*Article* **Evaluation of Suitable Water–Zeolite Coupling Regulation Strategy of Tomatoes with Alternate Drip Irrigation under Mulch**

**Xiaolan Ju, Tao Lei \*, Xianghong Guo, Xihuan Sun, Juanjuan Ma, Ronghao Liu and Ming Zhang**

College of Water Resource Science and Engineering, Taiyuan University of Technology, Taiyuan 030024, China; juxiaolan0757@link.tyut.edu.cn (X.J.); guoxianghong@tyut.edu.cn (X.G.); sunxihuan@tyut.edu.cn (X.S.); mjjsxty@163.com (J.M.); liuronghao@tyut.edu.cn (R.L.); zhangming0789@link.tyut.edu.cn (M.Z.) **\*** Correspondence: lcsyt@126.com

**Abstract:** The water (*W*; *W*50, *W*75, and *W*100)–zeolite (*Z*; *Z*0, *Z*3, *Z*<sup>6</sup> and *Z*9) coupling (*W*-*Z*) regulation strategy of high-quality and high-yield tomato was explored with alternate drip irrigation under mulch. Greenhouse planting experiments were used in monitoring and analyzing tomato growth, physiology, yield, quality, and water use efficiency (*WUE*). Suitable amounts of *W* and *Z* for tomato growth were determined through the principal component analysis (*PCA*) method. Results showed that tomato plant height (*Ph*), stem thickness (*St*), root indexes, leaf area index (*LAI*), photosynthetic rate (*Pn*), transpiration rate (*Tr*), stomatal conductance (*Gs*), organic acid (*OA*), and yield showed a positive response to *W*, whereas nitrate (*NC*), vitamin C (*VC*), soluble solid (*SS*), intercellular CO2 concentration (*Ci*), fruit firmness (*Ff*), and *WUE* showed the opposite trend. The responses of *Ci* and *Ff* to *Z* were first negative and then positive, whereas the responses of other indexes to *Z* showed an opposite trend (except yield under *W*50). The effects of *W*, *Z*, and *W*-*Z* on tomato growth, physiological, and quality indexes and yield were as follows: *W* > *Z* > *W*-*Z*; the effects on *WUE* were as follows: *Z* > *W* > *W*-*Z*. The two principal components of growth factor and water usage factor were extracted, and the cumulative variance contribution rate reached 93.831%. Under different treatments for tomato growth, the comprehensive evaluation score F was between −1.529 and 1.295, the highest treated with *Z*6*W*100, the lowest treated with *Z*0*W*50. The *PCA* method showed that under the condition of alternate drip irrigation under mulch, the most suitable *W* for tomato planting was 100% *<sup>E</sup>* (*<sup>E</sup>* is the water surface evaporation), and the amount of *<sup>Z</sup>* was 6 t·ha<sup>−</sup>1.

**Keywords:** alternate drip irrigation under mulch; water; zeolite amount; tomato growth; water use efficiency; principal component

#### **1. Introduction**

Tomatoes are popular worldwide and a high water-dependent horticultural crop, cultivated in open fields and greenhouses [1]. Water stress has negative effects on biochemical and physiological systems and affects the healthy development of crops [2]. Adding zeolite (*Z*) can improve soil water retention [3], improve the ability of crop roots to absorb and utilize soil water [4], and reduce the impact of water stress on crop growth and physiological development [5]. Alternate drip irrigation is an effective irrigation method that saves water and regulates soil quality. It has the advantages of low degree of soil evaporation, water conservation, high yield, and high irrigation efficiency [6]. On the basis of the coupling strategy of *Z* modifier and alternate irrigation, this paper is expected to further enrich the alternative irrigation theory, explore the potential of agricultural water conservation, and promote the efficient utilization of water resources and tomato yield and quality.

Previous studies mainly revealed the effects of *Z* on tomato growth [7], yield [8,9], and dry matter [10]. However, studies on the effects of *Z* on photosynthetic characteristics,

**Citation:** Ju, X.; Lei, T.; Guo, X.; Sun, X.; Ma, J.; Liu, R.; Zhang, M. Evaluation of Suitable Water–Zeolite Coupling Regulation Strategy of Tomatoes with Alternate Drip Irrigation under Mulch. *Horticulturae* **2022**, *8*, 536. https://doi.org/ 10.3390/horticulturae8060536

Academic Editors: Xun Li, Xiaohui Hu and Shiwei Song

Received: 1 May 2022 Accepted: 14 June 2022 Published: 16 June 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

153

*WUE*, and root growth of tomato during growth period are few. The effects of *W* on the yield of tomato under drip irrigation and furrow irrigation [11–13], root growth under alternate partial root-zone irrigation [14,15], nutritional quality under drip irrigation [11], *WUE* under furrow irrigation [12], and other indicators [16,17] have been studied more. However, the effect of coupled water (*W*) regulation strategy combined with alternate drip irrigation under mulch and *Z* improvement on the growth and development of tomato throughout the growth period is still unclear. Owing to significant spatial variability in soil water between the dry and wet areas of alternate drip irrigation under mulch, the resulting physiological stimulation effect of the rhizosphere, and excellent water absorption and water retention performance of *Z*, the response of soil wetting body rhizosphere and crop growth and physiology under this coupled *W* regulation strategy must be different from that under conventional drip irrigation and furrow irrigation. Although the effects of a single factor of *W* or *Z* on crop growth have been clarified [8,12,15], the primary and secondary relationships between the effects of *W* and *Z* on tomato are unknown. The *W*-*Z* strategy has a synergistic effect on potato yield [18], soybean growth, and seed quality [19], and bean yield and *WUE* [20]. However, effects on the root growth, photosynthetic characteristics, and nutrient quality of tomato remain unclear.

