The Response of Table Grape Growth, Production, and Ripening to Water Stress
Four year old Flame Seedless' grapevines located in a commercial vineyard subjected to increased water stress levels based on infrared canopy temperatures and the Crop Water Stress Index (CWSI) for two years. CWSI levels were approximately .18, .30 and .33 for the wet, medium and dry treatments. In the first year there were no significant differences in yield however, there was a significant reduction in the amount of water applied in both the medium and dry treatments when compared to the wet treatment. In addition, the wet treatment had significantly greater growth during the first growing season when comparing pruning weights.
Commercial cultivation and production of table grapes is increasing in Arizona. Decreasing water supplies, increasing water cost, and the passage of the 1980 Groundwater Management Act are forcing Arizona growers to be more water efficient if they are to remain competitive with other markets. Research was begun in 1989 to determine the effects of water stress on grape production, vine growth, and berry ripening based on thermal infrared canopy temperatures and the use of the Crop Water Stress Index to schedule irrigations.
"Flame Seedless" table grapes in their fourth leaf (in 1989) growing at Verde Grande Vineyards in Stanfield, Arizona, were subjected to increasing water stress levels based upon infrared canopy temperatures and the CWSI. Twelve drip irrigated (2 one-half gallon per hour emitters per plant) rows containing 65 plants, spaced 8 feet in row and 12 feet between rows, were randomized into three water treatments containing four replicates. Infrared temperature readings were taken with an Everest Interscience, Inc.* Model 112 infrared thermometer with a 15 degree field of view from the southeast and southwest side of each row on the 10th, 33rd, and 55th plants and averaged. The normal grower irrigation schedule was monitored as a fourth treatment in 1990.
Vapor pressure deficits were taken at 15 minute intervals near the 33rd plant with an aspirated psychrometer (Environmental Tectonics Corp.*, Psychro-Dyne model). Canopy temperature measurements were collected two to three times per week when skies were clear between 1100 and 1500 hours (Mountain Standard Time). Neutron moisture measurements were recorded three times per week. Irrigations were scheduled at increasing CWSI values to induce water stress levels ranging from well-watered to moderate water stress.
Soil moisture content was monitored to a 4 foot depth in 1 foot increments with a Campbell Pacific Nuclear Model 503DR neutron moisture meter between the 32nd and 33rd plant on each row. When irrigation s were scheduled, attempts were made to add only enough water to refill the soil profile to field capacity. Water applied was measured with a flow meter on each row.
Fruit characteristics (cluster weight, berry weight and diameter, brix, and titratable acidity) were measured on May 23, May 30, and June 5 in 1989 and May 30, June 6, June 12, June 16, June 25, and July 3 in 1990 from the three plants per row which were used for the infrared readings. Each row was harvested on June 5 in 1989, and June 19, June 29, and July 5 in 1990 by the grower. The number of packed boxes (grades 1 and 2) were recorded for each row in 1989. Packed boxes (grades 1, 2, and 3) and raisin clusters were recorded for each row in 1990. Vine growth was measured from total pruning weights per row on December 21, 1989, and January 4, 1990.
Results and Discussion
The CWSI at irrigation ranged from 0.18 to 0.33 units, and 0.12 and 0.24 units in 1989 and 1990, respectively. A ± 0.05 unit standard error is commonly associated with CWSI field measurements. In 1989 actual CWSI's at irrigation were 0.18, 0.30, and 0.33 for the wet, medium, and dry treatments respectively (Table 1). There were significant differences in total water applied (P = 0.05) and vine pruning weights (P = 0.01), but not in the number of packed boxes per acre. The wet treatment required more water at 44.7 inches and also produced the highest amount of growth at 4824 pounds of pruning per acre. The medium treatment had the highest number of packed boxes (grades 2 and 2) at 400 boxes per acre, although no significant differences between treatments were measured.
There was a very high linear correlation between the CWSI at irrigation and 1) total applied water r = 0.9996), and 2) vine pruning weight r = 1.00) indicating the CWSI is a useful tool to measure water stress effects on growth and production of table grapes. The relationship between the CWSI at irrigation and the number of packed boxes (grades 1 and 2) was lower with r = 0.7125.
