Abstract: The aim of the experiment was to examine the application of partial rootzone drying and deficit irrigation on growth and plant development of tomatoes. Potted fresh market tomatoes (Lycopersicon esculentum Mill.) in pots were subjected to partial root zone drying (PRD) and controlled deficit irrigation (CDI) under glasshouse conditions. Roots of plants were remained attached to plants and half the volume divided in one plant and the other half planted in the other adjacent pot. The treatments were: well-watered continually maintained close to field capacity in both pots (control), CDI50 (half the amount of water in control divided equally to both pots with each watering), PRD50 (half the amount of water in control applied to one pot while water was withheld from the other pot until soil water declined to 50-70% the field capacity and then water was applied to the other pot), PRD25 (half the amount of water in control was applied to one pot while water was withheld from the other pots until soil moisture declined to 25-50% field capacity and then water was applied to the other pot) and CDI25 (quarter amount of water in control divided equally to both pots with each watering). Imposing water deficit reduced fruit yield up to 18% in PRD50 and 33% in CDI50 which coincided with an impairment of fruit expansion. The percentage of fruit dry matter and osmotic potential increased in both PRD and CDI compared with the control. The incidence of blossom end rot increased in both CDI and PRD25 compared with the control and PRD50 treatments. Cell wall peroxidase in the epidermal layer of fruit may have a role in cessation of fruit expansion towards fruit maturity under reduced water availability.
INTRODUCTION
Reducing water use is important due to water scarcity and continuing huge demand from agriculture. Techniques to conserve water need to be explored by developing precision water management for sustainable water use in horticultural crop production. Partial rootzone drying (PRD) and control deficit irrigation are used as commercial practices to reduce water and fertilizer input in horticultural crop production. Partial rootzone drying to save water derived signals produced by the roots and transported to the leaves that alter growth and yield. Investigators attempted to establish responses of tomato (Lycopersicon esculentum Mill.) yield to PRD. Often there was no difference in yield response between PRD and well watered controls (Kirda et al., 2004; Stikic et al., 2003). However, Zegbe et al. (2003) found a significant reduction in fresh yield of tomatoes with PRD compared to a fully irrigated treatment. Fruit expansion rate can be an important determinant in yield responses to reduced water availability. Although yield per plant is not affected by reducing water added to plants in PRD; fruit expansion rate is sensitive to soil moisture depletion (Davies et al., 2000; Kirda et al., 2004). In retrospective, Andrews et al. (2002) indicated that cessation of fruit growth may reflect changes in the activity of enzymes that may rigidify cell walls in the fruit skin.
This research was undertaken to determine effects of PRD and control deficit irrigation on yield components, epidermal peroxidase activities, water relations and stomatal responses of tomatoes. The hypothesis was what differences exist by varying levels and method of root drying and that such differences are expressed in the yield components.
MATERIALS AND METHODS
Seeds of the fresh market tomato, cv. Ailsa Craig, were sown in 100 mm (0.37 L) volume ) pots filled with compost (John Innes No. 2, Arthur Bowers, Lincoln, UK) and seedlings were grown in the glasshouse at Lancaster Environment Centre, University of Lancaster, England. The experiment was conducted from October 2004-March 2005. The photoperiod in the glasshouse was adjusted to a 12 h photoperiod with metal halide lamps. Temperature was maintained between 25 and 18°C and relative humidity between 65 and 80%, day and night, respectively.
Seedlings with the first truss bud just visible were transferred to pots containing 5 L of John Innes No. 2 growing medium. During transplanting, roots were washed free from soil so that half of roots were in one pot and the other half was in a pot immediately adjacent. Half strength nutrient solution (Cooper, 1979) was added at 7 days after transplanting. Plants were: well-watered and maintained close to field capacity in both pots (control); CDI50 (half the amount of water in controls divided equally to both pots at each watering); PRD50 (half the amount of water in control applied to one pots while water withheld from the other pot until soil moisture declined to 50-70% of the field capacity at which point watering was provided to the other pot); PRD25 (half the amount of water in control applied to one pot while water was withheld from the other pot until soil moisture declined to 25-50% field capacity at which point watering was provided to the other pot) and CDI25 (quarter amount of water in control divided equally to both pots at each watering).
Changes in soil moisture content were monitored throughout the growing period using a soil moisture meter (HH2 Moisture Meter with Theta Probe type ML2X, Delta T-Devices, Cambridge, UK). Leaf water potential was measured on the youngest fully expanded leaf using a pressure chamber (Soil Moisture Equipment Corp, Santa Barbara, CA, USA). The amount of water added to treatments was adjusted depending on soil water probe readings, which were calibrated to the actual soil moisture content % (v/v) to indicate changes in soil water content.
