High-density planting of tomato cultivar's with early decapitation of growing point increased yield in a closed hydroponic system

Abstract

Decapitation of tomato growing point's results in a shortened growth season with no new flower trusses initiated thereafter. Since less flower trusses could lead to reduced yield, this study was conducted to investigate whether increased plant density could contribute to increase yield per unit area. The response of nine tomato cultivars with early decapitation of growing points at different plant densities (10, 16, 20 and 25 plants/m²) was investigated in a closed hydroponic system under a 40% white shadenet structure. The growing points of all plants were removed between the second and third inflorescence, with two leaves remaining above the second inflorescence. Fruit physiological disorders, number of fruits, as well as total and marketable yield were recorded. Cultivar ‘Miramar’, gave the highest marketable and total yield, followed by cultivars ‘Rodade’ and ‘Alfar’. Cultivars ‘Alexis’, ‘Star9006’ and ‘Zeal’ produced highest average fruit mass compared to other cultivars. There was no significant interaction effect between cultivars and plant densities on tomato yield. Fruit cracking was significantly higher for ‘Linares’ and ‘Star9006’, while raincheck was higher for cultivars ‘Alexis’ and ‘FA593’. ‘Rodade’ showed high incidence of zippering. Plant densities of 20 or 25 plants/m² produced significantly higher marketable and total yield, while plant densities of 10 or 16 plants/m² resulted in higher average fruit mass and highest incidence of fruit cracking. Nutrient uptake, as revealed in the mineral content of fruit, was not affected by plant density. Cultivar ‘Rodade’, an open-pollinated determinate cultivar, especially shows potential to compete with more expensive hybrid cultivars in such a production system in terms of yield. Results demonstrate that high-density planting of tomato with decapitated growing points increased yield per unit area in a closed hydroponic system.

Keywords:

• cultivars
• fruit cracking
• pruning
• trusses
• yield
• decapitating growth point

Introduction

Tomato (Solanum esculentum) is widely grown all over the world and is in demand all year-round in South Africa. However, tomato growers in South Africa are faced with many challenges, including a high energy expense to cool and heat the greenhouses, inconsistency in the availability of tomatoes throughout the year due to changes in environmental conditions, high costs of hybrid seeds and labour (Maboko et al. 2012).

Soilless production of vegetables, as compared with traditional field and greenhouse production in soil, allows the efficient use of water and nutrients by crops (Resh 1996). Although the majority of vegetable production in South Africa is still open field cultivation, soilless cultivation in a protected environment has gained popularity due to improved yield and quality (Niederwieser 2001). The two most popular commercial hydroponic systems applied in South Africa are closed and open bag systems (Maboko et al. 2011a). The open bag system is generally utilised for production of crops with an indeterminate growth habit, while the closed system using the gravel-film technique is utilised to produce leafy vegetables (Maboko et al. 2011a). In open bag systems, once the nutrient solution passes the root system, the excess nutrient solution runs to waste; while in closed systems, the excess nutrient solution is recovered, sometimes filtered, replenished and recirculated. Recently, soilless cultivation tended towards closed systems, thereby reducing nutrient losses and protection of the environment (Schwarz et al. 2009). Although the recirculation of the nutrient solution complicates the control of plant nutrition, as compared to open hydroponic systems, it has the advantage of saving both water and nutrients (40–50% of the total supply) (Dasgan & Ekici 2005). In addition, it is environment friendly due to recirculation of the nutrient solution, which allows for restriction or complete prevention of nutrients leaching into the ground water (Savvas 2002).

