Physiological Response of Tomato and Cucumber Plants to Micro-Spray in High-Temperature Environment: A Scientific and Effective Means of Alleviating Crop Heat Stress

Abstract

Drought and heat stresses are severe threats to crop production and food security in arid and semi-arid regions of the globe, especially during the summer. This study investigates the effects of micro-spray on micro-climatic and physiological characteristics of cucumber and tomato plants for four growing seasons (two growing seasons each (cucumber: 2017 and 2018 and tomato: 2021 and 2022)). The experiment was conducted in a Venlo-type greenhouse where plants were irrigated with a combined micro-spray and drip irrigation (MSDI) system (1 min for tomato (seven times a day) and 2 mins (four times a day) for cucumber) and drip irrigation (DI). Both plants’ growth and physiological parameters were recorded at the end of the experiment. Moreover, the experimental results showed that plant height and stem diameter increased under MSDI in all seasons. In addition, a micro-spray duration of 2 min decreased the average daily air temperature (T_a) and leaf temperature (T₁) by 0.8 °C and 4.9 °C, respectively, and increased the average daily relative humidity (RH) by 4.3%. However, a micro-spray duration of 1 min reduced the average daily T_a and T₁ by 0.76 °C and 4.6 °C, respectively, but the increase in RH (2.7%) was much lower than that in the single micro-spray duration of 2 min. In addition, the net leaf photosynthetic rate (P_n), the effective quantum yield of PSII (Φ_PSII), and the stomatal conductance (G_s) of both plants increased under MSDI compared with DI. A K-means analysis showed that MSDI could eliminate the adverse effects of sunlight stress on cucumbers and tomatoes. MSDI increased the yield of cucumber and tomato plants by 18.9% and 40.4%, respectively. The fruit weight of cucumber (2017) and tomato (2021) under MSDI did not increase significantly compared to DI, which indicates that MSDI mitigated heat stress, prevented flowers from being burned, and increased the number of fruits. The results also suggested that the total soluble solids in the cucumbers and tomatoes showed no significant differences (p > 0.05) between the two treatments in four seasons. In conclusion, the MSDI system can be an appropriate strategy for the irrigation needs and climate control of plants grown in greenhouses during the hot season.

1. Introduction

With the development of greenhouses in agriculture, plants can be cultivated by providing the required environmental conditions for better growth and sustainable food production [1]. Furthermore, greenhouses allow plants to be grown in any season that would otherwise inhibit their growth. Greenhouse technologies have promoted science-based solutions for optimal plant production by adjusting internal climatic factors, which creates more economic value [2].

In recent years, climate change and extreme weather events, such as increasingly frequent heat stress, have increased [3,4], and high-temperature stress on plants from the internal environments of greenhouses is expected to affect plant growth and development [5]. During the summer in South China, the air temperature inside greenhouses reaches 40 °C or more. This temperature-increasing trend may seriously affect average plant growth [6] because the air temperature increases the heat load on the plants. An extremely high temperature also increases plant water vapor demand, leading to higher leaf transpiration rates and stomatal closure, which results in higher leaf temperature and a decrease in net photosynthetic rate [7]. Thus, plant growth is inhibited by several factors, such as leaf wilting or curling, which can affect plant yield and quality [8,9]. In addition, because of the above issue, it is essential to cool the air temperature inside greenhouses, which requires an appropriate method to cool the air temperature around the plant canopy. The current common cooling methods used in greenhouses include natural ventilation, shade nets, and wet curtain fans, but these methods have several drawbacks. In addition, natural ventilation is ineffective when the temperature difference between the indoor and outdoor environment is significantly higher, sunshades will affect normal photosynthesis in crops, and wet curtain fans are costly and difficult to maintain. Evaporative cooling can effectively reduce the increasing heat of plants and drought stresses [10]. Micro-spray, with its evaporative cooling characteristics, is one of the most commonly used irrigation methods to improve the growing environment of crops under greenhouse conditions [11,12]. A micro-spray cooling system that applies water near the plant canopy can enhance evaporative cooling during periods of heat stress in plant growth [13]. Previously, the micro-spray cooling system was investigated in many studies worldwide on different plants.

Micro-spray reduces air temperature and increases the relative humidity [10], photosynthetic rate, and stomatal conductance [14,15] and increases the effective quantum yield of PSII plants [6]. However, the effects of micro-spray on plant transpiration rates can vary. Previously, it was found that it reduced maize and jujube transpiration rates by 0.75–1.03 mmol m⁻²s⁻¹ and 0.61–0.66 mmol m⁻²s⁻¹, respectively [16]. Still, one study showed that micro-spray increased globe artichoke transpiration rates by about 0.84 mmol m⁻²s⁻¹ [15].

Cucumbers (Cucumis sativus L.) and tomatoes (Solanum lycopersicum L.) are essential horticultural crops planted in greenhouses globally. In a greenhouse, fluctuations in air temperature can cause heat stress and induce leaf and stem wilting, which has a significant negative effect on average growth, quality, and photosynthesis in cucumbers (air temperature over 30 °C) and tomatoes (air temperature over 35 °C) [17,18]. The susceptibility of cucumbers and tomatoes to high temperatures varies according to genotype and growth stage [19]. So, studying the physiological effects of micro-spray on different crops is essential in designing a proper and scientific micro-spray system.

In a micro-spray system, the amount of water sprayed can be small in the form of droplets, as the aim is not to wet the rooting zone but to increase the evaporative cooling around the crops. These droplets can potentially reduce the amount of irrigation but not enough to meet a plant’s irrigation needs [20]. Therefore, micro-spray can be used as a supplementary irrigation method with drip irrigation for plants in greenhouses to avoid high-temperature stresses. Drip and micro-spray irrigation have recently been used in China to improve crop growth and regulate the micro-climate in greenhouses [21,22]. For this combined irrigation method, the operating time of micro-spray needs to be explored. Therefore, rapid feedback and accurate techniques, such as chlorophyll fluorescence technology, should be used to detect plant physiology responses to micro-spray under heat-stress conditions.

