Effect of Deficit Irrigation on Agronomic and Physiological Performance of Young Persimmon (Diospyros kaki Thunb.) Trees

Abstract

This article addresses the impact of deficit irrigation on the agronomic and physiological performance of “Rojo Brillante” persimmon trees in a Mediterranean climate. It compares the effect of a sustained deficit irrigation (SDI; imposing water deficit uniformly throughout the entire crop cycle) strategy and two regulated deficit irrigation (RDI; enforcing a water deficit during the phenological phases that are less sensitive to water stress) strategies. Field trials were conducted from 2022 to 2024 at the Cajamar Experimental Center in Paiporta, Valencia, Spain. The trees respond to mild water stress reducing transpiration through stomatal closure. RDI resulted in modest irrigation water savings (11–16%), minimizing fruit drop, leading to an increased number of fruits per tree and a higher marketable yield, although this came at the cost of a reduced unit fruit weight. SDI achieved a 30% reduction in irrigation water usage without impacting on the marketable yield, but it also caused a decrease in unit fruit weight. RDI increased water productivity (yield obtained per amount of water applied) primarily through higher yields, while SDI improved productivity mainly by lowering the amount of irrigation water applied. Both irrigation strategies are recommended for cultivating “Rojo Brillante” persimmons. RDI is especially advisable in years with lower fruit loads as more intensive thinning may be necessary in years with higher fruit loads. Conversely, SDI is recommended in situations where water availability is limited.

1. Introduction

1.1. Water Scarcity and Deficit Irrigation

Freshwater scarcity poses a significant challenge for agriculture, which accounts for over 70% of the world’s freshwater use. This water scarcity is significant in the Mediterranean regions as summers are long, hot, and dry, with rainfall varying greatly from year to year and season to season. This variability highlights the importance of irrigation for crop production, emphasizing the need for strategies to optimize irrigation water usage.

Deficit irrigation (DI) involves providing crops with less water than needed for optimal growth and development [1,2,3]. It includes sustained (continuous) deficit irrigation (SDI) and regulated (controlled) deficit irrigation (RDI). SDI applies a uniform water deficit throughout the entire crop cycle, ensuring that the crops do not experience a significant water stress in any phenological phase. RDI targets specific phenological phases when crops are less sensitive to water stress, ensuring a balanced approach to water management [4].

1.2. Development and Importance of Persimmon Cultivation

The term persimmon commonly refers to Diospyros kaki Thunb., which is also known as the Japanese persimmon, Oriental persimmon, or Chinese persimmon. In French, it is called “kaki”, while in Spanish, it can be referred to as “caqui,” “kaki,” “palo santo,” or “palosanto.” In Italian, it is known as “cachi.” Additionally, both D. kaki and D. lotus are often called “loto” in Italy and Greece, deriving from the word “lotus” [5].

The genus Diospyros is part of the Lytrhaceae family and thrives in tropical and subtropical regions. Only three species—D. kaki, D. lotus, and D. virginiana—can adapt to temperate climates. Most of the persimmon varieties cultivated for their fruit belong to the species D. kaki, while D. lotus and D. virginiana are primarily used as rootstocks. A significant number of persimmon varieties originated from Japan, where extensive breeding programs are currently in progress. Additionally, many varieties have emerged from spontaneous mutations in the Mediterranean region, resulting in numerous local cultivars across different countries [6].

The persimmon tree has been known since ancient times in the Mediterranean basin but has historically been considered a minor fruit tree. In Spain, these trees have traditionally been grown as solitary specimens for personal consumption or in small family plantations aimed at local markets. Starting in the 1950s, small commercial plantations emerged featuring the most common varieties: “Tomatero”, “Picudo”, “Cristalino”, and “Xato de Bonrepòs” [5,6]. Among these, the “Rojo Brillante” (meaning “Bright Red”) variety appeared in the Valencian Region, though its exact origin is unclear. The prevailing theory suggests it originated from a mutation of the “Cristalino” variety. Initially known as “Rojo Gordo” (translated as “Fat Red” or “Big Red”) due to its large size, it was propagated by several farmers and cultivated alongside other varieties. In the 1970s, a study was conducted to select the best persimmon cultivars, leading to the recognition of “Rojo Gordo” as having the best characteristics. It was subsequently renamed “Rojo Brillante” [7]. Due to the importance achieved by this variety of ‘“Rojo Brillante” as well as its differential characteristics, in 1998, the quality figure Council Regulation (EC) “Kaki Ribera del Xúquer” was obtained, which was recognized in 2002 by the European Union [8].

Persimmon cultivars are classified based on the astringency of their fruits, which can be either non-astringent (NA) or astringent (A). Non-astringent fruits can be eaten immediately after harvesting, while astringent fruits require over-ripening or specific processes to remove their astringency before they can be consumed [9]. Additionally, both groups are further divided based on their response to pollination into two categories: varieties with constant pollination (PC) and those with variant pollination (PV).

“Rojo Brillante” is an astringent cultivar at harvest time with variant pollination (PVA). The trees are highly productive, require little or no thinning, and produce large, attractive fruits. These fruits are appealing in shape and color, offer a good flavor and aroma, and are easy to process to eliminate astringency [7,9,10,11].

The “Rojo Brillante” persimmon was traditionally sold when overripe, which eliminated its astringency and caused the pulp to soften, resulting in a gelatinous consistency. While this fruit was valued by a niche market, it posed significant challenges for post-harvest handling due to its low firmness, which restricted its commercial potential. In 1997, the Valencian Institute of Agricultural Research (IVIA) developed a post-harvest technique that utilized high concentrations of carbon dioxide. This method removed the astringency while preserving the fruit’s firmness. Similar technologies had already been applied to other persimmon cultivars in Israel and Japan. The adaptation of this technique for the “Rojo Brillante” variety allowed it to be exported to new markets, leading to an increase in the area cultivated in Spain in recent decades, particularly in the province of Valencia [6].

Until the 1980s, the total area of cultivated land for persimmons worldwide did not exceed 200,000 ha. However, since the 1990s, this area has been increasing annually, reaching approximately 950,000 ha by 2012. In 2020, 2021, and 2022, more than one million ha of persimmon cultivation were established. Before the early 1990s, global production was around one million tons. Since then, production has also risen significantly, similar to the increase in cultivated area, exceeding four million tons per year [12].

In recent years, China has been the largest producer of persimmons, responsible for approximately 70% of global production and over 90% of the cultivated area. Spain follows with an 8.3% share of world production and 1.7% of cultivated area [12,13]. In 2023, Spain cultivated 15,562 ha of persimmons. The cultivated area had been increasing until 2017, when it peaked at over 18,000 ha. However, this area has declined from 2020 to the present [13].

1.3. Deficit Irrigation Techniques in Persimon Cultivation

Previous research highlights the growing interest in persimmon cultivation, with studies conducted in Spain [14,15,16,17,18,19] and various countries, including Israel [20], China [21], and Japan [22,23,24,25]. Persimmon trees have long been considered to be salinity-sensitive plants [26,27]. More recently, Visconti et al. [28] found that “Rojo Brillante” grafted on D. lotus is highly sensitive to soil salinity. Lately, the same research team [29] observed that leaf necrosis in “Rojo Brillante” is caused by chloride, rather than other ions such as sodium. A particular focus has been placed on deficit irrigation methods explored in different studies [16,22,30,31,32,33].

Griñán et al. [33] found that persimmon plants employ stress avoidance mechanisms; when water availability is restricted, the leaf conductance and the duration of maximum stomatal opening decreased, which helps to prevent leaf turgor loss, maintaining the stem water potential within narrow limits. Additionally, the gradual recovery of leaf conductance observed after rewatering the plants can be seen as a mechanism to promote leaf rehydration.

Badal et al. [30] noted that the irrigation regime’s deficit resulted in less fruit drop compared to the Control group. They found a significant relationship between the amount of fruit that dropped during the physiological fruit drop period and the average stem water potential (Ψstem) during water restrictions. However, the timing of water stress also influenced these relationships. Early-season water stress had a more pronounced effect on reducing fruit abscission during the second wave of fruit drop.

Buesa et al. [31] reported that persimmon fruit growth is sensitive to water stress. RDI reduced final fruit weight, with a more significant reduction observed in the most stressed treatments. RDI allowed for water savings of up to 20% without reducing yield, thereby significantly enhancing water use efficiency. However, these treatments increased the number of fruits harvested; thus, the economic return was adversely affected by deficit irrigation. They concluded that further research is needed to define an RDI strategy that increases water use efficiency without compromising economic returns.

1.4. The Importance of Conducting Research in Local Conditions

The effective application of RDI depends on understanding the specific periods when a crop is most sensitive to irrigation deficits. This sensitivity varies from one crop to another based on their unique agronomic and physiological characteristics [34]. Therefore, results from one study cannot be directly applied to another location with different conditions as variations in outcomes have been observed with different irrigation systems and crop cultivars in diverse areas [35]. This highlights the need for research conducted under local conditions.

In this sense, the objective of this research was to evaluate the agronomic behavior of a young persimmon plantation using an SDI and two RDI strategies. Trees subjected to SDI received 70% of their irrigation water requirements (IWRs) throughout the growing season, while trees subjected to RDI1 and RDI2 received 60% and 40% of their IWR, respectively, during the physiological fruit drop period and 100% of their IWR outside this period. This study aimed to establish an irrigation schedule for growers, and it analyzes various factors including yield (both weight and number of fruits), fruit set, soil and tree water status, water productivity, quality characteristics of the fruit, and the occurrence of physiological disorders.

