimates of irradiance, the ratio of photosystem II (PSII) to photosystem I (PSI) and leaf–specific photosynthetic absorptance (Krause and Weis, 1991). Leaf irradiance is generally evaluated in situ while the ratio of PSII to PSI is considered constant. Consequently, the accurate evaluation of ETR using fluorescence techniques depends of a precise estimation of leaf absorptance. Absorptance varies from 0.805 to 0.916 in several species; however, an average value of 0.84 is generally used in most ETR evaluations (Bjorkman and Demmig–Adams, 1987; Mohammed et al., 1995; Knapp and Carter, 1998). While absorptance varies within and among plant species, there are few studies that relate the effect of chlorophyll concentration and tissue age to leaf absorptance in Vitis vinifera. Consequently, the objective of this study was to evaluate the optical properties of intact leaves for estimating chlorophyll concentration in Cabernet Sauvignon, Merlot and Tempranillo varieties.

 

Materials and Methods

Shoots of approximately 1.2–1.5 m in length from Cabernet Sauvignon, Merlot and Tempranillo were tagged at a vineyard in San Antonio de las Minas, Baja California, México (32° 00' N, 116° 38' W). All grapevines were planted on a north–south two–wire trellis system with drip irrigation. Leaves of different age from each variety (n=30) were collected on the same day in August 2009, placed in plastic bags and transported immediately in an ice cooler to a laboratory nearby (Instituto de Investigaciones Oceanológicas, Universidad Autónoma de Baja California). Basal leaves from each shoot were considered senescent, fully developed leaves from the mid section of the shoot were considered mature and apical leaves were considered young. Pigment concentration and optical properties of the leaves did not change for 24 h after collection as verified through a time series; however, all analysis were conduced within 30 min of collection.

Chlorophyll extraction

Chlorophyll a and b were extracted from tissue of the same area of the leaf where optical properties were determined. Tissue samples (1 cm²) were cut from the leaves using a cork borer and were placed in 4 mL of dimethylformamide (DMF) for 24 h at 4 °C. Chlorophyll concentration was determined spectrophotometrically using the equations described by Porra et al., (1989). Chlorophyll content of the extract (mg L^–1) and the total one–sided area of the leaf disks were used to compute leaf chlorophyll concentration.

Optical characteristics

In grapevine leaves, R and T were determined with a Quantum Sensor (LI–190, 400–700 nm, LICOR, Lincon, NE, USA) attached to an integrating sphere (LI–1800–12; LICOR, Lincon, NE, USA). In addition, R and T were determined from 400 to 850 nm using a spectroradiometer (Fieldspec, ASD, Boulder, Colorado, USA) attached via a fiber optic cable to an integrating sphere (LI–1800–12). The spectroradiometer recorded R and T at 1 nm intervals while the adaxial surface of the leaf faced the sphere interior. Reflectance was calculated by subtracting stray spectral radiance from the spectral radiances reflected by the leaf and reference and by dividing leaf reflected radiance by reference reflected radiance. Transmittance was calculated by dividing the transmittance of the sample by the reference transmittance. Absorptance was determined as 1–T–R (Krause and Weis, 1991). The linear relationship of chlorophyll concentration vs. R, T and A at 1 nm intervals (400 to 700 nm) was used to estimate a regression coefficient of determination (r²). The optimal electromagnetic wavelength (λ) value obtained from the r² vs. wavelength for R, T and A was used as numerator when calculating simple band ratios. Reflectance, T and A values from 400 to 850 nm were divided by the optimal λ value and then regressing total chlorophyll concentration vs. ratio value. This ensured that values were strongly correlated to chlorophyll concentration (Carter and Spiering, 2002). Best–fit ratios were determined by dividing R or T at each λ by the reflectance or transmittance at 850 nm (R₈₅₀ or T₈₅₀) or A at each λ by the A at 400 nm (A₄₀₀). Leaf chlorophyll concentration was then regressed with the resulting ratio values (using the exponential equation), and r² was calculated relative to numerator λ.

