Improving Photosynthetic Characteristics and Enhancing Yield and Quality of Nanfeng Tangerine via Deep Irrigation in Red Soil Hilly Regions

Abstract

Seasonal drought and poor soil water retention are key constraints to citrus production in China’s red soil hilly regions. This study investigated the effects of deep-layer irrigation on photosynthetic characteristics, yield, and fruit quality in a typical Nanfeng Tangerine orchard in Jiangxi. A factorial experiment with three irrigation depths (D1: 25 cm, D2: 50 cm, D3: 100 cm) and three water levels (W1: severe deficit, W2: mild deficit, W3: full irrigation) was conducted. The D2W2 treatment was identified as optimal. It significantly enhanced the net photosynthetic rate by 88.14% and improved instantaneous water use efficiency by 25.93% compared to the poorest-performing treatments. Furthermore, D2W2 achieved the highest yield per plant (58.13 kg) and superior fruit quality (soluble solids: 34.37 °Brix; titratable acidity: 0.46%; sugar–acid ratio: 15.93), a result corroborated by its top TOPSIS score (0.95). In conclusion, deep-layer irrigation at 50 cm combined with a mild water deficit is the recommended strategy for synchronizing water conservation, yield increase, and quality improvement in Nanfeng honey tangerine orchard.

1. Introduction

As the most widely produced fruit in the world, citrus is extensively cultivated across Asia, North America, and Europe [1]. In 2019, the citrus yield in southern China reached 43 million tons [2]. Among these regions, red soil hilly areas constitute one of the major citrus-growing zones in southern China, where annual precipitation is abundant but exhibits pronounced seasonal unevenness [3]. Soils in this region exhibit typical red soil properties, with pH values generally ranging from 4.72 to 5.86, indicating acidic conditions. The soil texture is predominantly sandy loam, with sand contents ranging from 60.66% to 68.85%. This poor soil water retention capacity severely constrains citrus production. Moreover, recurrent extreme weather events exacerbate water scarcity, leading to reductions in stomatal conductance, transpiration, and photosynthetic capacity of citrus leaves, thereby exerting adverse impacts on plant growth and development [4,5]. Citrus is characterized as a water-demanding crop [6], and under the context of global climate change, the increasing frequency of seasonal droughts and shortage of agricultural water resources, coupled with the poor water retention capacity of red soils [7], have emerged as major limiting factors for yield improvement and quality enhancement. Consequently, the development of efficient irrigation technologies is urgently required to mitigate the adverse effects of seasonal droughts and to promote the sustainable and high-quality development of the citrus industry.

In order to ensure citrus yield, various water-saving irrigation methods have been gradually promoted in recent years, such as drip irrigation [8] and sprinkler irrigation [9]. However, drip irrigation systems are costly. They require extensive pipelines and filtration equipment, and have emitters that are prone to clogging when filtration is inadequate or maintenance is insufficient. Conversely, sprinkler irrigation may cause leaf damage due to evaporative cooling under low-temperature and windy conditions [10]. Moreover, these techniques are predominantly applied at the soil surface or shallow subsurface (irrigation depth < 30 cm), where water evaporates rapidly from the upper soil layer and deep percolation losses are difficult to avoid, resulting in considerable room for improvement in overall water use efficiency (WUE). Deep irrigation, by contrast, involves burying emitters directly within the crop root zone, allowing irrigation water to infiltrate slowly and continuously into the rhizosphere. This technique not only satisfies crop water requirements but also effectively reduces soil surface evaporation and deep percolation losses, thereby improving WUE. Furthermore, deep irrigation helps maintain relatively stable soil moisture conditions in the root zone, creating a more favorable water environment for plant growth. This approach is particularly suitable for regions facing water scarcity, poor soil water-holding capacity, and high evaporative demand.

Previous studies have demonstrated the advantages of deep irrigation in a wide range of crops. For example, Valentín et al. [11] reported that compared with sprinkler systems, deep irrigation reduced seasonal evapotranspiration by 39%, with comparable reductions in both evaporation and transpiration components. Similarly, Umair et al. [12] found that, relative to traditional surface flooding, deep irrigation in winter wheat decreased evapotranspiration by 26%, increased net photosynthesis by 10%, enhanced intrinsic WUE by approximately 36%, and reduced the transpiration rate by about 22%; compared with surface drip irrigation, deep irrigation further reduced evapotranspiration by approximately 15%. Moreover, suitable deep irrigation depths have been shown to optimize root distribution, enhance root activity, promote nutrient uptake, and improve both crop yield and WUE [13,14]. For instance, Moura et al. [15] reported that the optimal fruit quality of fig was achieved under irrigation depths corresponding to 85.19–95.16% of ETc., and Wang et al. [16] suggested that subsurface drip irrigation at a depth of 40 cm significantly improved the water use efficiency of pear trees. Also, Ahmed Mohammed et al. [17] reported that subsurface drip irrigation at a depth of 40 cm not only conserved water but also resulted in significantly higher crop water productivity and yield compared with surface drip and bubbler irrigation. Nevertheless, the optimal emitter burial depth varies considerably among crops and soil types. At present, there remains a lack of research concerning the effects of different irrigation depths and water application rates on the photosynthetic performance, yield, and fruit quality of citrus in the red soil hilly regions of southern China.