2. Materials and Methods

2.1. Experimental Site Description

The field studies were conducted at the Cajamar Experimental Centre in Paiporta, Valencia, Spain (39.4175 N, 0.4184 W), over three consecutive growing seasons: 2022, 2023, and 2024. The soil is deep and presents a uniform root zone depth of 50 cm and a silt loam texture. Following the USDA Soil Taxonomy, it is a Petrocalcic Calcixerepts soil. The soil is non-saline [EC 1:2.5 of 0.48 dS m⁻¹] and has a pH = 8.08 and an organic matter content of 1.3%; it contains phosphorus at 171.4 mg kg⁻¹ (Olsen method) and potassium at 581 mg kg⁻¹ (ammonium acetate). At field capacity, the averages volumetric soil water content (VSWC), measured at a depth of 25 cm, was 27.0% in 2022, 27.2% in 2023, and 26.9% in 2024. At the permanent wilting point, the VSWC was 18.08% for all three years.

Irrigation water poses no health risks, particularly for persimmon crops, except for potential issues related to salinity. The EC of the irrigation water used was 1.65 dS m⁻¹, and the EC of the saturated soil extract was 2.48 dS m⁻¹. According to the Maas–Hoffman formula, these data indicate a potential yield loss of approximately 16%. To mitigate this loss, the irrigation rate was increased by incorporating a 15% leaching requirement, determined according to the equation proposed by [36] for drip irrigation. Considering the contribution of this leaching requirement, no symptoms of salinity or chloride toxicity were observed.

2.2. Plant Material and Agronomic Details

The planting of the experimental field was carried out in 2019 with “Rojo Brillante” persimmon trees grafted onto D. lotus rootstock. This cultivar is highly regarded for its excellent adaptation to local conditions, high productivity, and wide market acceptance. The trees are arranged on raised beds that are 2 m wide and 0.2 m high, covered with polypropylene mesh. The planting pattern for the persimmon trees follows a spacing of 3 m × 5 m, and the trees are trained into a vase-shaped system.

Agricultural practices implemented in the orchard include covering the soil between the raised beds with perennial grass (Arbovert Perenne; 100 kg ha⁻¹; Intersemillas, Valencia, Spain), fertilization, and pruning techniques commonly used by local farmers. Nutrient incorporation is conducted through fertigation at a rate of 130-60-150 kg ha⁻¹ year⁻¹ for N–P₂O₅–K₂O, following guidance from [37].

2.3. Growth Stages

The duration of the growing season, defined as the period from the first leaf sprouting to harvest, was 258 days in 2022, 245 days in 2023, and 237 days in 2024. The initiation of the physiological phases for the three growing seasons is detailed in

Table 1. Dates of the initiation of the physiological phases and harvest for three growing seasons.

Growing Season	Leaves Sprouting	Anthesis	Fruit Setting	Fruitlet Drop	Harvest
2022	25-February	1-May	13-May	19-May	7-November
2023	5-March	20-April	27-April	1-May	30-October
2024	28-February	23-April	26-April	1-May	14-October

.

Table 2. Duration (d; days), reference evapotranspiration (ETo; mm), crop coefficient (Kc; range of values), and irrigation water applied (IWA; mm) during the different irrigation stages across the three growing seasons.

Strategy	Period	2022				2023				2024
		d	ETo	Kc	IWA	d	ETo	Kc	IWA	d	ETo	Kc	IWA
Control	-	258	1122	0.03–0.15	109	245	1165	0.06–0.37	188	237	1152	0.07–0.41	283
SDI	-	258	1122	0.03–0.15	74	245	1165	0.06–0.37	129	237	1152	0.07–0.41	198
RDI1	FI	83	226	0.03–0.07	2	58	251	0.06–0.18	21	63	241	0.07–0.20	17
	R	36	229	0.09–0.12	12	61	311	0.23–0.29	26	62	358	0.25–0.31	50
	FI	139	667	0.13–0.15	83	126	603	0.30–0.36	118	112	553	0.33–0.41	186
	-	258	1122	0.03–0.15	96	245	1165	0.06–0.36	165	237	1152	0.07–0.41	253
RDI2	FI	83	226	0.03–0.07	2	58	251	0.06–0.18	21	63	241	0.07–0.20	17
	R	36	229	0.09–0.12	8	61	311	0.23–0.29	18	62	358	0.25–0.31	33
	FI	139	667	0.13–0.15	83	126	603	0.30–0.36	118	112	553	0.33–0.41	186
	-	258	1122	0.03–0.15	92	245	1165	0.06–0.36	157	237	1152	0.07–0.41	235

FI: full irrigation; R: restriction period: (BBCH 71–77) from 19 May to 23 June in 2022 and from 1 May to 1 July in 2023 and 2024; -: entire cycle.

presents the duration, reference evapotranspiration (ETo), and crop coefficient (Kc) during the various irrigation stages across the three growing seasons.

2.4. Irrigation Strategies

Irrigation in each season and strategy began with the emergence of the first leaves (BBCH 10; [38]) on 25 February, 5 March, and 28 February in the years 2022, 2023, and 2024, respectively. It concluded with leaf drop (BBCH 93) on 9 November, 4 November, and 21 October in 2022, 2023, and 2024, respectively.

Irrigation for the Control strategy was conducted following the standard practices used by local growers [39]. This involved replacing the crop’s water needs while ensuring that soil moisture reached a depth of 50 cm, preventing water loss at that depth. For each irrigation event, SDI trees were irrigated with 70% of the water allocated to Control trees. For trees under RDI1 and RDI2, they were irrigated with 60% and 40% of their crop water needs, respectively, during the restriction period (Table 2). Outside this period, these trees were irrigated with 100% of their requirements.

The irrigation restriction phase for the RDI coincided with the physiological fruit drop period (BBCH 71–77), which occurred from 1 May to 1 July in 2023 and 2024. In 2022, this phase was from 19 May to 23 June due to a later anthesis, which resulted in nearly all fruits dropping by that date. Differential irrigation was initiated at the start of the 2022 growing season to avoid interfering with tree establishment.

For each event, the IWR was determined by means of the equation: IWR = (ETc − Pe)/Ef, where ETc (mm) represents crop evapotranspiration, Pe is effective precipitation (mm) determined from rainfall data using the US Bureau of Reclamation method [40], and Ef is irrigation efficiency. The Ef was estimated to be 0.825, considering both distribution uniformity (0.97, measured in situ) and a 15% leaching requirement (as mentioned in Section 2.1). Irrigation frequency varied from daily during peak water requirements in summer to every three days during lower requirements in spring.

The irrigation dose was determined to replenish the water needs from the prior period. ETc was calculated using the equation ETc = ETo × Kc [41], with Kc being the single crop coefficient, ranging from 0 to 0.72, as proposed by the Instituto Valenciano de Investigaciones Agrarias (IVIA; [42]) for the local conditions. The duration of each growth stage was adjusted according to the growing cycle. Daily ETo (mm day⁻¹) was determined using the FAO-56 Penman–Monteith equation.

$$\mathrm{ETo}={\displaystyle \frac{0.408 ∆\left({\mathrm{R}}_{\mathrm{n}}-\mathrm{G}\right)+\mathsf{\gamma }\frac{900}{\mathrm{T}+273}{\mathrm{u}}_2\left({\mathrm{e}}_{\mathrm{s}}-{\mathrm{e}}_{\mathrm{a}}\right)}{∆+\mathsf{\gamma }\left(1+0.34{\mathrm{u}}_2\right)}}$$

where Δ is the slope of the saturation vapor pressure–temperature relationship at mean air temperature (kPa °C⁻¹), Rn is the net radiation at the crop surface (MJ m⁻² d⁻¹), G is the soil heat flux (MJ m⁻² d⁻¹), T is the average air temperature (°C), u₂ is the wind speed at 2 m height (m s⁻¹), (es-ea) is the vapor pressure deficit (kPa), and γ is the psychrometric constant (kPa °C⁻¹).

The drip irrigation system consisted of two lateral lines (Netafim Standard Irrigation PE Pipes 16/4, Netafim, Valencia, Spain) per tree row. Each tree was irrigated with six self-compensating emitters (Netafim PCJ, 4.0 L h⁻¹) spaced 1.0 m apart. A water flow meter (MJ-LFC, Dn 22, NWM, Ningbo Water Meter Co., Ltd., Ningbo, China) recorded the amount of irrigation water applied for each strategy.

2.5. Volumetric Soil Water Content

Volumetric soil water content (VSWC; m³ m⁻³) was continuously monitored using ECH₂O EC-5 capacitance sensors connected to an Em50 data logger, managed through the ECH₂O Utility 1.6 software (Decagon Devices, Inc., Pullman, WA, USA). The sensors were factory calibrated (accuracy of ±3%).

Two sensors were placed per replicate, at depths of 0.25 m and 0.5 m below the emitter. The depth of 0.25 m corresponds to the depth where the maximum root density of persimmon trees occurred, as stated in previous experiments carried out at the same field. The sensor at 0.5 m confirmed that deep percolation was avoided.

VSWC was measured and recorded every 15 min. The variation in VSWC after irrigation was used to determine the in situ field capacity (FC). The VSWC data were presented as a percentage of FC, referred to as relative soil water content at 25 cm depth (RSWC), which minimized the impact of sensor calibration inaccuracies.

Irrigation for each strategy commenced when the RSWC in the Control group dropped to 90%. This criterion had previously been validated in preliminary studies.

