Correlations among Predawn Leaf, Midday Leaf, and Midday Stem Water Potential and their Correlations with other Measures of Soil and Plant Water Status in Vitis vinifera

A study was conducted to compare three measurements of determining water status of grapevines (Vitis vinifera L.) in the field. Predawn leaf water potential (ΨPD), midday leaf water potential (Ψl), and midday stem water potential (Ψstem) were measured on ‘Chardonnay’ and ‘Cabernet Sauvignon’ grapevines grown in Napa Valley, California late in the 1999 growing season. Both cultivars had been irrigated weekly at various fractions (0, 0.5, and 1.0 for ‘Chardonnay’ and 0, 0.5, 0.75, and 1.5 for ‘Cabernet’) of estimated vineyard evapotranspiration (ETc) from approximately anthesis up to the dates of measurements. Predawn water potential measurements were taken beginning at 0330 HR and completed before sunrise. Midday Ψl and Ψstem measurements were taken only between 1230 and 1330 HR. In addition, net CO2 assimilation rates (A) and stomatal conductance to water vapor (gs) were also measured at midday. Soil water content (SWC) was measured in the ‘Chardonnay’ vineyard using a neutron probe. Values obtained for ΨPD, Ψl, and Ψstem in this study ranged from about –0.05 to –0.8, –0.7 to –1.8, and –0.5 to –1.6 MPa, respectively. All three measurements of vine water status were highly correlated with one another. Linear regression analysis of Ψl and Ψstem versus ΨPD resulted in r2 values of 0.88 and 0.85, respectively. A similar analysis of Ψl as a function of Ψstem resulted in an r2 of 0.92. In the ‘Chardonnay’ vineyard, all three methods of estimating vine water status were significantly (P < 0.01) correlated with SWC and applied amounts of water. Lastly, ΨPD, Ψl, and Ψstem were all linearly correlated with measurements of A and gs at midday. Under the conditions of this study, ΨPD, Ψl, and Ψstem represent equally viable methods of assessing the water status of these grapevines. They were all correlated similarly with the amount of water in the soil profile and leaf gas exchange as well as with one another. vine YPD among three watering treatments but no differences in Ψl were found when measured at 1000 and 1600 HR. They concluded that ΨPD better reflected soil water availability than Ψl. van Zyl (1987) concluded that ΨPD detected the onset of water stress in grapevines earlier and more accurately than Ψl. Stem water potential is determined by enclosing a leaf in a plastic bag that is surrounded by aluminum foil, stopping transpiration, enabling that leaf to come into equilibrium with the water potential of the stem (Begg and Turner, 1970). The reported amount of time between enclosing the leaf in plastic and foil, and measuring Ψstem for trees and grapevines, has been from 45 to 120 min (Garnier and Berger, 1987; McCutchan and Shackel, 1992; Naor et al., 1997). Some have bagged leaves from 14 to 24 h before measuring Ψstem in grape (Liu et al., 1978; Stevens et al., 1995). Stem water potential has been shown to be less variable than Ψl and improved the ability to detect small, but statistically significant differences among treatments (McCutchan and Shackel, 1992). It was also found that a clear difference in Ψstem between two irrigation treatments occurred at an earlier date (1 week) during the growing season than differences in ΨPD and Ψl for the same treatments (Selles and Berger, 1990). In addition, Ψstem has been shown to be a linear function of applied water (Lampinen et al., 1995) and soil water availability (Stevens et al., 1995). Lastly, Ψstem has been highly correlated with tree (Olien and Lakso, 1986) and fruit (Naor et al., 1995) size in apple [Malus sylvestris (L.) Mill var. domestica (Borkh.) Mansf.]. It has been suggested that for a measure of plant water status (such as Ψl) to be a sensitive indicator of water stress, it must be responsive to differences in soil moisture status and/or resulting growth differences due to water applications (Higgs and Jones, 1990). It should also be closely related to shortand medium-term plant stress responses (Shackel et al., 1997) and less dependent upon changes in environmental conditions (Jones, 1990; Received for publication 25 May 2001. Accepted for publication 1 Jan. 2002. Corresponding author; University of California, Kearney Agricultural Center, 9240 S. Riverbend Ave., Parlier, CA 93648; e-mail [email protected]. Visiting scientist; Facultad de Agronomia, Universidad del Zulia. Nucleo Agropecuario, Av. Goajira, Maracaibo 4001, Estado Zulia, Venezuela. Since development of the pressure chamber (Scholander et al., 1965), measurement of leaf water potential (Ψl) has been used as a tool to assess the water status of plants (Jones, 1990; Koide et al., 1989). Accordingly, leaf Ψl has been used to monitor the water relations of grapevines (Vitis L. sp.) (Smart and Coombe, 1982; Williams et al., 1994). It has been correlated with various aspects of grapevine physiology (Naor et al., 1994; Williams et al., 1994), vegetative growth (Schultz and Matthews, 1988, 1993), and reproductive growth and yield (Greenspan et al., 1996; Grimes and Williams, 1990). Grapevine Ψl has been shown to be fairly consistent up and down the axis of the shoot of Vitis labruscana Bailey when leaves are