2015 Volume 84 Issue 1 Pages 37-45
In order to modify nutrient solution for tomatoes grown in extremely low-volume substrate (ELVS) combined with low-node-order pinching and high-density planting (LN&HD), the effects of nutrient solution concentration and supplemented K and P were investigated. Plant growth, nutrient uptake, and fruit yellow-shoulder disorder were measured in two experiments. Treatments included three nutrient solution concentration levels (0.6, 0.9, and 1.2 dS·m−1 EC) in Experiment 1 and added K or P (EC 0.9+K, EC 0.9+P) and P+K (EC 0.9+P+K) on the basis of Enshi nutrient solution at EC 0.9 dS·m−1 in Experiment 2. Tomatoes ‘CF Momotaro York’ were grown in a 250 mL pot filled with granular rockwool combined with LN&HD. A high-frequency and small-volume fertigation system was used based on the integrated solar radiation amount. In Experiment 1, mineral elemental uptake rate, plant growth and fruit yield increased with increasing nutrient solution concentration, and fruit yellow-shoulder disorder decreased. In Experiment 2, P supplementation treatments of EC 0.9+P and EC 0.9+P+K largely enhanced fruit yield, shoot weight and all nutrient uptake rates. In contrast, yield improvement was not observed at EC 0.9+K, and only K uptake was promoted in K-supplemented treatment. The fruit yellow-shoulder incidence and index tended to decrease in the following order: EC 0.9 > EC 0.9+P > EC 0.9+K > EC 0.9+P+K. In conclusion, it was effective to increase P and K concentrations of nutrient solution on the basis of Enshi nutrient solution formula (EC 0.9 dS·m−1), to maintain optimal tomato vegetative growth, achieve higher yield comparable to that with EC 1.2 dS·m−1, and produce fewer P- and K-deficiency symptoms in the case of ELVS combined with LN&HD culture by a high-frequency and small-volume application method.
Tomato (Solanum lycopersicum L.) is one of the most important horticultural crops in Japan and its year-round supply is desired by consumers. However, it is extremely difficult to produce normal tomatoes in summer due to high-temperature conditions. Recently, plant factories using sunlight were installed in many countries. In Japan, research on fully monitored and controlled culture systems integrated into natural-light-using plant factories for year-round plant production has also been conducted (Hashimoto, 2013; Masuda et al., 2013; Noguchi, 2012; Saito et al., 2008; Toma et al., 2013).
One of the culture systems applied in plant factories is the extremely low-volume substrate (ELVS) cultivation system that is known as the D-tray cultivation system (Fig. 1). For the last several years, a great deal of effort has gone into development of the D-tray cultivation system in combination with low-node-order pinching and high-density planting (LN&HD) in tomato production (Egishi et al., 2012; Endo et al., 2007; Kiriiwa, 2008). The planting density is 3.6 to 5.4 plants·m−2, which is about 2 to 2.5 times as dense as 2.1 to 2.5 plants·m−2 in conventional tomato production systems by high wire layering. The distances between plants are usually 24 cm and 17 cm at 3.6 and 5.4 plants·m−2 density, respectively. The structure of the D-tray (60 cm length, 20 cm width, 10 cm height) is characterized by 10 connected D-shaped pots (250 mL/pot) arranged in 2 rows, with each row including 5 pots. Owing to less water and nutrients being available in the restricted root zone, a high-frequency and small-volume fertigation system based on the integrated solar radiation amount should be required. Because the transportation of seedlings from the greenhouse in which they were raised, transplanting to the D-tray culture system and cleaning after harvesting are very easy to perform, labor-saving operations can be realized. The D-tray cultivation system therefore enhances economic efficiency for tomatoes, especially in the case of LN&HD, in which 3.5 croppings per year are achieved in the case of 3- to 4-node-order pinching (Tamai, 2014).
Structure of tomato growing system with extremely low-volume substrate.