Chlorophyll fluorescence technology has been used to study the distribution of excitation energy inside the photosynthetic apparatus. It is used to explore the mechanism and regulation of photosynthesis as an indicator to assess plant growth responses [23,24]. It is also applied to monitor the physiological status of crops and discover the effects of biotic and abiotic stresses on crops [25]. For abiotic stresses, such as heat stress, several studies have shown that chlorophyll fluorescence analysis can be used as a sensitive indicator of chloroplast membrane damage and changes in the function of the photosynthetic apparatus [26]. Chlorophyll fluorescence parameters, such as the effective quantum yield of PSII (Φ_PSII), reflect the photosynthetic efficiency measured under natural light conditions and are excellent parameters for monitoring heat stresses [27]. Studies have shown that photosynthetically active radiation (PAR) is the most critical factor influencing the Φ_PSII, and there is a negative and significant correlation between PAR and the Φ_PSII [6]. The K-means clustering algorithm is considered one of the research community′s most influential and popular data mining algorithms. It has the remarkable features of being fast and easy to implement [28,29]. The K-means clustering algorithm has been applied in agriculture, where it is used to determine the change in moisture content corresponding to different temperature ranges [30].

Therefore, the objectives of this study are as follows: (1) to investigate the effects of combined micro-spray and drip irrigation (MSDI) and drip irrigation (DI) on plant growth and fruiting; (2) to analyze the effects of MSDI and DI on daily changes in the net leaf photosynthetic rate (P_n), the transpiration rate (T_r), and stomatal conductance (G_s) at the leaf scale; (3) to study the effect of PAR on the effective quantum yield of PSII (Φ_PSII) of crops using K-means cluster analysis; and (4) to explore the specific effect of micro-spray on the Φ_PSII of crops under MSDI and DI.

2. Materials and Methods

2.1. Experimental Site

The experiment was conducted in a greenhouse in Zhenjiang, Jiangsu Province, China (32°20′ N, 119°45′ E), for two growing seasons of cucumber, in 2017 and 2018 (transplanted in April and harvested in July), and tomato, in 2021 and 2022 (transplanted in March and harvested in July). The greenhouse is 32 m in length, 20 m in width, and 3.8 m in height, with the longer side oriented east–west and covered with 4 mm float glass with a transmittance greater than 89%. The greenhouse has 5 soil troughs for growing plants (1500, 65, and 30 cm in length, width, and depth, respectively), with an aisle (0.85 m in width) between each soil trough. In the summer, the maximum air temperature outside the greenhouse exceeds 30 °C, and the air temperature inside reaches over 40 °C. The daily irrigation of plants in the four-year experimental treatments was based on the daily evaporation data measured with the evaporation pans (20 cm in diameter).

2.2. Experimental Design

The experiment was conducted using cucumber (Cucumis sativus L., variety Youliang 3–2) and tomato (Solanum lycopersicum L, variety Hezuo 908) in four growing seasons (two growing seasons each for cucumber (2017 and 2018) and tomato (2021 and 2022)) under two treatments: drip irrigation (DI) and micro-spray combined with drip irrigation (MSDI). We set up the experimental area to coincide with the cycle of local production activities so that the results of the trials could be applied to actual production activities. Also, the fertigation for cucumber and tomato plants under DI and MSDI was applied as per the local practice at four different growing stages.

Cucumber seeds were sown in transplant trays on 21 February 2017 and 25 February 2018, and then 32 seedlings were transplanted into soil troughs on 6 April 2017 and 11 April 2018, respectively. Moreover, tomato seeds were sown in transplant trays on 26 January 2021 and 13 January 2022. Then, 50 seedlings were transplanted into soil troughs on 20 March 2021 and 1 March 2022, respectively. Micro-spray was applied for cucumbers 30 days after transplanting (DAT) in 2017 and 25 DAT in 2018, and for tomatoes, 40 DAT in 2021 and 55 DAT in 2022. However, the plant densities were maintained at 3.3 plants/m² for cucumber and 5.1 plants/m² for tomato. In addition, the plants were separated by placing plastic film in the middle of the DI and MSDI treatments.

The drip irrigation system was used to supply water to tomato and cucumber plants throughout the growing seasons during four stages: seedling, flowering, fruiting, and fruit picking. The amount of applied water was 40 mm and was supplied in 2–5 days intervals after transplanting to ensure deeper roots and vigorous plant growth. The daily irrigation volume of the drip irrigation system was referred to as the cumulative pan evaporation (E_pan), measured using two evaporation pans (20 cm in diameter). The pans were placed inside the greenhouse and 5 cm above the plant canopy at two locations [31]. The evaporated volume from each pan was measured at 6:00 pm daily. Before the next reading, the pans were cleaned and refilled with fresh water up to 20 mm depth.

In addition, for MSDI treatment, atomization nozzles were used to micro-spray the water. The nozzles were placed 2.5 m above the soil surface. The droplet sizes of the selected nozzles were 65 µm, and the flow rate of the nozzles was 7.5 L/h. For MSDI, amount of drip irrigation was same with DI treatment. The nozzle spraying time and spraying interval time were divided into different levels: for cucumbers (2017 and 2018), the spraying time was 2 min, four times a day (after each hour spraying interval at 9:00, 10:00, 11:00, and 12:00). For tomatoes (2021 and 2022), the spraying time was 1 min, seven times a day (after each hour spraying interval at 9:00, 10:00, 11:00, 12:00, 13:00, 14:00, and 15:00. The nozzles were only opened when the air temperature was above 30 °C (named “high temperature”) [6].