uniformly exposed to solar radiation (Liu et al., 1978). Lastly, Ψl has also been used as a factor in a functional model of stomatal conductance of grapevines (Winkel and Rambal, 1990). There have been reports in which it was suggested that midday or diurnal measurements of Ψl did not provide a reliable estimate of plant water status. This was due to lack of correlation between Ψl with other physiological parameters, measures of growth, or amounts of applied water (Chone et al., 2001; Garnier and Berger, 1985; Higgs and Jones, 1990; Naor, 1998). Therefore, other methods of measuring plant water status in the field, such as predawn leaf water potential (ΨPD) and stem water potential (Ψstem) are being used. Measurements of ΨPD have been used in grape studies since it is assumed that before sunrise the vine is in equilibrium with the soil’s water potential (Correia et al., 1995; Schultz, 1996; Winkel and Rambal, 1993). Correia et al. (1995) found significant differences in 9056-SPW 3/13/02, 10:25 AM 448 449 J. AMER. SOC. HORT. SCI. 127(3):448–454. 2002. McCutchan and Shackel, 1992). The specific examples given above for grape would indicate that ΨPD, Ψl, or Ψstem may all be possible candidates. Only a few studies have actually compared one of the three methods of measuring Ψ with one another for determination of plant water status. Stevens et al. (1995) found that diurnal measures of Ψl and Ψstem of grape were highly correlated (r = 0.97) with one another. Conversely, Naor et al. (1995) found that the correlation between Ψl and Ψstem of apple resulted in a r of 0.35. Therefore, the purpose of this study was to measure ΨPD, Ψl, and Ψstem of two Vitis vinifera cultivars and compare the three with one another and with measures of leaf gas exchange, soil water content, and reproductive growth. Grapevines at two sites were chosen as they had been irrigated at various fractions of estimated vineyard evapotranspiration (ETc) from the initial irrigation of the season onward, providing plant material expected to exhibit large differences in soil and vine water status. Materials and Methods Two Vitis vinifera cultivars were used for the study, ‘Chardonnay’ and ‘Cabernet Sauvignon’. The 9-year-old ‘Chardonnay’ vineyard was located in the southern portion of Napa Valley (Carneros District), in California within 10 km of San Francisco Bay. The 10year-old ‘Cabernet Sauvignon’ vineyard was also located in Napa Valley, 3 km from Oakville (≈25 km from the Carneros site). Two rootstocks were used in the ‘Chardonnay’ vineyard, ‘5C Teleki’ (5C) and ‘110 Richter’ (110R). One rootstock was used in the ‘Cabernet Sauvignon’ vineyard, 5C. Vine and row spacings for the ‘Chardonnay’ and ‘Cabernet Sauvignon’ vineyards were 1.52 and 2.13 m and 1.0 and 1.83 m, respectively. The trellis system used in both vineyards was the vertical shoot positioned (VSP). Row directions in the ‘Chardonnay’ and ‘Cabernet Sauvignon’ vineyards were approximately east–west and north–south, respectively. The soil in the ‘Chardonnay’ vineyard was a Diablo fine, montmorillonitic, thermic Chromic Pelloxerert and that in the ‘Cabernet’ vineyard was a Bale fine-loamy, mixed, thermic Cumulic Ultic Haploxeroll. The soil pH of both vineyards was 5.5 and there were no apparent restrictions to root exploration of the profile. Both vineyards used for this research were also being used in an irrigation study investigating relationships among applied quantities of water, rootstock, and productivity. Three irrigation treatments were used in the ‘Chardonnay’ vineyard. Vines received applied amounts of water at 0, 0.5, and 1.0 times estimated ETc. The plot size of an individual irrigation–rootstock treatment consisted of 18 vines down the row using a single border vine and a border row receiving no applied water between plots. Vine water use was calculated as the product of potential ET (ETo) and the crop coefficient (kc). Potential ET was obtained from a California Irrigation Management Irrigation System (CIMIS) weather station located 8 km from the vineyard site. The seasonal crop coefficients (kcs) used were those developed by L.E. Williams in 1994 for a VSP trellis planted on 2.13-m row spacings (unpublished data) and expressed as a function of degree-days from budbreak using a base of 10 C. Four irrigation treatments were used in the ‘Cabernet Sauvignon’ vineyard: 0.0, 0.5, 0.75, and 1.5 times estimated ETc. The plot size of an irrigation treatment at this location was the entire row (78 vines). The kcs used to calculate ETc were similar to those in the ‘Chardonnay’ vineyard but were adjusted for the narrower row spacing (i.e., the kcs were ≈16% greater than for the 2.13 m row spacing). Potential ET for the ‘Cabernet’ vineyard was obtained from a CIMIS weather station located 3 km from the site. Differences in applied water amounts in both vineyards were obtained by using different numbers and/or sizes of in-row emitters using drip irrigation. Soil water content (SWC) was measured only in the ‘Chardonnay’ vineyard using a neutron probe (model 503 DR hydroprobe moisture gauge; Boart Longyear Co., Martinez, Calif.). Six access tubes were installed to a depth of 3 m in one quarter of an individual vine’s rooting volume. One tube was placed close to the trunk