Based on this, the present study was conducted in a typical Nanfeng Tangerine orchard within the red soil hilly region of Jiangxi, China, to evaluate combinations of different deep irrigation depths and water application rates through a field-controlled experiment. By systematically monitoring and analyzing key indicators, including leaf water status, photosynthetic performance, and fruit yield and quality, this study aims to elucidate the mechanisms by which deep irrigation influences Nanfeng Tangerine water use efficiency and productivity. The findings are expected to provide a theoretical basis and technical guidance for establishing scientific and efficient water-saving irrigation strategies in this region, thereby improving water use efficiency, ensuring fruit yield and quality, and promoting the sustainable development of orchards in areas prone to seasonal drought.

2. Materials and Methods

2.1. Overview of the Experimental Area

The experiment was conducted in 2024 in a standard Nanfeng Tangerine (NT, Citrus reticulata Blanco cv. Kinokuni) orchard located in Chating Village, Baisha Town, Nanfeng County, Jiangxi Province, China (27°05′ N, 116°27′ E). The experimental site lies within a subtropical monsoon climate zone, with an average annual temperature of 18.5 °C, annual precipitation of approximately 1845.5 mm, and a frost-free period of about 285 days, providing warm and humid conditions suitable for Nanfeng Tangerine growth. The orchard comprised 15-year-old trees planted at a density of 4 m × 3 m, exhibiting balanced vigor and high management standards. The soil is classified as typical red soil, acidic in nature, with a pH ranging from 4.72 to 5.86. The texture is sandy loam, containing 60.66–68.85% sand, 13.90–21.21% silt, and 17.76–27.13% clay, with relatively low water- and nutrient-holding capacity. Soil bulk density ranged from 1.26 to 1.70 g/cm³, and the field water-holding capacity was 0.28–0.33 m³/m³. Detailed physicochemical properties are presented in Table 1.

Table 1. Soil hydraulic parameters of Nanfeng Tangerine orchard.

Soil Depth (cm)	Soil Bulk Density (g/cm3)	Organic Carbon (g/kg)	pH	Field Capacity θf (m3/m3)	Saturated Water Content θs (m3/m3)	Sand (%)	Silt (%)	Clay (%)
10	1.26	32.13	5.86	0.33	0.48	68.85	13.04	18.10
30	1.34	20.88	5.79	0.29	0.44	65.47	13.90	20.63
50	1.46	14.45	5.06	0.31	0.39	66.83	13.40	19.77
70	1.64	11.38	4.88	0.28	0.36	68.43	13.81	17.76
100	1.70	5.44	4.72	0.32	0.34	60.66	12.21	27.13

2.2. Experimental Design

Observations were conducted on typical sunny days in July, August, September, and November 2024, corresponding to the vigorous photosynthetic period, fruit expansion period, sugar accumulation period, and pre-maturity period of Nanfeng Tangerine, respectively. This study employed a two-factor experimental design involving irrigation depth and water application rate. Preliminary root distribution analysis (Figure 1) indicated that up to 96% of Nanfeng Tangerine root biomass was concentrated within the top 1 m of soil. Based on this, three irrigation depth levels were established: 25 cm (D1), 50 cm (D2), and 100 cm (D3). Simultaneously, three water application gradients were set: W3 represented full irrigation corresponding to 100% field capacity (θ_f), which served as the control treatment in this study and represented the most widely adopted conventional irrigation regime for Nanfeng Tangerine cultivation in the red soil hilly regions, W2 represented mild deficit irrigation at 75% of W3, and W1 represented severe deficit irrigation at 50% of W3. A total of nine treatment combinations were arranged: D1W1, D1W2, D1W3, D2W1, D2W2, D2W3, D3W1, D3W2, and D3W3. Each treatment included seven Nanfeng Tangerine trees, with the three central trees selected as observation samples.

Figure 1. Diagrams of the deep irrigation system and distribution map of Nanfeng Tangerine root systems. (a) Shows the layout of the irrigation systems for different treatments in the experimental area; (b) shows a photo of laying rain shelter film on-site; (c) shows the structural diagram of the water emitter; (d) shows the distribution map of the Nanfeng Tangerine root systems).

The deep irrigation system used in this experiment consisted of a water reservoir, a pump, a filter, a main pipeline, lateral tubing, and field emitters (Figure 1). To eliminate potential confounding effects from water quality, a common water source from a dedicated well within the experimental orchard was utilized for all irrigation treatments. The system has been operating stably for three years. The design and installation of the emitters were as follows: holes were drilled using a soil auger in the four cardinal directions (east, south, west, and north) at a distance of 1 m from the tree trunk, with depths corresponding to the three preset irrigation levels (25 cm, 50 cm, and 100 cm). After soil removal, a 5 cm diameter PVC pipe was vertically inserted into each hole, with its base positioned at the target irrigation depth. Four drippers with a flow rate of 8 L/h were installed on the lateral tubing around each tree, connected to the PVC pipe via flexible tubing, to deliver water directly to the specified soil layer. For soil moisture monitoring, a TDR sensor was vertically installed 0.5 m west of the tree trunk (aligned with the center of the PVC pipe) to measure soil volumetric water content at five depths (0.1, 0.3, 0.5, 0.7, and 1.0 m). Additionally, every 15 days, soil samples were collected at 0.2 m intervals using a soil auger on the east side of the tree, and the soil water content of the 0–200 cm profile was determined via oven-drying.