On the other hand, no additional cooling methods were used during the experiment apart from natural ventilation, which ensured that the control variables were observed. The experimental setup is shown in Figure 1.

2.3. Measurements

2.3.1. Meteorological Data

The average daily air temperature (T_a) and relative humidity (RH) inside the greenhouse were collected using a data logger (CR1000, Campbell, Logan, UT, USA). The photosynthetically active radiation (PAR) was measured using a PQS 1 (Kipp & Zonen, Delft, The Netherlands), mounted 2 m above the soil surface. The air temperature and relative humidity probes (HMP155A, Campbell, Logan, UT, USA) were set at the same height as the plant canopy and adjusted over plant growth to measure the air temperature and relative humidity near the plants from both treatments. The accuracy of the data recording sensors was 0.1 °C for air temperature and ±1.7% for relative humidity. All data were collected and averaged for 10 min. The environmental factors recorded during the experimental period are shown in Figure 2.

2.3.2. Plant Growth, Yield, Total Soluble Solids, Irrigation Water Use Efficiency, and Fruit Shape Index

The plant height and stem diameter of plants grown under DI and MSDI were measured at the end of the experiments. Plant height (cm) was measured with a measuring tape. Stem diameter (mm) was measured at a marked place (5 cm above the soil surface) with a vernier caliper. The yield of plants was calculated using a weight balance (accuracy 0.1 g). For total soluble solids (TSS, degrees Brix), three fruits per plant from the same ripening batch with no external defects were selected and tested using the abbe refractometer (WYA-3S, Yidian, Shanghai, China) method. Before the test, the fruits were peeled and pressed, and the paste was filtered through four layers of fine cotton gauze. Each sample was measured three times to minimize observation errors. In addition, the irrigation water use efficiency was calculated using the following equation:

$$WUE=\frac{Y}{I}\times 100$$

(1)

where WUE is irrigation water use efficiency (kg/m³), $Y$ is tomato yield (t/ha), and $I$ is irrigation water amount (mm). The weight per fruit is the average value of each cucumber and tomato plant.

The fruit shape index is the ratio of the longitudinal to the transverse diameter of the fruit and was measured with an electronic vernier caliper (accuracy of 0.01 mm) and a meter ruler (accuracy of 1 mm).

2.3.3. Leaf Gas Exchange

During the micro-spray irrigation treatment, a photosynthesis system (GFS-3000, WALZ, Effeltrich, Germany) was used to measure the rate of photosynthesis (P_n), the rate of water loss (T_r), and the conductance of the stomata (G_s). The P_n, T_r, and G_s were measured on a clear, cloudless, sunny day during both the cucumber and tomato ripening and harvesting periods. For measuring leaf gas exchange parameters, six plants with uniform size and growth were randomly selected; further, three plant leaves were randomly chosen from the fifth functional leaf below the top leaf of each plant. Each leaf was measured three times, and the average values were taken every 2 h between 08:00 and 18:00. The RH and leaf T_l in the Walz leaf cuvette are the same as the environment in which the crop is located. It ensures that the photosynthetic rate is measured under actual conditions because the CO₂ variable in the greenhouse does not vary much during the day (around 466 ppm). To avoid interference by human movement, we used a CO₂ buffer tank to ensure that the ambient CO₂ variable is the same when measuring the gas exchange under MSDI and DI.

2.3.4. Chlorophyll Fluorescence, Photosynthetically Active Radiation, and Leaf Temperature

Chlorophyll fluorescence parameters, including the effective quantum yield of PSII (Φ_PSII and Equation (2)), were monitored using a PAM WinControl-3 controlled MONI-PAM multi-channel chlorophyll fluorometer (Monitoring-PAM, WALZ, Effeltrich, Germany). The chlorophyll fluorometer consists of 4 measuring heads, a data collection system, and a personal computer interface box. Two plants under DI and MSDI were selected to monitor the chlorophyll fluorescence parameters. The fifth functional leaf below the top leaf of two randomly chosen plants was used for the measurements. The selected plants were changed every morning before 9:00 am to eliminate individual differences. The photosynthetically active radiation (PAR) and leaf temperature (T_l) were also recorded with Φ_PSII. Φ_PSII can reflect the actual photosynthetic efficiency measured under natural light conditions. The MONI-PAM data collector was set up to collect data automatically every 10 min.

$${\Phi }_{PSII}=\frac{\left(F_m^{\prime }-F_s^{\prime }\right)}{F_m^{\prime }}$$

(2)

where Φ_PSII is the effective quantum yield of photosystem II, F’_m is the maximal fluorescence yield, and F’_s is the steady-state fluorescence yield.

2.3.5. K-Means Clustering Algorithm

The K-means clustering algorithm is an unsupervised algorithm usually used to classify an area. This study used the K-means clustering algorithm to classify changes in the Φ_PSII in different PARs. Clustering takes k points in the space as the centroid and minimizes the sum of squares of the distance from each data point to its assigned cluster center [32]. The centroid value of each cluster is updated successively using an iterative optimization procedure until the best clustering result is obtained.

The calculation process of the algorithm is as follows:

We randomly selected k points as the initial center C_i (i = 1, 2, 3, …, k). In this experiment, the k values were 2 and 3. Then, we calculated the distance from any sample point P_t (t = 1, 2, 3, …, n) to C_i, where n is the number of sample points.

$${\mathrm{D}}_{\left(t,i\right)}=\sqrt{{\left(P_t-C_i\right)}^2}$$

(3)

The sample point was classified into the cluster with the centroid and the smallest distance. The mean was used to update the centroid value of each group.

$$C_i=\frac{1}{n_i}\sum _{t=1}^{n_i}{P}_{it} i=\mathrm{1,2},\cdots k$$

(4)

Equations (2) to (4) were repeated until the centroid value of each cluster stopped changing.