Compared with a tomato production system by high wire layering characterized by a normal-volume rockwool slab substrate and nutrient solution of EC 1.6–2.0 dS·m−1 (Suzuki, 2006; Takakura, 2008), the considerable advantages of an ELVS cultivation system are the low costs of nutrients and substrate material as well as the labor-saving operations (Tamai, 2014). When using a high-frequency and a small-volume, the use of a low concentration of nutrient solution below EC 1.2 dS·m−1 was suggested to be feasible for tomatoes grown in ELVS and LN&HD in our preliminary experiment (Egishi et al., 2012). Although the best fruit production was recorded at EC 1.2 dS·m−1, we assumed that, if the fruit production could be improved up to the level of that at EC 1.2 dS·m−1 by increasing P and K concentrations on the basis of EC 0.9 dS·m−1 without increasing vegetative growth, the plants could be kept in a compact state to accommodate an extremely small planting distance (17 to 24 cm).
Moreover, potassium deficiency (yellow-shoulder fruit and leaf margin chlorosis) and phosphorus deficiency (purple color of the veins and stem, thin stem and poor cluster development) symptoms often appeared in the plants after they had experienced continuous rainy or cloudy weather caused by both low integrated solar radiation and a low nutrient solution fertigation amount. The tomato fruit yellow-shoulder disorder is characterized by sectors of yellow or green tissue under the peel, near the shoulder of the fruit. These areas will never ripen properly, and the tissue is often hard even when the rest of the tomato is ripe (Francis et al., 2000). Unfortunately, the cause of this disorder is complex and not completely understood. Several lines of evidence have showed that nutritional status, weather, plant genetics and their interactions are important factors influencing this condition (Hartz et al., 1999; Sacks and Francis, 2001). K content within the plant usually decreases one or two weeks prior to the appearance of yellow-shoulder, and fruit exhibiting yellow-shoulder disorder often contains a lower K concentration than normal, suggesting a role for K (Picha, 1987; Winsor and Massey, 1958). Ikeda (2005) showed that the fertilizer utilization efficiency was enhanced by high-frequency and small-volume application in the case of soilless culture and also even in soil culture. In ELVS culture, the utilization efficiency of mineral elemental nutrients is considered to be much greater than in high-volume substrate culture. If this is true, tomatoes should be grown and produce a comparable yield by applying a somewhat lower concentration of nutrient solution. However, there was no evidence on this issue. Terabayashi et al. (2004b) confirmed that the average rate of nutrient element uptake serves as a good criterion for estimating the amount of nutrient necessary for achieving the maximum yield of tomatoes grown hydroponically (DFT&NFT). It is essential to examine how the nutrients are absorbed and utilized for tomatoes grown in ELVS, so as to establish an adequate fertigation method by using low-concentration solution with few physiological disorders. Furthermore, there is little information about the nutrient uptake characteristics, especially the uptake patterns for tomato plants that are characterized by low-node-order pinching.
In the present study, the effects of nutrient solution concentration and supplemented K and P on plant growth, nutrient uptake and yellow-shoulder disorder of fruit were investigated in ELVS. In order to modify the nutrient solution in the ELVS with LN&HD system, two experiments were conducted by using Enshi formula nutrient solution. Plants grown at three different concentrations and under four treatments combined with K or P supplementation were compared in Experiments 1 and 2, respectively.
The treatments consisted of 3 concentration levels of diluted Enshi formula nutrient solution in Experiment 1 (0.6, 0.9, and 1.2 dS·m−1 EC), and K and/or P were supplemented up to 5.8 and 4.6 me·L−1 on the basis of EC 0.9 in Experiment 2, as shown in Table 1. Ca concentration in EC 0.6 in Experiment 1 was increased to 3.0 me·L−1, which was the same as in EC 0.9 for avoiding blossom-end rotted (BER) fruit. An experimental design with three replicates for each treatment was adopted, and each replicate contained 10 plants. Thus, 90 and 120 plants were used in Experiments 1 and 2, respectively.
Nutrient solution composition and concentration in Experiments 1 and 2.
