Location and environmental conditions of the place

The present work was carried out in a tunnel type greenhouse of 120 m², with passive climate control, in which a closed hydroponic system was established, in which three irrigation times were measured with one and two drippers, to obtain six percentage of drainage of the SN from the pots to a reservoir for recycling, the greenhouse is located in the Academic Unit of Agronomy of the Autonomous University of Zacatecas, at the coordinates: 22.72° north latitude and 102.68° west longitude, and altitude of 2 440 m.

Treatments and experimental design

Six SN drainage treatments were evaluated, resulting from the combination of three irrigation duration times (1, 2 and 3 min) and two streams of the emitters (2 and 4 L h^-1), the drainage percentages (%D) of the SN in the irrigation times mentioned and the flow of 2 L h^-1 were: 25, 30 and 40, which correspond to treatments 1, 2 and 3, respectively; the combinations with the same irrigation times and the flow of 4 L h^-1, the% D were 45, 50 and 60, corresponding to treatments 4, 5 and 6, respectively. The frequencies of the irrigation varied between 20 and 30 min, depending on the stage of development of the plants.

Each treatment consisted of five repetitions, where each experimental unit consisted of six pots with 25 liters of fine perlite as a substrate and two tomato plants in each pot, having a density of 3 plants per square meter. Each row of six pots had an SN capture system that by gravity drove it from the pots to a container to determine the volume of the SN drained, with this value and the volume of SN applied per day, the was determined %D (^{Massa et al., 2010}), the drained SN of each treatment was conducted to a deposit with a capacity of 1 000 L in which electrical conductivity and pH were measured, to restore SN in these parameters.

Plant material and production system

The variety of tomato (Lycopersicum esculentum Mill.) Used was the hybrid "El Cid", of indeterminate growth habit, of the saladette type. The transplant was performed 35 days after sowing. The pearlite was reused from four previous cycles for the production of tomato, cucumber (Cucumis sativus L.), tomato and tomato again. Perlite was disinfected before being used in each culture cycle with sodium hypochlorite at a concentration of 4 mg L^-1 (^{Cayanan et al., 2009}).

Handling the nutrient solution during the experiment

The SN Steiner was used in 75% of the first irrigations (between 28 and 32 irrigations) of each day, the drained SN of each treatment was stored in its respective reservoir, before the last eight irrigations of each day the SN drained was restored adding the necessary water to reduce the CE to 1.7 ±0.1 dS m^-1 and the necessary amount of a 1 N solution of potassium nitrate to achieve an CE of 2 ±0.1 dS m^-1 and the required amount of a 1 N solution of sulfuric or phosphoric acid to re-establish the pH at 5.5 ±0.1. Phosphoric acid was only used when the phosphorus content in the SN dropped to 35 ppm.

The water used to prepare the SN had: pH= 7.23, CE= 0.55 dS m^-1, concentration of soluble ions in mol_c m^-3: NO₃= 0.21, P-PO₄= 0.02, SO₄= 0.70, HCO₃= 3.60, Cl= 0.80, Ca= 1.85, Mg= 1.48, K= 0.26, Na= 2.09 and in mg L^-1 the micronutrients: Fe= 0.03, Mn= 0.02, Zn= 0.01, Cu= 0 and B= 0.1.

The fertilizers used to prepare SN were calcium nitrate (Ca(NO₃)₂·0.2NH₄NO₃·H₂O), potassium nitrate (KNO₃), potassium sulfate (K₂SO₄), magnesium sulfate (MgSO₄·7H₂O), monopotassium phosphate (KH₂PO₄), phosphoric acid (H₃PO₄) and sulfuric acid (H₂SO₄). To provide the micronutrients, a fertilizer was used that contains a concentration of: 6.6, 2.6, 1.1, 0.9, 0.3 and 0.2%, Fe, Mn, Zn, B, Cu and Mo, respectively, whose concentration in the SN was: 1.98, 0.78, 0.33, 0.27, 0.09 and 0.06 mg L^-1, for the respective micronutrients; Fe, Mn, Zn and Cu were applied as a chelate with EDTA. The water used was directly from deep well, the presence in it of bacteria and fungi was negative.

Variables measured in the plant

In a plant of each experimental unit the following measurements were carried out.

Performance. It was expressed in kilograms produced per square meter; three fundamental performance components were also evaluated, which were:

Number of fruits per plant. The fruits harvested during the cycle that reached physiological maturity were counted and finally the average of the harvested fruits per plant was obtained.

Size of fruits. It was obtained at the time of harvest with a ScalaTM digital vernier in millimeters measuring the equatorial and polar diameters.

Weight of fruits. The fruits were weighed when harvested, the weighing was performed with a digital scale EK-200i with a precision of 0.01 g. The measurement of fruit size and weight allowed to classify the fruits in categories according to the following scale: 1a (>130 g), 2a (100-130 g), 3a (60-100 g) and 4a (<60 g).