The value of the contour coefficient is in the range of (−1, 1). The closer the samples of the same category are to each other, and the further the samples of different types are from each other, the higher the score and the better the clustering effect.

2.4. Data Analysis

Statistical analyses were performed using SPSS 22 (SPSS Inc., Chicago, IL, USA) to detect outliers and to perform the least significant difference (LSD) test to determine the statistically significant difference between two irrigation treatments at a significance level of p < 0.05. Figures were produced using Origin 2022 (OriginLab Inc., Hampden, MA, USA).

3. Results

3.1. Effects of MSDI on Plant Growth

Plant height and stem diameter are the most visible parameters that can be measured to show the difference between the two treatments (MSDI and DI). In all four planting seasons, MSDI improved plant height and stem diameter compared with DI for cucumbers and tomatoes throughout the measuring periods. Figure 3 shows the variations in plant height and stem diameter of cucumbers and tomatoes planted over four years. Significant differences (p < 0.05) in the stem diameter (Figure 3a, 2017) and plant height (Figure 3b, 2018) of cucumbers and in the plant height and stem diameter of tomatoes (Figure 3d, 2022) under the two different treatments were detected. MSDI increased cucumber and tomato plant height by 7.1% and 19.2%, and stem thickness by 7.0% and 6.5%, respectively. Comparing the data from four planting seasons, there was no certain pattern in the MSDI treatment indicating significant differences in plant height and stem diameter between the two crops in different seasons. However, plant height and stem diameter were greater in the MSDI treatment than the DI treatment, indicating a positive effect of MSDI on plant growth.

3.2. Response of Air Temperature, Relative Humidity, Leaf Temperature, and the Effective Quantum Yield of PSII to Micro-Spray

In this study, air temperature (T_a), relative humidity (RH), and leaf temperature (T_l) were continuously monitored from 08:00 to 18:00. The measured data from 42 DAT in 2017 and 47 DAT in 2021 were used as examples for analysis. T_a, RH, and T_l were basically the same under the two irrigation treatments when the micro-spray was not turned on at 08:00, but then T_a, RH, and T_l showed differences with an increase in solar radiation. When the micro-spray was switched on, T_a and T_l both decreased and RH increased. The arrows in Figure 4 show the peaks after the micro-sprays; the cooling effect peaked about 10 min after the end of a single micro-spray and then diminished over time.

Figure 4a,b shows the effect of a single micro-spray duration of 2 min on T_a and RH of the environment and T_l of cucumbers and tomatoes, respectively. The change in T_a and RH of the MSDI treatment for cucumbers and tomatoes was related to the initial T_a; the higher the T_a, the smaller the change. Figure 4a showed that a single micro-spray duration of 2 min decreased the average daily T_a and T_l by 0.8 °C and 4.9 °C, respectively, and increased RH by 4.3%. Figure 4b shows that a single micro-spray duration of 1 min had a similar cooling effect to the single micro-spray duration of 2 min, reducing the average daily T_a and T_l by 0.76 °C and 4.6 °C, respectively. However, a single micro-spray duration of 1 min increased RH by 2.7%, which was less than the effect of the single micro-spray duration of 2 min.

Changes in the effective quantum yield of PSII (Φ_PSII) of the crop in the daytime under different irrigation treatments are shown in Figure 5. After the micro-spray was turned on, the increase in Φ_PSII was 40–53% and 6–27% for cucumbers when T_a was less than or more than 33 °C, respectively. The increase in Φ_PSII was 49–62% and 12–29% for tomatoes when T_a was less than or more than 35 °C, respectively. The Φ_PSII of cucumbers and tomatoes under MSDI gradually decreased before 10:00 and increased after 10:00, and the lowest Φ_PSII were 0.418 and 0.384, respectively, because the crop stomatal conductance decreased to avoid excessive evaporation of water from the plants. The Φ_PSII of cucumbers and tomatoes under DI increased gradually from 8:00 to 18:00. Before the micro-spray was turned on, the Φ_PSII of cucumbers and tomatoes differed under different irrigation treatments, and the Φ_PSII of the MSDI treatment was higher than that of the DI treatment. At the end of the day of micro-spraying, the variability in Φ_PSII for cucumbers and tomatoes decreased over time with different irrigation treatments. But the differences in Φ_PSII between the two treatments still existed, which proved that the effect of MSDI was not transient during the time when the micro-spray was turned on. There was a potential for the Φ_PSII of cucumbers and tomatoes to be affected when the micro-spray was turned off.

3.3. Net Photosynthetic Rate, Transpiration Rate, and Stomatal Conductance

Figure 6 shows the variation in the transpiration rate (T_r), the photosynthetic rate (P_n), and the stomatal conductance (G_s) of cucumbers and tomatoes in four seasons. Comparing the T_r of cucumbers and tomatoes, the changes in T_r under two irrigation treatments showed increasing and then decreasing trends. MSDI increased the average daily T_r by 73.0% and 46.2% for cucumbers in 2017 and 2018, respectively (Figure 6a,d), but did not increase the T_r of tomatoes much in 2021 and 2022 (Figure 6g,j). The T_r of cucumbers and tomatoes reached their maximum at 12:00.