Tomato seedlings ‘CF Momotaro York’ with 5 fully expanded true leaves were planted into a D-tray (250 mL) filled with granular rock wool on Sep. 13, 2012, in Experiments 1 and 2. All of the plants were supplied with 1/2 strength Enshi formula nutrient solution from germination until the beginning of the treatments. Treatments were initiated on Sep. 28 at the flowering stage of the first inflorescence. The experiments were conducted in a heated plastic greenhouse in which the minimum (heating) and ventilation air temperatures were 15°C and 23°C, respectively, and the average measured temperature was 20°C throughout the entire growing period. Flowers were vibrated manually to ensure pollination during anthesis. Periodic operations of binding, fruit thinning, and lateral bud and basal leaf pruning were carried out in a timely manner. Plants were trained vertically with a single stem and topped at the 2nd upper leaf above the 3rd truss on Oct. 29. Each truss was adjusted to have four fruit. Harvesting started on Nov. 27 and finished on Dec. 24. Nutrient solution distributed through the drip fertigation system to each plant was recirculated with a pump at a flow rate of approximately 25 mL·min−1. The fertigation frequency was controlled using a solar radiation controller for 90 s each time 1.0 MJ·m−2 solar radiation had accumulated. As a result, fertigation occurred approximately 20–25 times per day on completely sunny days. Groups of 10 plants were fertigated with the nutrient solution from each 35 L reservoir. The solution in each reservoir was refilled with an original solution to compensate for crop water consumption every day and replaced by a new one every week.
Measurements and statistical analysesDaily measurements of the refilled volume of each reservoir, EC and pH of the residual solution in the reservoir, and the nutrient solution fertigation volume were conducted. Plant water uptake rate was determined as shown in equation (1).
| (1) |
The concentration of nutrient in each reservoir was determined weekly. Spectrophotometry was used for the analysis of NO3-N and PO4-P. The levels of K, Ca, and Mg were determined by an atomic absorption method (iCE 3000 Series; Thermo Fisher Scientific Ltd., Cambridge, UK). Owing to ELVS and high-frequency and small-volume nutrient solution application, the factors of evaporation and nutrient accumulation in medium were negligible, so nutrient absorption was measured just by the change of residual solution concentration.
The weekly plant nutrient uptake rate was expressed by the difference between the sum of the nutrient amount included originally and the weekly refilled nutrient solution, and the residual amount left after one week, as shown in equation (2).
| (2) |
Fertilization (NO3-N) efficiency (kg·e−1) was calculated in marketable yield (kg/plant) divided by uptake amount of NO3-N (e/plant) (10 plants, 3 replications; n = 30).
Total fruit yield, marketable fruit yield, fruit number and fruit weight were assessed over the course of the ripening stage for each truss. The cracked and BER fruit were not included in marketable yield. The soluble solid content (SSC) of fruit juice in each truss squeezed by hand with a cheesecloth was determined with a hand refractometer (Model FR-100; Atago Co. Ltd., Tokyo, Japan).
Fruit yellow-shoulder disorder was recorded at every harvest time. The severity was classified into 4 levels as follows: Severe = 3, Medium = 2, Light = 1, and None = 0, in which the yellow-colored area as a proportion of the total fruit shoulder area was more than 50%, 50–20%, less than 20%, and 0%, respectively (Fig. 2). ‘Severe’ status was regarded as representing unmarketable fruit, and “Medium”, “Light”, and “None” were regarded as marketable fruit. Fruit yellow-shoulder incidence and the index (YSI) were calculated as shown in equations (3) and (4), respectively.
| (3) |
| (4) |
Levels of severity of fruit yellow-shoulder disorder classified as None, Light, Medium, and Severe from left to right.
All data were subjected to ANOVA and Scheffe’s multiple range test, as appropriate.
Weekly variations of nutrient uptake rate are shown in Fig. 3. Nutrient uptake rates were closely related to the fertigation concentration, although the extent varied depending on the kind of nutrient element. Weekly K uptake rate tended to increase during flowering and the expansion of upper leaves until the pinching stages, and was maintained at a high level during fruit enlargement and ripening stages. The rate was reduced with decreasing fruit number during the harvesting stage, and was finally reduced to the same level in the last week under all treatments (Fig. 3). The uptake rate of K was greatest at EC 1.2, followed by EC 0.9 and then EC 0.6. The patterns of weekly N and Ca uptake rates during the growing period showed almost the same trend as K, except for the rates being maintained during the harvesting stage. In addition, Ca uptake rate in EC 0.6 fluctuated at nearly the same level as EC 0.9, because the concentration was up-regulated to the same level as in EC 0.9 for the avoidance of BER fruit. The weekly uptake rates of PO4-P and Mg were relatively stable during the growing stage.