In South Africa, hydroponically grown tomatoes are generally produced in an open bag hydroponic system at a plant density of 2.5–3 plants/m² and plants are allowed to grow to a height of 2–2.5 m bearing 7–9 fruit trusses for a period of five months from transplanting to the termination of the plants. The marketable yield of tomatoes grown in an open bag hydroponic system in a temperature-controlled plastic tunnel ranges from 6 to 7 kg/plant (Maboko et al. 2012), while in a closed hydroponic system using the gravel-film technique under a shadenet structure, yields of 7–8 kg/plant are obtained (Maboko et al. 2011a). Improved yield and quality of tomato cultivars grown in a closed hydroponic system using the gravel-film technique has been reported by Maboko et al. (2011a). Okano et al. (2001) reported that tomato plants often reach a length of 10–20 m and 25–30 fruit trusses can be harvested under careful crop management and environmental control. Growing indeterminate tomatoes require high and strong infrastructure to carry the weight of tomato fruits, as well as the plant material, which consequently will require high costs. Removal of the growing point and limiting flower trusses will shorten the growth season, and requires less maintenance and input costs, including less pesticide and fertiliser use, as compared to the standard system, which allows plants to grow taller at lower densities for a longer growing period.

Many studies have shown a linear increase in tomato yield when plant population was increased (Ara et al. 2007; Maboko et al. 2011b), while other studies showed that higher yields are dependent on cultivar selection, crop management and growing season (Jensen & Malter 1995; Lorenzo & Castilla 1995; Rodriguez et al. 2007; Maboko et al. 2012). Most plants appear to increase yield per unit area as plant density increases, but only up to a certain plant density, after which yield per unit area declines (Akintoye et al. 2009). Early marketable yield, total marketable yield, fruit quality, fruit size, harvesting date and length of the harvesting period are affected by the number of fruit trusses per plant and plant population (Heuvelink 1995; Saglam & Yazgan 1995; Maboko et al. 2011b). Plant density plays an important role to optimise yield, while a shortened growth season will result in less input costs (Saglam & Yazgan 1995; Okano et al. 2001). However, there is little or no information on the performance of tomato cultivars pruned back to two trusses, combined with different plant densities when grown in a closed hydroponic system using the gravel-film technique in a shadenet structure. The objective of this study was to evaluate the response of tomato cultivars with decapitated growing points to high-density planting in an effort to increase yield, while shortening the growing season in a closed hydroponic system.

Materials and methods

Experiments were conducted under a 40% white shadenet structure at the Agricultural Research Council– Vegetable and Ornamental Plant Institute (ARC–VOPI), Roodeplaat (25°59′S; 28°35′E and at an altitude of 1200 m above sea level) from 17 October 2011 to 10 January 2012. The minimum and maximum temperatures in the shadenet structure, 9.2 and 37°C, respectively, were measured with Tinyview data-loggers (Gemini Data Loggers, UK) placed under a Stevenson-type screen.

The plantlets were transplanted 35 days after seeding, utilising a closed (gravel-film technique) hydroponic system. Thirty-six treatment combinations were used, namely, four plant densities (10, 16, 20 and 25 plants/m² at a plant spacing of 25×40, 25×25, 20×25 and 10×20 cm, respectively), combined with nine tomato cultivars (‘FA593’, ‘Linares’, ‘Alexis’, ‘Alfar’, ‘Miramar’, ‘Rodade’, ‘Star9006’, ‘Star9011’ and ‘Zeal’). Tomato cultivars with indeterminate growth habit were ‘FA593’, ‘Linares’, ‘Alexis’, ‘Alfar’ and ‘Miramar’, while tomato cultivars with a determinate growth habit were ‘Rodade’, ‘Star9006’, ‘Star9011’ and ‘Zeal’. A randomised complete block design was used with three replicates.