2.6. Plant Water Status

To measure the midday stem water potential (Ψstem), a pressure chamber (Soil Moisture Equipment Corp. model 5100A, Santa Barbara, CA, USA) was used [31]. Mature, fully expanded leaves from the north-facing side and middle third of each tree were covered with silver foil and enclosed in plastic bags two hours prior to the measurements. Midday leaf conductance (gs) was measured using a porometer (Delta T AP4, Delta-T Devices, Cambridge, UK) on the abaxial surface of the leaves. Both measurements were conducted simultaneously, before the corresponding irrigation event, on twelve leaves per treatment, consisting of two leaves per tree, two trees per block, and three replication blocks. Between fourteen and sixteen measurements were taken each growing season, with each measurement occurring before the corresponding irrigation event.

2.7. Tree Performance Determinations

Tree size was described by the maximum height and canopy diameter in two perpendicular orientations (north–south and east–west) on the four trees per replicate at the end of each growing season (after leaf abscission). The trunk perimeter was measured at the beginning of the experiment and at the end of each growing season, at 0.2 m above the ground. The relative growth in these parameters was calculated as the ratio of the increase in each parameter to its initial value for each growing season.

The growth of the three branch types—(i) Spurs: Short branches that have a high number of mixed buds; (ii) Mixed-Fruit-Bearing Branches: Branches of variable vigor that contain mixed buds predominantly at their basal areas; (iii) Woody Branches: High-vigor branches with vertical growth that only bear woody buds [43,44]—was measured in 2024 using a tape measure on two branches of each type per tree. Dropped fruitlets per tree were collected and recorded weekly to help determine the number of flowers. Since flower drop was null, the number of flowers was considered identical to the number of fruitlets (including both dropped and harvested fruits).

Sampling for leaf nutrient analysis was conducted on 12 July of 2023 and 2024 [37,45], collecting thirty-two fully expanded leaves per treatment and replication block, 1.5 m above ground level, from branches with mixed-fruit-bearing branches [45], pooling them for analysis by treatment and replication block. Leaf analyses were performed in the Las Palmerillas laboratory in Cajamar, following the official analysis methods [46].

2.8. Chill and Heat Accumulation

For each season, chill accumulation was determined, from leaf abscission to bud break, by the chilling hours model [43,47,48]. Heat accumulation was assessed as growing degree hours (GDHs), from bud break to flowering, considering a base temperature of 4 °C, an optimum temperature of 25 °C, and a critical temperature of 36 °C, which is the temperature above which no significant growth occurs [49].

2.9. Yield, Water Productivity, and Fruit Assessments

The fruits were manually harvested on 7 November 2022, 30 October 2023, and 14 October 2024, when they reached their commercial maturity, which was determined by the criteria of the Council Regulation (EC) “Kaki Ribera del Xúquer”, which requires a color value of 4 on its color chart [50].

Total and marketable yields were measured in both weight and number, with any rejections visually assessed due to fruit alterations such as sunburn, scratches, deformations, and small size. Fruits were considered marketable only when their maximum equatorial diameter was at least 40 mm [51]. Water productivity was calculated as the ratio of marketable yield to IWA [4].

In 2022, the low yield obtained did not allow its analysis to be carried out. In the 2023 and 2024 harvests, 48 representative marketable fruits were selected for each treatment (four fruits per tree, four trees per block, and three replication blocks). The analyses were conducted at the Agronomy Laboratory at the Universitat Politècnica de Valencia. Maximum equatorial diameter (D) and length from calyx to style (L) were measured using a digital caliper (Powerfix Profi+, Ovibell GmbH & Co. KG, Mülheim an der Ruhr, Germany). Fruit was weighted with a precision balance (Mettler Toledo AB204–S, Mettler-Toledo S.A.E., Cornellà del Llobregat, Spain). Fruit size is evaluated based on either the maximum diameter of the equatorial section or the individual unit weight [51], both of which were assessed in this study.

The skin color coordinates (L*, a*, and b*) were measured using a Minolta CR-300 chroma meter (Konica Minolta Sensing Inc., Tokyo, Japan). Measurements were taken in triplicate at two equidistant points from the equatorial region of each fruit, and the average values were considered. The hue angle (h*) and color index were calculated using the equations h* = arctan (b*/a*) [52] and CI = 1000 × a*/(L* × b*), respectively [53].

Each fruit was divided into two equal parts; prior to this partition, one half was peeled and used to measure flesh firmness at six locations (three near the calyx and three in the equatorial section) using a penetrometer (FT-327; T.R. Turoni Srl, Forli, Italy) equipped with an 8 mm diameter probe. The average value was recorded and expressed in Newtons (N). After this, the fruit was divided, and the flesh of the peeled part was blended using a blender (Taurus Liquafruit Pro Compact, 600 W; Taurus, Oliana, Spain) to extract the juice [54]. The other half was weighed using an analytical balance and then dried at 65 °C in a forced air oven (Selecta 297; Barcelona, Spain) until reaching a constant weight to determine dry matter content.

The soluble solids content (SSC, measured in °Brix) of the juice was evaluated using an Atago digital refractometer (model No. 20; Atago, Bellevue, WA, USA) at a laboratory temperature (20 °C). The titratable acidity (TA) was assessed with an acid–base potentiometer (877 Titrino Plus; Metrohm Ion Analysis, Herisau, Switzerland), titrating the sample up to a pH of 8.1 using 0.1 N NaOH. The results were reported as g of malic acid L⁻¹. The maturity index (MI) was determined as the SSC/TA ratio.

At the time of harvest, all fruits were carefully inspected to identify any physiological disorders, such as sunburn, scratches, deformities, small size, and, when applicable, cracking. Fruits with blemishes were classified as non-marketable. The proportion of non-marketable fruit was presented as a percentage of the total number of marketable and non-marketable fruits, along with their distribution across various types of blemishes.

2.10. Experimental Design and Statistical Analysis

The experiment was designed as a randomized complete block design with three replicates. Each experimental plot consisted of four trees that were very similar in appearance (height, ground shaded area, and trunk cross-sectional; [55]).

The Statgraphics Centurion 19 software [56] was used to evaluate the results by analysis of variance (ANOVA). Percentage data were arcsine transformed, and the least significant difference (LSD) at a 0.05 probability level was used as the mean separation test.

3. Results and Discussion

3.1. Climatology

Figure 1 shows maximum and minimum temperatures, precipitation, ETo, and radiation for the three experimental seasons. The total ETo values throughout the season showed only a slight difference. However, this variation becomes more significant when focusing on ETo recorded during the initial phase of the crop cycle (from sprouting to the point when water restriction began in RDI). During this period, ETo values were 226 mm in 2022, 251 mm in 2023, and 241 mm in 2024. In terms of precipitation, 2022 experienced average rainfall with a total of 505 mm occurring in most of the rainfall in the spring. In contrast, 2023 was characterized by high temperatures and relatively low precipitation at 352 mm (half of this rainfall occurred in September). Meanwhile, 2024 was quite dry, recording only 226 mm of rainfall before harvest and 177 mm afterward. The effective precipitation (Pe) varied across the seasons, measuring 364 mm in 2022, 244 mm in 2023, and 130 mm in 2024 but also seasonally. Spring 2022 was particularly rainy, with 467 mm recorded from March through May. In contrast, the springs of 2023 and 2024 were rather dry. Additionally, rainfall in September was low in both 2022 and 2024, while 2023 saw a more substantial amount of 136 mm. There is a relationship between precipitation and recorded radiation. In 2022, radiation levels remained low until 5 May. In 2023, low radiation was observed during the last week of May and the first week of June. In 2024, low radiation occurred during the second half of March, the last week of April, and the second week of June (Figure 1).

3.2. Soil and Plant Water Status

The irrigation water applied (IWA) in the Control strategy for the crop cycle in 2024 was 283 mm (Table 2), which is 51% higher than the 188 mm used in 2023. This amount was in turn 73% greater than the 109 mm applied in 2022. The primary factors contributing to these increases were the rainfall recorded in September 2023 and the delayed start of irrigation in spring 2022 due to the rainfall during that period. The crop was three years old at the beginning of the study, and its water requirements were significantly lower than those of a mature crop. These requirements increase in parallel with plant size, which corresponds with the annual increase in crop coefficient (Kc) values. Among the established irrigation strategies, the SDI led to a 30% reduction in irrigation water compared to the Control strategy. In terms of the RDI, RDI1 and RDI2 achieved savings of 11% and 16%, respectively, when averaged over the three years.

Figure 2 illustrates the seasonal variations in relative soil water content (RSWC, the VSWC expressed as a percentage of FC), air vapor pressure deficit (VPD), stem water potential (Ψstem), and stomatal conductance (gs). The objective of initiating irrigation when the VSWC reached 90% of field capacity was successfully achieved. The VPD values ranged from 1.33 to 5.26 kPa. In 2022, the maximum annual value reached 4.65 kPa on 23 June; in 2023, it increased to 5.26 kPa on 23 August; and in 2024, it was 3.29 kPa on 31 July.

The average values recorded for RSWC, Ψstem, and gs in the SDI strategy were 84.8%, −1.18 MPa, and 330 mmol m⁻² s⁻¹, respectively, before the evaluated irrigation events. In contrast, the values for the Control strategy were notably higher, with RSWC at 92.1%, Ψstem at −0.99 MPa, and gs at 370 mmol m⁻² s⁻¹. Midday Ψstem declined slightly as the season progressed, which aligns with findings from previous studies on persimmon [15,22,30,31] and other fruit crops such as pomegranate [57]. During the water restriction phases, the values associated with RDI were lower than those of SDI, as expected due to the higher level of water restriction, particularly in the RDI2 treatment (76.6%, −1.47 MPa, 293 mmol m⁻² s⁻¹). These values were close to those of the Control strategy for the remainder of the crop cycle (88.7%, −1.04 MPa, 352 mmol m⁻² s⁻¹). In both RDI strategies, a rapid recovery of Ψstem was observed when full irrigation was resumed, consistent with observations from previous studies [30,31].

