Effects of Different Cultivation Methods on Yield and Quality of Greenhouse-Grown Tomatoes in the Gobi Desert

Abstract

Optimizing cultivation methods is crucial for enhancing the productivity and sustainability of protected agriculture in water-limited regions. This study systematically evaluated the effects of three cultivation methods—bucket, non-woven bag, and underground trough cultivation—on the root-zone environment, growth, yield, and quality of greenhouse-grown tomatoes (Solanum lycopersicum L. ‘Provence’) in a Gobi Desert facility. Key root-zone parameters (substrate temperature, moisture, and electrical conductivity) and plant agronomic traits (plant height, stem diameter, leaf number, and SPAD) were monitored throughout the growth cycle. Final yield and fruit quality indicators (lycopene, soluble sugars, and vitamin C) were analyzed. The results demonstrated that bucket and underground trough cultivation created a more stable root-zone environment, with better moisture retention and temperature regulation than bag cultivation. These methods significantly improved plant growth, with yield per plant increasing by 23.18% and 18.18% under bucket and underground trough cultivation, respectively, alongside enhanced fruit quality metrics. In conclusion, bucket and underground trough cultivation effectively optimize the root-zone environment, leading to superior tomato growth, yield, and quality compared to traditional non-woven bag cultivation. These methods show significant potential for application in arid and semi-arid regions to support sustainable and efficient greenhouse production.

1. Introduction

Tomato (Solanum lycopersicum L.) is a globally important vegetable crop, and greenhouse cultivation has become crucial for a stable year-round supply, especially in adverse climatic regions [1,2,3,4]. The Gobi Desert, characterized by severe water scarcity, high evaporation, poor soils, and large diurnal temperature fluctuations [5,6], poses great challenges to conventional tomato cultivation, making substrate-based soilless systems increasingly necessary.

In substrate-based cultivation, the root-zone environment (e.g., substrate temperature, moisture, and electrical conductivity (EC)) affects root activity, nutrient uptake, and ultimately tomato yield and quality [7,8,9]. Cultivation structures (buckets, non-woven bags, and underground troughs) differ significantly in shaping root-zone conditions: non-woven bags are low-cost but prone to rapid environmental fluctuations, while buckets and underground troughs may provide more stable conditions [10,11,12,13].

Previous studies have shown that cultivation systems influence tomato growth, but most were conducted under moderate climates; their effectiveness in the extremely arid Gobi Desert remains unclear [14,15,16,17,18]. Additionally, comprehensive evaluations linking cultivation methods, root-zone dynamics, plant growth, and fruit quality are limited.

This study evaluated three cultivation methods (bucket, non-woven bag, and underground trough) in a Gobi Desert greenhouse to: (i) clarify their effects on root-zone stability under arid conditions, (ii) assess impacts on tomato growth, yield, and quality, and (iii) provide a scientific basis for suitable cultivation system selection in arid regions.

2. Materials and Methods

2.1. Experimental Site Overview and Climate Conditions

The experiment was conducted from August 2024 to February 2025 at the Shuixigou Gobi Experimental Base (87°28′ E, 43°27′ N, altitude 1641 m) in Urumqi, Xinjiang, China, which has a typical temperate continental arid climate (mean annual temperature 5–7 °C, precipitation 250–300 mm, potential evaporation 1800–2000 mm). The trial was performed in a single-span greenhouse (110 m × 12 m × 4.5 m) with an integrated environmental control system (monitoring and regulating air temperature and humidity). During tomato growth, the average daily greenhouse air temperature and relative humidity were 22.21 °C and 82.25%, respectively.

2.2. Growing Substrate

All treatments adopted a soilless substrate composed of 60% fine coconut coir (0–6 mm) and 40% coarse coconut coir (10–20 mm). Before transplanting, the initial chemical properties of the substrate were analyzed and were consistent across all experimental units (see

Table 1. Physicochemical properties of substrate.

Bulk Density (g/cm3)	EC (mS/cm)	pH	Total Nitrogen (g/kg)	Organic Matter (g/kg)	Alkali-Hydrolyzable Nitrogen (mg/kg)	Available Phosphorus (mg/kg)
0.18 ± 0.04	0.39 ± 0.03	6.16 ± 0.02	6.94 ± 0.15	528.56 ± 89.57	372.86 ± 53.12	180.75 ± 12.57

for details).