Under PRD, switching watering regimes between left and right pots was based on depletion of soil moisture. Supplementary nutrient was added to the growing medium using a full strength of Cooper Formulation (Cooper, 1979) to all plants on days when watering of PRD treatments was switched to the other half of the root zone. The experiment was conducted in a completely randomized design with six replications with two plants/replication. Plants were grown to three leaves above the third fruit truss and side shoots removed to develop a single stemmed tomato plants.
Changes in fruit diameter were used as an indicator for fruit growth. Flowers from the first flower truss were tagged and labeled to determine fruit age after anthesis. Fruit diameter was determined with a caliper. Total yield per plant was assessed from the three fruit trusses developed on the plants. Numbers of fruit, final fruit diameter, dry matter concentration and fruit fresh weight were recorded at final harvest. All fruit were examined for incidence of blossom end rot (BER) and numbers of fruit affected recorded. Fruit at the same maturity stage were sampled, frozen with liquid nitrogen and stored at -20°C until analysis to determine osmotic potential of the fruit sap. The frozen fruit was thawed and centrifuged at 3,000 g for 5 min and sap was placed in 1.5 mL Eppendorf tubes. Osmotic potential of tomato sap was determine using an osmometer (5500 Vapour Pressure Osmometer, Wescor, Logan, Utah, USA).
At each sampling date of fruit growth, strips of fresh epidermis (100 mg) were removed using a clean razor blade, placed in a mortar and thoroughly ground in the presence of liquid nitrogen. The tissue was then removed and placed directly into another mortar containing 1 mL of 10 mM sodium succinate/calcium chloride buffer using the guaiacol method (Thompson et al., 1998). Absorbance from spectrophotometer was translated into activity for peroxidase using extinction co-efficient for the product formed (tetraguaiacol) which is 2660 m2 mol-1.
RESULTS AND DISCUSSION
Throughout the experimental period, soil water content remained relatively constant for the control treatment at approximately 0.25-0.32 g cm-3. Effects of alternating soil drying and re-watering on one side of the PRD treatments are shown in Fig. 1a and b. The decline in soil water content in the drying side of the PRD system was relatively rapid from about 0.28 g cm-3 to about 0.15 g cm-3 over
| Fig. 1a: | Changes in volumetric soil moisture content (10-1 g cm-3) of plants exposed to well watered (•), PRD50 left pot (� ), PRD50 right pot (� ) and CD150 (� ). Bars represent means ±SE |
| Fig. 1b: | Changes in volumetric soil moisture content (10-1 g cm-3) of plants exposed to well watered (•), PRD 25 left pot (� ), PRD 25 right pot (▼) and CD125 (� ). Bars represent means ±SE |
| Fig. 2: | Changes in leaf water potential (ψw) and stomatal conductance (g) of plants exposed to well watered (•), PRD50 (� ), CD150 (� ), PRD25 (▼) and CD125 (� ). Bars represent means ±SE |
the first 5 to 6 days. The well-watered side of the PRD system also experienced a significant decline in soil water content as plants were forced to absorb more water to compensate for the loss of water in the drying side of the split pots. Re-hydration of soil on the previously dry side of the PRD system took approximately 3 to 4 day. The soil water content of deficit irrigation remained relatively constant at approximately 0.15-2.0 g cm-3 and 0.06-0.15 g cm-3 in CDI50 and CDI25, respectively. Despite the decrease in soil water content in PRD50, for the first 24 days of drying and re-watering, there were no differences in leaf water potential between the well watered and PRD50 plants. Leaf water potential of plants in CDI treatments was consistently lower than well watered and PRD50 (Fig. 2a).
The basis underlying root signaling regulating plant growth was translated on the use of PRD in trials on yield responses in various crops species. In tomatoes, reports on yield comparison of plants grown with full irrigation and PRD or CDI accumulated with mixed results (Davies et al., 2000; Stikic et al., 2003; Zegbe et al., 2003, 2004; Kirda et al., 2004). Fruit fresh weight was significantly reduced in both PRD and CDI compared with well-watered plants, the reduction was less than that of the others. Consistent with the evidence from other deficit irrigation experiments (Mingo et al., 2003; O`Connell and Goodwin, 2007), fruit size of plants submitted to deficit irrigation was smaller than those of well-watered plants. We assumed that the reduction in fruit size placed a significant impact on reducing fresh weight yield in PRD and CDI treatments (Table 1, Fig. 3a). PRD50 produced larger fruits that all other deficit irrigation treatments. In water stress fruits, compensatory growth is one of the phenomenons of accelerated growth rate after removal of water stress resulted in greater fruit volume at harvest relative to the non-stressed fruits (Huang et al., 2000). Although fruit size of PRD50 never exceeded fruits from well watered plants, there was a tendency that PRD50 may experienced this compensatory growth better than other deficit irrigation treatments.
