Introduction
Worldwide, it is estimated that the total extraction of fresh surface water and groundwater for agricultural purposes is 69%, including irrigation, livestock and aquaculture (FAO, 2014). Around 50% of this water consumption corresponds to crop evapotranspiration (Kohli, et al., 2010) including fruit trees which are highly dependent on irrigation (Naor, 2006).
In Colombia, Boyacá is considered the most important area in production of deciduous fruit crops (pear, peach, plum and apple) About 5.382 ha were grown in 2010, which 31% relates to pear production. These crops continue their expansion, thus, it demands technological advances in the improvement of plant material, sanitary conditions, and irrigation techniques (Miranda et al., 2013).
The growers are conditioned by the available water that is scarce in some areas (De la Rosa et al., 2015), which makes it necessary to employ strategies to improve efficiency and optimize its use (Cui et al., 2009). Those strategies are based on the detection of plant response to water deficit (De Swaefet al., 2008) by means of indicators of plant and soil water status (water potential, sap flow, and trunk diameter variation) to determine the best irrigation scheduling (Rallo were 5.06%, 78.2 and 23.9, respectively and pH was 4.6 et al., 2017). (Molina et al., 2015).
Regulated deficit irrigation (RDI) has been widely used in deciduous and other fruit tree crops, such as citrus and grapevine (McCarthy, 2005) and consists of reducing water supply only in periods when fruit growth is less sensitive. The trees have shown various mechanisms of adaptation to water deficit, which usually imply osmotic adjustment, changes in tissue elastic properties, stress evasion by stomatal closure, and change in leaf area, amongst others (Torrecillas et al., 2001).
Several plant indicators have been considered as a tool for irrigation scheduling; one of the most used is the midday stem water potential (Ψ t ) (Intrigliolo and Castel, 2006). Based on this indicator, it is possible to know the plant water condition and its response to RDI. Techniques such as the Pressure-Volume (PV) curve are useful to determine if there is an osmotic adjustment due to water restriction (Mellishoet al., 2011).
The aim of this study was to evaluate the response of pear trees to water restriction and the influence of RDI during the rapid fruit growth stage on yield and quality of fruits in order to develop strategies to optimize the use of water on deciduous crops.
Materials and Methods
Location
The experiment was conducted from October2015 to April 2016, in an open field pear orchard (Pyrus communis L. cv. Triumph of Vienna) at the plot known as "Finca San Benito", located in Sesquilé, Cundinamarca (5°02' N and 73°47' W, elevation 2,595 m a.s.l.). Grafted pear trees of the cv. "Triumph of Vienna" with a well-developed root system were planted in 1998, in a spacing pattern of 4x4 m. All of them had similar management of pest and disease control, edaphic fertilization was applied twice a year and fertigation through the drip irrigation system every 15 d. The total amount applied was 60, 44, 100 k of N, and, respectively, using N32 liquid nitrogen (32%, N); 83% phosphoric acid (53%,); potassium nitrate (46% N, 13%). Additional applications of phosphorus, potassium and boron were performed.
The soil was identified as Histosol, with loamy and clay texture. The volumetric moisture at field capacity and at permanent wilting point was 26.9% and 15.3%, respectively. The content of organic matter, potassium and phosphorus were 5.06%, 78.2 and 23.9, respectively and pH was 4.6 (Molina et al., 2015).
Climate and Irrigation
The crop irrigation needs (ETc) were calculated according to Penman-Monteith method using a maximum crop coefficient (Kc = 0.8) at full canopy growth (Allen et al., 2006). The irrigation was applied through a drip system with two lines per row of trees and six 8 drippers per plant. Three treatments were applied from 60 to 140 d after full bloom (DAFB), corresponding to the rapid fruit growth stage (December 26, 2015 to March 9, 2016): (T1) Control in which the plants were irrigated at 100% of Etc, (T2) irrigation at 25% of T1 and (T3) no irrigation. During the rest of the season all plants were irrigated at 100% ETc. Weather conditions (temperature, relative humidity, precipitation, wind speed, solar radiation) were measured using an automatic weather station WS-GP1 (AT delta-T Devices Ltda., Cambridge, UK) located on the plot. The volume of irrigation water applied was measured with 12 mm volumetric counters Controlagua® (F.F. Soluciones S.A, Colombia). The accumulated volume of irrigation water applied during the treatments (60 to 140 DAFB) was 1460, 394, and 0 L per tree respectively in T1, T2, and T3.
Plant water relations
The midday stem water potential (Ψ t ) was measured every 8 d on two leaves of two trees in each replicate (16 leaves per treatment) selected from the inner part of the canopy, enclosed in an aluminum covered plastic bag during 90 min before the measurement with a pressure chamber (Model 600 Pressure Chamber Instrument, PMS Instrument Company, Oregon, USA) (Naor et al., 1995).