Midday Ψstem values recorded in previous studies on “Rojo Brillante” Control plants ranged from −0.4 to −1.1 MPa [30,31]. The values obtained in this study are considered extreme as they were measured just before irrigation. Consequently, the Ψstem of the Control trees indicates an adequate water status for the plants [15]. In the case of SDI, the minimum midday Ψstem value recorded is higher (less negative) than the lowest values found in comparable studies [15,22,30,31]. This suggests that the water stress experienced by these plants was less severe, indicating a mild level of stress. On the other hand, the Ψstem values for RDI trees indicate a moderate level of stress [15].

The gs decreased as the season progressed (Figure 2), as reported by Griñán et al. [33], who stated gs values oscillated between 335 and 387 mmol m⁻² s⁻¹ for the Control trees and between 311 and 344 mmol m⁻² s⁻¹ for the SDI trees. During the water restriction period, the RDI trees exhibited lower gs values compared to the SDI ones, dropping to 278 mmol m⁻² s⁻¹ for RDI1 and 247 mmol m⁻² s⁻¹ for RDI2. This decline was due to more severe water restrictions, with contributions of only 60%, 40%, and 70% of water needs for RDI1, RDI2, and SDI, respectively. In the initial and final phases of the season, the RDI trees behaved similarly to the Control plants. However, the recovery of gs values was slower than that of Ψstem. According to the literature, gs tends to respond less quickly than Ψstem to water stress [58]. The gs values observed in this study were higher than those reported by [32], which ranged from 250 to 300 mmol m⁻² s⁻¹ for Control trees and decreased to 100 mmol m⁻² s⁻¹ for SDI plants. Similarly, Ballester et al. [59] recorded an average gs of 151 mmol m⁻² s⁻¹ for Control trees and 111 mmol m⁻² s⁻¹ for plants experiencing water stress. It was observed (Figure 3) that both Ψstem and gs decrease as the RSWC decreases. Significant positive linear correlations (p ≤ 0.01) were found between Ψstem and gs with RSWC when considering all four strategies across the three growing seasons. The correlation coefficients were 0.94 for Ψstem and 0.82 for gs.

Additionally, a significant positive linear correlation (p ≤ 0.01; r = 0.88) was observed between gs and Ψstem (Figure 4). This suggests that “Rojo Brillante” trees effectively manage their water status by reducing transpiration through stomatal closure, consistent with findings from [60]. These trees maintain Ψstem within narrow limits during periods of water restriction. Thus, the “Rojo Brillante” persimmon trees exhibit a typical anisohydric behavior, as noted by previous studies [61]. Griñán et al. [32] reported that persimmon trees respond to mild water stress by developing a stress avoidance mechanism. When water availability is restricted, the trees reduce gs to control water loss through transpiration and prevent leaf turgor loss. Correlations have been also found between Ψstem and gs with VDP; however, these correlations have a lower value of r than those obtained for RSWC. This lower correlation may be due to two main factors: first, edaphic-based stress is more significant than atmospheric stress, as noted by [59] and [31], who pointed out that persimmons are not highly sensitive to VPD. Second, there is considerable tree-to-tree variability in Ψstem and gs, along with notable heterogeneity in leaf water status and stomatal conductance. This variability arises from differences in hydraulic resistance among various parts of the tree [62]. Figure 2 and Figure 3 illustrate the tendency towards anisohydric behavior in persimmons, which is further supported by the slope shown in Figure 4, according to [60].

3.3. Tree Performance

As expected, all three size-related parameters (canopy height and diameter and trunk perimeter) showed significant increments (p ≤ 0.01;

Table 3. Effect of growing season and irrigation strategy on size [canopy height (H; cm) and diameter (D; cm) and trunk perimeter (TP; cm)].

	H		D		TP
Growing season (GS)
2022	228.9	c	83.2	c	19.3	c
2023	283.2	b	132.6	b	22.0	b
2024	315.2	a	163.8	a	23.7	a
Irrigation strategy (IS)
Control	295.1	a	126.8		22.6	a
RDI1	266.2	b	126.3		21.7	b
RDI2	272.0	b	126.6		21.0	c
SDI	269.8	b	126.5		21.3	bc
GS × IS
2022—Control	241.7		83.5		19.9
2022—RDI1	221.5		83.2		19.4
2022—RDI2	222.8		81.5		19.2
2022—SDI	229.6		84.7		18.9
2023—Control	305.8		132.8		23.2
2023—RDI1	272.5		128.3		22.1
2023—RDI2	280.6		136.3		21.1
2023—SDI	273.9		133.0		21.5
2024—Control	337.8		163.9		24.8
2024—RDI1	304.5		167.4		23.6
2024—RDI2	312.6		162.2		22.8
2024—SDI	305.9		161.8		23.4
Source (df)	% sum of squares
GS (2)	60.39	**	90.18	**	58.08	**
IS (3)	6.14	**	0.00	ns	7.01	**
GS × IS (6)	0.58	ns	0.39	ns	1.05	ns
Residual (132)	32.89		9.43		33.86
Standard deviation	27.47		11.21		1.42

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. df: degrees of freedom; ns: no significant difference; **: significant differences at p ≤ 0.01.

) over three years, with the average height of the trees rising from 171 cm to 338 cm for the Control trees. The irrigation strategy affected canopy size, with all three deficit irrigation strategies resulting in reduced canopy height (p ≤ 0.01) and trunk perimeter (p ≤ 0.01) compared to the Control. DI did not significantly affect (p ≤ 0.05) canopy diameter. Additionally, no significant differences (p ≤ 0.05) were observed in relative growth for any of the analyzed parameters. Canopy height, and consequently relative growth, could have been influenced by pruning, although it was executed using consistent criteria across all cases. These findings align with the limited published studies on this topic in the area. For instance, Buesa et al. [31] examined the potential impact of various DI strategies on trunk perimeter growth and found no significant effects.

3.4. Flowering

The total number of flowers (fruitlets) was affected by the growing season (p ≤ 0.01;

Table 4. Effect of growing season and irrigation strategy on the total number of fruitlets per tree and their distribution (%) between dropped fruitlets and harvested fruits.

	Total Fruitlets		Dropped Fruits		Harvested Fruits
Growing season (GS)
2022	79.00	c	98.74	a	1.17	c
2023	212.25	b	83.01	b	16.99	b
2024	311.04	a	43.34	c	56.66	a
Irrigation strategy (IS)
Control	188.22		80.06	a	19.94	c
RDI1	225.39		75.38	b	24.62	b
RDI2	188.58		68.28	c	31.72	a
SDI	200.86		76.40	ab	23.49	bc
GS × IS
2022—Control	84.25		98.69	a	1.32	f
2022—RDI1	88.67		98.73	a	1.27	f
2022—RDI2	66.50		98.68	a	1.32	f
2022—SDI	76.58		98.87	a	0.78	f
2023—Control	188.00		86.49	b	13.51	e
2023—RDI1	249.50		84.72	b	15.28	e
2023—RDI2	206.17		76.93	c	23.07	d
2023—SDI	205.33		83.90	b	16.10	e
2024—Control	292.42		55.02	d	44.99	c
2024—RDI1	338.00		42.69	e	57.32	b
2024—RDI2	293.08		29.23	f	70.77	a
2024—SDI	320.67		46.43	e	53.57	b
Source (df)	% sum of squares
GS (2)	63.38	**	85.34	**	85.38	**
IS (3)	1.60	ns	2.86	**	2.87	**
GS × IS (6)	0.64	ns	2.35	**	2.34	**
Residual (132)	34.39		9.44		9.41
Standard deviation	73.15		8.10		8.09

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. df: degrees of freedom; ns: no significant difference; **: significant differences at p ≤ 0.01.

). In 2024, trees produced an average of 311 fruits tree⁻¹, which was an increase compared to 212 fruits tree⁻¹ in 2023. This production level was also much higher than the 79 fruits tree⁻¹ recorded in 2022, which aligns with expectations due to the increase in canopy size as trees mature (Table 3).

Flowering is triggered by both endogenous and exogenous signals [63]. Exogenous signals include factors such as photoperiod, temperature, and stress, while endogenous signals include aspects as plant age and fruit load, as well as nutritional and hormonal status. In this experiment, the photoperiod was consistent across the three seasons observed.

Lang et al. [64] defined dormancy as the temporary suspension of visible growth in any plant structure containing a meristem. They distinguished three phases: para-, endo- and ecodormancy. The paradormancy refers to growth suppression caused by other parts of the tree (such as apical dominance) due to the influence of “inhibitory molecules”. During the endodormancy phase, growth is not possible even under appropriate temperature conditions. This is because the buds require a period of low temperatures (chilling requirements) during the winter rest. After this endodormancy phase is completed, the buds must then be exposed to warmer temperatures during the eco-dormancy phase (heat needs) to initiate flowering.

Chill accumulations were 416, 468, and 450 chill hours (with temperatures above 7.2 °C) in 2022, 2023 and 2024, respectively. In all cases, these quantities exceeded the threshold considered necessary for astringent cultivars, considered between 200 and 400 h, and particularly for “Rojo Brillante”, which requires relatively little chilling hours, typically less than 200 h [6,43,48,65]. The slight variations in chill accumulation recorded in this study do not account for differences in flowering intensity. The delay in bud break during 2023 resulted in delayed flowering, which decreased the GDH value to 12,907, compared to 14,672 in 2022 and 14,572 in 2024.