*PCA* uses the idea of dimensionality reduction to convert multiple indicators with many correlations into few comprehensive indicators, so that the comprehensive performance of each indicator can be accurately determined. *PCA* has been widely applied to agriculture. For instance, it was used in establishing a comprehensive evaluation model of plant height (*Ph*), stem thickness (*St*), yield, and *WUE* for green pepper growth [21] and a comprehensive evaluation model of yield, *WUE*, and nutritional quality for potato growth [22]. However, these previous models did not involve crop root growth and physiological indicators. The root systems of plants are the main organs that absorb water and mineral nutrients, and their form and configurations largely determine the ability of plants to acquire nutrients [23]. Photosynthesis is the material basis for crop yield formation and improving crop light energy utilization efficiency is one of the important ways to improve crop yield [24]. Therefore, incorporating tomato root growth, photosynthetic physiology, and other indicators into the comprehensive evaluation model may be reasonable and objective.

The objective of this paper is to study the effects of *W*-*Z* effect of alternate drip irrigation under mulch on tomato to construct a comprehensive evaluation model of tomato growth on the basis of growth, physiology, quality, yield, and *WUE*. The suitable W and Z amount for tomato growth were clarified to provide theoretical guidance for tomato planting under the condition of alternate drip irrigation under mulch.

#### **2. Materials and Methods**

#### *2.1. Experimental Site*

The test was carried out in the greenhouse of the High-efficiency Water-saving Demonstration Base of Shanxi Water Conservancy and Hydropower Research Institute. The average altitude of the area is 763–780 m, and the site has warm temperate continental climate. The annual average precipitation is about 468.4 mm, annual average evaporation is 1812.7 mm, and annual average temperature is 9.5 ◦C. The test soil was clayey loam with a saturated water content of 0.44 cm3·cm−3, and the field water holding rate was 0.28 cm3·cm<sup>−</sup>3. The experimental irrigation water source was the fresh water well in the base.

#### *2.2. Experimental Design*

This experiment studied the growth characteristics of tomato under different *W* and *Z* amounts. Here, *W* was set at three levels: *W*50, *W*75, and *W*100, which were 50%, 75%, and 100% water surface evaporation (*E*), respectively. Meanwhile, *Z* was set at four levels: *<sup>Z</sup>*0, *<sup>Z</sup>*3, *<sup>Z</sup>*6, and *<sup>Z</sup>*9, which were 0, 3, 6, and 9 t·ha−1, respectively. The experiment adopted

a comprehensive experimental design with 12 treatments in total, and each treatment was repeated three times. The experimental tomato (Ao-guan No. 8) was planted in a film-mulched ridge planting mode. The ridge length was 6 m, the ridge surface width was 0.8 m, the ridge height was 0.15 m, and the ridge width was 0.6 m. A total of 30 tomato plants were planted in each row with a row spacing of 0.4 m × 0.4 m. According to local planting habits, all plots were managed in the same field. The irrigation method was alternate irrigation under mulch. Two drip irrigation belts were laid in each ridge, the distance between drip heads was 40 cm, and the working flow was 1.2 L·h<sup>−</sup>1. Irrigation was performed once every 4 days, and drip irrigation was conducted with one-side irrigation. The type of zeolite used in this study was the 4A zeolite purchased from Shanxi Taiheng Technology Co., Ltd. (Shanxi, China). Before tomato planting, zeolite was mixed into the soil to a depth of 30 cm. Tomato plants were transplanted on 4 June 2020, and the end time was 7 October 2020.

#### *2.3. Measurement Parameters and Methods*

(1) *Ph* and *St*: three tomatoes with similar growth rates at the seedling stage were selected and *Ph* was measured from the stem base of each tomato to the growth point with a tape measure. The accuracy was 1 mm. *St* below the first lateral branch of tomato stem base was measured with a digital vernier caliper with an accuracy of 0.01 mm.

(2) Root characteristic parameters: in the harvest period, three tomato plants with the same growth rates were randomly selected from each treatment. With the plants as the center, the whole root was excavated in the quadrats with an area of 40 cm × 40 cm and depth of 60 cm, and the whole root was removed and washed with water for the removal of debris. A V700 scanner (EPSON company, Nagano, Japan) was used for root scanning, and Win RHIZO 2003 software was used in analyzing and obtaining root characteristic parameters, such as total root length (*RL*), total root surface area (*RS*), and total root volume (*RV*).

(3) *LAI*: in each growth period, three representative plants were selected from each treatment, the leaf area was measured, and the average value was taken.

(4) Physiological indexes: during the expansion period of tomato fruit, clear and cloudless weather was selected, and photosynthesis was measured with an Li-6400 portable photosynthetic instrument. The measurement time was 9:00–11:00. The leaf selection principle was at the same position and the same leaf age.

(5) *Ff*: *Ff* was measured with GYJ-4 hardness tester.

(6) Nutritional quality: determination of *OA* by NaOH titration [25], determination of *VC* with molybdenum blue colorimetry [26], determination of *NC* by sulfuric acid salicylic acid method [27], and determination of *SS* with PAL-1 handheld refractometer [28]. The average of three measurements was obtained as the final measurement.

(7) Yield: yield was measured with an electronic scale and had an accuracy of 0.01 kg.

#### *2.4. Data Processing and Statistical Analysis*

Microsoft Office 2020 was used for data calculation, IBM SPSS statistics 25 was used for two-way ANOVA and principal component analysis, GSTA V7.0 for gray correlation analysis, and Origin 2018 was used for drawing.