The water application for each treatment was precisely controlled using electromagnetic flow meters installed on the lateral tubing. Irrigation frequency was dynamically adjusted based on soil moisture measurements from TDR sensors and oven-drying, as well as prevailing meteorological conditions, to ensure that the applied water did not exceed the preset irrigation limits. Full irrigation (W3) served as the reference treatment, with irrigation triggered when the soil volumetric water content fell below 65% of field capacity (θ_f). The irrigation quota was calculated using the following formula [18]:

$$I=P\ast H\left({\theta }_{max}-\theta \right)\gamma$$

(1)

where I = irrigation amount per tree (mm), P = planned wetting ratio (set to 0.3), H = planned wetting layer depth (set to 1 m), γ = soil bulk density (set to 1.48 g/cm³), θ_max = upper irrigation limit (i.e., 100% θ_f, field capacity), and θ = soil volumetric water content before irrigation (m³/m³).

2.3. Index Determination

2.3.1. Soil Water Content

Soil samples were collected from the 0–200 cm soil profile at 20 cm intervals using a soil auger. The samples were immediately placed into pre-labeled aluminum containers and sealed for storage. The sealed samples were oven-dried at 105 °C for 10 h until a constant weight was reached to determine the gravimetric soil water content, which was then multiplied by the soil bulk density to calculate the soil volumetric water content (θ). The soil volumetric water content was calculated using the following formula [19]:

$$\theta ={\theta }_m\times r$$

(2)

where θ = soil volumetric water content, with the unit of m³/m³, θₘ = soil mass water content; and r = soil bulk density (set to 1.48 g/cm³).

2.3.2. Leaf Relative Water Content

Leaf relative water content (LRWC) is closely associated with Nanfeng Tangerine growth, and monitoring its dynamics can provide early indications of growth trends and insights into the plant’s water physiological characteristics. In this study, LRWC was determined using the oven-drying method. From July to November 2024, mature leaves from representative, healthy, and uniformly growing one-year-old shoots were collected from the east, south, west, and north directions of each treatment tree, ensuring consistent leaf positions. For each tree, 5–8 leaves were sampled 2 h after sunset. Immediately after detachment, their fresh weight (FW) was recorded. The leaves were then immersed in 500 mL of distilled water in a covered dish and kept in the dark; after 24 h, the saturated weight (SW) was measured. Finally, the leaves were oven-dried at 75 °C for 48 h to obtain the dry weight (DW). Each treatment was replicated three times, and the average value was used to calculate the leaf relative water content for that sampling event.

2.3.3. Photosynthetic Characteristics

During the experimental period, leaf gas exchange parameters were measured monthly in each plot. According to the experimental design, measurements were conducted on the 15th day after each irrigation event, during clear mornings from 9:30 to 11:30 using a LI-6800 portable photosynthesis system (LI-COR, Lincoln, NE, USA). In each observation area, leaves from one-year-old or older shoots at the same leaf positions were selected, with five leaves per position. The net photosynthetic rate (P_n), stomatal conductance (G_s), transpiration rate (T_r), and intercellular CO₂ concentration (C_i) were recorded for each selected leaf.

2.3.4. Yield

All fruits from four citrus trees in each plot were harvested and classified according to the NY/T 1190-2006 Citrus Grading Standard. Non-standard fruits were defined as those with a single fruit weight of less than 80 g or a transverse diameter smaller than 50 mm, which do not meet the commercial specification. Damaged fruits were defined as those with more than 5% of the peel surface mechanically injured, or those affected by pests, diseases, or decay. During harvest, fruits from each sample tree were examined and classified individually. The weights of non-standard and damaged fruits were excluded from the calculation of single-tree yield.

2.3.5. Meteorology

A field-installed meteorological station was used to directly monitor near-surface weather conditions. The recorded meteorological variables included relative humidity (RH), solar radiation (SR), air temperature (T), atmospheric pressure (P), and land surface temperature (LST). Meteorological data were collected using an automatic agricultural weather station (model DZZ5; Huayun Shengda Meteorological Technology Co., Ltd., Beijing, China), which was installed at the experimental base (27°05′41″ N, 116°27′48″ E).

2.4. Date Analysis and Processing

Experimental data were organized using Excel 2021 and subjected to analysis of variance (ANOVA) in SPSS 27.0. Differences among treatments were evaluated using Duncan’s multiple range test at a significance level of p < 0.05. Figures and charts were generated using Origin 2021. Multi-objective comprehensive analysis was performed using SPSSAU based on the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS).

3. Results and Analysis

3.1. Water Status of Nanfeng Tangerine Leaves

Changes in the relative water content (RWC) of citrus leaf under different treatments are shown in Figure 2. Under W1 conditions, RWC decreased with increasing irrigation depth across months, but differences among treatments were not significant (p > 0.05). Under W2 conditions, RWC exhibited an increasing trend over the months, yet differences among treatments remained nonsignificant (p > 0.05). Under W3 conditions, RWC showed a single-peak pattern in July and September, with treatment D2W3 in September being significantly higher than D1W3 by 17.95%. In contrast, RWC in August and November increased with irrigation depth. The minimum RWC in July and August occurred in D2W2, at 69.75% and 72.01%, respectively, which were 17.41% and 17.99% lower than D2W3 (87.16%) and D1W1 (90%). In July, the maximum RWC was observed in D2W3 (87.46%) and the minimum in D2W2 (69.75%), with D2W3 significantly higher than D2W2 by 25.39%, while differences with other treatments were not significant. In August, no significant differences were observed among treatments. In September, D1W3 differed significantly from D1W1 and D2W3, with D1W1 showing the highest value (92.43%), 18.06% higher than D1W3 (78.29%). In November, D3W1 differed significantly from D1W2, D2W2, and D3W3, with D1W2 achieving the maximum RWC (97.37%) and D3W1 the minimum (75.06%); this represents a 29.72% increase in D1W2 compared to D3W1. Overall, irrigation depth, irrigation level, and their interaction had no significant effect on RWC in July and August, while irrigation depth and irrigation level significantly affected RWC in September and November, respectively.