At mid-water restriction season (February 19, 2016), PV curves were performed according to the free transpiration technique (Tyree and Hammel, 1972) in order to determine the osmotic potential at the turgor loss point(Ψ oppt), modulus of elasticity (ε), apoplastic relative water content (RWC a ) and relative water content at the turgor loss point (%RWC ppt ) (Mellishoet al., 2011). To obtain the curve, 20 leaves were cut per replicate and the petioles were immersed in distilled water for 24 h. Once saturated, the fresh weight of each leaf was measured with a precision electronic balance 0.1 mg (Precisa XT2202A, Dietikon, Switzerland) and then the leaf water potential (Ψ) was determined using the Scholander pressure chamber, finally, leave fresh weight was measured again. The procedure was performed successively at regular intervals as the leaves were dehydrated under ambient conditions until the minimum weight loss was found (Corcuera, 2003), then leaves were oven dried at 60° C for 72 h to determine the dry weight.
To find the osmotic potential at full turgor (f 100) four saturated leaves were taken from each replicate and wrapped in aluminum foil and immediately frozen in liquid nitrogen in order to halt their metabolic activity (Abril, 2015). 72 h after, the tissue was macerated and centrifuged for 10 min at 10,000 rpm in order to extract the cellular fluids and determine f 100 using an Osmometer (Vapro®, Wescor, USA). The osmotic adjustment (∆O) was estimated as the difference between f100 of the plants of T2 f100 and T3 and of the T1 Control treatment (Ruiz et al., 2000).
Yield and quality
All the fruits at physiological maturity were harvested from all the trees. Two fruits were randomly picked by treatment and replicate (8 fruits per treatment), to which they were determined: The higher (a) and minor (d) equatorial diameter and length (L) using a manual calibrator; fresh weight of each fruit using a precision balance 0.1 mg (Precisa XT2202A, Dietikon, Switzerland); the volume from the displacement of distilled water in a graduated 1,000 mL container; the color of the pulp and the peel using a colorimeter Chromameter CR-400 (Konica Minolta®, Japan); titratable total acids (TTA) using Titroline® 6000 (SI Analytics, Japan) automatic titrator with 0.1 N sodium hydroxide (NaOH); total soluble solids (TSS) with manual optical refractometer (Kikuchi Precision Optics, Tokyo, Japan); maturity index (IM) as the relationship: IM = , (Rodríguez et al., 2010) and firmness using 2 mm test with CT3 TextureAnalyzer (Brookfield Engineering Labs., USA) in the equatorial direction and at the poles of each fruit.
Statistical analysis
The experiment was carried out in a randomized block design with four replicates (12 plots with 12 to 20 trees each). Measurements were taken on two adjacent trees per replicate, considering the edge effect. Data were analyzed with ANOVA, and means were compared using Duncan's test at P≤0.05 using the software InfoStat 2016 from InfoStat, FCA, Universidad Nacional de Córdoba, Argentina (Di Rienzo et al., 2016).
Results and Discussion
Climate and irrigation
During the study, the weather variables were measured. The minimum temperature, in January 2016 was 2.9°, and the maximum was 27.8°C, the mean was 14°C. The mean relative humidity was 81%, the highest was in April (98%) and the lowest was in January (60%). Mean ETo was 3.1 mm d-1, a 14% lower than the maximum ETo recorded in the year (3.6 mm d-1). The average vapor pressure deficit (VPD) was in the range of 0.62 (in January) to 0.03 (in April). The total precipitation during the restriction period was 218 mm (Fig. 1).
Plant water relations
The was influenced by precipitation. From 80 to 140 DAFB a significant difference was observed between T3 and other treatments according to Duncan's test (P≤0.05), indicating that Ψ t was a good indicator of the irrigation regimes applied. The Ψ t had the same values for all three treatments at 150 and 170 DAFB due to rainfall. During the treatments, there was an increase of Ψ t represented by more negative values in T3 (mean -1.17 MPa) than T2 (mean -0.9 MPa) and T1 (mean -0.9 MPa). T3 had a maximum of -1.78 MPa at 126 DAFB while there were no significant differences between T1 and T2, which means that the amount of water applied in T2 allowed to maintain water conditions of the plant similar to the Control (Fig. 2). Similar results in the same crop were reported by Abril (2015), with Ψ t between -0.3 MPa and -1.22 MPa.
In other studies in pear, Caspari et al. (1994) reported more negative values of in trees variety 'Hosui' under water restriction (-2.0 MPa to -1.7 MPa), while Morandi et al. (2014) found values of Ψ t between -0.6 MPa and -1.25 MPa for the 'AbbeFettel' variety, very similar to that reported by Marsal et al. (2002) (-0.5 MPa to -1.5 MPa) in pear 'Blanquilla' and Naor (2001) (-1.2 MPa and -3 MPa) in pear variety 'Spadona' under RDI treatments.