Comparing P_n and G_s of the two crops, the changes in P_n and G_s tended to be the same, reaching their maximum around 10:00 and then gradually decreasing. Cucumbers and tomatoes closed parts of their stomata for their own protection in order to prevent their leaves from losing water severely due to the increased air temperature, which led to the rapid decrease in P_n and G_s. MSDI had the greatest effect on improving the P_n and G_s of cucumbers and tomatoes at 10:00. With the two types of irrigation, the P_n (Figure 6b,e) and G_s (Figure 6c,f) of cucumbers converged after 14:00. For tomatoes, the improvement in P_n (Figure 6h,i) and G_s (Figure 6k,l) with MSDI also converged after 18:00. The increase in operation time of the micro-spray throughout the day prolonged the time limit of the improvement in P_n and G_s under MSDI. The daily mean P_n was increased by 84.6% in 2017 (Figure 6b), 82.7% in 2018 (Figure 6e), 55.3% in 2021 (Figure 6h), and 29.1% in 2017 (Figure 6k) for cucumbers and tomatoes under MSDI compared with DI. The daily mean G_s increased by 112.3% in 2017 (Figure 6c), 148.1% in 2018 (Figure 6f), 45.1% in 2021 (Figure 6i), and 46.6% in 2022 (Figure 6l) under MSDI compared with DI. The changes in the P_n and G_s of cucumbers with high air temperatures showed a single peak trend of increasing and then decreasing under the two irrigation treatments, but the changes in the P_n and G_s of tomatoes showed a double peak, with a small increase in P_n and G_s at 16:00 after experiencing a photosynthesis midday break. Photosynthesis in cucumbers is closely related to stomata. The photosynthetic rate of cucumbers showed a single peak because cucumbers are less heat-tolerant than tomatoes. The T_a of the greenhouse decreased after 16:00. At this time, the stomata of tomato leaves partially opened for photosynthesis, but the stomata of cucumber leaves opened to a lesser extent, and the overall photosynthesis was in the form of a single peak. This proved that the spray irrigation time periods of 9:00–12:00 for cucumbers and 9:00–15:00 for tomatoes are scientifically effective at high temperatures.

The relationships between Φ_PSII and T_r, P_n, and G_s are shown in Figure 7. As Φ_PSII increased, the T_r, P_n, and G_s of cucumbers and tomatoes showed decreasing trends during the day. The T_r and P_n of cucumbers (Figure 7a,b) and tomatoes (Figure 7d,e) had strong linear negative correlations with Φ_PSII (p < 0.05). The enhancement in T_r and P_n of cucumbers and tomatoes under the MSDI treatment decreased with the increase in Φ_PSII, which indicated that micro-spray is effective in enhancing crop physiology when heat stress is high.

3.4. Comparison of Φ_PSII at Different PARs

PAR is the environmental factor that most influences Φ_PSII. Figure 8 shows that PAR is the most important factor influencing Φ_PSII, and there is a negative and significant correlation between PAR and Φ_PSII. The Φ_PSII of cucumbers (Figure 8a) and tomatoes (Figure 8b) under different PAR levels in the MSDI treatment was higher than in the DI treatment, which meant that the plants were less affected by stress from the outside world. The average increase in the Φ_PSII of cucumbers was 13.36% for a single micro-spray duration of 2 min and 9.74% for tomatoes for a single micro-spray duration of 1 min. With the increase in PAR, the Φ_PSII of cucumbers and tomatoes decreased sharply and then slowly. Under the same degree of light stress, the Φ_PSII of tomatoes was smaller than that of cucumbers, indicating that the response of tomatoes to light was more intense than that of cucumbers.

The efficient use of light energy (EUL) is the amount of energy that crops can absorb for photosynthesis. The EUL was positively correlated with PAR for both cucumbers and tomatoes. As PAR increased, light stress and the EUL also increased, which suggested that mitigating heat stress by lowering PAR could affect the EUL. Micro-spray increased the EUL of cucumbers and tomatoes, which suggested that MSDI can mitigate heat stress without affecting the normal growth of crops.

The contour coefficients of different clusters for cucumbers and tomatoes obtained using the K-means cluster analysis are shown in

Table 1. Contour coefficient of different clusters for cucumbers and tomatoes obtained using K-means cluster analysis.

Plants	Cucumber				Tomato
K-means cluster	DI K = 2	DI K = 3	MSDI K = 2	MSDI K = 3	DI K = 2	DI K = 3	MSDI K = 2	MSDI K = 3
Contour coefficient	0.815	0.717	0.783	0.649	0.663	0.596	0.649	0.619

. When K = 2, the contour coefficients of cucumbers and tomatoes were greater than when K = 3 under the two irrigation treatments. The analysis based on the K-means clustering algorithm (K =  2 and 3) is shown in

Table 2. Cluster centroid coordinates of cucumbers and tomatoes in the K-means cluster analysis.

Plant	Irrigation	K = 2	PAR	ΦPSII	K = 3	PAR	ΦPSII
Cucumber	DI	1	29.5	0.639	1	12.26	0.689
		2	305.13	0.366	2	114.72	0.418
					3	351.56	0.363
	MSDI	1	34.92	0.692	1	12.26	0.730
		2	330.27	0.46	2	114.72	0.525
					3	351.56	0.455
Tomato	DI	1	59.11	0.38	1	40.98	0.411
		2	255.38	0.215	2	153.01	0.25
					3	309.45	0.197
	MSDI	1	71.26	0.404	1	29.18	0.496
		2	299.15	0.255	2	146.32	0.277
					3	351.59	0.237

Note: When K = 2 or 3, we divided the effect of PAR on the Φ_PSII of cucumbers and tomatoes into two and three parts, respectively.