Figure 2. Dynamics of relative water content (RWC) in Nanfeng Tangerine leaves under different irrigation treatments. Notes: In the figure, D represents the irrigation depth factor; W, the irrigation amount factor; and D × W, their interaction. * indicates a significant difference between single factors or interaction effects, while “ns” indicates no significant difference. Different letters in the figure indicate significant differences among treatments at the same time point, whereas the same letters indicate no significant differences among treatments at the same time point. The error bars indicate the standard error (mean ± SE). The same conventions apply to subsequent figures.

3.2. Photosynthetic Characteristics of Nanfeng Tangerine Leaves

Figure 3 illustrates the effects of irrigation depth and water application rate on Nanfeng Tangerine leaf’s net photosynthetic rate (P_n), transpiration rate (T_r), stomatal conductance (G_s), and intercellular CO₂ concentration (C_i) in July, August, September, and November 2024. Overall, P_n, T_r, and G_s in July and November were significantly lower than in August and September. In July and November, differences among treatments for P_n, T_r, and G_s were relatively small, whereas significant differences were observed among treatments in August and September (p < 0.05). Furthermore, intercellular CO₂ concentration (C_i) showed minor differences among treatments from July to September, but significant differences appeared in November (p < 0.05). Analysis revealed that irrigation level and its interaction with depth had significant effects on P_n and T_r in August (p < 0.05), while irrigation depth, level, and their interaction significantly affected Tr in July and November (p < 0.05). Specifically, under W1 and W2 conditions in August, P_n, T_r, and G_s increased with irrigation depth. Under W1, treatment D3W1 exhibited a significantly higher P_n (10.64 μmol·m⁻²·s⁻¹) than D1W1 (5.65 μmol·m⁻²·s⁻¹; p < 0.05), representing an 88.32% increase; T_r was also significantly higher in D3W1 (3.50 mmol·m⁻²·s⁻¹) than in D1W1 (1.53 mmol·m⁻²·s⁻¹), with an increase of 128.76%. Under W2, differences among treatments were not significant. In contrast, under W3 (full irrigation), P_n and T_r decreased with increasing irrigation depth, with D1W3 showing higher P_n (13.34 μmol·m⁻²·s⁻¹) and T_r (5.42 mmol·m⁻²·s⁻¹) than D2W3 (8.30 μmol·m⁻²·s⁻¹ and 3.34 mmol·m⁻²·s⁻¹), with increases of 60.72% and 62.28%, respectively. G_s showed no significant differences under W1; under W2, D3W2 (0.24 mmol·m⁻²·s⁻¹) was significantly higher than D1W2 (0.17 mmol·m⁻²·s⁻¹), with an increase of 41.18%, while under W3, no significant differences were observed. In September, under W1 and W2, P_n, T_r, and G_s exhibited an initial increase followed by a decrease with irrigation depth. Under W2, D2W2 showed significantly higher P_n (9.15 μmol·m⁻²·s⁻¹) than D1W2 (7.45 μmol·m⁻²·s⁻¹) and D3W2 (8.14 μmol·m⁻²·s⁻¹), with increases of 22.82% and 13.27%, respectively; T_r was also significantly higher in D2W2 (4.06 mmol·m⁻²·s⁻¹) than in D1W2 (1.68 mmol·m⁻²·s⁻¹) and D3W2 (1.98 mmol·m⁻²·s⁻¹), with increases of 141.67% and 105.05%. Under W3, P_n increased with irrigation depth, with D2W3 (11.08 μmol·m⁻²·s⁻¹) being significantly higher than D1W3 (6.71 μmol·m⁻²·s⁻¹), exhibiting a 65.13% increase; T_r showed no significant difference, while G_s increased with irrigation depth, with D3W3 reaching 0.20 mmol·m⁻²·s⁻¹ compared to 0.16 mmol·m⁻²·s⁻¹ in D1W3, which is an increase of 25%. In November, C_i decreased significantly with increasing irrigation depth (p < 0.05), with the highest value observed in D1W1 (340.26 μmol·mol⁻¹), which was significantly higher than in D3W3 (59.85 μmol·mol⁻¹).

Figure 3. Effects of irrigation depth and amount on temporal dynamics of photosynthetic characteristics in Nanfeng Tangerine trees (a), (b), (c), and (d) represent the net photosynthetic rate, transpiration rate, stomatal conductance, and intercellular CO₂ concentration of Nanfeng Tangerine leaves, respectively.