2.3. Experimental Design and Cultivation Treatments

A single-factor completely randomized design with three biological replications was adopted in this study, with a total of 9 experimental plots (3 cultivation methods × 3 replications). Each plot was a 12-m-long single row containing approximately 60 tomato plants, with 180 plants per cultivation method. For the determination of yield and quality, 1 m at both ends of each row was discarded to eliminate edge effects. The three cultivation treatments were as follows:

• Bucket cultivation (TK): Plastic buckets (30 cm inner diameter × 30 cm height).
• Non-woven bag cultivation (DK): Cylindrical non-woven fabric bags (30 cm inner diameter × 30 cm height).
• Underground trough cultivation (CK): Trapezoidal sunken trenches (top width 30 cm, bottom width 25 cm, depth 30 cm), lined with geotextile to isolate substrate from native soil (see Figure 1 and Figure 2 for schematics).

Five-leaf stage tomato seedlings (Solanum lycopersicum L. cv. ‘Provence’) were transplanted on 15 August 2024, with 1.5 m row spacing and 20 cm plant spacing. Treatments were randomly assigned within each block to minimize greenhouse environmental variability effects.

2.4. Irrigation, Fertigation, and Crop Management

Each treatment was equipped with pressure-compensating drip irrigation tape (flow rate: 1.0 L h⁻¹; emitter spacing: 20 cm). All treatments adopted the same daily fixed-time and fixed-quantity irrigation schedule with consistent fertigation frequency and equal nutrient solution supply throughout the growing period. The total fertilizer application rates were 340 kg/hm² for nitrogen (N), 180 kg/hm² for phosphorus (P), and 300 kg/hm² for potassium (K). Fertilizers were uniformly applied through drip fertigation to ensure consistent nutrient input across all cultivation treatments. All other agronomic practices, including pruning, pollination and pest management, were consistently conducted following standard commercial greenhouse management protocols for high-wire tomato production.

2.5. Data Collection and Measurements

2.5.1. Substrate Physicochemical Properties

Before transplanting, five random substrate samples per plot were collected from each plot, mixed, air-dried, and analyzed for physicochemical properties including total nitrogen (TN), soil organic matter (SOM), alkali-hydrolyzable nitrogen, available phosphorus (AP), pH, electrical conductivity (EC), and bulk density, following conventional analytical methods [19,20].

2.5.2. Root-Zone Environment Monitoring

Throughout the entire tomato growing season, fixed sensors were used for all-day continuous monitoring of root-zone substrate moisture content, temperature and electrical conductivity. In each experimental plot, three measuring points were arranged in an S-shaped pattern at a uniform soil depth of 15 cm. The collected data were finally used to calculate daily average values and treatment average values for statistical analysis.

2.5.3. Plant Growth and Physiological Measurements

At 30, 45, 60, 75, and 90 days after transplanting (DAT), five uniform plants per plot were tagged. Plant height (cotyledonary node to apical meristem), stem diameter (1 cm above cotyledonary node), and number of fully expanded main stem leaves were measured. Chlorophyll relative content (SPAD) was determined on the youngest fully expanded leaf using a SPAD-502 (Konica Minolta, Tokyo, Japan).

2.5.4. Yield and Fruit Quality Analysis

Marketable fruits were harvested multiple times at commercial ripeness (full red color). Total fresh weight per plot was recorded and converted to yield per hectare; average fruit weight was calculated from 10 random fruits per plot. At peak harvest, 15 fruits per treatment (5 per replication) were sampled for quality analysis: lycopene (spectrophotometric method after acetone-hexane extraction), vitamin C (2,6-dichlorophenol-indophenol titration), nitrate (ultraviolet spectrophotometry after water extraction), and soluble sugar (anthrone-sulfuric acid method). All analyses were conducted at the certified analysis and testing center of the College of Horticulture, Tarim University.

2.6. Statistical Analysis

Data were organized using Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA, USA) and analyzed with SPSS Statistics 25.0 (IBM Corp., Armonk, NY, USA). One-way ANOVA was used to evaluate differences among treatments, and Fisher’s LSD test (p < 0.05) was used for mean comparisons. All figures were generated with OriginPro 2025 (OriginLab Corporation, Northampton, MA, USA). All growth parameters were analyzed separately at each sampling date, and the time effect was not included in the one-way ANOVA model.

3. Results

3.1. Daily Mean and Hourly Variations in Substrate Temperature, Moisture Content, and Electrical Conductivity (EC) Under Different Cultivation Methods