. When K =  3, the average Φ_PSII of the second and third clusters were about (0.019–0.033)/100 μmol m⁻²s⁻¹ in four seasons, but the difference was so small that it was negligible. When K =  2, the hierarchy was clear, and the classification effect was the best [30]. Therefore, we used the divided PAR ranges and the changes in Φ_PSII when K =  2 as a guide to study the improvement following MSDI in plants under heat stress. For the variable PAR, significance is presented at the level of K =  2, indicating that the variable PAR is significantly different between the categories delineated with the cluster analysis when K =  2.

Cluster centroid coordinates can represent the data in a group; analyzing the change in cluster centroid coordinates can help determine the change in data. We analyzed the relationship between PAR and Φ_PSII and opted to use the K-means clustering algorithm for analysis. The classification level was clear when K =  2 (Table 2), which was consistent with the actual application [33]. So, we classified Φ_PSII at different PARs into two categories, i.e., category 1 and category 2, respectively, based on the effect of different PARs on Φ_PSII.

As shown in Table 2, MSDI has an enhancement rate of 8.29% for the Φ_PSII of cucumbers in category 1 and 25.7% in category 2. The improvement rate of MSDI for the Φ_PSII of tomatoes is 6.32% in category 1 and 18.6% in category 2. Category 2 is for high light conditions, and high temperatures usually occur in high light conditions, which proves that MSDI improves the Φ_PSII of cucumbers and tomatoes more in high-temperature environments. The cluster centers in the cucumber and tomato cluster analyses were both improved in the category 1 case by 5.42 μmol m⁻²s⁻¹ and 12.15 μmol m⁻²s⁻¹ of PAR, respectively, proving that MSDI could eliminate the negative effects of light stress for cucumbers and tomatoes.

The K-means clustering analysis results are shown in Figure 9. The Φ_PSII response to light stress was reduced when PAR exceeded 180 μmol m⁻²s⁻¹ for cucumbers and 160 μmol m⁻²s⁻¹ for tomatoes under DI. For cucumbers under DI, we classified PAR into low light (PAR < 180 μmol m⁻²s⁻¹) and high light (PAR > 180 μmol m⁻²s⁻¹), and for tomatoes under DI, we classified PAR into low light (PAR < 160 μmol m⁻²s⁻¹) and high light (PAR > 160 μmol m⁻²s⁻¹) (Figure 9). The Φ_PSII of cucumbers and tomatoes was more sensitive to light stress under low light, and the Φ_PSII range of variation was 0.22/100 μmol m⁻²s⁻¹ (cucumber) and 0.33/100 μmol m⁻²s⁻¹ (tomato). The enhancement in PAR under high light had less effect on Φ_PSII (0.03/100 μmol m⁻²s⁻¹, cucumber) (0.02/100 μmol m⁻²s⁻¹, tomato). The response of Φ_PSII to PAR was greater for tomatoes than cucumbers, but the maximum light energy conversion efficiencies of both tomatoes and cucumbers tended to be stable in the absence of light at around 0.8.

3.5. Response of Yield, Weight per Fruit, Total Soluble Solids, Water Use Efficiency, and Fruit Shape Index Changes to Micro-Spray

The total fruit yield (Y_i), weight per fruit (W_f), and total soluble solids (TSS) during the four growing seasons are illustrated in Figure 10. The Yi of cucumbers and tomatoes under DI were 1115.6 g/plant and 994.3 g/plant, respectively. MSDI increased the Yi of cucumbers and tomatoes by 18.9% and 40.4%, respectively, compared with DI. In contrast, the W_f of cucumbers (2017) decreased under MSDI compared with DI. The W_f of cucumbers (2018) and tomatoes (2021, 2022) increased by 16.7%, 9.9%, and 20.9% under MSDI compared with DI, respectively. In 2017, the W_f of cucumbers did not increase under MSDI, but Y_i increased, which indicated that MSDI mitigated heat stress, prevented flowers from being burned, and increased the number of fruits. Total soluble solids (TSS) is an essential parameter for judging fruit quality. The results indicated that the TSS of cucumbers and tomatoes showed no significant difference (p > 0.05) between the two treatments. The TSS of cucumbers was 3.4% under the DI and MSDI treatments, and the TSS of tomatoes was 5.1% and 5.0% under DI and MSDI, respectively. The TSS under DI was greater than MSDI, but the difference was not significant.

Irrigation water use efficiency (WUE) as well as the fruit shape index for cucumbers and tomatoes under two different treatments in four years are shown in

Table 3. Irrigation water use efficiency and the fruit shape index for cucumbers and tomatoes under two treatments in four growing seasons.

Plant	Year	Irrigation	Yield (kg/m2)	Water Consumption (m3/m2)	Water Use Efficiency (kg/m3)	Fruit Shape Index
Cucumber	2017	DI	3.63 ± 1.71 a	0.10	36.89 ± 17.41 a	8.43 ± 0.91 a
		MSDI	4.28 ± 1.14 a	0.16	26.85 ± 7.19 a	8.71 ± 1.04 a
	2018	DI	3.73 ± 1.20 a	0.11	34.41 ± 11.11 a	8.17 ± 0.86 a
		MSDI	4.47 ± 1.42 a	0.17	27.07 ± 8.58 a	8.28 ± 0.91 a
Tomato	2021	DI	4.46 ± 1.60 a	0.13	35.59 ± 12.77 a	0.95 ± 0.10 a
		MSDI	5.85 ± 2.21 a	0.20	29.08 ± 11.00 a	0.94 ± 0.07 a
	2022	DI	5.74 ± 1.47 a	0.14	40.22 ± 10.33 a	0.92 ± 0.10 a
		MSDI	8.47 ± 0.89 b	0.25	33.64 ± 3.41 a	0.94 ± 0.09 a

Note: Bars with a different letter are significantly different between MSDI and DI in the same years at p < 0.05.