The Ψ t was slightly higher in comparison to other species such as apricot (-1.08 MPa to -1.87 MPa) (Pérez et al., 2014) and jujuba (-1 MPa and -4 MPa) (Cruz et al., 2012). On the other hand, Ψ t had a behavior similar to that reported by Samperio et al. (2015) in 'Red Beaut' plum (between -0.59 MPa and -0.70 MPa) and Podestá et al. (2010) who found a maximum Ψψ of -1.53 MPa in cherry trees under water deficit.
At 119 DAFB (59 d of water restriction, the day on which the PV curve was performed) there was a significant difference that showed osmotic adjustment in Ψ100 between T3 and T1 (-3.1and -2.53 MPa, respectively). The greatest difference respect to Control was observed in the treatment T3 (0.57 MPa) that coincides with that reported in the apricot tree by Ruizet al. (2000). The decrease in the osmotic potential can be attributed to the active solute accumulation on the leaf tissues which has been considered as a mechanism of osmotic adjustment in mature peach plants and depends on the species, variety, severity of water restriction, leaf maturity and the time in which the restriction was applied (Marsal and Girona, 1997; Cruz et al., 2012).
The other parameters derived from the PV curve did not present a significant difference according to the Duncan's test (P≤ 0.05) due to the treatments, similar to the results reported by Mellisho et al., 2011 in peaches under water restriction where were no differences respect the Control in the turgor loss point (Ψ ρρρ WRC ppt ), or in de modulus of elasticity (ε), however they found osmotic adjustment of 0.18 MPa, that is lower than the values founded on this study.
It is important to note that these plants were under RDI treatments in previous years, which may have been preconditioned or affected (Ruizet al., 2000) causing their water status to be maintained. It is also observed that the increase of in Ψ100 T3 could contribute to maintain the turgor in the plant as observed in the values of Ψ ορρt (-5.9 MPa) and ε (11.5 MPa) slightly higher for T3.
Yield and quality
The harvested fruits were classified into three size categories according to diameter: category I larger than 65 mm, category II between 65 and 50 mm and category III less than 50 mm. Fruit yield per tree (kg/tree) was higher for T1 (23.59 kg) without significant difference with T2 and T3 according to the Duncan's test (P≤0.05). The mean weight of the fruits was homogeneous between treatments with 160g, 170 g and 170 g for I, II and III respectively, it was similar to data found by Molina (2014) (140 g) and slightly lower than the reported by Abril, (2015) (200 g) in the same orchard. The number of fruits per tree varied between treatments without significant differences according to Duncan's test (P≤0.05). More fruits were obtained per tree in T1 (113 fruits) than in T2 (92 fruits) and T3 (102 fruits). The highest number of fruits corresponded to categories I and II in all treatments, with no significant differences, meaning that the majority of the fruit corresponded to fruits larger than 50 mm coinciding with Pérez et al. (2014); De la Rosa et al. (2015), who did not find significant differences in peach yield under moderate RDI treatments.
Although there was no significant difference in yield, there is a slight variation of T1 considering that the water status of the tree (Ψt) did not show significant differences between T1 and T2. The results suggest that the T2 represents water saving up to 75% during the rapid fruit growth stage (obtaining similar yield results).
As the plant water relations did not present significant differences between treatments, there was no difference in the quality parameters: volume, density, sphericity, color index (CI) of pulp and peel, firmness and TSS according to Duncan's test (P≤0.05) (Tab. 3) similar to that found by De la Rosa et al. (2015) in nectarines, indicating that the fruits in the three treatments present similar quality characteristics.
The sphericity of the fruits was 0.96 which is very close to one (1) indicating an almost spherical shape typical of the Triumph of Vienna pears what may be a desired attribute of quality for consumers. The titratable total acidity (TTA) and maturity index (MI) were the only parameters with significant difference according to the Duncan's test (P≤0.05) between Control and T2 and T3. Because the MI is derived from the relation between TSS and TTA, its variation is due to the reduction of malic acid quantity for T1. Parra et al. (1998) reported mean TSS values similar to those found (12.67 °Brix) indicating that as the fruit grows and develops in the plant, °Brix increases, while the acid content decreases. The MI had values higher than those reported by Arenas (2012) for the same species under normal irrigation conditions and lower than those obtained by Abril (2015) (41, 48 y 46 °Brix gL-1 for T1, T2 y T3 respectively) and Molina (2014) (61, 65 y 61 °Brix gL-1 for T1, T2 y T3 respectively).
Conclusions
Pear trees were resistance to the water deficit as they performed osmotic adjustment, but osmotic potential was similar in all treatments. The application of RDI did not affect the yield or quality of the fruits in relation to the well irrigated trees, indicating that a similar production can be achieved with savings of water up to 100% during the rapid fruit growth stage.