Figure 4 and Figure 5 present the mean values of photosynthetic parameters across different months and the instantaneous leaf water use efficiency (WUEi) under each treatment. Treatment D2W2 exhibited significantly higher mean P_n, T_r, and G_s compared to the other treatments. Under the W1 and W3 conditions, P_n increased with irrigation depth, whereas under W2, it exhibited a single-peak pattern, reaching its maximum in D2W2 (12.09 μmol·m⁻²·s⁻¹), significantly higher than in D1W1 (7.98 μmol·m⁻²·s⁻¹) and D1W2 (9.46 μmol·m⁻²·s⁻¹). Similarly, T_r and G_s under W1 and W2 also displayed a single-peak trend, with D2W2 achieving the highest values (T_r = 4.25 mmol·m⁻²·s⁻¹; G_s = 0.23 mmol·m⁻²·s⁻¹), significantly greater than D1W1 (T_r = 2.51 mmol·m⁻²·s⁻¹; G_s = 0.13 mmol·m⁻²·s⁻¹). At the same irrigation depth, WUEi exhibited a single-peak trend with increasing water application, reaching the maximum under W2. Specifically, D1W2 (0.363) showed a 35.45% increase over D3W3 (0.268). Correlation analysis (Table 2) indicated that environmental factors—including relative humidity (RH), solar radiation (SR), air temperature (T), and land surface temperature (LST)—were significantly correlated with photosynthetic parameters (p < 0.05). RH was highly negatively correlated with P_n, T_r, C_i, and G_s (p < 0.01) and significantly positively correlated with WUEi (p < 0.05). In contrast, SR, T, and LST were highly positively correlated with the photosynthetic parameters (p < 0.01) and highly negatively correlated with WUEi (p < 0.01). Soil volumetric water content (VSWC) was significantly positively correlated with P_n and T_r (p < 0.05), whereas WUEi was significantly negatively correlated with T_r, C_i, and G_s (p < 0.05).

Figure 4. Mean values of photosynthetic factors in Nanfeng Tangerine leaves across different treatment groups ((a), (b), (c), and (d) refer to the net photosynthetic rate, transpiration rate, intercellular CO₂ concentration, and stomatal conductance of Nanfeng Tangerine leaves, respectively). Notes: In the figure, different letters indicate significant differences between different treatment groups, while the same letter indicates no significant differences between different treatment groups. The error bars indicate the standard error (mean ± SE). The same applies to the figure below.

Figure 5. Variation in instantaneous water use efficiency (WUEi) of Nanfeng Tangerine leaves across irrigation treatments.

Table 2. Correlation analysis between photosynthetic parameters and environmental factors in Nanfeng Tangerine leaves (* and ** denote statistical significance at p < 0.05 and p < 0.01, respectively).

	RH	SR	T	P	LST	VSWC	Cr	Gs	Pn	Tr	WUEi
RH	1
SR	−0.998 **	1
T	−0.802 **	0.819 **	1
P	0.134	−0.18	−0.667 **	1
LST	−0.832 **	0.841 **	0.992 **	−0.583 **	1
VSWC	−0.226	0.245	0.311	−0.306	0.281	1
Cr	−0.547 **	0.556 **	0.496 **	−0.204	0.492 **	0.034	1
Gs	−0.929 **	0.935 **	0.704 **	−0.114	0.717 **	0.291	0.588 **	1
Pn	−0.918 **	0.919 **	0.755 **	−0.162	0.777 **	0.331 *	0.445 **	0.954 **	1
Tr	−0.838 **	0.850 **	0.715 **	−0.245	0.714 **	0.380 *	0.550 **	0.932 **	0.917 **	1
WUEi	0.341 *	−0.366 *	−0.587 **	0.592 **	−0.549 **	−0.046	−0.860 **	−0.359 *	−0.279	−0.405 *	1

3.3. Nanfeng Tangerine Yield

Figure 6 shows the per-tree Nanfeng Tangerine yield under different irrigation treatments. Overall, treatment D2W2 produced significantly higher yields than the other treatments (p < 0.05), averaging 58.13 kg tree⁻¹, which represents a 77.88% increase compared with the lowest-yielding D3W1 treatment (32.68 kg tree⁻¹). Regarding irrigation levels, W2 generally resulted in higher yields than W1 across most depths and was comparable or even superior to W3. For instance, D1W2 and D1W3 yielded 45.77 kg tree⁻¹ and 48.02 kg tree⁻¹, respectively, which are increases of 28.24% and 34.55% over D1W1 (35.69 kg tree⁻¹). Notably, D2W2 outyielded D2W3 by 15.03%, indicating that at 50 cm depth, moderate water deficit can satisfy crop water demand, and additional irrigation does not further increase yield, reflecting the diminishing marginal effect of water utilization. Considering irrigation depth, D2 (50 cm) outperformed D1 (25 cm) and D3 (100 cm), as this depth corresponds to the main root distribution layer of citrus (20–80 cm), facilitating efficient water uptake. Specifically, D2W2 produced 21.26% and 32.01% higher yields than D1W2 (46.23 kg tree⁻¹) and D3W2 (39.52 kg tree⁻¹), respectively. In contrast, deep irrigation at 100 cm performed poorly, especially under W1, and even increasing irrigation to W3 could not surpass D2W2. This suggests that excessive irrigation depth places water beyond the effective root zone, reducing water use efficiency and negatively affecting yield. Linear regression analysis between Nanfeng Tangerine yield and photosynthetic parameters (Figure 7) showed that yield (Y) had R² values of 0.62 and 0.59 with P_n and G_s, respectively, with a significant positive correlation with P_n (p < 0.05) and an extremely significant positive correlation with G_s (p < 0.05). This indicates that these are key controlling factors for Nanfeng Tangerine yield. In contrast, Y had lower R² values with Tr (0.34) and Ci (0.39), and their effects on yield were not significant (p > 0.05). In summary, Nanfeng Tangerine yield was highest under D2W2 (58.13 kg tree⁻¹), and leaf P_n and G_s significantly contributed to yield improvement.

Figure 6. Nanfeng Tangerine tree yield (per plant) across irrigation treatments.

Figure 7. Linear regression analysis of Nanfeng Tangerine yield against photosynthetic parameters (a), (b), (c), and (d) represent the net photosynthetic rate, transpiration rate, stomatal conductance, and intercellular CO₂ concentration of Nanfeng Tangerine leaves, respectively.