of the vine and another midway between vines within the row. Two access tubes were placed midway between rows, in line (perpendicular) with the two in-row tubes. The last two access tubes were placed midway between the four tubes, mentioned previously (i.e., 1/4 the distance between rows). There was one access tube site per irrigation treatment–rootstock combination. Measurements of SWC began at a depth of 0.15 m from the soil surface and at each 0.3-m depth, thereafter. The neutron probe was calibrated with the vineyard’s soil type and expressed as percentage volumetric water content. Soil water content used in the study was the mean of all access tubes at an individual site and at all depths measured. Vine water status and leaf gas exchange were measured on two dates (24 Aug. and 21 Sept. 1999) in the ‘Chardonnay’ vineyard and one date (25 Aug. 1999) in the ‘Cabernet’ vineyard on randomly selected vines only in block 1 of the larger irrigation study at both locations. Soil water content was also measured only in block 1 of the ‘Chardonnay’ vineyard both days. All dates were cloud free. Water potential readings were conducted according to the procedures of Padgett-Johnson et al. (2000) and Koide et al. (1989). Specifically, predawn Ψ measurements began at ≈0330 HR and were finished before sunrise using a pressure chamber (PMS Instruments Co., Corvallis, Ore.). Midday measurements of Ψl and Ψstem occurred between 1230 and 1330 HR, Pacific Daylight Time. Leaf blades for ΨPD and Ψl determinations were covered with a plastic bag, quickly sealed, and petioles then cut within 1 to 2 s. The time between leaf excision and chamber pressurization was generally <10 to 15 s. Leaves, chosen for midday Ψl determinations, were fully expanded, mature leaves exposed to direct solar radiation. These leaves were located on the south side of east–west rows and the west side of the north–south rows. About 90 to 120 min before midday measurements, leaves for determination of Ψstem were enclosed in black plastic bags covered with aluminum foil. Leaves chosen for Ψstem measurements were of similar age and type as those used for Ψl but were located on the north side of the vines in east– west rows and the east side of vines in north–south rows to minimize any possible heating effects. Leaves for midday determinations of Ψl and Ψstem were taken from the same vine and simultaneously measured. One leaf from an individual vine was used for each measurement. In Aug. 2001, midday Ψl was measured on the cultivar Merlot grown in the San Joaquin Valley, comparing leaves covered with a plastic bag before excision, covered with a plastic bag just after excision, and leaves not covered with plastic. All other procedures were as described above for midday Ψl. A single leaf replication of each method to measure Ψl was taken from the same vine using six different vines. Vines were irrigated at 40% and 120% of estimated vineyard ET, weekly. Measurements of net CO2 assimilation rates (A) and stomatal conductance (gs) were taken subsequent to the measurements of midday leaf Ψ and completed by 1400 HR. Both measures of gas exchange were made with a portable infrared gas analyzer, LCA2 (Analytical Development Co., Hoddeson, United Kingdom) using the broad leaf chamber. Leaves chosen for gas exchange were similar to those used for Ψl. Solar radiation, net radiation, photosynthetic photon flux (PPF), ambient temperature and, relative humid9056-SPW 3/13/02, 10:25 AM 449 450 J. AMER. SOC. HORT. SCI. 127(3):448–454. 2002. ity were measured 1 m above the canopy and averaged hourly with a datalogger. Canopy temperature (to calculate canopy to air vapor pressure difference) was measured hourly with a hand-held infrared thermometer (model 39650-04; Cole-Parmer Inst. Co., Chicago, Ill.). Data were analyzed via regression analysis using linear, quadratic, and cubic terms. Since there were no improvements using either quadratic or cubic terms for analysis of any of the relationships obtained herein only linear regressions are presented. The relationships between midday measurements (Ψl and Ψstem) and ΨPD were analyzed using the means of an individual treatment (scion– rootstock combination, irrigation treatment, and date, n = 16). This was due to the fact that measurement of ΨPD was not necessarily determined on the same vines within the plot as done for Ψl and Ψstem. The relationship between Ψl and Ψstem was of individual leaf replicates (n = 6 for each scion–rootstock combination, irrigation treatment in the ‘Chardonnay’ vineyard on 24 Aug. while n = 5 for each treatment in the ‘Chardonnay’ vineyard measured on 21 Sept. and for the ‘Cabernet Sauvignon’ vines measured on 25 Aug.; total n = 86). The relationships between A and gs and water potentials were also determined using treatment means as A and gs were not necessarily determined on the same leaves and/or vines as Ψ measurements were within block 1 at each location. Differences in water potential among irrigation treatments at either site were analyzed via analysis of variance and means separated using Duncan’s multiple range test. An analysis of covariance was used to test for heterogeneity of slopes for the relationship between Ψstem and Ψl among the three different measurement dates.