Preliminary results, through harvest only, for 1990 are presented in Table 2. The highest number of packed boxes (grades 1, 2, and 3) were attained when irrigations were scheduled at a CWSI of 0.24 units. Lower CWSI at irrigation values decreased packed boxes linearly. In 1989 highest packed boxes grades 1 and 2 were attained at a CWSI at irrigation of 0.30 units. The results from both years data indicate if irrigations are scheduled at a CWSI value between 0.24 and 0.30 units, optimum production may be obtained. When irrigations were scheduled at a CWSI value lower than 0.24, the number of packed boxes per acre decreased in both years. The CWSI at irrigation in 1990 was highly correlated with 1) applied water r = 0.960) and 2) packed boxes (grades 1, 2, and 3) r = 0.9795) as in 1989. No significant difference between treatments was measured in residual number of raisin clusters left per row. As the CWSI at irrigation increased, the average water deficit of the four foot soil profile increased linearly and the number of irrigations required also decreased.
The number of and distribution of packed boxes grades 1, 2, and 3 per acre for each treatment at each harvest date in 1990 are presented in Table 3 The wet or medium treatments had the highest number of grade #1 boxes on all three harvest dates indicating earlier and greater sizing. The dry treatment produced the highest total number of packed boxes over all grades on the June 19 and June 29 harvests. Excluding the grower treatment, where irrigation schedules were not based on CWSI values, the number of grade #1 boxes decreased linearly with increasing CWSI at irrigation values. However, the number of packed boxed grades #2 and #3 increased linearly with increasing CWSI at irrigation values.
The change in cluster weight over time for 1989 and 1990 are shown in Figures 1 and 2, respectively. The wet treatment achieved the highest cluster weight and at an earlier date in 1989, while the medium treatment had a higher cluster weight in 1990 which accounts for the medium treatment having the highest number of grade #1 packed boxes on the June 19, 1990 harvest. The dry treatment lagged behind other treatments in cluster weight gain in both years. Cluster weight decreased for all treatments after June 25, 1990 (Figure 2). This could possibly be due to overripening and raisin formation or because the nicest berries had been removed in earlier harvests as indicated in the reduced number of packed boxes for all treatments on the July 5, 1990 harvest (Table 3). The wet treatment sized earlier with a greater and earlier berry weight (Figures 3 and 4) and berry diameter (Figures 5 and 6) than the other treatments in 1989 and 1990, respectively. Berry diameter (Figure 6) is linear with the total harvest of grade #1 boxes (Table 3). The earlier berry sizing and weight of the wet treatment is probably due to increased vine vigor, which resulted in higher plant sugar and water accumulation when compared to the less vigorous medium and dry treatments based on pruning weights in 1989 (Table 1), although this must still be verified from 1990 pruning weights.
Figures 7 and 8 depict small insignificant differences in degree brix between treatments in 1989 or 1990, respectively. The slight drop in degree brix between June 25 and July 3, 1990 for the wet, medium, and dry treatments (Figure 8) may indicate the onset of souring and raisin formation as previously mentioned. The grower treatment degree brix increases during this period which may explain the significantly higher number of total packed boxes for the grower treatment on the July 5, 1990 harvest over the other three treatments (Table 3). As we would be expected, titratable acidity decreased over time as degree brix increases (Figures 9 and 10) with no significant differences between treatments for both years. In addition the degree brix:acid ratio increase over time (Figures 11 and 12), and no significant differences were measured between treatments in both years. The lack of water treatment effects on the degree brix, titratable acidity and the brix to acid ratio indicates water stress had a lesser effect on these fruit characteristics.
The CWSI is a useful tool to determine the effects of water stress on table grape growth and ripening as indicated by the very high linear correlations between the CWSI at irrigation and 1) total applied water, r = 0.9996, and 2) vine pruning weights, r = 1.00, in 1989. In 1990, the CWSI at irrigation was highly linear with 1) total water applied, r = -0.960 and 2) packed boxes (grades 1, 2, and 3), r = 0.9795. Lower water stress levels (CWSI = 0.18 units at irrigation) promoted earlier berry sizing, increased berry weight, and increased cluster weight over drier treatments, although yield was decreased. Significantly higher growth (P = 0.01), based upon pruning weights, also was attained at the lower water stress level in 1989. Highest production was attained when irrigations were scheduled between 0.24 and 0.30 CWSI units in both years. The water stress treatments n this test had no effect on degree brix, titratable acidity, or the degree brix:acid ratio fruiting characteristics. The test should be repeated for at least another year in order to better define the response of table grape growth, production, and ripening to water stress.
* Trade names are included for th e benefit of the reader, and do not imply endorsement of the authors, Cooperative Extension, or the University of Arizona.
This is a part of publication AZ1051: "1998 Citrus and Deciduous Fruit and Nut Research Report," College of Agriculture, The University of Arizona, Tucson, Arizona, 85721.
This document located at http://ag.arizona.edu/pubs/crops/az1051/az105119.html
Return to Table of Contents