#### **3. Results**

#### *3.1. Effects of W-Z on the Soil Moisture Dynamics*

Figure 1 shows the effects of *W*-*Z* on soil moisture dynamics subjected to alternate drip irrigation under mulch. As shown in Figure 1, soil moisture content with different treatments, ranging from 16.84% to 33.65%, showed sawtooth fluctuations with time growth. In the range of seedling stage and fruit expansion stage (19 July to 12 September 2020), which belongs to the vegetative and reproductive growth process, soil moisture content decreased with increased time, attributed to increased water transpiration consumption, caused by the rapid development of the rhizosphere system and functional leaves. In the range of the fruit harvest stage (12 to 30 September 2020), soil moisture content stabilized with the increased time, attributed to the stabilization of plant growth and transpirational water consumption. As shown in Figure 1, soil moisture content increased by an average of 0.69–10.38% with the increased *W* from *W*<sup>50</sup> to *W*100, and increased by an average of 0.98–5.79% with the increased *Z* from *Z*<sup>0</sup> to *Z*9, respectively. This indicates that the increase in *W* and *Z* have different degrees of promoting effects on soil water content.

**Figure 1.** Effects of *W*-*Z* on variations of soil moisture during alternate drip irrigation under mulch.

#### *3.2. Effects of W-Z on the Growth and Physiological Characteristics of Tomato* 3.2.1. Effects of *W*-*Z* on Tomato Growth

Figure 2 shows the effect of *W*-*Z* on tomato growth indicators during alternate drip irrigation under mulch. Figure 2 shows that under the conditions of *Z*0, *Z*3, *Z*6, and Z9, when *W* increased from *W*<sup>50</sup> to *W*100, tomato *Ph*, *St*, *RL*, *RS*, *RV*, and *LAI* increased monotonically by 19.78–31.30%, 13.57–20.43%, 29.10–50.84%, 35.46–47.88%, 53.91–77.96%, and 14.55–35.67%, respectively. This result showed that an increase in *W* can significantly (*p* < 0.01) promote tomato growth between *W* and *Z* and have synergistic effects on tomato growth. Figure 2 also shows that under the conditions of *W*50, *W*75, and *W*100, when *Z* increased from *Z*<sup>0</sup> to *Z*6, *Ph*, *St*, *RL*, *RS*, and *RV,* these increased monotonically by 8.32–16.55%, 5.14–8.42%, 3.51–24.70%, 9.08–16.98%, and 4.67–20.13%, respectively. When *Z* increased from *Z*<sup>6</sup> to *Z*9, *Ph*, *St*, *RL*, *RS*, and *RV* decreased by 2.08–4.91%, 1.13–3.85%, 0.07–11.64%, 2.49–11.04%, and 2.19–9.84%, respectively. However, under the conditions of *W*50, *W*<sup>75</sup> and *W*100, when the amount of zeolite changed, the change in *LAI* was not regular. This result showed that the effects of increased *Z* on *Ph* (*p* < 0.01), *St*, and root system growth (*p* < 0.01) were first accelerated and then suppressed and the response trends and intensities of *Ph* and *St* to *Z* were relatively affected by *W* factors. In the *Z*0–*Z*<sup>6</sup> range, the amounts of *W* and *Z* had a synergistic effect on tomato growth but exerted antagonistic effects in the *Z*6–*Z*<sup>9</sup> range. Two-way ANOVA showed that the *W*-*Z* effect had a significant effect (*p* < 0.01) on *Ph*, *RL*, and *RS*, had a significant effect (*p* < 0.05) on *LAI*, but had no significant effect on *St* and *RV*. We found that the sum of the squares was for *Ph* (2778.95, 557.89, 170.54), *St* (20.38, 3.09, 0.54), *RL* (5,290,316.26, 348,892.52, 157,083.72), *RS* (170,236.82, 16,600.85, 3818.61), *RV* (28.87, 1.54, 0.55), and *LAI* (136.28, 16.10, 3.94) under the effects of *W*, *Z*, and *W*-*Z*, respectively. The effects of *W*, *Z*, and *W*-*Z* on *Ph*, *St, RL*, *RS*, *RV*, and *LAI* of tomato were as follows: *W* > *Z* > *W*-*Z*. The *W* factor played a leading role in tomato growth (Table S1).

**Figure 2.** Effects of *W*-*Z* on tomato growth during alternate drip irrigation under mulch. *Ph*, *St*, *RL*, *RV*, *RS*, and *LAI* represent plant height, stem thickness, root length, root volume, root surface area, and leaf area index, respectively.

#### 3.2.2. Effects of *W*-*Z* on Tomato Physiological Characteristics

Figure 3 shows the effect of *W*-*Z* on the physiological indexes of tomatoes subjected to alternate drip irrigation under mulch. Under the conditions of *Z*0, *Z*3, *Z*6, and *Z*9, when *W* increased from *W*<sup>50</sup> to *W*100, tomato *Ci* decreased monotonically by 8.96–12.57%, *Pn*, *Tr*, and *Gs* monotonically increased by 16.67–28.43%, 22.82–44.51%, and 20.37–26.16%, respectively. This result showed that an increase in *W* can significantly inhibit *Ci* (*p* < 0.01) and promote *Pn*, *Tr*, and *Gs*. *W* and *Z* exerted antagonistic effects on *Ci* and synergistic effects on *Pn*, *Tr*, and *Gs*. Under the conditions of *W*50, *W*75, and *W*100, when *Z* increased from *Z*<sup>0</sup> to *Z*6, *Ci* decreased monotonically by 2.11–9.02%, whereas *Pn*, *Tr*, and *Gs* monotonically increased by 6.49–15.20%, 8.95–27.34%, and 5.22–16.51%, respectively. When *Z* increased from *Z*<sup>6</sup> to *Z*9, *Ci* increased by 0.60–2.01%, whereas *Pn*, *Tr*, and *Gs* decreased by 0.43–3.45%, 1.44–9.18%, and 0.16–5.77%, respectively. This result showed that in the *Z*0–*Z*<sup>6</sup> range, *W* and *Z* had an antagonistic effect on *Ci* and a synergistic effect on *Pn*, *Tr*, and *Gs*, and opposite results were obtained in a range of *Z*6–*Z*9. Two-way ANOVA calculation showed that *W*-*Z* had a significant effect (*p* < 0.05) on *Ci*, *Tr*, and *Gs* but had no significant effect on *Pn*. As follows, the sum of the squares was for *Ci* (7994.00, 2138.08, 812.00), *Pn* (127.48, 28.14, 4.19), *Tr* (20.14, 4.58, 1.00), and *Gs* (0.10, 0.02, 0.01) under the effects of *W*, *Z* and *W*-*Z*, respectively. The effects of *W*, *Z*, and *W*-*Z* on tomato physiological indicators were as follows: *W* > *Z* > *W*-*Z*. The *W* factor played a leading role in tomato growth (Table S1).