Variables measured as fruit

Fruit samples were taken in two stages of production, with the same maturity index (color), in the fourth and in the eighth bunches, the samples were constituted by five fruits considering the same position of them in cluster. The fruits were crushed and ground in a Krups 267 juice extractor. After extracting the juice with an 80 mesh fiber, it was used to determine the following quality variables:

Total soluble solids (SST). A drop of the juice of the obtained sample was taken and placed in the sensor refractometer Atago, model N-1E, with a reading range of 0 to 32 °Brix.

pH and electrical conductivity. They were measured with a LAQUAtwin B-712 potentiometer for the pH test and a Laquatwin B-771 conductivity meter for electrical conductivity in a 50 mL sample of juice extracted from the fruits, the measurement accuracy was 0.01 in both properties.

Titratable acidity. It was determined by the method used by ^{San Martin et al. (2012}). This variable was expressed as a percentage (AT%) as a percentage of citric acid.

Maturity index (IM). For the determination of this parameter, it was calculated with the method used by ^{San Martin et al. (2012}), obtaining the relationship between SST and AT.

Variables measured in the nutrient solution

CE and pH. These parameters were measured with a PCE-CM 41 meter for electrical conductivity and a PCE-PH 22 meter for pH. These parameters were measured to regulate the nutrition of the crop and have a normal and balanced development and were measured both in the applied SN and in the drained.

To restore the pH, sulfuric and phosphoric acids were used, the amount applied was that required to lower the pH again to a value of 5.5.

Volume of solution applied. An extra dropper was installed in each experimental unit in a container to capture the solution applied in one day, at the end of the day it was measured with a graduated cylinder, obtaining the volume applied per day per plant.

Volume of drained solution. In each container (drainage tray) installed in each repetition the drained SN was collected and its volume was measured with the help of a graduated cylinder, per day and per experimental unit. This parameter also allowed to control the irrigation (frequency, between every 30 and 40 minutes, and number of irrigations, between 30 and 50) and the restoration of the pH and the CE of the SN.

Drainage percentage (%D). It was obtained by the relationship between the volume applied and the volume drained.

Water productivity. A record was made of the amount of SN applied for each treatment during the crop cycle. Subsequently, the ratio in kilograms per cubic meter of the amount of water applied and the amount of fruit produced with such quantity was obtained.

Statistical analysis

The data obtained were analyzed in an experimental design in randomized complete blocks and subjected to an analysis of variance (Anova) and a means test using the Tukey criterion, with (α≤ 0.05). This analysis was carried out with the statistical program SAS (Stadistical Analysis System) version 9.4.

Variables measured in the plant

Performance. The yield was greater in the treatments with 40 and 45% D with respect to the other treatments (Table 1), these yields are higher 14% than those obtained by ^{Sánchez et al. (2014)}, where they evaluated open and closed hydroponic systems, the present investigation provided results a little lower than those obtained by ^{Oztekin et al. (2008)} and ^{Nakano et al. (2010)} since they did not find significant differences using different and new substrates and different open and closed hydroponic systems in tomato culture.

Number of fruits. The number of fruits per plant was not different among the treatments evaluated (Table 1). The number of fruits in the treatments oscillated with greater frequency between 40-50 fruits per plant during the 18 cuts made throughout the cycle (may to november), these results are similar to those obtained by ^{Urrieta et al. (2012)}. The reason for not having differences is probably due to the fact that the number of fruits is strongly influenced by some morphological characteristics of the plants, among them the type of inflorescence and the number of flowers per bunch (^{Rivas et al., 2012}).

Size of fruits. This variable was greater in the treatments with 40 and 45% D with respect to the other treatments (Table 1); however, the results of the six treatments are similar to those of ^{Ucan et al. (2004)}. The size of the fruit is strongly related to the number of fruits, since, to a smaller number of fruits, the greater number of photoassimilates destined to each fruit, favoring larger fruit size (^{Ucan et al., 2004}; ^{Rivas et al., 2012}).

Weight of fruits. The weight of the fruits was strongly related to the size, to larger size, greater weight, for this reason the weight of the fruits was also greater in the treatments with 40 and 45% D (Table 1), these treatments presented an average weight of superior fruit 23% with respect to the treatment with the lowest average weight of fruit, corresponding to the treatment with 25% D, which may be due mainly to that in this treatment there was greater stress in the plant affecting its yield (^{Sengupta and Banerjee, 2012}; ^{Sánchez et al., 2014}).

The variables mentioned above (size and weight of fruits) it was determined that the fruits that predominated had weights of 60 to 100 g for the treatments with 25, 30, 50 and 60 %D and of category 2 (100-130 g) for treatments with 40 and 50 %D (Table 1), according to the scale established for the sizes (polar and equatorial diameters) and weights of the fruits.