| Table 1: | Effects of different irrigation regimes on number of fruit, fruit diameter, fruit fresh weight, fruit dry matter, number of fruit with blossom end rot fruit (BER) and fruit osmotic potential |
| Zdata was not transformed. YMeans
with the same letter(s) within columns are not significantly different
using LSD at p≤0.05 |
|
| Fig. 3: | Changes in fruit diameter and peroxidase activities (PER) during exposure to well watered (•), PRD50 (� ), CD150 (� ), PRD25 (▼) and CD125 (� ). Bars represent means ±SE |
Tomato fruit growth followed a sigmoidal curve with rates of fruit expansion (Ho and Hewitt, 1986). The results demonstrated that the restriction of tomato fruit growth in both CDI and PRD plants at phase I and/or at transition phase 1 to phase 2 (phase of cell division and the transition to cell expansion) was likely not due to the changes in cell wall peroxidase. Cell water relations may play a fundamental role in the early stage of fruit expansion and turgor–driven expansion may affect expanding tomato fruit. Similar role of hydraulic resistance in early fruit development was reported by Van leperen et al. (2003) . Furthermore, Abassi et al. (1998) found a decrease in peroxidase activity with the onset of bud break and during early fruit development in apple. They reported that an increase in peroxidase activity only occurred at the later stage of fruit development. Although there were inconsistencies with regards to nature of isoenzymes and bioassay determinations for peroxidase activity, there was a tendency of elevation of peroxidase activity in epidermal layer of fruits grown in both PRD and CDI treatments in phase 2 (cell expansion) of fruit growth . Cell wall peroxidase in fruit epidermis in both PRD and CDI treatments increased when measured on day 16 after anthesis. Given that fruit growth depends mainly on the mechanical properties of the skin and that peroxidase may stiffen cell walls (Fry, 1986), this enzyme could be involved in the cessation of fruit growth at that stage of fruit development (Andrews et al., 2002). They also suggested that the late appearing isoenzymes were not associated with fruit ripening or softening and are probably not ethylene-induced; this peroxidase may act to control fruit growth by cross-linking wall polymers within the fruit skin, thus mechanically stiffening and terminating growth.
The similarity in percentage of fruit dry matter in response to both PRD and CDI may indicate a lower water influx into fruit tomato grown with reduced water availability. This lead to a higher percentage of fruits with blossom end rot in both treatments compared to those of control plants. The finding confirmed earlier reports that PRD50 did not enhance blossom end rot (Zegbe et al., 2004; Davies et al., 2000) for glasshouse trials, but control deficit irrigation aggravated this disorder in tomatoes. Saure (2001) indicated that BER seems to occur when amount of stress exceeds stress tolerance, most frequently in young fruit at the beginning of cell enlargement. It was suggested that mild stress may increase stress tolerance by temporarily promoting the formation of ABA which reduces GA activity and promotes Ca2+ import. This hypothesis may be applicable with plants grown in PRD50 which showed a lesser incidence of BER compared with other deficit irrigation treatments. Present data on BER follows a similar trend as observed in hot pepper (Capsicum annum L.) plants grown in PRD and CDI (Dorji et al., 2005). They assumed that localized irrigation of soil as in CDI leaves a greater proportion of rhizosphere to dry and greater reduction in water availability to fruit development. The resistance with to water influx in the xylem to the growing fruit tissue with CDI and PRD25 may restrict Ca influx which induces BER.
The application of PRD by withholding water from the half of the root system to decline to 50-70% of the field capacity (PRD50) at which point watering was reversed can be a feasible technique in saving water under condition of marginal water supply as an alternative to CDI. The potential of PRD as a management tool needs further refinement especially in relation to assimilate partitioning in fruit. PRD can be potentially used to create the phenomenon of compensatory fruit growth that can benefit tomato yield.
ACKNOWLEDGMENTS
Mohd Razi Ismail is affiliated with the Department of Crop Science, Faculty of Agriculture, University Putra Malaysia, 43400 UPM Serdang, SELANGOR, Malaysia. We thank Association Commonwealth University for providing grant to conduct research at Lancaster Environmental Centre, University of Lancaster, United Kingdom and to University Putra Malaysia for granting sabbatical leave to the first author. We also thank Professor W.J. Davies for advice on PRD treatments. Mrs. Maureen Harrison for plant preparation and maintenance.