3.4. Fruit Quality

Figure 8 illustrates the effects of different irrigation treatments on citrus fruit quality. Regarding fruit soluble sugar content, treatment D2W2 was significantly superior to all W1 and W3 irrigation levels (p < 0.05), with an average soluble solid content of 34.37 °Brix, the highest among all treatments, followed by D3W2 (33.14 °Brix). Compared with the lowest value, D1W3 (25.95 °Brix), these represent increases of 32.45% and 27.71%, respectively. This indicates that under deep irrigation (D2 and D3), implementing mild deficit irrigation (W2) promotes sugar synthesis and accumulation, showing a clear advantage. At the same irrigation level, irrigation depth had a significant regulatory effect on sugar accumulation. Under W2, both D2W2 and D3W2 had significantly higher sugar content than D1W2, increasing by 18.64% and 14.39%, respectively. Under W1, D3W1 (31.76 °Brix) was also higher than D1W1 (29.06 °Brix) and D2W1 (29.52 °Brix), indicating that deeper irrigation layers favor sugar accumulation. Under W3, differences among irrigation depths were not significant, and overall sugar content was lower than in W1 and W2 treatments, suggesting that excessive irrigation may inhibit sugar synthesis. At the same irrigation depth, water application had a pronounced effect on sugar accumulation. For example, under D2, sugar content in D2W2 was 16.43% and 25.53% higher than in D2W1 and D2W3, respectively; under D3, D3W2 exceeded D3W1 and D3W3 by 4.35% and 20.77%, respectively. In contrast, under D1, differences among the three irrigation levels were not significant, indicating limited regulatory capacity of shallow irrigation on sugar accumulation. Overall, D2W2 was optimal for enhancing fruit sugar content, demonstrating that a 50 cm irrigation depth combined with mild water deficit is most favorable for sugar synthesis and enrichment.

Figure 8. Nanfeng Tangerine quality of different treatment groups ((a), (b), and (c) represent the sugar content, titratable acidity, and sugar–acid ratio of Nanfeng Tangerine, respectively). The error bars indicate the standard error (mean ± SE).

Regarding fruit acidity, W2 treatments generally exhibited lower acidity, contributing to improved fruit flavor. Among them, D2W2 showed the lowest acidity (0.46%), significantly lower than D1W2 (0.49%) and D3W2 (0.48%). This indicates that mid-layer irrigation is more favorable for the metabolism and conversion of organic acids, likely by optimizing the root-zone water environment and thereby inhibiting acid accumulation. Under W3 conditions, the acidity of D1, D2, and D3 all exceeded 0.50%, significantly higher than in the W1 and W2 treatments, with no significant differences among depths (p > 0.05), suggesting that under high irrigation, water amount has stronger influence on acidity than irrigation depth. Under W1, acidity ranged from 0.49% to 0.50%; although slightly lower than W3, no significant improvement was observed. This indicates that severe water deficit can suppress acid accumulation, but its effect is less pronounced than in W2, possibly because excessive stress limits metabolic activity.

The sugar–acid ratio (SAR), an important indicator of fruit flavor balance, was sensitive to irrigation treatments in this study. W2 treatments generally exhibited higher SAR, enhancing fruit flavor. Among them, D2W2 had the highest SAR (15.93), significantly superior to all other treatments (p < 0.05) and representing a 19.33% increase over the lowest treatment, D1W3 (13.35). Under W2, D2W2′s SAR was 11.55% higher than that of D1W2, and although not significantly different from D3W2, it showed overall better performance, indicating that mid-layer irrigation is more favorable for coordinated regulation of sugar and acid, thereby improving fruit flavor. W1 treatments displayed intermediate SAR values, while W3 treatments were generally lower, suggesting that full irrigation may suppress sugar synthesis and delay acid degradation, ultimately being unfavorable for enhancing flavor quality.

3.5. Comprehensive Analysis of Yield, Water Use Efficiency, and Fruit Quality of Each Treatment

To systematically evaluate the combined effects of different irrigation treatments on Nanfeng Tangerine yield and fruit quality, the yield, water use efficiency (WUE), and sugar–acid ratio (SAR) were selected as key decision indicators, and a multi-objective comprehensive assessment was conducted using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The results are presented in Table 3. Analysis indicated significant differences in TOPSIS scores among treatments, highlighting clear differentiation in the overall effectiveness of deep irrigation for simultaneously improving yield and quality. Among the nine treatments, D2W2 exhibited the optimal comprehensive performance, with an average yield of 58.13 kg tree⁻¹, WUE of 0.34%, SAR of 15.93, and TOPSIS score of 0.95, ranking first and indicating a notable advantage in balancing yield enhancement and quality optimization. D2W3 and D3W3 ranked second and third with TOPSIS scores of 0.66 and 0.61, respectively, reflecting relatively favorable comprehensive performance and demonstrating a certain level of synergy among yield, WUE, and SAR. Treatments D1W2, D2W1, D3W2, and D1W3 had intermediate TOPSIS scores ranging from 0.28 to 0.53. In contrast, D1W1 and D3W1 scored the lowest (0.13 and 0.07), ranking eighth and ninth, revealing that shallow irrigation treatments struggle to achieve coordinated improvements in yield and quality. Overall, a mid-layer irrigation depth (50 cm) combined with mild deficit irrigation is most conducive to achieving the dual objectives of high yield and superior fruit quality in Nanfeng Tangerine.

Table 3. TOPSIS comprehensive analysis results.