Plant stress is the last exogenous signal noted by [63], but no stress was recorded during bud dormancy or after this period. The only variable that differed was the irrigation treatment, which did not significantly affect (p ≤ 0.05; Table 4) the number of fruitlets (flowers) produced.

Concerning endogenous signals, a high number of fruits negatively affects flowering induction/differentiation [66]. However, if natural fruit drop occurs in immature fruits 30 days after full bloom (as occurred in the present study), or if the thinning of immature fruits takes place during this period, it can prevent fruit alternation. In 2022, the total yield from the Control trees was much lower at 0.34 kg tree⁻¹, compared to 11.1 kg tree⁻¹ in 2023, and even much lower than 35.8 kg tree⁻¹ in 2024. Thus, the fruit load on the trees each season was independent of subsequent flowering, indicating that there was no biennial or alternate bearing, as reported [67]. All trees involved in the study received the same fertilization treatments. As will be shown afterwards (

Table 5. Leaf macronutrient (%) and micronutrient (mg L⁻¹) concentrations (12 July 2023 and 2024) from persimmon fruits “Rojo Brillante”.

	Macronutrients										Micronutrients
	N		P		K		Ca		Mg		Fe		Cu		Mn		Zn
Growing season (GS)
2023	1.81	b	0.21	a	1.50	b	0.73	b	0.35	b	72.58	b	4.67	b	154.25	b	23.25	b
2024	2.09	a	0.15	b	1.84	a	1.86	a	0.83	a	77.08	a	9.25	a	322.33	a	57.58	a
Irrigation strategy (IS)
Control	1.97		0.19		1.67		1.39		0.57		70.83		7.17		250.67		39.33
RDI1	1.92		0.18		1.68		1.34		0.57		76.67		6.67		252.50		43.50
RDI2	1.93		0.16		1.64		1.41		0.55		66.67		7.00		232.33		42.83
SDI	1.98		0.18		1.70		1.05		0.69		85.17		7.00		217.67		36.00
GS × IS
2023—Control	1.80		0.21		1.47		0.73		0.41		64.00		4.67		168.33		24.00	e
2023—RDI1	1.81		0.22		1.50		0.71		0.40		76.33		4.33		158.33		24.67	e
2023—RDI2	1.80		0.19		1.52		0.81		0.32		55.33		5.00		150.33		22.33	d
2023—SDI	1.82		0.22		1.52		0.66		0.28		94.67		4.67		140.00		22.00	e
2024—Control	2.15		0.16		1.86		2.05		0.72		77.67		9.67		333.00		54.67	c
2024—RDI1	2.03		0.15		1.85		1.96		0.73		77.00		9.00		346.67		62.33	b
2024—RDI2	2.05		0.14		1.76		2.00		0.78		78.00		9.00		314.33		63.33	a
2024—SDI	2.15		0.15		1.88		1.45		1.10		75.67		9.33		295.33		50.00	b
Source (df)	% sufm of squares
GS (1)	69.81	**	64.40	**	68.35	**	79.65	**	48.83	**	1.60	ns	85.77	**	82.67	**	86.00	**
IS (3)	2.64	ns	4.76	ns	1.16	ns	5.06	ns	2.51	ns	15.23	ns	0.54	ns	2.39	ns	2.63	ns
GS × IS (3)	2.24	ns	2.87	ns	1.92	ns	2.61	ns	8.83	ns	19.37	ns	0.54	ns	0.44	ns	1.99	ns
Residual (16)	25.30		27.97		28.57		12.68		39.83		63.80		13.16		14.51		9.38
SD	0.11		0.02		0.13		0.28		0.27		17.41		1.10		43.12		6.94

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. df: degrees of freedom; ns: no significant difference; **: significant differences at p ≤ 0.01.

), the nutrient content in the leaves is within normal ranges, indicating that it is unlikely to be the cause of the varying flowering intensities observed between growing seasons. The age of the trees ranged from 3 to 5 years during the study, leading to increases in both height (37.7%) and diameter (96.9%). This growth enhanced their flowering capacity but also resulted in a decrease in their overall vigor. It is evident that the age of the plantation is the primary endogenous factor contributing to the different flowering intensities observed across the three growing seasons.

3.5. Physiological Fruit Drop

Traditionally, physiological fruit drop has been considered as a self-regulatory mechanism that adjusts the number of fruits to match the tree’s capacity for metabolite supply [68]. This interpretation is supported by a correlation between physiological fruit drop and carbohydrate levels in leaves and fruitlets observed at the end of the physiological fruit drop period in trees where competition was altered by changing the number of fruits [69]. Numerous studies have demonstrated that ethylene induces abscission in various tree species, including persimmon [70]. Ethylene is recognized as the primary hormonal factor in abscission, especially under various stress conditions. There is a close relationship between ethylene levels and abscission; likewise, inhibitors of ethylene biosynthesis can prevent ethylene’s effects and reduce fruit drop in several species [70].

The percentage of dropped fruitlets, and, consequently, the percentage of harvested fruits, was significantly affected (p ≤ 0.01) by both the growing season and the irrigation strategy, as well as their interaction (Table 4). Specifically, the number of dropped fruitlets decreased with the increasing age of the trees. The highest drop rates were observed in the Control trees, while the lowest drop rates were found in the RDI2 trees. In 2022, fruit dropping was particularly pronounced, primarily due to the excessive vigor of young trees (three years old) and the unfavorable weather conditions during spring. Indeed, an important abscission of fruitlets occurred in spring (p ≤ 0.01; 98.7% on average), which resulted in a lack of fruits maturing for harvest, making it unviable to assess the impact of irrigation strategies on yield. This notable physiological drop in fruit production was widespread throughout the region that season, resulting in an average production reduction of 70% reported in the Valencian Community (Spain). In 2023, fruit abscission remained considerably high (averaging 83%), albeit lower (p ≤ 0.01) than the previous year. This drop was also associated with the vigor of young trees (then four years old) and varying weather conditions, including rainfall, low radiation, and cooler temperatures recorded in May. In 2024, when the trees were five years old entering their adult phase [44], the registered climate data were close to average values. As a result, the rate of fruitlet abscission was significantly lower, averaging 43% (p ≤ 0.01). These results are similar to those reported by [71] for their control treatment of “Rojo Brillante” and those by [72] for “Fuyu” (non-astringent persimmon). Interaction analysis (p ≤ 0.01) indicated that the RDI2 strategy resulted in the lowest percentage of fruit drop during the 2023 and 2024 growing seasons, leading to the highest percentage of fruit harvested. In 2022, as expected due to the large number of dropped fruitlets, there were no significant differences observed.

This result aligns with findings from other researchers: Intrigliolo et al. [73] reported that applying a 50% water restriction strategy during spring led to a lower fruit drop, necessitating more thinning compared to Control trees and those subjected to the same water restriction during the summer or autumn months. Similarly, [31] found that implementing a 50% water restriction in June and July reduced fruit drop, suggesting that moderate stress can decrease fruit drop as water stress increases. They also identified a correlation between the tree’s water status (measured by the average Ψstem during the water restriction period) and fruit drop in two of the three years studied. Furthermore, Buesa et al. [31] observed that the same water restriction applied in spring and, particularly in summer, resulted in reduced fruit drop and an increased number of harvested fruits compared to the Control treatment. However, George et al. [74] highlight that, in studies conducted in Japan, severe water stress, specifically at leaf water potentials below −1.8 MPa, has been shown to increase fruitlet drop. It is believed that fruitlet abscission occurs due to heightened ethylene production, which is triggered by the blockage of photosynthate transport [70]. They reported that the transport of photosynthates, along with water to the fruit, helps inhibit fruit drop by preventing the induction of ethylene synthesis in young persimmon fruits. The apparent contradiction between these findings may arise from the fact that [70,74] focus on severe water stress, whereas the present study, as well as studies by [30,31,73], addresses moderate or mild water stress conditions.

The physiological fruit drop observed in this study aligns with the early drop described by [75]. During this early fruitlet drop, an abscission layer develops between the peduncle and the parent branch [74,75]. At this stage, the persimmon pedicel detaches along with the young fruit [76]. The young fruits fall from the connection points between the stalk and the persimmon calyx because the vascular bundles are present only in the stalk, not in the calyx [76]. The late fruit drop typically occurs from mid-August to September, during which an abscission layer forms between the peduncle and the flesh [75]. However, this late drop has not been recorded in our experiments.

There are two types of physiological fruit drop that have been linked to various causes [75]. For seedless cultivars like “Rojo Brillante”, key factors include insufficient accumulation of assimilates due to low sunlight and excessive rainfall, reduced root activity caused by excessive soil moisture, and inadequate nutrition for all the fruits when there is a high fruit set. In March 2022, the maximum temperature was significantly below average, coinciding with a period of heavy rainfall, with up to 342 mm recorded in March. This substantial rainfall contributed to reduced solar radiation, which was about 50% of the average solar radiation value. Consequently, this decline in solar radiation may have led to decreased photosynthesis, resulting in a lower supply of carbohydrates for the developing fruits, which could promote their abscission. Although to a lesser extent, the reduced solar radiation and higher rainfall compared to average levels continued into April 2022. In contrast, in March and April 2023, maximum temperatures were higher than average. During these months, rainfall was negligible (Figure 1), leading to sunny days that doubled the solar radiation recorded in March 2022. However, in May 2023, rainfall occurred, which lowered temperatures and radiation. In 2024, weather conditions returned closer to average, resulting in normal levels of fruitlet drop.