Variables measured as fruit

Total soluble solids (SST). The treatment with 25 %D was statistically greater than the others in both stages of measurement; likewise, the treatment with 30 %D was also greater in the second stage, the results obtained varied from 4.85 to 5.99 °Brix (Table 2), these values are higher than those found by ^{Rivas et al. (2012)} from 4.57 to 5.10 °Brix in different tomato cultivars and ^{Urrieta et al. (2012)} from 4.81 to 5.33 °Brix using native varieties as plant material.

The SST are a parameter of fruit quality that varies with the electrical conductivity of the SN and the hydric stress of the fruit during its development (^{Cespedes et al., 2004}), the greatest water stress condition in this investigation was presented in the treatments of 25 and 30 %D, the tomato fruits of plants under this type of stress store mainly ions and organic molecules such as fructose and glucose (^{Munns, 2002}), which increases the SST. The results of the present investigation are in agreement with those found by ^{Chang et al. (2012)} who found that with SN drainage percentages lower than 30% or close, SST increase and as %D increases, SST decreases.

pH of juice. The pH in the juice did not have a significant difference in any of the treatments evaluated. The pH is an indicator of quality, the best values in tomatoes of quality for this variable are between 4 and 5 (^{Aguayo et al., 2004}) and the results obtained in this investigation agree with the mentioned thing (Table 2), a juice of Fruits with lower acidity have greater acceptance, since this way their sweet and pleasant taste stands out more when consumed fresh (^{Urrieta et al., 2012}).

Electric conductivity. In this variable values were found from 5.30 (treatment with 60 %D, Stage 1) to 6.32 (treatment with 25 %D, stage 2) (Table 2). The treatments with 25 and 30 %D presented higher CE, due to the accumulation of solutes (salts) in the substrate, the results found in this

variable in all the treatments agree with those observed by ^{San Martin et al. (2012}), who found that there is a direct relationship between the accumulated salinity in the substrate and the electrical conductivity of the fruit; likewise, as %D is decreased, the accumulation of salts increases the CE in the root and also in the juice of the fruit (^{Sengupta and Banerjee, 2012}).

Acidity titulable. The treatments tested showed no statistical differences in the two production stages (Table 2). The six treatments presented values from 0.15 to 0.32% acidity (Table 2), similar to those obtained by ^{Bugarin-Montoya et al. (2002}) (0.19-0.32%) with different concentrations of potassium in a closed hydroponic system, and slightly lower than those found by ^{Rivas et al. (2012)} (0.25-0.43%) using diverse cultivars.

Maturity index. This index was higher in the treatment with 55 %D in the first stage and with 55 and 60 %D in the second stage, which is due to the combination of high values of SST with low AT, the results obtained in the six treatments varied between 16.9 and 33.4 for the IM (Table 2) with an average of 26.7. With values obtained in a range of 12 to 18, the maturity index is the one indicated for many tomato varieties (^{Bugarin-Montoya et al., 2002}). According to ^{Kader et al. (1978)} when the IM is greater than 10 the tomato fruits are considered of better quality.

Variables measured in the nutrient solution

It is important to emphasize that these variables in the nutrient solution were only measured in order to carry an adequate plant nutrition and correct application of the same, in order to have a balance in the development of the plants.

CE and pH. In the applied SN, values were maintained between 1.7 and 2.6 dS m^-1 and between 5.5 and 6 respectively, depending on the stage of development in which the crop was found as recommended by ^{Putra and Yuliando (2015)}; ^{Baca et al. (2016)}. The EC and the pH of the drained SN were increased during the cycle (Results not reported), with averages of 4.28, 4.19, 4.11, 4.04, 3.82, 3.77 and 4.32 dS m^-1 CE for the six treatments and means of 6.65, 6.62, 6.79, 6.37, 6.58, and 6.5.

In the CE, these results are probably due to the fact that the plant absorbs more water than nutrients in proportion, and taking into account the amount of water lost through evaporation, the drained solution will have a concentration of solute (nutrients) greater than of water (solvent), thus increasing the electrical conductivity. Results similar to the present investigation found ^{Chang et al. (2012)} with different percentages of drainage and percentage of zero drainage, since the electrical conductivity increased when decreasing the percentage of drainage obtaining from 2-6 dS m^-1 with percentages of 30 to 50%.

Water productivity The productivity obtained in the six treatments ranged between 18 and 43 kilograms of fruit produced per cubic meter of water applied (Table 1), although there was no difference in the number of fruits, if there was a significant difference for productivity , presenting the highest values with the lowest drainage percentages (25 and 30%) since they were in which the amount of water applied was lower and as the percentage of drainage increased the productivity of the water decreased. The productivity reflects an efficient use of water, in hydroponic systems and especially closed ones, water savings can be achieved from 30 to 40% (^{Komosa et al., 2011}) or up to 46% (^{Dasgan and Ekici, 2005}).