Statistical analysis

The r² values were used to evaluate linear and non–linear relationships of leaf chlorophyll concentration vs. R, T or A at 1 nm intervals from 400 to 850 nm. Non–linear relationships were modeled by an exponential curve (y=a+be^cx). Significant differences of chlorophyll concentration among leaves of different ages and among varietals was determined by a one–way ANOVA after testing for normality and homoscedasticity of the data (Sokal and Rohlf, 1994). All pairwise multiple comparisons were conducted using Tukey's test (p<0.05). Best–fit linear and exponential regressions were drawn in the Figures using SigmaPlot (SPSS Inc., USA) and Tablecurve 2D (ver. 5.0, SPSS Inc., USA). All results are shown as mean ± standard deviation.

 

Results and Discussion

Chlorophyll content in all three varietals fluctuated as a function of leaf age. In general, chlorophyll a content in Merlot (14.6±4.6μg cm^–2) and Cabernet Sauvignon (14.4±3.1 μg cm^–2) leaves was approximately 20 % greater (p<0.05) than in the Tempranillo leaves (Table 1). In contrast, the concentration of chlorophyll b in the leaves of all three varietals was similar (approx. 3.8 μg cm^–2, p > 0.05). The chlorophyll a/b ratio was approximately 7 % greater (p<0.05) in Merlot and Cabernet Sauvignon (approx. 3.6± 0.3) than in Tempranillo (3.3±0.4). Leaf age plays a significant role on the leaf chlorophyll concentration in several species, including grapevines (Poni et al., 1994; Waldhoff et al. , 2002). Mature grapevine leaves generally have the greatest chlorophyll concentration while apical, youngest leaves tend to have the lowest levels (Poni et al., 1994). The increase in chlorophyll content in the photosynthetic apparatus is triggered by the exposure of the leaf tissue to light (Aleith and Richter, 1991). Sangiovese grapevines, for example, reach maximum chlorophyll levels approximately 80 d after the leaf is developed and maintain these levels until reaching the harvest period (Poni et al., 1994). Consistent with findings reported by Schultz (1996), the chlorophyll a + b concentration in our study ranged from 0.8 mg cm^–2 in young leaves to more than 30 mg cm^–2 in mature leaves.

In all cases, the chlorophyll a + b concentration vs. R, T and A relationship (evaluated from 400 to 700 nm) was better explained by an exponential curve rather than by a linear relationship (Figure 1). The r² for these relationships was below 0.90 and as low as 0.59. Pooled A values for Merlot (0.80±0.07), Cabernet (0.82± 0.05) and Tempranillo (0.78± 0.08) were relatively similar. However, younger leaves in all three varieties had the lowest chlorophyll levels and the lowest A, while mature leaves had the greatest chlorophyll concentration and highest A values. These findings are consistent with results from studies where absorption increased in the leaves due to increasing levels of chlorophyll concentration (Carter and Knapp, 2001; Carter and Spiering, 2002). Furthermore, Cabernet had the narrowest A values while Tempranillo had the greatest A range, spanning from approximately 0.5 in young leaves to approximately 0.9 in mature leaves. These differences in chlorophyll content as a function of leaf development suggest that leaf age must be standardized when evaluating physiological characteristics such as photosynthesis, chlorophyll and nitrogen content, as well as optical properties.

Leaf A is a key optical characteristic for evaluation of photosynthetic ETR in vascular and non–vascular plants (Krause and Weis, 1991; Mercado et al., 1996). It is assumed that approximately 84 % of the incident quanta are absorbed by the leaf pigments; however, A is species specific and can fluctuate as a function of irradiance history of the plant, chlorophyll levels and nitrogen concentration in the tissue (Carter and Spiering, 2002). In our study, when A from young and mature leaves were pooled together, the mean A values of the Merlot (0.80±0.07), Cabernet (0.82±0.05) and Tempranillo (0.78±0.08) were below the assumed A value of 0.84 for vascular plants. While Cabernet leaves had the narrowest A values indicating a relative homogeneity in pigment concentration among leaves, Tempranillo leaves had an A range that spans from approximately 0.5 in young leaves to approximately 0.9 in mature leaves. However, if A of young leaves is not considered, A of the three varieties falls close to the 0.84 value observed in other plant species (Knapp and Carter, 1998). Our observations suggest that while the 0.84 A value is acceptable for estimating absolute ETR in mature grapevine leaves, relative ETR values must be reported in young leaves unless the specific A of the leave is evaluated.