**Figure 3.** Effects of W-Z on tomato physiological indexes during alternate drip irrigation under mulch. *Pn*, *Gs*, *Ci*, and *Tr* represent photosynthetic rate, stomatal conductance, intercellular CO2 concentration, and transpiration rate, respectively.

#### *3.3. Effects of W-Z on Tomato Quality*

Figure 4 shows the effect of *W*-*Z* on tomato quality during alternate drip irrigation under mulch. Under the conditions of *Z*0, *Z*3, *Z*6, and *Z*9, when *W* increased from *W*<sup>50</sup> to *W*100, tomato *OA* monotonically increased by 13.51–16.90%, and *NC*, *VC*, *SS*, and *Ff* monotonically decreased by 25.12–29.11%, 14.44–17.81%, 17.24–28.57%, and 23.16–33.69%, respectively. These results showed that an increase in *W* had a significant effect (*p* < 0.01) on tomato quality and *W* and *Z* had synergistic effects on *OA* and antagonistic effects on *NC*, *VC*, *SS*, and *Ff*. Under the conditions of *W*50, *W*75, and *W*100, when *Z* increased from *Z*<sup>0</sup> to *Z*6, *NC*, *VC*, *SS*, and *OA* monotonically increased by 6.42–16.23%, 3.65–10.18%, 3.57–20.00%, and 6.80–8.10%, respectively, and *Ff* decreased monotonically by 4.05–17.20%. When *Z* increased from *Z*<sup>6</sup> to *Z*9, *NC*, *VC*, *SS*, and *OA* decreased by 4.00–7.23%, 2.69–3.40%, 1.72–12.50%, and 1.52–4.23%, *Ff* increased by 4.55% and 6.27% under *W*<sup>50</sup> and *W*<sup>75</sup> treatments, respectively, whereas it decreased by 2.82% under *W*<sup>100</sup> treatment. These results showed that under the conditions of *W*<sup>50</sup> and *W*75, the significant effect (*p* < 0.01) of increasing *Z* on the quality of tomatoes was first suppressed and then enhanced, and under the *W*<sup>100</sup> condition, the effects of *NC*, *VC*, *SS*, and *OA* were first accelerated and then suppressed, and only *Ff* was suppressed. In a range of *Z*0–*Z*6, *W* and *Z* exerted synergistic effects on tomato quality and antagonistic effects in a range of *Z*6–*Z*<sup>9</sup> (except *W*<sup>100</sup> *Ff*). The two-way ANOVA calculation showed that the *W*-*Z* effect had no significant effect on *OA* and *Ff* but had a significant effect (*p* < 0.01) on *NC*, *VC*, and *SS*. The sum of the squares was for *NC* (244.37, 22.07, 5.26), *VC* (44.32, 6.00, 2.04), *SS* (10.20, 1.15, 0.58), *OA* (0.01, 0.01, 0.00), and *Ff* (1.36, 0.13, 0.08) under the effects of *W*, *Z* and *W*-*Z*, respectively. The effects of *W*, *Z*, and *W*-*Z* on tomato physiological indicators were as follows: *W* > *Z* > *W*-*Z*. The *W* factor played a leading role in tomato quality (Table S1).

**Figure 4.** Effects of W-Z on tomato quality during alternate drip irrigation under mulch. *NC*, *VC*, *SS*, *OA* and *Ff* represent nitrate, vitamin C, soluble solid, organic acid, and fruit firmness, respectively.

#### *3.4. Effects of W-Z on Tomato Yield and WUE*

Figure 5 shows the effect of *W*-*Z* on tomato yield and *WUE* during alternate drip irrigation under mulch. Under the conditions of *Z*0, *Z*3, *Z*6, and *Z*9, when *W* increased from *W*<sup>50</sup> to *W*100, tomato yield monotonically increased by 16.56%, 16.31%, 16.25%, and 8.48%, respectively, whereas *WUE* monotonically decreased by 5.80%, 6.77%, 7.47%, and 16.17%, respectively. An increase in *W* had a significant promoting effect (*p* < 0.01) on the yield, and the strength of this promoting effect gradually decreased with increasing *Z*, and *WUE* showed opposite results. Under the conditions of *W*50, *W*75, and *W*100, when *Z* increased from *Z*<sup>0</sup> to *Z*6, the yield increased by 9.21%, 15.06%, and 8.91%, respectively, whereas *WUE* increased by 45.20%, 47.06%, and 42.63%, respectively. When *Z* increased from *Z*<sup>6</sup> to *Z*9, the yield under *W*<sup>50</sup> treatment increased by 3.68%, the yield under *W*<sup>75</sup> and *W*<sup>100</sup> treatments decreased by 2.25% and 3.24%, respectively, and *WUE* decreased by 16.60%, 20.68%, and

24.44%. These results showed that an increase in *Z* had a significant effect (*p* < 0.01) on tomato yield and *WUE*. In the *Z*0–*Z*<sup>6</sup> range, *W* and *Z* had synergistic effects on yield and *WUE*. In the *Z*6–*Z*<sup>9</sup> range, *W* and *Z* exerted antagonistic effects on yield and *WUE*. The two-way ANOVA calculation showed that the *W*-*Z* had no significant effect on yield and *WUE*. The effects of *W*, *Z*, and *W*-*Z* on the yield were 2574.23, 231.50, and 76.92, respectively. The effects of *W*, *Z*, and *W*-*Z* on *WUE* were 57.05, 830.44, and 11.44, respectively. The effects of *W*, *Z*, and *W*-*Z* on tomato yield were *W* > *Z* > *W*-*Z*, whereas the effect on WUE was as follows: *Z* > *W* > *W*-*Z*. The *W* factor played a dominant role in tomato yield, whereas *Z* factor played a dominant role in tomato *WUE* (Table S1).