The RSWC in the Control strategy ranged from 90% to 100% of filed capacity (Figure 2). Consequently, there was never any excess soil moisture. Regarding the nutritional status of the trees, all received the same fertilization treatment. The nutrient content of the leaves (Table 5) fell within normal ranges [37,45]. Significant differences (p ≤ 0.01; Table 5) were observed between the growing seasons, with higher nutrient levels recorded in 2024 compared to 2023, except for phosphorus (P). These differences between the two growing seasons align with findings reported by [45]. Additionally, Ref. [77] correlated an unusual and significant fruit drop of the “Loutian” persimmon tree in Luotian County, China, with low levels of available P in the soil, specifically noting values of 5.80 mg kg⁻¹. However, at the experimental station where the study was conducted, the level of available P in the soil was high, exceeding 170 mg kg⁻¹. The P content in the leaves ranged from 0.14% to 0.22%, which is slightly above the range considered normal (0.08% to 0.14%) by experts for interpreting foliar analysis, as well as the maximum level observed in high-yield orchards [37,45]. Therefore, it does not appear to be the cause of the fruit drop between the growing seasons.

A positive linear correlation (p ≤ 0.01; r = 0.74; Figure 5) was found between the number of dropped fruitlets and the total number of fruitlets over the three years of the study. However, due to variations in the age and size of the trees, as well as differences in weather conditions during each growing season, separate analyses were conducted for each growing season. Coefficients for 2022 (r = 1.00) and 2023 (r = 0.99) were highlighted, given that in those years fruit drop was notably high, with rates of 98.7% and 83.0%, respectively. A high rate of fruit set indicates significant competition for photoassimilates among the fruits; when fruit set is abundant and nutritional resources are insufficient for all fruits, it leads to substantial fruitlet drop [75]. This physiological fruit drop is understood as a self-regulatory mechanism that adjusts the number of fruits to remain within the tree’s capacity to supply necessary metabolites [68].

Fruitlet drop may not have a single cause. Reig et al. [71] found a peak in ethylene production at anthesis, and they propose that this increase in ethylene at that stage likely is a cause to physiological fruitlet abscission as there is no shortage of hormones or carbohydrates present at that time. The elevated ethylene production is linked to increased abscission in both vegetative and reproductive organs [71]. However, Gómez-Cadenas et al. [78] suggest that the abscission following anthesis, in the transition from ovary to fruit, is primarily hormonal in nature and is largely driven by the action of gibberellic acid (GA), which starts the process of fruit growth. The subsequent development of fruit requires nutrient availability, particularly the supply of carbon, as has been indicated for certain crops [78].

Figure 6 illustrates the seasonal variations in the weekly rates of fruitlet drop. The physiological fruitlet drop observed in this study displayed two waves, which is typical for seedless fruits [30,74,75]. This phenomenon generally occurs from anthesis to early July [74,75], coinciding with maximum shoot growth and starch depletion [79]. In 2022, there was only one wave of fruit drop, with nearly all fruit (98.7%) falling during this single wave. The onset of fruitlet drop began 8 to 18 days after anthesis, occurring mainly over a 35-day period in 2022, when only one wave was present. In contrast, during 2023 and 2024, the fruitlet drop extended over 60 days, reflecting the presence of two waves. The delay in fruit abscission relative to anthesis is attributed to the time required for the synthesis and secretion of hydrolytic enzymes induced by ethylene [71].

Figure 7 illustrates the seasonal variations in fruit growth during the 2023 and 2024 growing seasons. The data show that “Rojo Brillante” persimmon fruits exhibit a sigmoid growth curve in both seasons, as indicated by [31]. Therefore, fruit diameter growth does not stop at any point during growth, which does occur in fruits with a double sigmoid growth curve. Water restriction during the growth stop period in fruits with double sigmoid growth is crucial to ensure optimal growth and prevent unit weight loss, but in fruits with a sigmoid growth curve, the determination of the ideal timing for the application of RDI is more complicated [31].

The fruits harvested in 2024 were smaller than those from 2023. This difference is attributed to the significantly higher number of fruits collected in 2024 compared to 2023. No significant differences in growth were observed among the different irrigation strategies. However, during the fruit maturation phase, noted in the final measurements, the Control group exhibited a higher value compared to the deficit irrigation (DI) strategies, as shown in

Table 6. Effect of the growing season and the irrigation strategy on the total and marketable yield, unit weight (UW; g), size [diameter of the equatorial section (D; mm)], and shape index (D/L, being L the length, mm) of the marketable fruits and water productivity (WP).

	Total Yield				Marketable Yield										WP
	(kg tree−1)		(No. tree−1)		(kg tree−1)		(No. tree−1)		UW (g)		D		D/L		(kg m−3)
Growing season (GS)
2023	12.67	b	34.77	b	12.35	b	33.85	b	379.76	a	87.52	a	0.99		1.95	b
2024	37.84	a	174.13	a	36.47	a	166.40	a	228.56	b	70.19	b	0.98		6.27	a
Irrigation strategy (IS)
Control	21.80	c	77.17	c	21.29	c	75.13	c	337.57	a	80.75	a	0.96		3.34	b
RDI1	27.28	ab	116.75	ab	26.43	ab	112.50	ab	293.94	b	78.49	b	1.00		4.21	a
RDI2	28.22	a	125.46	a	26.73	a	117.00	a	286.89	b	78.23	b	0.99		4.53	a
SDI	23.72	bc	98.42	b	23.19	bc	95.88	b	298.22	b	77.95	b	0.99		4.36	a
GS × IS
2023—Control	9.65		24.92	d	9.53		24.58	d	405.16		88.10		0.97		1.43
2023—RDI1	13.74		38.25	d	13.31		37.00	d	368.41		87.83		1.03		2.03
2023—RDI2	15.97		43.75	d	15.41		42.08	d	372.70		87.31		0.99		2.42
2023—SDI	11.32		32.17	d	11.15		31.75	d	372.77		86.83		1.00		1.90
2024—Control	33.96		129.42	c	33.05		125.67	c	269.99		73.40		0.96		5.24
2024—RDI1	40.81		195.25	a	39.55		188.00	a	219.46		69.15		0.98		6.39
2024—RDI2	40.47		207.17	a	38.05		191.92	a	201.09		69.15		0.99		6.64
2024—SDI	36.12		164.67	b	35.23		160.00	b	223.68		69.06		0.98		6.82
Source (df)	% sum of squares
GS (1)	78.54	**	76.04	**	78.44	**	77.00	**	80.09	**	79.08	**	2.10	ns	79.42	**
IS (3)	3.36	**	5.38	**	2.79	**	4.74	**	5.44	**	1.30	**	4.79	ns	3.60	**
GS × IS (3)	0.15	ns	2.11	*	0.24	ns	1.81	*	0.60	ns	0.63	ns	2.72	ns	0.66	ns
Residual (88 y/376 z)	17.94		16.47		18.53		16.45		13.86		18.99		90.39		16.32
Standard deviation	6.28		33.87		6.12		32.00		32.85		4.34		0.06		1.02

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. Degrees of freedom (df): z for D and D/L; y for TY, MY, UW, and WP. ns: no significant difference; ** (*): significant differences at p ≤ 0.01 (p ≤ 0.05).

(p ≤ 0.01).

In the 2024 growing season, shoot growth was measured on all three branch types (spurs, mixed-fruit-bearing, and woody branches; Figure 8). Based on visual observation rather than an accurate count, it can be stated that the majority of the harvested fruit came from the spurs, while a smaller amount came from branches with mixed fruit. Both mixed-fruit-bearing and woody branches exhibited strong growth from 3 May to 13 May. This growth likely reduced the carbohydrate supply to the fruitlets, potentially causing stress that could lead to increased ethylene biosynthesis in the fruit and subsequent fruit drop. Sun et al. [70] noted that persimmon fruit does not produce ethylene while still attached to healthy trees. However, when the transport of photosynthates from the parent tree is blocked, ethylene synthesis begins in the fruit. This process can lead to fruit drop, which may explain the highest weekly rate of fruitlet drop observed between 3 May and 13 May.

3.6. Yield and Fruit Characteristics

Both total and marketable yields were affected by the growing season and irrigation strategy (p ≤ 0.01; Table 6). Specifically, the marketable yield in 2024 reached 36.5 kg tree⁻¹, which is three times greater than the 12.4 kg tree⁻¹ recorded in 2023. The average yield in 2024 aligns with the average reported for that season by commercial plantations [80]. However, it is lower than the yields reported in other seasons by [30,31] for the same “Rojo Brillante” variety, which were from plantations that were eight years old at the start of their studies. Given that the plantation in question was only five years old in 2024, the yield can be considered normal [44].

In terms of irrigation strategies, the RDI2 strategy produced significantly higher total and commercial yields compared to the Control strategy, with p-values lower than 0.01. Specifically, the RDI2 strategy yielded 12.9% and 62.6% more in kg and in the total number of fruits, respectively. This improvement is largely attributed to a lower physiological fruit drop observed in the RDI2 strategy. While the fruit drop was also reduced in the RDI1 strategy, it was not as pronounced as in RDI2 (Table 4).

Due to the varying number of total fruits harvested per tree in different growing seasons (on average 34.8 in 2023 and 174.1 in 2024), the average unit weight of the marketable fruits obtained in 2024 was significantly lower (p ≤ 0.01) at 228.6 g compared to the 2023 harvest, which averaged 379.8 g. Additionally, the unit weight of fruits from the Control strategy was higher (p ≤ 0.01; 337.6 g) than that of the deficit irrigation (DI) strategies, which averaged 293.0 g. This difference is related to the number of marketable fruits harvested, with an average of 75.1 in the Control strategy compared to 108.5 in the DI strategies, since the greater the number of fruits per tree, the greater competition between them for carbohydrates [81].