Figure 3 presents the daily mean and hourly variations in substrate temperature, moisture content, and electrical conductivity (EC) under three cultivation methods (bucket cultivation, TK; non-woven bag cultivation, DK; underground trough cultivation, CK). For the daily mean values (Figure 3A–C), substrate temperature (Figure 3A) differed among the systems: TK maintained relatively lower and more stable temperatures, DK exhibited greater fluctuations, and CK maintained relatively higher temperatures with moderate variation. Moisture content (Figure 3B) showed a similar pattern: TK maintained relatively high moisture levels during most of the monitoring period, DK exhibited more pronounced variation, and CK displayed intermediate and relatively stable moisture levels. EC values (Figure 3C) also varied among treatments: TK showed relatively higher EC levels, DK exhibited greater temporal variability, and CK maintained lower and more stable EC levels. Overall, the three cultivation systems showed distinct patterns in substrate temperature, moisture, and EC, reflecting differences in their root-zone environmental conditions. For the hourly variations (Figure 3D–F), substrate temperature followed a typical diurnal pattern, with increases during the daytime and decreases at night across all treatments. CK maintained the highest temperatures, followed by TK, while DK showed the lowest temperatures with greater fluctuations. Moisture content responded clearly to irrigation events, with TK maintaining relatively high levels, DK showing lower levels with greater variability, and CK exhibiting intermediate fluctuations. EC remained relatively stable during most daytime periods but increased after irrigation, with TK showing higher EC peaks, DK exhibiting more pronounced variation, and CK maintaining lower levels.

Overall, the three cultivation systems exhibited distinct daily mean and hourly variation patterns, highlighting differences in root-zone environmental conditions.

3.2. Effects of Different Cultivation Methods on Tomato Agronomic Traits

Figure 4 shows the dynamic changes in tomato growth traits under different cultivation methods (TK, DK, CK). As shown in Figure 4A, tomato plant height increased with days after transplanting (DAT), and cultivation methods significantly affected plant height, especially during 60–90 DAT. During this period, plant heights under TK and CK were significantly greater than under DK, with increases of 4.80–10.95% and 5.88–13.27%, respectively.

As shown in Figure 4B, tomato stem diameter also increased with DAT, with significant differences among cultivation methods, particularly during 60–90 DAT. Stem diameter grew rapidly in the first 60 DAT, then slowed down. During 60–90 DAT, stem diameters under TK and CK were significantly greater than under DK, with no significant difference between TK and CK. Compared to DK, TK increased stem diameter by 15.00%, 11.47%, 23.03%, 17.78%, and 20.25% at 30, 45, 60, 75, and 90 DAT, respectively; CK increased stem diameter by 16.29%, 18.89%, 16.10%, 12.18%, and 19.80% at the same time points.

As shown in Figure 4C, tomato SPAD values decreased with DAT, with significant differences among cultivation methods, especially during 45–60 DAT. No significant differences were observed among treatments in the first 30 DAT. From 60 to 90 DAT, SPAD values stabilized across all treatments. During 45–60 DAT, SPAD values under TK and CK were significantly higher than under DK (no significant difference between TK and CK), with increases of 6.26–8.44% and 5.45–9.32%, respectively, compared to DK.

As shown in Figure 4D, tomato leaf number increased with DAT, with significant differences among cultivation methods, particularly during 60–90 DAT. Leaf number increased markedly in the first 75 DAT, then slowed down and stabilized during 75–90 DAT. During 60–90 DAT, leaf numbers under TK and CK were significantly greater than under DK (no significant difference between TK and CK). Compared to DK, TK increased leaf number by 9.90%, 9.56%, and 7.33% at 60, 75, and 90 DAT, respectively; CK increased leaf number by 5.94%, 13.04%, and 15.59% at the same time points.

3.3. Effects of Different Cultivation Methods on Tomato Yield

As shown in

Table 2. Effects of different cultivation methods on fruit number per plant, yield per plant and total yield (t/ha) of tomato.

Treatment	Fruit Number per Plant	Yield per Plant (kg)	Yield (t/ha)
TK	19.33 ± 0.47 a	2.71 ± 0.06 a	87.36 ± 1.96 a
DK	18.00 ± 0.82 a	2.20 ± 0.12 b	73.23 ± 3.98 b
CK	19.00 ± 0.82 a	2.60 ± 0.05 a	86.81 ± 1.77 a

, different cultivation methods significantly affected tomato yield per plant and total yield per unit area but had no significant effect on the number of fruits per plant. There was no significant difference in the number of fruits per plant among bucket cultivation (TK), non-woven bag cultivation (DK), and underground trough cultivation (CK); compared with DK, TK and CK increased the number of fruits per plant by 7.39% and 5.56%, respectively.

Yield per plant was significantly higher in TK (2.71 kg) and CK (2.60 kg) than in DK, with increases of 23.18% and 18.18%, respectively, and no significant difference was found between TK and CK. Similarly, total yield per unit area was highest in TK (87.36 t) and CK (86.81 t), which were significantly higher than DK by 19.31% and 18.55%, respectively, with no significant difference between TK and CK.

3.4. Effects of Different Cultivation Methods on Tomato Quality

As shown in

Table 3. Effects of different cultivation methods on the contents of lycopene, vitamin C, nitrate, soluble sugar and organic acid in tomato.