**Figure 5.** Effects of *W*-*Z* on tomato yield and *WUE* during alternate drip irrigation under mulch.

#### *3.5. Comprehensive Evaluation and Analysis of Tomato*

Our goal is to achieve high yield and water efficiency while ensuring normal growth of tomato plants and high fruit quality. Therefore, the factors of *W* and *Z* have different effects on tomato growth, physiology, quality, yield, and *WUE*. Determined based on a single index, the analysis results of the optimal *W* coupling strategy were inconsistent, and objectively and comprehensively meeting the goals of high-quality and high-efficiency planting and cultivation of tomatoes was difficult. Therefore, a comprehensive evaluation of various tomato indicators based on principal component analysis is necessary. A total of 17 indexes was used in this study, larger than the number of experimental treatments by 12, and forms a non-positive definite matrix in statistical analysis, making it difficult to carry out comprehensive evaluation. The gray correlation analysis method can be used to screen various tomato indicators [29]. The gray correlation analysis results (Table S2) showed that the gray correlation ranking was *RV* > *St* > *Tr* > *LAI* > *VC* > *NC* > *Pn* > *SS* > *WUE* > *Ff* > yield > *Ph* > *Gs* > *Ci* > *OA* > *RS* > *RL*. However, because the *KMO* statistic of the first 11 indicators was 0.390, when the principal component analysis was performed, it did not meet the principal component analysis standard. Therefore, these 11 indicators were selected for the comprehensive evaluation of principal components: *RV*(X1), *St*(X2), *Tr*(X3), *VC*(X4), *NC*(X5), *Pn*(X6), *WUE*(X7), *SS*(X8), *Ff*(X9), yield(X10), and *Ph*(X11). After statistical calculation, the *KMO* statistic in this study was 0.575, and *p* < 0.001 for Bartlett test. Therefore, the data samples in this study were suitable for *PCA*. Based on the extraction criteria with eigenvalues of ≥1 [30], two principal components were obtained in this study (Tables S3 and S4). The contribution rate of the variance in the first principal component *F*<sup>1</sup> (named as growth factor) was 77.521%, which was a comprehensive reflection of tomato growth and quality. The contribution rate of the variance in the second principal component *F*<sup>2</sup> (named as water use factor) was 16.310%, which was a comprehensive reflection of tomato water use. The score function of each principal component was as follows:

*F*1= 0.335X1+0.333X2+0.326X3−0.291X4−0.301X5+0.329X6−0.047X7−0.293X8−0.335X9+0.286X10−+0.324X11 (1) *F*2−= 0.053X1+0.153X2−+0.180X3+0.352X4+0.333X5−+0.184X6+0.660X7+0.364X8−0.084X9+0.282X10+0.135X11 (2)

After the proportion of the variance contribution rate corresponding to each principal component was taken to the cumulative contribution rate of the principal component variance as the weight, the comprehensive model of the principal component score F was obtained using Formula (3).

$$F = (0.775F\_1 + 0.163F\_2) / 0.938 \tag{3}$$

The comprehensive score *F* can be used as an objective evaluation index for evaluating the pros and cons of different *W*-*Z* strategies. Submitting each parameter under different treatments into Formula (3) can generate a comprehensive evaluation score of each treatment, as shown in Figure 6. The *Z*0*W*<sup>50</sup> had the lowest comprehensive evaluation score at −1.529, whereas *Z*6*W*<sup>100</sup> had the highest result of 1.295. The coupling planting strategy of *<sup>Z</sup>* of 6 t·ha−<sup>1</sup> and *<sup>W</sup>* of 100% *<sup>E</sup>* was recommended.

**Figure 6.** Comprehensive evaluation score of tomato planting under different *W*-*Z* conditions.

#### **4. Discussion**

#### *4.1. Effects of Water Level on Tomato Growth*

The present results showed that when *W* increased from 50% *E* to 100% *E*, tomato *Ph* and *St* increased by 19.78–31.30% and 13.57–20.43%, respectively. Wei et al. [31] found that tomato *Ph* under 4950, 4750, and 4500 m3·ha−<sup>1</sup> irrigation amount increased by 3.3%, 5.5%, and 5.7%, whereas the *St* increased by 1.9%, 11.4%, and 7.2%, respectively, compared with the 4050 m3·ha−<sup>1</sup> irrigation treatment. *Ph* and *St* increased first and then decreased with increasing irrigation amount, and the response intensity of *Ph* and *St* to *W* was lower than that in this study. In previous studies, soil texture was sandy soil and sandy clay, and the soil moisture content was maintained at 5.5–34.5% [31]. In this study, the soil texture was clay loam, and the soil moisture content was maintained at 16.84–33.65% (Figure 1), which was higher than that in previous reports. More sufficient soil moisture will help accelerate root water absorption and improve water absorption and utilization and then increase plant growth accumulation.