Tomato seedlings were transplanted 6 cm deep into gullies filled with crushed granite rocks of irregular shape with a diameter ranging from 16 to 19 mm. A gravel-film technique hydroponic system was used to perform the trial, as described by Maboko et al. (2011a). The nutrient solution was renewed with fresh nutrient solution on a weekly basis. The composition and chemical concentration of fertilisers used in the nutrient solution were: Hygroponic® (Hygrotech (Pty). Ltd., South Africa) comprising of N (68 mg/kg), P (42 mg/kg), K (208 mg/kg), Mg (30 mg/kg), S (64 mg/kg), Fe (1.254 mg/kg), Cu (0.022 mg/kg), Zn (0.149 mg/kg), Mn (0.299 mg/kg), B (0.373 mg/kg) and Mo (0.037 mg/kg), as well as calcium nitrate [Ca(NO₃)₂] comprising of N (117 mg/kg) and Ca (166 mg/kg). The fertilisers applied, as from transplanting until the plants were three weeks old, were 600 g Hygroponic and 600 g Ca(NO₃)₂ in 1000 L water. Thereafter, 900 g Hygroponic and 900 g calcium nitrate were applied per 1000 L water. The electrical conductivity (EC) and pH of the nutrient solution was maintained within a range of 1.9–2.3 mS/cm and 5.8–6.1, respectively.

The growing points of all plants were removed between the second and third inflorescence at 30–35 days after transplanting, with two leaves remaining above the second inflorescence. Each plant was supported by a twine to keep it upright.

Data collection and statistical analysis

The performance of the tomato cultivars was evaluated for total yield, marketable and unmarketable yield, as well as physiological disorders occurrence. Fruits were regarded as unmarketable when they exhibited cracking, zippering, rotting, blossom-end rot, raincheck, catface or fell into the extra small size category (less than 40 mm diameter).

Four ripe fruits of medium size (60–70 mm diameter) were harvested randomly from the second truss of four plants per replicate, per cultivar for nutrient analysis (P, K, Ca, Mg, Mn, Zn, B, Cu and Fe). Nutrient concentrations were expressed per gram fruit dry mass. Tomato fruits were oven-dried at 70^°C and ground using a mill with a 1 mm of sieve. Nitrogen was determined on dry-milled material using a Carlo Erba NA 1500 C/N/S Analyzer. An aliquot of the digest solution was used for ICP-OES-69 (Inductively Coupled Plasma Optical Emission Spectrometry) for determination of Mg, P, K, Fe and Zn concentrations.

Data were subjected to analysis of variance (ANOVA) using the statistical program GenStat® version 11.1 (Payne et al. 2008). Treatment means were separated using Fisher's protected T-test least significant difference (LSD) at the 5% level of significance (Snedecor & Cochran 1980).

Results

There were no significant interactions among the treatments and, therefore, only the main factors are discussed.

Effect of cultivar

Tomato cultivars had a significant effect on marketable yield, unmarketable yield, number of marketable fruits, total yield and average fruit mass (Table 1). Cultivar ‘Miramar’ gave the highest marketable yield, number of marketable fruits and total yield, followed by cultivars ‘Rodade’ and ‘Alfar’ in terms of marketable yield and number of marketable fruits (Table 1). The highest unmarketable yield was produced on determinate tomato cultivars, i.e. ‘Star9006’, ‘Star9011’, ‘Rodade’ and ‘Zeal’, and indeterminate tomato cultivars, ‘FA593’ and ‘Linares’. Cultivar ‘Alfar’ and ‘Miramar’, followed by ‘Alexis’ resulted in the lowest percentage unmarketable yield and highest percentage marketable yield (Table 1). Cultivars ‘Zeal’, ‘Star9006’ and ‘Alexis’ produced the largest average fruit mass compared with other cultivars (Table 1).

Table 1. Effect of cultivar on tomato yield per metre square.