Treatment	Lycopene (g/100 g)	VC (g/100 g)	Nitrate Content (g/100 g)	Soluble Sugar Content (g/100 g)	Organic Acid Content (g/100 g)
TK	6.76 ± 0.19 b	22.30 ± 0.29 c	33.34 ± 0.03 a	43.77 ± 0.49 b	3.65 ± 0.03 b
DK	11.92 ± 0.10 a	25.80 ± 0.24 b	10.95 ± 0.19 c	51.47 ± 0.37 a	3.69 ± 0.01 b
CK	5.05 ± 0.10 c	32.53 ± 0.08 a	32.10 ± 0.58 b	33.51 ± 0.29 c	3.92 ± 0.02 a

, cultivation methods significantly affected all measured tomato quality indicators. Non-woven bag cultivation (DK) had the highest lycopene and soluble sugar contents, intermediate vitamin C (VC) and organic acid contents, and the lowest nitrate content. Underground trough cultivation (CK) had the highest VC and organic acid contents, while bucket cultivation (TK) had the highest nitrate content.

Specifically, DK’s lycopene content was significantly higher than TK’s (76.33% increase) and CK’s (136.04% increase). CK’s VC content was significantly higher than DK (26.09% increase) and TK (45.87% increase). TK’s nitrate content was significantly higher than CK’s and DK’s; DK had the lowest nitrate content, 67.16% and 65.89% lower than TK and CK, respectively. DK’s soluble sugar content was significantly higher than TK (17.60% increase) and CK (53.60% increase). CK’s organic acid content was significantly higher than that of DK and TK, while no significant difference was found between DK and TK.

4. Discussion

The root-zone environment is critical for crop performance, and our results show that different cultivation methods significantly affect the physical and chemical stability of the root zone in Gobi Desert greenhouses: underground trough cultivation (CK) provides the most stable root-zone environment with minimal fluctuations in temperature, moisture, and EC; bucket cultivation (TK) maintains high moisture levels but tends to accumulate salt; and non-woven bag cultivation (DK) has the highest variability in root-zone parameters due to limited substrate volume and poor buffering capacity. These differences in root-zone environment may contribute to variations in tomato growth, with CK and TK supporting superior plant growth in terms of plant height, stem diameter, and SPAD values, while DK constrains growth due to unstable root-zone conditions.

In terms of fruit quality, CK and TK improve vitamin C and organic acid contents, while DK has the highest lycopene and soluble sugar contents with the lowest nitrate concentration. On the other hand, it also indicates that the substrate water retention capacity and root-zone environmental buffering performance of DK are weaker than those of TK and CK. Although DK inhibits plant growth to a certain extent and reduces yield, the moderate rhizosphere microenvironmental stress formed under this condition is conducive to promoting the accumulation of flavor and nutritional substances such as lycopene and soluble sugar in fruit, and meanwhile reduces the nitrate content.

Overall, underground trough cultivation (CK) is the most robust system for greenhouse tomato cultivation in arid Gobi regions, bucket cultivation (TK) is a viable high-yield alternative requiring proper irrigation management to avoid salt accumulation, and non-woven bag cultivation (DK) is less suitable for long-term stable production due to its poor environmental buffering capacity, despite its advantage in improving intrinsic fruit quality via moderate rhizosphere stress.

5. Conclusions

This study clarified the effects of different cultivation systems on the root-zone environment and growth of greenhouse tomatoes in the arid Gobi Desert. The results confirm that cultivation structure significantly shapes root-zone temperature, moisture, and salinity, thereby affecting tomato growth and yield. Underground trough cultivation maintained stable root-zone conditions, supporting balanced plant growth and higher yield. Bucket cultivation promoted vigorous plant growth and high productivity but required reasonable irrigation management to mitigate salt accumulation. Non-woven bag cultivation exhibited greater variability in root-zone environmental parameters. Cultivation systems also differed markedly in regulating fruit quality at the peak harvest stage: underground trough and bucket cultivation improved yield and conventional nutritional indexes, while non-woven bag cultivation obtained higher lycopene and soluble sugar contents in harvested fruits. This study highlights the importance of cultivation system design for optimizing root-zone microenvironments in arid protected agriculture and provides a scientific basis for selecting and optimizing tomato cultivation patterns in the Gobi Desert and similar arid regions. From a practical production perspective, underground trough cultivation is recommended as the priority mode for large-scale stable production in Gobi greenhouses. Bucket cultivation can be adopted as a high-yield alternative, with strict irrigation and salt regulation management. Non-woven bag cultivation is not suitable for long-term continuous cropping but can be appropriately applied for high-quality fruit production under a limited planting scale.

This study was conducted with a single tomato cultivar and only in one growing season, and economic benefit analysis was not included. Future research should expand to multiple cultivars and growing seasons and further combine economic evaluation to provide more comprehensive theoretical and technical support for the sustainable development of facility agriculture in arid Gobi regions.