Changes in nutrient uptake rate of tomatoes grown in extremely low-volume substrate (Expt. 1). Nutrient uptake rate per plant per week (me/plant/week) = [original concentration (me·L−1) × (35 (L/container) + weekly nutrient solution refilled amount (L/container)) − concentration after one week (me·L−1) × 35 (L/container)]/10 (plants/container). Values are the means ± SE (n = 3).
Changes in nutrient uptake rate of tomatoes grown in extremely low-volume substrate (Expt. 2). Values are the means ± SE (n = 3).
Daily variation of solution EC decreased regularly from the beginning to the end of each week along with refilling, and pH was maintained at a mean of around 7 in all treatments throughout the entire cultivation period (data not shown).
Plant growth, and fruit production and quality as influenced by nutrient solution concentration are shown in Table 2. The shoot weight, stem diameter and total yield were greatly improved with a higher-concentration solution, while the fruit yellow-shoulder incidence and index decreased (Table 2; Fig. 5a). For the shoot weight, that for EC 1.2 was significantly higher than those for EC 0.9 and EC 0.6, which was attributable to the greatest expansion of upper leaves with the former (data not shown). Owing to the high yield at the 2nd truss, the total yield in EC 1.2 was significantly higher than the others. The stem diameter below the 3rd truss and the 3rd truss yield in EC 0.9 were significantly higher than in EC 0.6 (Table 2). The YSI was 0.67 ± 0.04 in EC 0.6, which was double those in EC 0.9 and EC 1.2 (Table 2). The incidence showed the same tendency as the YSI. “Severe” fruit classified as being unmarketable accounted for 12% in EC 0.6, but only 3% in both EC 0.9 and EC 1.2 (Fig. 5a). There were no K- and P-deficiency symptoms of leaves in both experiments (data not shown). The fertilization (NO3-N) efficiency levels were 7.3, 6.1, and 4.9 kg·e−1 at EC 0.6, EC 0.9, and EC 1.2, respectively, and there was a significant difference among them (P < 0.05, n = 30) (data not shown).
Tomato plant growth and fruit production as influenced by nutrient concentration in Experiment 1.
Fruit yellow-shoulder incidence as influenced by nutrient concentration or composition (a and b in Experiments 1 and 2, respectively). Values are the means ± SE (n = 3). Fruit yellow-shoulder incidence (%) = (no. of yellow-shoulder fruit/total no. of fruit) × 100.
As shown in Figure 4, the uptake rates of all nutrients examined in this study were improved in the P supplementation treatment. Except for K, the uptake patterns of other nutrients in EC 0.9+P+K were similar to those in EC 0.9+P, while those in EC 0.9+K were similar to those in EC 0.9. EC 0.9+P and EC 0.9+P+K showed larger fluctuations than EC 0.9 and EC 0.9+K. In contrast, the uptake rate of nutrients was not influenced by enhanced K concentration, except for K itself. The K uptake rate was enhanced by P supplementation. The average weekly K uptake rates throughout the entire period in EC 0.9, EC 0.9+P+K, EC 0.9+P, and EC 0.9+K treatments were 9.2, 14.5, 12.2, and 11.2 me/plant/week, respectively. For the P-supplemented treatments, the average increment of weekly K uptake rate was 3.1 me/plant/week in EC 0.9+P and EC 0.9+P+K compared with those of EC 0.9 and EC 0.9+K, while the K-supplemented treatments of EC 0.9+K and EC 0.9+P+K showed an average increment of 2.2 me/plant/week compared with those of EC 0.9 and EC 0.9+P (Fig. 4).
Tomato plant growth and fruit production as affected by P and K are shown in Table 3. In comparison with EC 0.9 and EC 0.9+K, the plant growth and fruit production were all improved in P supplementation treatments (EC 0.9+P and EC 0.9+P+K). Especially for EC 0.9+P, the shoot weight, stem diameter, 2nd truss yield, and total yield were all significantly higher than in EC 0.9. In contrast, the effect of EC 0.9+K on tomato plant growth and yield was not obvious. The YSI was reduced to 0.20 ± 0.09 at EC 0.9+K and 0.15 ± 0.06 at EC 0.9+P+K in contrast to 0.31 ± 0.11 in EC 0.9 and 0.25 ± 0.06 in EC 0.9+P (Table 3). Furthermore, the incidences of unmarketable fruit (Severe status) from high to low were 3% in EC 0.9, 2.2% in EC 0.9+K, 1.1% in EC 0.9+P, and 0.3% in EC 0.9+P+K (Fig. 5b). Consequently, the reduction of fruit yellow-shoulder disorder tended to be greater in P- or K-supplemented treatment, or upon their combined usage.