In terms of size, as expected (given that there is a significant positive linear relationship between average diameter (D) and the UW; D = 0.1105x + 45.244, r: 0.99, p ≤ 0.01), the fruits harvested in 2023 were larger (p ≤ 0.01; D = 87.5 mm) than those harvested in 2024 (D = 70.2 mm). Furthermore, the Control fruits exhibited a greater equatorial diameter (p ≤ 0.01; average D = 80.8 mm) compared to fruits from the deficit irrigation strategies (average D = 78.2 mm). A quadratic relationship was established between the diameter (Y; mm) and the number of fruits harvested per tree (X), as shown in Figure 9. This finding is consistent with results obtained by [81] in peach trees. According to Agustí et al. [81], this inverse relationship is likely due to competition among fruits for carbohydrates, which reduces their availability when the number of fruits is excessively high.

The growing season and irrigation strategy did not significantly affect the shape of the fruit (Table 6). All the fruits maintained a similar shape, with the shape index (calculated as D/L, L denotes the length from the calyx to the base, both in mm) ranging from 0.96 to 1.03.

In 2024, the average WP (Table 6) was higher (p ≤ 0.01) than in 2023, with values of 6.27 kg m⁻³ in 2024 compared to 1.96 kg m⁻³ in 2023. Additionally, the WP achieved with DI strategies was also higher (p ≤ 0.01; averaging 4.37 kg m⁻³) than that of the Control strategy, which had an average of 3.34 kg m⁻³. The differences between the growing seasons can be attributed to the substantially higher marketable yield in 2024 compared to 2023, with yields of 12.4 kg tree⁻¹ in 2023 and 36.5 kg tree⁻¹ in 2024. Regarding irrigation strategies, the Control strategy resulted in both the lowest marketable yield (9.5 kg tree⁻¹ in 2023 and 33.1 kg tree⁻¹ in 2024) and the highest IWA, which was 188 mm in 2023 and 283 mm in 2024. These findings align with the results reported by [31], who observed higher WP with deficit irrigation strategies than with the Control, noting no significant differences among the DI treatments. However, Intrigliolo et al. [73] found that water productivity only increased when water restrictions were applied during the latter part of the growing season.

Various indices have been utilized to assess the skin color of persimmons, based on the CIE L*C*h* color space parameters L*, a*, and b* (or from Hunter L, a, and b coordinates). These indices include the a/b ratio [50,82]; Chroma [18,83,84,85]; hue angle [18,83,84,85,86,87]; color index (CI), which is likely the most commonly used index for both “Rojo Brillante” [10,18,88,89,90,91,92] and other cultivars [86]. Additionally, color charts such as the Japanese standard color chart [93] and the chart used by the Council Regulation “Kaki Ribera del Xúquer” are also employed in evaluations. Asakuma and Shiraishi [87] discovered a strong positive correlation between hue angle and the color chart. For this reason, the hue angle and CI values are presented in

Table 7. Effect of growing season and irrigation strategies on skin color [hue angle and color index (CI)], dry matter content (DM; %), firmness (N) of flesh, soluble solids content (SSC, °Brix), titratable acidity (TA, g malic acid L⁻¹) of flesh juice, and maturity index (MI).

	Hue Angle		CI		DM		Firmness		SSC		TA		MI
Growing season (GS)
2023	74.66	a	4.40	b	18.84		46.32	a	18.10	a	0.25	a	71.79	b
2024	68.84	b	6.59	a	18.19		40.85	b	17.16	b	0.22	b	78.61	a
Irrigation strategy (IS)
Control	72.74	a	5.13	c	17.18	b	44.61	a	17.46	b	0.24		73.77
RDI1	72.32	ab	5.26	bc	18.14	b	44.81	a	17.57	b	0.23		75.80
RDI2	71.47	b	5.65	ab	19.56	a	42.59	b	17.82	a	0.24		75.59
SDI	70.46	c	5.93	a	19.17	a	42.33	b	17.69	ab	0.23		75.65
GS × IS
2023—Control	74.04	b	4.59	c	16.89	c	45.98	a	18.19	a	0.26		71.11
2023—RDI1	74.38	b	4.48	c	17.88	bc	46.35	a	18.12	a	0.25		72.04
2023—RDI2	75.85	a	4.06	c	20.67	a	46.49	a	18.04	a	0.25		72.31
2023—SDI	74.35	b	4.47	c	19.91	a	46.44	a	18.07	a	0.25		71.72
2024—Control	71.45	c	5.68	b	17.47	bc	43.23	b	16.72	e	0.23		76.44
2024—RDI1	70.26	c	6.04	b	18.40	b	43.26	b	17.02	d	0.22		79.56
2024—RDI2	67.09	d	7.24	a	18.44	b	38.70	c	17.59	b	0.23		78.88
2024—SDI	66.56	d	7.40	a	18.44	b	38.22	c	17.31	c	0.22		79.58
Source (df)	% sum of squares
GS (1)	24.16	**	18.30	**	1.47	ns	16.43	**	59.56	**	27.76	**	13.39	**
IS (3)	2.19	**	1.53	**	12.06	**	3.12	*	4.74	**	1.33	ns	0.79	ns
GS × IS (3)	4.61	**	3.00	**	5.28	**	11.40	**	9.81	**	0.93	ns	0.28	ns
Residual (376)	69.03		77.16		81.19		69.05		25.89		69.98		85.54
Standard deviation	4.94		2.26		2.46		4.74		0.33		0.02		9.00

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. ns: no significant difference; ** (*): significant differences at p ≤ 0.01 (p ≤ 0.05).

.

Significant differences (p ≤ 0.01) were observed between the color indices of the fruits harvested in the two growing seasons. Fruits from 2024 exhibited a lower hue angle (68.8, indicating they were redder) and a higher CI value (6.6, indicating they were more orange) compared to those harvested in 2023, which had hue angles of 74.7 and CI values of 4.4, respectively. Regarding irrigation strategies, the SDI, and to a lesser extent the RDI2, resulted in the lowest (p ≤ 0.01) average hue angle (70.5, indicating a redder color) and the highest average CI value (5.9, indicating a more orange color). The analysis of the interaction between both factors (p ≤ 0.01) shows that the impact of the irrigation strategy was more pronounced in 2024 than in 2023. Specifically, fruits associated with the SDI strategy and RDI2, which imposed more intense water restrictions than RDI1, exhibited the lowest hue angle values (ranging from 66.6 to 67.1, indicating they were more reddish) and the highest CI values (ranging from 7.4 to 7.2, indicating they were more orange). These CI values are consistent with those reported by [10,94] for “Rojo Brillante”. These results suggest that the fruits harvested in 2024, particularly those from the SDI and RDI2 strategies, were more mature than those harvested in 2023.

The ripening of persimmon fruit involves a change in color from yellow to deep orange or red, passing through light orange. This color change results from the degradation of chlorophyll and the biosynthesis of carotenoids [95]. Lutein is the primary carotenoid present in the skin during the green stage, while β-cryptoxanthin and zeaxanthin accumulate during the coloring stage [83,96]. Plaza et al. [83] found that the main carotenoids present in the fruit of the “Rojo Brillante” persimmon were β-cryptoxanthin and β-carotene, along with lutein and lycopene, and that the lycopene content increased with ripening, particularly when the fruit showed a reddish-orange color.

The levels of secondary metabolites, including carotenoids, can be influenced by different factors (temperature, water availability, ultraviolet radiation, nutrient levels, atmospheric pollution, mechanical stress, pathogen attack, cultivar, ripening stage, and, when applicable, de-astringency treatment) that may vary seasonally [83,97]. Variations in these factors can lead to differences in skin color, which may have occurred in this study. It should be noted that, following the Council Regulation “Kaki de la Ribera del Xúquer” recommendations, the fruits should have been harvested when the color chart value was 4. The yield obtained in this study was intended for commercial use, and, according to the marketing company, the harvest was initially scheduled for Friday, October 11th. However, due to company issues, it was postponed until Monday, October 14th, resulting in a higher level of ripeness.

The irrigation strategy and its interaction with the growing season influenced the flesh dry matter content (DM) of the fruits (p ≤ 0.01). Fruits from the most severe DI strategies, namely, RDI2 and SDI, exhibited higher DM values of 19.6% and 19.2%, respectively, compared to those from the Control and moderate deficit irrigation strategy (RDI1), which had DM values of 17.2% and 18.1%, respectively. Interaction analysis indicated that these differences were significant (p ≤ 0.01) in 2023.

Estimating the DM of fruits is essential for optimizing harvest times, particularly for climacteric fruits that continue to ripen after harvest. These fruits must be harvested with a minimum DM content to ensure adequate starch accumulation and the desired post-harvest sugar content. If harvested too early, the fruits may not achieve good quality later on. Conversely, if the fruits ripen excessively, they may become overly soft, complicating transportation and storage.

The firmness values of the fruits ranged from 38.2 N for SDI fruits harvested in 2024 to 46.0 N for Control fruits harvested in 2023. These values align with those reported by [98] and are considered optimal for marketing. Changes in firmness, and consequently texture, during fruit ripening are mainly the result of the degradation of the primary cell wall [98]. While there is no universally accepted minimum degree of firmness, values below 10 N after storage and marketing are generally considered commercially inadequate [98]. The lowest firmness values correspond to the ripest fruits, which have the highest CI. A negative linear correlation was found between firmness and the CI (p ≤ 0.01; r = −0.63; Figure 10), which is very close to the correlations reported by [98] for the same cultivar. Although the correlation coefficient is not particularly high, as indicated by [90], the relationship between firmness and color varies depending on the harvest date and treatment. The equation provided was derived from fruits harvested on different dates in 2023 and 2024. This correlation supports the validity of the CI as a criterion for determining harvest timing.