The present results showed that with an increase in *W*, tomato *RL*, *RV*, *RS*, and *OA* increased by 29.10–50.84%, 53.91–77.96%, 35.46–47.88%, and 13.51–16.90% respectively, whereas *Ff*, *NC*, *VC*, and *SS* decreased by 23.16–33.69%, 25.12–29.11%, 14.44–17.81%, and 17.24–28.57%, respectively. Cabello et al. [32] found that when the amount of irrigation increased from 60% to 140% *ETc* (crop evapotranspiration), the firmness of melon decreased by 7.27%, and the response intensity of hardness to *W* was significantly lower than that in this study. This result may be due to the different reduction ranges in cell wall pressure and cell wall swelling pressure when crops were subjected to different degrees of water stress [33], which affected the change in *Ff*.

Yang Hui et al. [15] showed that under different nitrogen application conditions, with an increase in irrigation amount, tomato *RL*, *RS*, *RV*, *VC*, *OA*, and *NC* first increased by 30.21%, 66.70%, 47.07%, 9.72%, 21.90%, and 22.04% on average, respectively, and then decreased by 10.48%, 7.53%, 10.13%, 12.73%, 6.38%, and 29.89% on average, respectively, whereas *SS* decreased by 10.81% on average. The response intensity and trend of *RL*, *RS*, *RV*, *NC*, *SS*, *VC*, and *OA* to *W* were different from the results of this study. This difference may be caused by different basic experimental conditions, such as experimental design and soil texture. Previous studies reported that the pot experiment was adopted, the soil texture was heavy loam, the soil total nitrogen content was 0.81 g·kg−1, and the nitrogen application rate was 0.18–0.42 g·kg−<sup>1</sup> (equal to 565–1320 kg·ha−1) [15]. The greenhouse experiment was used in this study, the soil texture was clay loam, the soil total nitrogen content was 1.12 g·kg−1, and the nitrogen application rate was 350 kg·ha−1. In previous studies, the amount of nitrogen application was 1.61–3.77-times that in this study. Excessive nitrogen input significantly inhibits root growth while causing waste of resources [34]. In addition, the water-holding performance of soil in the present study was better than that in previous reports. Appropriate soil nutrients and sufficient water can stimulate root growth, improve the ability of roots to absorb water and nutrients, and improve tomato quality.

The present results showed that with increasing Irrigation, the *Pn*, *Tr*, and *Gs* of tomatoes increased by 16.67–28.43%, 22.82–44.51%, and 20.37–26.16%, respectively, and *Ci* decreased by 8.96–12.57%. Guo et al. [35] showed that with the weakening of drought degree, the *Pn*, *Tr*, *Gs*, and *Ci* of drought-resistant soybeans increased by 91.06%, 90.30%, 89.01%, and 50.71%, respectively. The response intensity of *Pn*, *Tr*, and *Gs* to *W* was significantly higher than that in this study, and the response trend of *Ci* to *W* was the opposite to that in this study. This finding may be attributed to different degrees of water stress and restrictions on photosynthesis. The previous irrigation levels were 40%, 70%, and 85% of the field water capacity, and the water stress intensity was higher than that in this study. The water potential in the cells increased with the water absorption ability of tomato guard cells because of high water stress intensity, enabling guard cells to absorb more water, increasing the degree of stomata opening and stomatal conductance [36], accelerating water loss in leaves, and increasing the values of *Tr* and *Pn*. The photosynthesis of tomatoes was limited by nonstomatal factors because of water stress [37], which increased the *Ci* of leaves.

Water significantly affected the yield and *WUE* of tomatoes throughout the growth period. The present results showed that after irrigation increased from 50% *E* to 100% *E*, the tomato yield increased by 8.48–16.56%, whereas the *WUE* decreased by 5.8–16.17%. Xia et al. [38] found that as the relative water content of the substrate increased from 50 ± 5% to 95 ± 5%, the yield of tomato per square meter increased by 48.05% and *WUE* decreased by 28.23%. Abdel-Razzak et al. [11] found that with an increase in irrigation amount from 50% *eTc* to 100 *eTc*, tomato yield increased by 89.09% and *WUE* decreased by 14.80%. Xia et al. reported that the response intensity of yield and *WUE* to *W* were higher than that in this study possibly because of a difference in substrate. The substrate was organic matter mixed with mushroom residue and peanut shell mixed fermentation material and sheep manure, which contained 185.38 mg·kg−<sup>1</sup> N, 122.46 mg·kg−<sup>1</sup> P2O5, and 3.92 g·kg−<sup>1</sup> K2O (equal to 180 kg N·ha−1, 119 kg P2O5·ha−1, and 3812 kg K2O·ha−1) [38]. In this study, natural soil was used for cultivation, which had 350 kg N·ha−1, 200 kg of P2O5·ha−1, and 400 kg K2O·ha−1. Compared with organic matrix cultivation, the application amount of potassium fertilizer was significantly reduced, and, thus, the amounts of nutrients required for crop growth were reduced to a certain extent. However, the response intensity of yield to water in the Abdel-Razzak study [11] was significantly higher than that in this study, which may be because the soil texture in the previous study was sandy loam [11], and the soil texture in this study was clay loam, and the soil water-holding capacity was better than the predecessors. Therefore, the soil osmotic pressure increased significantly when the predecessors were irrigated in deficit, inhibiting crop growth and, thus, reducing yield [39].