Cultivar	Marketable yield (kg/m^2)	Unmarketable yield (kg/m^2)	Number of marketable fruits	Total yield (kg/m^2)	Average fruit mass (g)	Percentage of unmarketable yield	Percentage of marketable yield
Alexis	18.07^bc	2.22^de	130.2^cd	20.29^bc	140.3^ab	10.84^cd	89.16^ab
Alfar	20.06^b	1.50^e	151.7^bc	21.56^bc	133.3^bcd	6.92^d	93.08^a
FA593	17.99^bc	3.80^ab	145.6^bc	21.78^bc	125.0^cde	17.88^ab	82.12^cd
Linares	14.72^c	4.29^ab	127.3^cd	19.01^c	117.6^e	22.97^a	77.03^d
Miramar	27.05^a	2.57^cd	228.3^a	29.62^a	120.6^e	8.57^d	91.43^a
Rodade	20.87^b	3.27^bc	172.2^b	24.14^b	122.3d^e	15.07^bc	84.93^bc
Star9006	15.28^c	4.43^a	109.5^d	19.71^bc	140.8^ab	22.92^a	77.08^d
Star9011	16.97^bc	3.98^ab	126.7^cd	20.95^bc	135.5^bc	19.49^ab	80.51^cd
Zeal	15.24^c	3.47^abc	103.6^d	18.70^c	147.7^a	18.85^ab	81.15^cd
LSD0.05	4.19	1.04	34.69	4.47	10.96	5.22	5.22

Note: Values in a column followed by the same letter are not significantly different (p > 0.05), using Fishers' protected t-test. ns: not significant; LSD: least significant difference.

Fruit cracking was significantly higher for cultivar ‘Linares’, followed by ‘Star9006’, while raincheck was higher for cultivars ‘Alexis’, ‘FA593’ and ‘Linares’(Table 2). There were no significant differences among the cultivars with regard to blossom-end rot, fruit rot and incidence of fruits exhibiting catface. Higher number and mass of fruits exhibiting zippering were found in cultivars ‘Rodade’, followed by ‘Star9006’ and ‘Zeal’ in this regard (Table 2). The highest number and mass of extra-small sized fruits were found in cultivars ‘Rodade’ and ‘Miramar’.

Table 2. Effect of cultivar on fruit physiological disorders and extra-small sized fruits per metre square.

Cultivar	Fruit cracking		Blossom-end rot		Catface		Raincheck		Fruit rot		Zippering		Extra-small sized fruit (>40mm)
	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)
Alexis	4.00^d	0.56^c	0.83	0.11	0.42	0.07	5.88^a	0.86^a	0.92	0.08	1.12^d	0.12^d	7.8^c	0.43^b
Alfar	2.63^d	0.35^c	1.21	0.11	0.75	0.09	1.17^b	0.13^bc	0.46	0.07	0.71^d	0.11^d	20.5^b	0.64^b
FA593	13.58^bc	1.87^b	0.42	0.05	0.75	0.14	5.75^a	0.84^a	1.33	0.15	0.71^d	0.08^d	17.1^bc	0.66^b
Linares	22.12^a	2.85^a	1.25	0.13	0.75	0.09	3.96^a	0.43^b	0.38	0.05	2.50^cd	0.21^d	16.0^bc	0.53^b
Miramar	3.79^d	0.48^c	0.67	0.09	1.17	0.16	1.54^b	0.24^bc	0.62	0.06	1.00^d	0.12^d	34.4^a	1.42^a
Rodade	4.92^d	0.574^c	1.00	0.10	1.46	0.16	0.29^b	0.04^c	0.88	0.10	10.29^a	1.03^a	38.1^a	1.27^a
Star9006	15.46^b	2.18^ab	0.79	0.127	1.96	0.25	0.75^b	0.12^bc	1.71	0.24	7.67^b	0.85^ab	15.2^bc	0.67^b
Star9011	13.96^bc	2.07^b	2.00	0.346	1.33	0.19	0.42^b	0.06^c	1.21	0.16	4.12^c	0.45^c	16.6^bc	0.70^b
Zeal	10.00^c	1.494^b	1.29	0.129	1.38	0.29	0.67^b	0.10^bc	1.04	0.14	7.33^b	0.70^b	18.2^b	0.61^b
LSD0.05	4.98	0.74	ns	ns	ns	ns	2.23	0.34	ns	ns	2.05	0.24	9.80	0.42

Note: Values in a column followed by the same letter are not significantly different (p > 0.05), using Fishers' protected t-test. ns: not significant; LSD: least significant difference.