Treatment	Mean Yield	Mean Water Use Efficiency	Mean Sugar–Acid Ratio	TOPSIS Score	Ranking
D2W2	58.13	0.34	15.93	0.95	1
D2W3	50.28	0.28	13.83	0.66	2
D3W3	48.90	0.27	13.93	0.61	3
D1W2	45.77	0.36	14.28	0.53	4
D2W1	41.98	0.29	14.48	0.36	5
D3W2	39.52	0.35	15.75	0.30	6
D1W3	40.02	0.27	13.35	0.28	7
D1W1	35.69	0.30	14.58	0.13	8
D3W1	32.68	0.30	15.48	0.07	9

4. Discussion

4.1. Effects of Subsurface Irrigation on Nanfeng Tangerine Photosynthetic Characteristics

Photosynthesis underpins the existence and proliferation of almost all life forms. As a process that captures and stores light energy through a series of biochemical reactions, its core function lies in converting light energy into free energy that drives physiological metabolism, providing the fundamental energy for all life activities [20]. In this study, W2 irrigation conditions were generally conducive to enhancing Nanfeng Tangerine leaf’s net photosynthetic rate (P_n), stomatal conductance (G_s), and transpiration rate (T_r), while instantaneous water use efficiency (WUEi) was higher than under the W1 and W3 treatments. Previous studies have shown that severe water deficit can cause chloroplast swelling, disorganized arrangement, blurred stroma lamellae, loosening of grana connections, and thylakoid swelling or disintegration, resulting in damage to the ultrastructure of photosynthetic organs [21]. Such structural disruptions impair electron transfer from the plastoquinone pool to PSI terminal acceptors, induce stomatal closure, reduce maximum carboxylation rate, weaken RuBP regeneration capacity, and limit ATP and NADPH supply, thereby further reducing RuBP regeneration and net photosynthesis [22]. Consequently, severe water deficit significantly diminishes leaf P_n and can even damage the photosynthetic system, hindering photosynthetic efficiency. Conversely, mild water deficit has been reported to enhance photosynthetic performance. For example, light water stress applied during stem elongation and tillering in winter wheat increased the maximum photochemical efficiency of PSII, actual photochemical efficiency, maximum carboxylation rate, and electron transport rate, while reducing non-photochemical quenching, resulting in higher P_n [23]. Similar findings were reported in kiwifruit, where irrigation at 55–85% of full supply significantly improved instantaneous WUE compared with fully irrigated controls, with 85% irrigation during fruit expansion showing statistically significant differences (p < 0.05) [24]. Citrus also responds positively to moderate water deficit, with WUEi increasing by 13–15% under mild deficit conditions [25]. In the present study, the 50 cm irrigation depth (D2) exhibited optimal performance. During the peak photosynthetic period in summer and autumn (July–September), D2 matched the primary Nanfeng Tangerine root distribution layer (0–100 cm), while maintaining a moderate soil moisture gradient that stabilized G_s and promoted coordinated increases in P_n and Tr; for example, in August, D2W3′s P_n was 135.69% higher than that of D1W1, with Tr reaching the seasonal maximum. Similarly, Yang et al. [7] reported that a 40 cm irrigation depth increased wheat canopy light interception by 14.5% and the population-level photosynthetic rate by 14.6%. Wang et al. [26] found that compared with surface drip irrigation, subsurface drip irrigation increased leaf chlorophyll content, promoted root growth, and thus enhanced crop photosynthesis.

4.2. Effects of Subsurface Irrigation on Yield

This study demonstrated that subsurface irrigation had a significant impact on Nanfeng Tangerine yield. Under W2 conditions, Nanfeng Tangerine yield was not reduced; instead, the D2 (50 cm) treatment achieved a 15.03% increase compared with W3. This can be attributed to the alignment of the 50 cm irrigation depth with the main root distribution layer. When combined with mild deficit irrigation, D2 avoided the high evaporation losses of shallow irrigation (D1) while overcoming the limited root access and deep percolation losses associated with deep irrigation (D3), resulting in significantly higher yields than D1, D3, and fully irrigated treatments at the same depth. These results confirm that 50 cm represents the optimal irrigation depth for yield improvement. Similar findings have been reported by other studies. Li et al. [27] found that compared with full irrigation, Nanfeng Tangerine trees receiving 55%, 70%, and 85% of full irrigation exhibited yield increases of 3.0%, 5.1%, and 3.1%, respectively. Cai et al. [28] reported that subsurface drip irrigation at a 40 cm depth significantly improved apple yield, while Guo et al. [29] observed that irrigation at 70% root-zone depth effectively promoted crop growth. Furthermore, this study revealed that Nanfeng Tangerine yield was highly significantly correlated with photosynthetic parameters including the net photosynthetic rate (P_n), transpiration rate (T_r), and stomatal conductance (G_s) (p < 0.05). These findings are consistent with those of Long et al. [30] and Parry et al. [31], showing that increased leaf photosynthesis enhances yield. This is because carbon assimilation integrates over the entire growing season and throughout the canopy, such that even modest increases in net photosynthesis can translate into substantial gains in biomass and yield through cumulative spatial and temporal effects. Similar conclusions have been reported by the authors of [32], who found that the drought stress at different growth stages reduced the leaf area index and chlorophyll content in summer maize, suppressed gas exchange, decreased photosynthetic performance, and ultimately impeded the accumulation and distribution of photosynthates, leading to significant yield reduction.