Growing season, irrigation strategy, and their interaction affected the soluble solids content (SSC) of persimmon fruit flesh (p ≤ 0.01; Table 7). Fruits harvested in 2023 had a higher average SSC, 18.1° Brix, compared to those harvested in 2024, which averaged 17.2° Brix. These values are notably higher than those reported by [18] for the same cultivar, which ranged between 14.7 and 15.9° Brix across two maturity stages. Among the irrigation strategies, RDI2 produced a higher average SSC, of 17.8° Brix (p ≤ 0.01), compared to RDI1 and Control, which averaged 17.6° and 17.5° Brix, respectively. Analyzing the interaction further (p ≤ 0.01), there were no significant differences between the irrigation strategies in 2023; however, in 2024, the RDI2 strategy resulted in a higher SSC value, of 17.6° Brix, than the SDI strategy, which averaged 17.3° Brix, and both of these values exceeded those of the RDI1 and Control strategies, averaging 17.0° and 16.7° Brix, respectively.

The differences observed in SSC, both in comparison to the study conducted by [99] with the same cultivar (ranging between 14.7 and 15.9 °Brix in two maturity stages) and across different growing seasons (with an average difference of 0.94° Brix), are important as an increase of just 1° Brix is considered a notable enhancement in the perception of the fruit’s flavor [100]. The predominant sugars in the flesh of the “Rojo Brillante” persimmon are sucrose (a non-reducing sugar), glucose, and fructose (reducing sugars), in that order [101]. This pattern is typical for PVA cultivars. Notably, sucrose hydrolyzes into fructose and glucose as the fruit ripens [50].

Titratable acidity was influenced only by the growing season, not by the irrigation strategy. This may be due to the significant variability indicated by the high residual sum of squares (70.0%). The main organic acids present in persimmon fruits are malic acid, followed by citric and fumaric acids [75,102].

Acidity measurements provide a reliable estimate of acidity intensity, but SSC is not a good predictor of the sweetness perceived by consumers, particularly in astringent persimmon cultivars, because in addition to sugars and organic acids, SSC includes soluble tannins [101]. The presence of both acidity and tannin content in astringent fruits, where soluble tannin levels remain high, complicates the relationship between SSC and sweetness perception. The maturity index (MI) interprets SSC in conjunction with acidity as a de-astringency treatment can eliminate astringency before consumption. The MI ratio balances SSC and acidity values, showing similar values in different strategies. The MI was only influenced by the growing season (p ≤ 0.01), being the highest MI that obtained in 2024 (78.6) compared to 2023 fruits (71.8). Notably, the same harvesting criteria were applied in both seasons. In 2024, the harvest date was advanced by 16 days compared to 2023, although it was still delayed by three days compared to the initially proposed date.

During the ripening process of fleshy fruits, a series of physical, biochemical, and physiological changes occur. These changes include alterations in color, flavor, texture, and the production of aroma, organic acids, and polyphenols [103]. Our study found that the application of DI resulted in an increase in color and a decrease in fruit firmness, suggesting that its effects on fruit flavor and aroma should not be overlooked. While fruit flavor is associated with non-volatile compounds, aroma is linked to volatile compounds [103], particularly flavonoids [104,105], where researchers have examined the chemical characteristics of various persimmon cultivars, including Fuyu, Hachiya, Chocolate, and Sharon. Their analysis revealed unique combinations of compounds, such as alcohols, aldehydes, esters, ketones, and terpenes for each cultivar. Consequently, it would be interesting to analyze, using liquid chromatography–mass spectrometric analysis, the potential effects of DI strategies on these components in the “Rojo Brillante” cultivar.

3.7. Fruit Physiological Disorders

Non-marketable yield (

Table 8. Effect of growing season and irrigation strategy on non-marketable yield (% yield by weight) and its partitioning into sunburned, scratched, deformed, and small fruits.

	Total		Sunburned		Scratched		Deformed		Small Fruits
Growing season (GS)
2023	2.32	b	0.20	b	0.00	b	2.12	a	0.00	b
2024	3.52	a	0.62	a	1.56	a	0.34	b	0.99	a
Irrigation strategy (IS)
Control	1.82	b	0.42		0.71		0.62		0.07	b
RDI1	2.94	b	0.46		0.61		1.39		0.48	b
RDI2	4.69	a	0.35		1.19		1.88		1.26	a
SDI	2.24	b	0.42		0.61		1.03		0.17	b
GS × IS
2023—Control	1.08		0.00		0.00		1.08		0.00	b
2023—RDI1	2.71		0.39		0.00		2.32		0.00	b
2023—RDI2	3.45		0.15		0.00		3.30		0.00	b
2023—SDI	2.06		0.27		0.00		1.78		0.00	b
2024—Control	2.56		0.84		1.41		0.17		0.14	b
2024—RDI1	3.16		0.53		1.22		0.46		0.95	b
2024—RDI2	5.92		0.54		2.39		0.46		2.53	a
2024—SDI	2.43		0.57		1.23		0.27		0.35	b
Source (df)	% sum of squares
GS (1)	4.37	*	6.08	*	43.63	**	13.25	**	12.25	**
IS (3)	14.68	**	0.24	ns	4.20	ns	3.59	ns	10.93	**
GS × IS (3)	2.26	ns	2.34	ns	4.20	ns	2.06	ns	10.93	**
Residual (88)	78.68		91.33		47.98		81.10		65.89
Standard deviation	2.65		0.85		0.86		2.30		1.20

Mean values followed by different lowercase letters in each column indicate significant differences at p ≤ 0.05 using the LSD test. df: degrees of freedom; ns: no significant difference; ** (*): significant differences at p ≤ 0.01 (p ≤ 0.05).

), although generally low, was significantly affected (p ≤ 0.01) by growing season, irrigation strategy, and their interaction. The highest percentage was recorded in 2024, with an overall non-marketable yield of 3.52%, compared to 2.32% in 2023. This increase was observed in each section of non-marketable yield, with the exception of deformed fruits, which accounted for 2.12% in 2023 and only 0.34% in 2024.

The irrigation strategy had an impact on the non-marketable yield, with the RDI2 strategy resulting in the highest average percentage of 4.69% (p ≤ 0.01). Notably, the incidence of small fruits was highlighted, averaging 1.26% and peaking at 2.53% in 2024 (p ≤ 0.01). This increase in the percentage of small fruits can be attributed to heightened competition for photoassimilates among the fruits. This competition arose due to the greater number of fruits harvested in 2024 under the RDI2 strategy, which totaled 207, compared to the other irrigation strategies, where the numbers ranged from 129 to 195.

Machuca et al. [106] indicate that malformations in fruits can include asymmetry, double fruits, or cracking, which can affect up to 30% of the yield. High temperatures and severe water stress have been identified as contributing factors to the development of these disorders. These authors hypothesize that the deformations are due to the disruption of the early stages of fruit development. In the present experiment, higher temperatures (and VPD) were recorded in March–April and July–August (the two critical periods associated with the incidence of these deformities [106]) in 2023 than in 2024 (Figure 2)). The water stress levels achieved with the tested strategies were moderate or mild, so their impact on fruit deformation was likely not statistically significant. However, in 2023, the Control strategy resulted in the lowest percentage of deformed fruits, at 1.1%, while the RDI2 strategy, which involved the most significant water restriction from 1 May to 1 July, resulted in the highest percentage of deformed fruits, at 3.3%.

The incidence of sunburned fruit was negligible (<1%) during the 2023 season despite the high summer temperatures (with a maximum of 46.8 °C and radiation levels reaching up to 28.4 MJ m⁻²). This was likely because the fruits were protected from direct sunlight by the foliage.

Although persimmons are known to be very sensitive to wind—since the fruits can rub against leaves and branches, leading to spots on the skin that reduce their commercial value [65]—the incidence of scratches was non-existent in 2023. Additionally, the irrigation strategy did not significantly affect this outcome as the average scratch incidence was low at 1.56% in 2024. During both growing seasons, wind speeds were moderate, not exceeding 10 km h⁻¹ in 2024 and 8 km h⁻¹ in 2023.

4. Conclusions

As soil moisture decreases, both stem water potential and leaf conductance decline. This indicates that young “Rojo Brillante” persimmon trees effectively manage their water status by closing their stomas, reducing, therefore, their transpiration, and maintaining their stem water potential within narrow limits during periods of water stress. Both stem water potential and leaf conductance recover after the period of water stress, indicating no long-term physiological damage.

The regulated deficit irrigation (RDI) strategies tested resulted in water savings of 11% to 16% of the total irrigation water requirement. These strategies reduced fruit drop, increasing the number of fruits per tree and improving the marketable yield, although they did lead to a decrease in fruit size. The sustained deficit irrigation (SDI) achieved water savings of 30%, without affecting marketable yield, but also resulted in a reduction in fruit size.

Both RDI2 and SDI strategies resulted in earlier fruit ripening. The fruits exhibited a redder color and a more orange appearance, along with lower flesh firmness and higher soluble solid content. These changes could potentially allow for earlier harvest dates. RDI2 resulted in a higher non-marketable yield, a consequence of the higher incidence of small fruits harvested per tree, due to the lower fruit drop.

In terms of water productivity, RDI strategies increased mainly due to higher yields, while SDI improved it primarily by reducing the amount of irrigation water applied. Therefore, all irrigation strategies are recommended for growing “Rojo Brillante” persimmons. RDI strategies are always recommended, while the SDI strategy is recommended when water availability is limited.