There was a clear difference in R, T and A (evaluated from 400 to 700 nm) spectra among leaves from different ages in all three varieties studied (Figure 2). Reflectance values at λ >750 nm increased as the age of the leave increased while l values at approximately 550 nm remained relatively constant. Transmittance values increased slightly at λ >750 nm, however, there was a greater increase in T values at approximately 550 nm as the age of the leaf increased. The range of R values was 3– to 6–fold lower than the range of T and A values for all three varieties. In contrast to R values, A in the leaves of Merlot, Cabernet and Tempranillo was very low at λ >750 nm; however, there was a great increase in A at approximately 550 nm as the age of the leaf increased. The exponential relationship between chlorophyll a + b concentration vs. R, T and A (evaluated from 400 to 700 nm) in the leaves of the three varieties is consistent with the relationship observed in the leaves of Vitis rotundifolia and other broad–leaf plants using monochromatic spectra (approx. 700 nm) (Carter and Spiering, 2002). Furthermore, the total chlorophyll concentration vs. R had lower r² relative to T and A in the three varieties studied. This suggests that T and A are better predictors of total chlorophyll in the leaves of Cabernet, Merlot and Tempranillo relative to R.

Wavelengths >720 nm have been suggested as having the greatest potential as an aid for the distcriminatin of grapevine varieties (Lacar et al., 2001. Howver, differences in R in mature leaves at wavelengths >720 were not observed among the three varieties studied here. Besides, strong differences in R values (wavelengths >720 nm) were observed in young leaves among all threes varietals. This suggests that optical properties of young leaves might be a better descriptor of grapevine variety than those observed in mature leaves. Thus, hyperspectral imagery for vineyard varietal mapping might increase their sensitivity at the onset of leaf development.

The r² from the total chlorophyll concentration vs. T and A showed similar patterns among varieties, however, the total chlorophyll concentration vs. R varied significantly among varieties (Figure 3). In general, the exponential model provided an approximately 10 % better fit of the chlorophyll a + b vs. wavelength relationship than the linear model. Best fit regression for the linear or exponential chlorophyll vs. R relationship was found at approximately 695 nm for all varietals. The best fit regression for the linear or exponential chlorophyll vs. T and A showed a wide band from 450 to 700 nm, however, there was a slight peak at approximately 700 nm.

In general, the exponential model provided an approximately 10 % better fit of the chlorophyll a+b vs. wavelength relationship than the linear model (see Figure 3). This is consistent with the observed exponential relationship between R, T and A vs. chlorophyll concentration observed in a number of other species (Carter and Knapp, 2001; Carter and Spiering, 2002). There was a clear peak of the r² in the red/far–red spectrum (approx. 700 nm) when evaluating the concentration of chlorophyll a + b vs. R. This is also consistent with observations by Carter and Spiering (2002) in a number of species. However, the r² was relatively stable between 450 and 700 nm when T or A were used to evaluate chlorophyll a+b concentration in the leaves of all three varieties in our study. This suggests that a wide–band sensor in the visible light spectrum (i.e. PAR radiometer) might provide similar results as a narrow–band spectroradiometer for the evaluation of chlorophyll in the leaves of grapevines.

The optimum band ratios were generated by dividing R at the best fit λ (indicated in Figure 3) by R at each λ through the 400 to 850 nm range and regressing total chlorophyll concentration vs. ratio value (Figure 4). Similar band ratios were generated for T and A. Consistently, maximum values of r² from the total chlorophyll content vs. R and T maxima (λ ≈ 700 nm) were obtained when these values were divided by λ at 850 nm. In contrast to R and T, values of r² were highest for A when maximum values of A were divided by λ at 400 nm and then regressed to total chlorophyll content using the exponential equation of the three varieties studied. Best–fit ratios (r²) determined by dividing R or T at each λ by R₈₅₀ or T₈₅₀, or A at each λ by A₄₀₀ and regresses to leaf chlorophyll concentration were approximately 5 to 35 % greater than those obtained the non–normalized λ maxima for R, T, and A.