Effect of plant density

Plant densities of 20 or 25 plants/m² produced significantly higher marketable and total yield, while plant densities of 10 or 16 plants/m² resulted in higher average fruit mass (Table 3). Number of marketable fruits was the highest at a plant density of 25 plants/m². Plants subjected to a density of 10 plants/m² produced the highest percentage of unmarketable yield (Table 3).

Table 3. Effect of plant density on yield of tomatoes per metre square.

Plant density/m^2	Marketable yield (kg/m^2)	Unmarketable yield (kg/m^2)	Number of marketable fruits	Total yield (kg/m^2)	Average fruit mass (kg)	Percentage of unmarketable yield	Percentage of marketable yield
10	15.90^c	3.72	116.0^c	19.62^b	138.1^a	20.16^a	79.84^b
16	17.80^bc	3.04	133.2^bc	20.83^b	135.9^a	15.05^b	84.95^a
20	19.37^ab	2.89	151.4^b	22.26^ab	130.8^b	13.60^b	86.40^a
25	20.81^a	3.48	175.1^a	24.30^a	121.1^b	14.97^b	85.03^a
LSD0.05	2.79	ns	23.13	2.98	7.31	3.48	3.48

Note: Values in a column followed by the same letter are not significantly different (p > 0.05), using Fishers' protected t-test. ns: not significant; LSD: least significant difference.

The incidence of fruit cracking decreased with an increase in plant density (Table 4). The incidence of fruit cracking was significantly higher at a low plant density of 10 plants/m², although not significantly different to 16 plants/m². Closer plant spacing of 20 and 25 plants/m² produced fruits exhibiting zippering, as well as extra-small sized fruits. Plant density did not have a significant effect on mineral uptake by tomato plants as reflected by mineral concentration in the fruit (Table 5).

Table 4. Effect of plant density on fruit physiological disorders and extra-small sized fruits per metre square.

Plant density/m^2	Fruit cracking		Blossom-end rot		Catface		Rain-check		Fruit rot		Zippering		Extra-small sized fruit (>40mm)
	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)	Number	Mass (kg)
10	12.57	1.85^a	1.76	0.27	1.22	0.26	2.63	0.41	1.09	0.13	2.91^c	0.31^c	13.4^b	0.49^c
16	9.89	1.38^ab	0.63	0.07	0.91	0.11	2.91	0.42	0.81	0.11	3.41^bc	0.35^abc	16.9^b	0.59^bc
20	8.04	1.08^b	1.06	0.10	1.09	0.14	1.30	0.15	0.89	0.10	4.43^ab	0.47^ab	24.0^a	0.85^b
25	9.70	1.20^b	0.76	0.09	1.20	0.14	2.24	0.28	1.00	0.12	5.02^a	0.50^a	27.5^a	1.15^a
LSD0.05	ns	0.49	ns	ns	ns	ns	ns	ns	ns	ns	1.37	0.15	6.53	0.28

Note: Values in a column followed by the same letter are not significantly different (p > 0.05), using Fishers' protected t-test. ns: not significant; LSD: least significant difference.

Table 5. Effect of cultivar and plant spacing on mineral concentration of tomato fruit (DM).

Plant density/m^2	Mg (mg/kg)	P (mg/kg)	K (%)	Fe (mg/kg)	Zn (mg/kg)
10	92.3	321.0	0.20	9.48	5.64
16	89.3	300.6	0.20	10.36	5.53
20	93.9	306.2	0.20	9.43	5.15
25	97.2	315.6	0.21	10.09	5.13
LSD0.05	ns	ns	ns	ns	ns
Cultivar
Alexis	84.0^d	324.8^ab	0.22	9.56	4.38
Alfar	93.6^abcd	291.0^bc	0.21	10.45	5.33
FA593	103.3^a	318.2^ab	0.21	10.15	5.50
Linares	89.8^bcd	262.3^c	0.20	10.75	5.64
Miramar	91.4^bcd	314.1^abc	0.20	11.50	6.15
Rodade	89.8^bcd	334.2^ab	0.19	9.80	5.15
Star9006	97.8^abc	354.2^a	0.21	9.46	5.71
Star9011	100.8^ab	311.0^abc	0.51	9.21	6.20
Zeal	88.4^cd	287.7^bc	0.20	7.70	4.18
LSD0.05	11.55	53.22	ns	ns	ns