Tomato plant growth and fruit production as influenced by nutrient composition in Experiment 2.
An appropriate balance between vegetative and reproductive growth is an indispensable consideration for accommodating LN&HD cultivation of tomatoes grown in ELVS. A previous experiment showed that total N concentration of 6.5 me·L−1 in 0.9 dS·m−1 EC of nutrient solution was sufficient for plant vegetative growth because of the optimal expansion of upper leaves (Egishi et al., 2012), and the yield once reached up to 2 kg/plant (3 trusses) by using EC 0.9 dS·m−1 nutrient solution in our preliminary experiment. As stated in the introduction, the planting distance is only 17 to 24 cm in the D-tray cultivation system in the case of LN&HD culture. It is necessary to keep plants in a compact state to accommodate a small planting distance. In this study, fruit yield was expected to reach up to the same level as in EC 1.2 dS·m−1 by adding P or K, or upon their combined usage, based on EC 0.9 dS·m−1. As a result, leaf and stem fresh weight, and stem diameter in EC 1.2 were significantly higher than in EC 0.9 and EC 0.6, and total yield was greatest at EC 1.2, which was the same as in previous research (Egishi et al., 2012). However, marketable yield and YSI were similar between EC 0.9 and EC 1.2 treatments, and fertilization (NO3-N) efficiency per marketable fruit at EC 0.9 (6.1 kg·e−1) was significantly higher than at EC 1.2 (4.9 kg·e−1). Therefore, EC 0.9 dS·m−1 was considered as a comparable basis for fertilization reduction and avoiding expanded and overlapped foliage.
During the tomato reproductive phase of growth, fruit and stem are the major sink organs for K and carbon assimilates from the source (Bar-Tal and Pressman, 1996; Kanai et al., 2007). Moreover, K is the most abundant cation present in the phloem sap (almost 80% of the total cations) as a consequence of sugar charging and transport mechanisms/processes through the phloem into sink organs (Cakmak, 2005). In the present study, the pattern of K uptake rate rose up to the highest level during the fruit enlargement and ripening stages, was then reduced with decreasing fruit number during the harvesting stage, and was finally reduced to the same level at the end of harvest in Experiments 1 and 2 (Figs. 3 and 4). These findings are the same as reported by Ishihara et al. (2007). In addition, weekly averages of 3.1 me/plant and 2.2 me/plant in terms of K uptake rate were increased in response to P and K supplementation, respectively. K uptake seems more likely to be influenced by nutrient solution concentration or P plays an indirect role in the process of K uptake. Not only K uptake but also the uptake of other nutrients including P itself was greatly strengthened by enhancing P concentration, but this enhancement was not observed in K supplementation treatment.
There is evidence suggesting that high-P-concentration solution improved the uptake rate of other nutrients (N, K, Ca, and Mg) in tomato DFT culture system (Terabayashi et al., 2004a). The effects of P on early root growth, expansion and contraction of stem, and fruit diameter are mainly attributable to the regulation of CO2 assimilation in leaf photosynthesis (Jacob and Lawlor, 1992). However, P affected plant growth directly through regulating the hydraulic conductivity of water in the roots (Clarkson et al., 2000; Fujita et al., 2003; Papadopoulos, 1991; Radin and Matthews, 1990). This change in hydraulic conductivity of the roots affected leaf expansion, which depends on cell multiplication and elongation of the newly formed cells in plants, and turgor pressure is a crucial factor for cell expansion (Hsiao, 1973; Munns et al., 2000). The partitioning of solutes to the fruit from the source leaf is dependent on adequate phloem turgor (Fujita et al., 2003). In the present study, supplemented P largely promoted sink organ growth and expansion of upper leaves, which were not observed in the EC 0.9+K treatment. The total yields in EC 0.9+P and EC 0.9+P+K reached up to the same level as in EC 1.2 treatment in Experiment 1 (Table 2).