Absorptance in grapevine leaves, measured with a simple wide–band spectro–radiometer sensor, varieties approximately 0.60 to more than 0.90 as a function of leaf plastochron index (Schultz, 1996). Band ratios have been developed using R measurements to optimize the evaluation of canopy area, canopy water status or canopy chlorophyll concentration (Barbar et al., 2006). Optimum band ratios were generated by dividing R at the best fit λ (indicated in Figure 3) by R at each λ through the 400 to 850 nm range and regressing total chlorophyll concentration vs. ratio value (see Figure 4). Similar band ratios were generated for T and A. Consistently, maximum values of r² from the total chlorophyll content vs. maximum R and T (λ =700 nm) were obtained when these values were divided by λ at 850 nm. In contrast to R and T, values of r² were highest for A when maximum values of A were divided by λ at 400 nm and then regressed to total chlorophyll content using the exponential equation of the three varieties studied.

The chlorophyll a + b concentration vs. R (R₇₀₄, R_704/850), T (R₇₀₅, R_705/850) and A (A₇₀₇, A _707/400) the leaves of the three varieties studied followed an exponential relationship (Figure 5). In general, there was a similar or better relationship between the chlorophyll content vs. R, T, or A when the normalized values were used. Furthermore, the R_705/850 vs. total chlorophyll relationship had greater r² values than when the R was used to estimate (R_400–700) chlorophyll levels. In contrast to R, the relationship of T and A vs. chlorophyll concentration was relatively similar when using the 400–700 nm measurements or the single wavelength or the ratio. Best–fit ratios were determined by dividing R or T at each λ by A₄₀₀. Chlorophyll concentration was then regressed with the resulting ratio values (using the exponential equation), and r² was calculated relative to numerator λ (Figure 6). In general, r² values were approximately 5 to 35 % greater than those obtained from the non–normalized λ maxima for R, T, and A. Similar to previous analysis (Figure 3), r² minima occurred at approximately λ 680 and 750 nm. This further suggests that the use of a narrow waveband around 700 nm divided by a far red signal (≈850 nm) would yield the most accurate estimates of leaf chlorophyll concentration based on R and T.

 

Conclusions

The results of this study indicate that the use R, T and A evaluated using broad wavelength sensors (400–700 nm) can provide a reliable estimate of a wide range of chlorophyll content in the leaves of Vitis vinifera. However, the estimation of chlorophyll content in the leaves can be enhanced using monochromatic wavelengths of approximately 700 nm for the evaluation of R, T and A or their ratio. No significant differences in optical properties were observed among the mature Cabernet Sauvignon, Merlot and Tempranillo grapevines studied here. Strong differences in R values (>720 nm), however, were observed in young leaves among all threes varietals studied here. This suggests that young leaf reflectance (λ >720) might be a better descriptor of grapevine than those observed in mature leaves.

 

Literature Cited

Aleith, F., and G. Richter. 1991. Chloroplast differentiation in somatic embryos of carrot; Efficiency of blue and red light irradiance on gene expression. J. Plant Physiol. 138:304–308.         [ Links ]

Augenstein, E. W., D. A. Stow, and A. S. Hope. 1991. Evaluation of SPOT HRV–XS data for kelp resource inventories. Photogramm. Eng. Rem. Sens. 57: 501–509.         [ Links ]

Barbar, M., M. Reynolds, M. Ginkel, A. Klatt, W. Raun, and M. Stone. 2006. Spectral reflectance to estimate genetic variation for in–season biomass, leaf chlorophyll, and canopy temperature in wheat. Crop Sci. 46: 1046–1057.         [ Links ]

Bica, D., G. Gay, A. Morando, and E. Soave. 2000. Effects of rootstock and Vitis vinifera genotype on photosynthetic parameters. Acta Hort. 526: 373–380.         [ Links ]

Bjorkman, O., and B. Demmig–Adams. 1987. Photon yield of O₂ evolution and chlorophyll fluorescence characteristics at 77 K among vascular plants of diverse origins. Planta 170: 489–504.         [ Links ]