4.3. Effects of Subsurface Irrigation on Fruit Quality

This study found that the D2W2 treatment, combining a 50 cm irrigation depth with mild deficit irrigation, significantly outperformed other treatments in key fruit quality indicators, including soluble sugar content, titratable acidity, and the sugar–acid ratio (p < 0.05). Mild water deficit (W2) can optimize fruit flavor through a dual regulatory mechanism: on one hand, by upregulating the activity of enzymes related to sugar synthesis [33], and on the other hand, by moderately suppressing organic acid accumulation. Other studies have similarly shown that mild water deficit positively influences crop quality. For instance, Hu et al. [34] reported that the T1 (deficit irrigation) treatment saved 31.58% of water compared with the control (CK) and significantly increased soluble sugars, titratable acidity, total phenolics, tannins, and total anthocyanins in grape berries. In contrast, excessive irrigation (W3) did not enhance fruit sugar content and instead increased citric acid, consistent with Dong et al.’s [35] findings in jujube. Severe water deficit (W1) exhibited a dual inhibitory effect on both sugar accumulation and citric acid reduction, with similar findings reported for other crops; for example, Lobos et al. [36] found that severe deficit irrigation significantly decreased soluble solids in blueberries. In the present study, fruit under the D2 irrigation depth exhibited higher sugar content, lower acidity, and a more favorable sugar–acid ratio than D1 and D3. Comparable results have been reported for other crops: Li et al. [37] observed that when irrigation pipes were buried at 40 cm, vitamin C content and the sugar–acid ratio in greenhouse tomatoes increased under all ventilation levels, likely because the soil medium buffered air effects, limiting root exposure compared with the shallower 15 cm pipe placement. Similarly, Wildman et al. [38] found that fruits irrigated at greater depths (76–122 cm) exhibited significantly higher sugar content and lower acidity than those irrigated at shallower depths (36–43 cm).

4.4. Significance of Subsurface Irrigation for Yield and Quality Improvement

The red soil hilly regions have long been constrained by a combination of natural factors, including fragmented terrain, rapid water infiltration, and weak soil water retention, which, compounded by frequent seasonal droughts, make precise water management a critical bottleneck for achieving high yield and fruit quality in orchards [39,40]. This study demonstrated that deep irrigation at a 50 cm depth (D2) combined with mild deficit irrigation (W2) can effectively enhance the drought resilience of Nanfeng Tangerine orchards in these areas. This irrigation depth precisely targets 96% of Nanfeng Tangerine root distribution within the top 1 m of soil (main roots at 20–80 cm), simultaneously avoiding the surface evaporation losses associated with shallow 25 cm irrigation (due to the high sand content and rapid water loss in red soils under high temperatures) and circumventing the root inaccessibility and deep leakage issues associated with 100 cm deep irrigation, thereby significantly improving water availability during drought periods. Moreover, this irrigation regime maintained relatively high net photosynthetic rates and stomatal conductance during the high-drought summer and autumn seasons, mitigating drought-induced damage to the photosynthetic system. Deep irrigation can significantly influence Nanfeng Tangerine photosynthetic characteristics (P_n, T_r, G_s, C_i), instantaneous water use efficiency, and yield by regulating both irrigation depth and water volume. Aligning the wetted soil depth with root spatial distribution satisfies crop water demand while minimizing surface evaporation and deep percolation, ensuring precise water delivery to the root zone [11]. Future research should further optimize deep irrigation parameters by considering variations in Nanfeng Tangerine cultivars and regional soil types, thereby enhancing the applicability and universality of this technology.

5. Conclusions

The effects of deep irrigation, with different combinations of irrigation depth and water volume, on Nanfeng Tangerine photosynthetic characteristics, fruit yield, and quality were systematically investigated. The main conclusions are summarized below.

Regarding photosynthetic performance, treatment D2W2 exhibited the highest P_n, T_r, and G_s, which significantly increased by 51.5%, 69.32%, and 76.92% compared with D1W1 (p < 0.05). Environmental factors, including relative humidity (RH), solar radiation (SR), air temperature (T), and land surface temperature (LST), were significantly correlated with photosynthetic parameters (p < 0.05), with RH and SR showing the strongest influence on P_n. For leaf water use efficiency, W2 (mild deficit irrigation) outperformed both W1 (severe deficit irrigation) and W3 (full irrigation), with D1W2 reaching a maximum of 0.36%, which was significantly higher than D1W3 and D3W3 by 33.95% and 35.45%, respectively, while differences among other treatments were not significant (p > 0.05). Regarding fruit yield, D2W2 significantly outperformed all other treatments (p < 0.05), averaging 58.13 kg/tree. In terms of fruit quality, W2 generally surpassed W1 and W3 at most irrigation depths, with D2W2 showing the best performance: soluble sugar content of 34.37 °Brix, titratable acidity of 0.46%, and a sugar–acid ratio of 15.93. Multi-objective evaluation using TOPSIS, based on yield, water use efficiency, and the sugar–acid ratio, ranked the treatments as D2W2 > D2W3 > D3W3 > D1W2 > D2W1 > D3W2 > D1W3 > D1W1 > D3W1, with D2W2 achieving the highest score (0.946). In summary, a 50 cm irrigation depth combined with deficit irrigation (D2W2) represents the optimal strategy for conserving water, increasing yield, and improving fruit quality in Nanfeng Tangerine orchards of the red soil hilly regions. Translation of this regime to other soil or climatic zones must be tailored to local root distributions and soil moisture dynamics. While the D2W2 approach shows significant potential for water-saving and yield enhancement in the study area, its large-scale adoption should be preceded by small-scale trials that account for local economic constraints and management systems to ensure long-term viability.