. Turning on the micro-spray increased water consumption, but there was no significant difference in WUE between the two treatments for cucumbers and tomatoes. The yield of cucumbers and tomatoes under the MSDI treatment was higher than the DI treatment. There was no significant change in the fruit shape index of cucumbers and tomatoes under the two irrigation treatments. MSDI reduced WUE compared with DI but increased yields and improved economics, which suggests that micro-spray can be used in greenhouses and provides a basis for field applications.

4. Discussion

In the context of climate change, extreme events are listed among the primary abiotic stresses that negatively affect crop growth and yield, and the plant response to this stress varies according to the length and size of the temperature rise and the crop stage at which the stress occurs [34]. The heat stress problem has always been a complicated problem that needs to be solved in daylight conditions in greenhouses. Therefore, an appropriate cooling method is necessary for successful crop production in greenhouses. Evaporative cooling is the most effective cooling method among various cooling technologies [10] that is applied with roof evaporative cooling, condensing fan cooling systems, micro-spray cooling systems, etc. It has been pointed out that the effects of different micro-spray water volumes and frequencies on crops under high-temperature stress are not the same [14] and also depend on the effectiveness of different crops in responding to the micro-spray [16]. This indicates that the appropriate water volume and frequency of micro-spray are also what we need to study at present. For this reason, the discussion is structured into three sub-sections. The first sub-section discusses the effects of MSDI on crop growth as well as fruit quality. The second sub-section discusses how MSDI on a daily scale affects the crop micro-climate as well as leaf temperature (T_l) during multiple irrigations. The third sub-section discusses the effects of MSDI on short-term leaf gas exchange and chlorophyll fluorescence on a daily scale and investigates the effects of MSDI on cucumbers and tomatoes during the growth stage based on physiological conditions.

4.1. Effects of Micro-Spray on Crop Growth and Fruit Quality

Plant height is a function of the number of nodes and the length of each internode, both of which are strongly influenced by greenhouse temperatures [35], and the overgrowth of plants at high temperatures occurs [36]. Micro-spray changes the air temperature (T_a) during the day, thus affecting internode spacing. Stem diameter has been considered the best predictor of plant growth and is usually discussed with water status or drought stress [37,38]. In this study, it was found that the plant height and stem thickness of cucumbers and tomatoes increased under MSDI in four planting seasons. Comparing plant height and stem thickness under the two irrigation treatments, there was a significant difference (p < 0.05) in the plant height of cucumbers in 2018 and tomatoes in 2022, and in the stem thickness of cucumbers in 2017 and tomatoes in 2022. The stem diameter of cucumbers and tomatoes under MSDI was greater than it was under DI, probably due to the improved environment caused by the micro-spray, which reduced limitation due to heat stress.

Crop yield is most affected by fertilizer and irrigation. In this study, the same fertilizer was applied to cucumbers and tomatoes under both treatments and other field management was carried out in the same way. The effects of micro-spray on plant yield and quality have been extensively studied, and it was found that micro-spray increased grape and globe artichoke yields by 27.3% and 51.9%, respectively [10,15]. MSDI increased the yield of cucumbers and tomatoes by 18.9% and 40.4%, respectively, compared with DI in this study, which showed that MSDI could improve the growing environment of cucumbers and tomatoes. Although micro-spray increased water consumption, it improved crop yield under MSDI, which improved the economic efficiency of the crops. In general, crop yield can be increased by promoting leaf photosynthesis. In this study, weight per fruit (W_f) did not increase, but total yield increased in cucumbers (2017) and tomatoes (2021) under MSDI compared to DI, suggesting that MSDI mitigated heat stress, avoided flower scorching, and increased fruit number.

Total soluble solids (TSS) have been used by many researchers to estimate the quality of fruits [39]. TSS is mostly made up of sugars and acids and has been linked to water stress [40] (Zhang et al., 2017). It was found that there was a decreasing trend in TSS with increasing water use [41], which can be used as an index to quickly evaluate the comprehensive quality of fruits [30]. In this study, TSS was higher under DI than it was under MSDI, but there was no significant difference (p > 0.05) between the two treatments over the four seasons. This suggests that the tomato plants did not really absorb the water from the micro-spray because it mostly evaporated.

4.2. Effects of Micro-Spray on Crop Micro-Climate and Leaf Temperature in Daily Scaled Blades

Suitable air temperature (T_a) and relative humidity (RH) are important prerequisites for ensuring the photosynthetic capacity of crops, and high temperatures interfere with vegetative development, thereby adversely affecting the growth stage of crops [42]. The high temperature received by different crops varies with phenological stage, organ, and involved tissue. The degree of change in T_a and RH is not the same for different sprinkler durations or sprinkler frequencies in a single pass. Different amounts of water in a single pass are set for different crops with different maximum temperatures that they can tolerate so that the T_a at which different crops are exposed reaches their acceptance level. Prolonged evaporative cooling in greenhouses, such as micro-spray, also increases air humidity [43]. High humidity inhibits plant transpiration, making plants more susceptible to disease. So, we need to study how to set the right duration of sprinkler irrigation to achieve the best cooling effect while saving water. Changes in environmental factors outside the crops are the indirect effects of micro-spray on the crops. A plant’s response is a direct estimate of the cooling method, and leaf temperature (T_l) is a better way to determine the status of a crop than air temperature because it is more directly related to the plant’s physiological response. Micro-spray cooling has been shown to have a positive effect on changes in T_l [15].