#### *4.2. Effects of Zeolite Amount on Tomato Growth*

This paper showed that after *Z* increased from *Z*<sup>0</sup> to *Z*6, tomato *Ph*, *St*, *NC*, *VC*, *SS*, and *OA* increased by 8.32–16.55%, 5.14–8.42%, 6.42–16.23%, 3.65–10.18%, 3.57–20%, and 6.8–8.1%, respectively, whereas *Ff* was reduced by 4.05–17.20%. After *Z* increased from

*Z*<sup>6</sup> to *Z*9, *Ph*, *St*, *NC*, *VC*, *SS*, and *OA* were reduced by 2.08–4.91%, 1.13–3.85%, 4.0–7.23%, 2.69–3.4%, and 1.72–12.50%, respectively, whereas *Ff* increased by 4.55–6.27%, except for *W*100. Obregon-Portocarrero et al. [40] found that after *Z* increased from *Z*<sup>0</sup> to *Z*15, maize *Ph* and *St* increased by 2.96% and 6.06%, respectively. When *Z* increased from *Z*<sup>15</sup> to *Z*35, *Ph* decreased by 10.07%, whereas *St* showed no significant difference. Petropoulos et al. [41] found that compared with the treatment without *Z*, the addition of *Z* increased *Ff* and *OA* by 2.16% and 78.13%, respectively, and *SS* was reduced by 6.29%. This is due to the fact that adding *Z* can improve soil structure by improving the availability of fertilizer and the buffer capacity of soil [42], promoting crop root development, and improving crop growth and quality. However, the excessive application of *Z* caused a large amount of sodium ions from *Z* to invade the soil, which poisoned the roots of tomato and inhibited the normal growth of tomato [43]. The response intensities of *Ph* and *St* to *Z* in the study of Obregon-Portocarrero [40] and the response trends of *Ff*, *OA*, and *SS* to *Z* in the study of Petropoulos et al. [41] were different from those in this study, possibly because of differences in application amount and period of nitrogen fertilizer addition. In the study of Obregon-Portocarrero, nitrogen application was carried out at two time points: 30 kg·ha−<sup>1</sup> at 15 days and 70 kg·ha−<sup>1</sup> at 70 days after sowing. In the study of Petropoulos, conventional fertilizer (N:P:K = 21:0:0) was applied twice. In this study, nitrogen was applied three times: half of the nitrogen was applied at the base fertilizer stage, and one-fourth of nitrogen was applied with a drip irrigation system at the first and third ear fruit expanding stages. The amount of nitrogen was 350 kg·ha<sup>−</sup>1. Sufficient nitrogen fertilizer and rational distribution of nitrogen fertilizer can supply nitrogen for crop growth effectively and continuously and then promote crop growth accumulation.

This paper showed that when *Z* increased from *Z*<sup>0</sup> to *Z*6, tomato *RL*, *RV*, and *RS* increased by 3.51–24.70%, 4.67–20.13%, and 9.08–16.98%, respectively. When *Z* increased from *Z*<sup>6</sup> to *Z*9, *RL*, *RV*, and *RS* decreased by 0.07–11.64%, 2.19–9.84%, and 2.49–11.04%, respectively. Wu et al. [44] found that compared with treatment without *Z*, adding *Z* increased *RL* and *RV* by 9.90–15.54% and 5.45–15.04%, respectively. The response trends of *RL* and *RV* to *Z* were different from those in this study, possibly because of differences in test conditions, such as application amount of *Z* and crop type. Previous studies set *Z* at two levels: 0 and 10 t·ha−1, and the test crop was rice [44]. In the present study, *<sup>Z</sup>* was set at four levels: 0, 3, 6, and 9 t·ha<sup>−</sup>1, and the test crop was tomato. Previous studies revealed that adding *Z* had a promoting effect on crop growth. The present study set multiple levels, which accurately described the effects of different *Z* levels on crops. Different crop types have different levels of response to *Z* levels. The current paper showed that after *Z* increased from *Z*<sup>0</sup> to *Z*6, tomato *Ci* decreased by 2.11–9.02%, whereas *Pn*, *Tr*, and *Gs* increased by 6.49–15.20%, 8.95–27.24%, and 5.22–16.51%, respectively. When *Z* increased from *Z*<sup>6</sup> to *Z*9, Ci increased by 0.60–2.01%, whereas *Pn*, *Tr*, and *Gs* decreased by 0.43–3.45%, 1.44–9.18%, and 0.16–5.77% respectively. Zheng et al. [45] showed that when *Z* increased from *Z*<sup>0</sup> to *Z*15, the *Pn*, *Tr*, and *Gs* of rice increased by 0.52%, 17.13%, and 35% respectively, whereas *Ci* decreased by 17.02%. Chi et al. [46] showed that when *Z* increased from *Z*<sup>0</sup> to *Z*10, the *Pn*, *Tr*, and *Gs* of rice increased by 0.95%, 3.93%, and 5.56% respectively, whereas *Ci* decreased by 4.58%. The response intensities of *Pn*, *Tr*, *Gs*, and *Ci* to *Z* were different from those in this study, possibly because of differences in irrigation methods and *Z* burial depths. Zheng et al.'s studies showed that *Z* was mixed into the soil to a depth of 5 cm under continuous flood irrigation [45]. Chi et al.'s studies showed that *Z* was applied to the soil along with the base fertilizer at one time and mixed with the soil evenly [46]. The present study adopted alternate drip irrigation under mulch, and *Z* was mixed into the soil to a depth of 30 cm. The apparent morphological characteristics of the crops were promoted as available nitrogen content in deep soil was reduced due to the increased buried depth of *Z*, increasing the chlorophyll content and leaf area index of crops and then promoting the net photosynthetic rate [47].

This paper showed that compared with no-*Z* treatment, the application of *Z* increased the yield and *WUE* of tomatoes by 0.52–15.06% and 7.77–47.06%, respectively. Previous studies showed that the application of *Z* significantly saved water by 4.8–11.4% and increased production by 9.7% [44]. The response intensities of yield and *WUE* to *Z* were less than those in this study, which may be caused by different crop types, irrigation methods, and soil texture. Previous studies adopted alternative irrigation, the test crop was rice, and the soil texture was sandy loam. The soil was affected by wind and sand in the entire year, poor fertilizer, and water-holding capacity [44]. This study adopted alternate drip irrigation under mulch, the test crop was tomato, and the soil texture was clay loam. The soil moisture content was 16.8–33.7%. The soil water-holding capacity in this study was better than that reported by previous studies, and sufficient soil water will help accelerate tomato growth. Compared with conventional alternate irrigation, alternate drip irrigation under mulch can stimulate and strengthen the root absorption compensation function [48] and thereby, improves root activity and *WUE*, and then promotes tomato growth and yield. These features were the reasons for the difference between the present study and previous reports.