Fujita et al. (2003) reported that low P impaired sink activity, directly reducing the size of the fruit, and resulted in the suppression of leaf photosynthesis and reduced stomatal conductance. Thus, both source and sink organs are parts of a single inseparable system and an effect on one part is bound to have a consequential and concurrent influence on the other. In this study, stronger sink organs in P supplementation probably needed sufficient phloem turgor to maintain a good source-sink balance cycle in the plant, resulting in higher uptake of nutrient elements (Fig. 4) and nutrient content in leaves (data not shown).
Fruit yellow-shoulder disorder was reduced in P and K supplementation treatment, as described previously in the results (Tables 2 and 3; Fig. 5). There was a direct relationship between higher K nutrition and a lower incidence of yellow-shoulder disorders in the greenhouse environment (Picha, 1987; Picha and Hall, 1981). Soil-exchangeable K was reported to have a significant influence on yellow-shoulder disorders (Hartz et al., 1999). Alba et al. (2007) showed that higher exchangeable K and P nutrient status had a positive correlation with tomato fruit color. In addition, fruit tissue affected by coloring disorder contains less than half the P and K levels when compared with red tissue (Picha, 1987), suggesting that both P and K are involved in tomato fruit coloring. It can be concluded that yellow-shoulder disorder was reduced to different levels by supplementation of P or K or both of them. As described earlier in this paper, the improvement of K uptake was greater in EC 0.9+P (3.1 me/plant/week) than in EC 0.9+K (2.2 me/plant/week) throughout the entire period of study (Fig. 4). The fact that K nutrition has a greater role in tomato fruit coloring disorder than P has previously been reported. Evidence indicated that the application of K fertilizers can enhance the levels of carotenoids in tomato, especially lycopene (Helyes et al., 2009; Ramírez S. et al., 2009; Trudel and Ozbun, 1971). Furthermore, K may be involved in the activities of one or more enzymes, such as phytoene synthase and phytoene desaturase, that synthesize phytoene from geranylgeranyl diphosphate, which is the first committed step in the carotenoid biosynthetic pathway (Taber et al., 2008). Consequently, it is necessary to clarify the relationship between yellow-shoulder disorder and K and P uptake, and fruit carotenoid metabolism in further research.
In the case of ELVS culture, plant roots are restricted to 250 mL volume, so the substrate conserves less water or nutrient elements unless nutrient solution is applied at high-frequency and a small-volume. The available mineral elemental nutrients for plants are markedly less than for the conventional NFT or DFT system. Terabayashi et al. (2004a) showed that tomatoes can be cultured in a DFT system with a limited supply of nitrate and phosphorus at 50 and 20 me/plant/week, and the initial concentrations of NO3-N, PO4-P, and K were 8.0, 3.2, and 6.0 me·L−1, respectively. Moreover, Otsuka-A formula nutrient solution was used for tomato cultivation by high wire layering in a closed hydroponic system with substrate, in which NO3-N, P, and K concentrations were 7.0, 2.0, and 3.7 me·L−1, respectively (Ishihara et al., 2007). In our preliminary experiment, P- or K-deficiency symptoms easily occurred when we used the Enshi formula nutrient solution at EC 0.9 dS·m−1 for tomatoes grown in ELVS. In the present study, P and K concentrations were regulated up to 4.6 and 5.8 me·L−1, respectively, based on EC 0.9 dS·m−1, but NO3-N was kept at 6.0 me·L−1. Compared with conventional NFT or DFT, the nutrient solution concentrations for tomatoes grown in ELVS culture are characterized by lower N and higher P and K concentrations.
As a result of the present study, it was shown to be effective to increase the P and K concentrations of nutrient solution on the basis of Enshi nutrient solution formula at EC 0.9 dS·m−1 in order to maintain optimal tomato vegetative growth, achieve higher yield comparable to that for EC 1.2 dS·m−1, and produce fewer P- and K-deficiency symptoms in the case of ELVS combined with LN&HD culture by a high-frequency and small-volume application method.