In this study, two sets of single micro-spray irrigation water volumes were set for different crops, and the changes in T_a and T_l for the single micro-spray duration of one minute and two minutes did not differ much. A single micro-spray duration of 1 min reduced T_a and T_l by 0.76 °C and 4.6 °C, respectively. A single micro-spray duration of 2 min reduced T_a and T_l by 0.8 °C and 4.9 °C, respectively. Both spray durations of the micro-spray were off for ten minutes to maximize the cooling effect. However, the difference in RH between one minute and two minutes of spray was only 2.7% and 4.3%, respectively, and the RH in the greenhouse after micro-spray did not exceed 60%, which proved that micro-spray showed an excellent ability to improve greenhouse heat stress and avoid high humidity.

4.3. Effects of Micro-Spray on Leaf Gas Exchange and Φ_PSII

Crop physiological conditions such as the transpiration rate (T_r), photosynthetic rate (P_n), and stomatal conductance (G_s) can directly reflect the real condition of crops, and micro-spray can increase the P_n and G_s of the crop [14,15]. However, the response of T_r to micro-spray is not clear, and it has been pointed out that a change in the micro-climate leads to a reduction in water transpired by the leaves. G_s prevents an increase in T_l and a decrease in photosynthesis [7]. There are studies that point out that micro-spray increases the T_r of crops. In this study, a single micro-spray with different amounts of water increased the T_r of cucumbers and tomatoes but did not show significant differences. MSDI increased the daily mean P_n by 84.6% and 82.7% (2017 and 2018) and by 55.3% and 29.1% (2021 and 2022) in cucumber leaves (2017, 2018) and tomato leaves (2021, 2022), and increased the daily mean G_s by 112.3% and 148.1% (2017 and 2018) and by 45.1% and 46.6% (2021 and 2022), respectively. The changes in the P_n and G_s for cucumber leaves with high air temperature showed a single peak trend of increasing and then decreasing under the two irrigation treatments, but the changes in the P_n and G_s of tomato leaves showed a double peak, with a small increase in P_n and G_s at 16:00 after experiencing a photosynthesis midday break. This proved that the spray irrigation periods of 9:00–12:00 for cucumbers and 9:00–15:00 for tomatoes are scientifically effective at high temperatures.

Φ_PSII is widely used as a parameter to reflect the light use efficiency of plants and can reflect the degree of stress suffered by plants [44,45]. Some studies have shown that PAR is the most influential parameter on Φ_PSII. In this study, the values of Φ_PSII varied with PAR, which is similar to previous studies [46]. The daily average Φ_PSII of cucumbers and tomatoes increased under MSDI. Before the micro-spray irrigation was turned on, the Φ_PSII of cucumbers and tomatoes under two irrigation treatments were different, and the Φ_PSII of MSDI was higher than that of DI. The variability in the Φ_PSII of cucumbers and tomatoes decreased with time under the two irrigation treatments, but the difference still existed at the end of one day of micro-spray irrigation. This demonstrated that the effect of MSDI on enhancing Φ_PSII was not transient when the spray irrigation was switched on. The Φ_PSII of cucumbers and tomatoes was also affected when the spray irrigation was not switched on.

The K-means clustering algorithm is considered to be one of the most influential and popular data mining algorithms in the research community. It has the remarkable feature of being fast and easy to implement and can be widely used in the field of data mining [28,29]. Changes in temperature and substrate water content have been explored using the K-means clustering method [30]. In this study, we analyzed the relationship between the PAR and the change in Φ_PSII and opted to use the K-means clustering method for analysis. When K = 2, the classification level was clear (Table 2), and there were significant differences in the PAR variables between the categories classified with the cluster analysis, proving that it is scientifically feasible to classify the effect of PAR on Φ_PSII into two parts. The PAR and Φ_PSII corresponding to the center coordinates of the clusters of cucumbers and tomatoes under MSDI were increased. It was demonstrated that MSDI increased the Φ_PSII of cucumbers and tomatoes and alleviated light stress. The first cluster of cucumbers and tomatoes had a smaller temperature range but the largest change in Φ_PSII, indicating that the Φ_PSII of cucumbers and tomatoes responded more to PAR. The second cluster of cucumbers and tomatoes floated less with the change in PAR, and the trend line was very flat (Figure 9). This result suggests that PAR is not the main factor affecting the Φ_PSII of cucumbers and tomatoes in the second cluster.

5. Conclusions

In this study, two irrigation treatments (DI and MSDI) were designed for cucumbers (2017 and 2018) and tomatoes (2021 and 2022) grown in a greenhouse for four seasons. The experiments demonstrated the following:

MSDI increased the plant height and stem diameter of cucumbers and tomatoes in the four seasons, indicating a positive effect of the MSDI treatment on plant growth.

A single micro-spray duration of 2 min decreased T_a and T_l by 0.8 °C and 4.9 °C, respectively, and increased RH by 4.3%. A single micro-spray duration of 1 min decreased the average daily T_a and T_l by 0.76 °C and 4.6 °C, respectively, but the increase in RH was only 2.7%, which was much less than the single micro-spray duration of 2 min.

The P_n and G_s of cucumbers and tomatoes increased under MSDI, with the greatest enhancement at 10:00. At the same time, MSDI also improved the Φ_PSII of cucumbers and tomatoes. The K-means analysis showed that MSDI could eliminate the negative effects of light stress on cucumbers and tomatoes.

MSDI increased the yield of cucumbers and tomatoes by 18.9% and 40.4%, respectively. MSDI mitigated heat stress, prevented flowers from being burned, and increased the number of fruits. The result showed that the TSS of cucumbers and tomatoes showed no significant difference (p > 0.05) between the two treatments in the four seasons.

In conclusion, MSDI can be an appropriate strategy for the irrigation needs and climate control of plants grown in greenhouses during the hot season. What needs to be investigated in the future is the adjustment of the micro-spray running times and the frequency of micro-sprays to cope with more severe environmental changes under different environmental factors.