#### *4.3. Comprehensive Evaluation Analysis*

Two traditional comprehensive evaluation models based on *Ph*-*St*-yield-*WUE* indicators and yield-*WUE*-quality indicators were used for validation in this study: *PCA*Bi and *PCA*Wang models [21,22]. Comprehensive evaluation results of tomato growth based on *PCA*Bi, *PCA*Wang, and the *PCA*Jv model proposed by this paper are shown in Table 1. *Z*6*W*<sup>100</sup> was the best treatment, whereas *Z*0*W*<sup>50</sup> was the worst in the *PCA*Bi and *PCA*Jv models, and differences in the ranking of other treatments were observed. The *PCA*Wang model showed that *Z*6*W*<sup>50</sup> and *Z*0*W*<sup>100</sup> were the best and worst treatment for tomatoes, respectively, in contrast to the *PCA*Jv model. Considerable differences in ranking results were found between the *PCA*Wang and *PCA*Jv models. These differences may be due to differences in index factors among the evaluation methods and subsequent changes in principal component loads and contribution rates.

**Table 1.** Comprehensive evaluation results of tomato growth based on *PCA*Jv, *PCA*Bi, and *PCA*Wang models.


Note: *F*Jv, *F*Bi, and *F*Wang were the comprehensive principal component scores of *PCA*Jv, *PCA*Bi, and *PCA*Wang models, which can be calculated by Formulae (1)–(3), respectively.

In the *PCA*Bi model, *Ph*, *St*, and yield in *F*<sup>1</sup> had large loads with values of 0.975, 0.979, and 0.949, respectively. In the *PCA*Wang model, *NC*, *VC*, *SS*, and *OA* in *F*<sup>1</sup> had large loads with values of 0.954, 0.974, 0.963, and 0.848, respectively. The load values of *Ph*, *St*, yield, *NC*, *VC*, and *SS* in the *PCA*Jv model *F*<sup>1</sup> were 0.945, 0.971, 0.836, 0.879, 0.851, and 0.857, respectively. In the *F*<sup>2</sup> of these three models, *WUE* dominated the largest load, with values of 0.998, 0.810, and 0.884. The variance contribution rates of *F*<sup>1</sup> and *F*<sup>2</sup> were 70.290% and 25.023% in the *PCA*Bi model, 68.821% and 23.067% in the *PCA*Wang model, and 77.521% and 16.310% in the *PCA*Jv model. These results indicated that the variance contribution rate of *F*<sup>1</sup> increased and that of *F*<sup>2</sup> decreased compared with those in previous studies, which was

due to the changing of the index load. The *PCA*Jv model focuses much more on the effects of growth, physiological, yield, and quality indicators on the growth and development of tomatoes throughout the growth period. The *PCA*Wan*<sup>g</sup>* model showed that some treatments were ranked in the order of *Z*0*W*<sup>50</sup> > *Z*0*W*<sup>75</sup> > *Z*0*W*<sup>100</sup> (Table 1), indicating under a condition without *Z*, tomato growth characteristics improve with decreasing moisture content. This finding contradicts the general law of tomato growth response to water content. It showed that the rationality and applicability of the model are closely related to the selection of evaluation indexes and crop types. Therefore, the typical indicators that can objectively reflect the growth status of crops should be reasonably selected in combination with specific crop types in the modeling process. These indicators can help build a reasonable model evaluation system and obtain reasonable evaluation results.

#### **5. Conclusions**

(1) Tomato *Ph*, *St*, root indexes, *LAI*, *Pn*, *Tr*, *Gs*, *OA*, and yield showed a positive response to *W*, whereas *NC*, *VC*, *SS*, *Ci*, *Ff*, and *WUE* showed opposite trends. The response of *Ci* and *Ff* to *Z* was first negative and then positive, whereas that of other indexes to *Z* showed an opposite trend (except *W*<sup>50</sup> yield).

(2) The effects of *W*, *Z*, and *W*-*Z* on tomato growth physiological indexes, quality indexes, and yield were as follows: *W* > *Z* > *W*-*Z*; the effects on *WUE* were as follows: *Z* > *W* > *W*-*Z*.

(3) The two principal components of growth quality factor and water usage factor were extracted through analysis, and the cumulative variance contribution rate reached 93.831%. According to the comprehensive score evaluation of principal components, the optimum *<sup>W</sup>* of tomatoes was 100% *<sup>E</sup>*, and the amount of *<sup>Z</sup>* was 6 t·ha−<sup>1</sup> during alternate drip irrigation under mulch.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/horticulturae8060536/s1, Table S1: Results of two-way ANOVA for tomato growth, physiology, quality, yield and WUE, Table S2: Grey relational analysis calculation, Table S3: Eigenvalues and Cumulative Variance Contribution Rates of Tomato Evaluation Factors, Table S4: Factor loading matrix of principal components on each index.

**Author Contributions:** Conceptualization, X.J.; methodology, X.S.; software, M.Z.; validation, X.J., T.L. and X.G.; formal analysis, R.L.; investigation, J.M.; resources, T.L.; data curation, X.G.; writing—original draft preparation, X.J.; writing—review and editing, T.L.; data analysis and visualization, X.G.; supervision, T.L.; project administration, T.L.; funding acquisition, T.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China (51809189; 51909184), the China Postdoctoral Science Foundation (2020M670693), and the Scientific and Technologial Innovation Programs of Higher Education Institutions in Shanxi (2019L0136).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** We appreciate that postgraduates from the university of the first author, who investigated and collected data. We are also grateful to the editors and anonymous reviewers for their suggestions and comments.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