Note: Values in a column followed by the same letter are not significantly different (p > 0.05), using Fishers' protected t-test. ns: not significant; LSD: least significant difference; DM: dry matter.

Discussion

The differences in yield among the cultivars (Table 1), confirm the findings reported by Maboko et al. (2012) that cultivar selection is an important criteria to be considered in any cultivation system. Reduced average fruit mass and increased extra-small sized fruits at higher plant densities can be linked to increased competition for light. In this study, the decapitation of growing points resulted in a shortened growing period of 12 weeks. This compares very favourably to the 20-week growth season under normal cultivation practices in an open bag hydroponic system (Maboko et al. 2012). The high yield per metre square at higher plant density is a direct result of the higher number of plants per plot. The higher marketable yield at high plant densities can be linked to smaller fruit, less prone to fruit cracking. It should be noted that hybrid seeds of indeterminate tomato cultivars are expensive and increases in plant density will lead to higher input costs due to the increased number of seedlings per unit area. However, open-pollinated cultivars of determinate growth habit are of low cost and easily accessible.

In this study, severity of fruit cracking seems to have been influenced by plant spacing and cultivar (Tables 2 and 4). High plant population density resulted in high canopy coverage, and fruits were less exposed to sunlight compared to lower plant density. Fruit on the second truss of widely spaced plants/low plant density were more subjected to direct sunlight and high temperature, which resulted in the higher incidence of fruit cracking. Dorais et al. (2001) reported that high temperature and irradiance favour the pulp expansion towards the interior of the fruit and consequently, a weakening of the cuticle which results in fruit cracking. This is in agreement with the results of Peet and Willits (1995) and Cheryld et al. (1997). The difference in yield and susceptibility to fruit cracking could be due to the genetic make-up of the hybrid cultivars. Pruning back the plants to two trusses can also play a major role in producing tomatoes during a shorter growth season in order to avoid unfavourable weather conditions.

Raincheck, a physiological disorder, resulting from rainwater hitting fruit shoulders, causing them to be rough, is often more prevalent under shadenet structures because rainwater can pass through such structures (Maboko et al. 2011a). The cause of raincheck is unknown, but heavy rain might alter fruit development and/or water uptake, which could disrupt an epidermal development, causing cracks on fruit shoulders (Huang & Snapp 2004; Maboko et al. 2011a). Results are in agreement with Masarirambi et al. (2009) that cultivars differ in susceptibility to raincheck. According to Masarirambi et al. (2009), raincheck can be alleviated by use of resistant cultivars with good leaf coverage protecting fruit from rain. Previous reports attributed tomato fruit physiological disorders, including fruit cracking, catface, raincheck and zippering to cultivar variation and environmental conditions (Dorais et al. 2001; Saure 2001; Masarirambi et al. 2009; Peet 2009; Maboko et al. 2013).

The results show that high-density tomato plantings with growing point decapitated to two trusses, show tremendous potential to achieve high yield during a shortened growth season of 12 weeks. Cultivar ‘Miramar’, followed by cultivars ‘Rodade’ and ‘Alfar’ gave the highest marketable and total yield in this experiment. Cultivar ‘Rodade’, although an open-pollinated determinate cultivar, especially shows potential to compete with more expensive hybrid cultivars in such a production system with high-density plantings combined with early decapitation of growth points. Results demonstrate that high-density planting of tomato with decapitated growing points increased yield per unit area in a closed hydroponic system.