Blackmer, T. M., J. S. Schepers, G. E. Varvel, and E. A. Walter–Shea. 1996. Nitrogen deficiency detection using reflected shortwave radiation from irrigated corn canopies. Agron. J. 88: 1–5.         [ Links ]

Cabello–Pasini, A., and F. L. Figueroa. 2005. Effect of nitrate concentration on the relationship between photosynthetic oxygen evolution and electron transport rate in Ulva rigida (Chlorophyta). J. Phycol. 41: 1169–1177.         [ Links ]

Cabello–Pasini, A., R. Muñiz–Salazar, and D.H. Ward. 2004. Biochemical characterization of the seagrass Zostera marina at its southern end of distribution in the North Pacific. Cien. Mar. 30: 21–34.         [ Links ]

Carter, A., and B. Spiering. 2002. Optical properties of intact leaves for estimating chlorophyll concentration. J. Environ. Qual. 31: 1424–1432.         [ Links ]

Carter, G., and A. Knapp. 2001. Leaf optical properties in higher plants: linking spectral characteristics to stress and chlorophyll concentration. Am. J. Bot. 88: 677–684.         [ Links ]

Corp, L., J. McMurtrey, E. Middleton, C. Mulchi, E. Chappelle, and C. Daughtry. 2003. Fluorescence sensing systems: in vivo detection of biophysical variations in field corn due to nitrogen supply. Rem. Sens. Environ. 86: 470–479.         [ Links ]

Knapp, A., and G. Carter. 1998. Variability in leaf optical properties among 26 species from a broad range of habitats. Am. J. Bot. 85: 940–946.         [ Links ]

Krause, G. H., and E. Weis. 1991. Chlorophyll fluorescence and photosynthesis: the basics. Ann. Rev. Plant Physiol. Plant Mol. Biol. 42: 313–349.         [ Links ]

Lacar, F., M. Lewis, and I. Grierson. 2001. Use of hyperspectral reflectance for discrimination between grapevine varieties, Proceedings of the Geoscience and Remote Sensing Symposium, Sydney, Australia. pp: 2878–2880.         [ Links ]

Lang, N. S., R. L. Wample, R. Smithyman, and L. Mills. 1998. Photosynthesis and chlorophyll fluorescence in blackleaf–affected concord leaves. Am. J. Enol. Vitic. 49: 367–374.         [ Links ]

Mercado, J. M., C. Jimenez, F. X. Niell, and F. L. Figueroa. 1996. Comparison of methods for measuring light absortion by algae and their applications to the estimation of the package effect. Sci. Mar. 60: 39–45.         [ Links ]

Mohammed, G., W. Binder, and S. Gillies. 1995. Chlorophyll fluorescence: a review of its practical forestry aplications and instrumentation. Scand. J. Forest. Res. 10: 383–410.         [ Links ]

Peñuelas, J., and I. Filella. 1998. Visible and near–infrared reflectance techniques for diagnosing plant physiological status. Trends Plant Sci. 3: 151–156.         [ Links ]

Poni, S., C. Intrieri, and O. Silvestroni. 1994. Interactions of leaf age, fruiting and exogenous cytokinins in Sangiovese grapevines under non–irrigated conditions. II. Chlorophyll and nitrogen content. Am. J. Enol. Vitic. 45: 278–284.         [ Links ]

Porra, R. J., W. A. Thomson, and P. E. Kriedemann. 1989. Determination of accurate extinction coefficients and simultaneous equations for assaying chlorophylls a and b extracted with for different solvents: verification of the concentrations of chlorophyll standards by atomic absorption spectroscopy. Biochim. Biophys. Acta 975: 384–394.         [ Links ]

Schultz, H. 1996. Leaf absorptance of visible radiation in Vitis vinifera L.: estimates of age and shade effects with a simple field method. Sci. Hort. 66: 93–102.         [ Links ]

Sokal, R. R., and F. J. Rohlf. 1994. Biometry W. H. Freeman and Company, New York 960 p.         [ Links ]

Waldhoff, D., B. Furch, and W. J. Junk. 2002. Fluorescence parameters, chlorophyll concentration, and anatomical features as indicators for flood adaptation of an abundant tree species in Central Amazonia: Symmeria paniculata. Environ. Exper. Bot. 48:225–235.         [ Links ]