Study area

The study took place in a zenithal greenhouse located in the premises of the Colegio de Postgraduados, Campus Montecillo, in the Municipality of Texcoco, Estado de Mexico (19° 29’ N, 98° 54’ W and 2250 meters above sea level).

Plant material

Twelve-month old blueberries cv. Biloxi were planted in a hydroponic system with tezontle sieved in a 1 cm² mesh in 20-L plastic bags. Differences between treatments -as a result of the effect of the substrate- were distinguished at the beginning and end of the experiment, based on their physical properties: granulometry, porousness, water retention, and airing. Plants were irrigated using a modified Steiner universal solution. The Biloxi variety was chosen for its unfettered propagation and adaptability for production in Mexico.

Experimental design

A complete randomized block design was used for the experiment and the following treatments were applied: 1) control, irrigated until the point of drainage (100 % of the daily demand of water) and an osmotic potential of -0.027 MPa (3.4 mmolc.L^-1 of Ca⁺⁺, 1.5 mmolc.L^-1 de Mg⁺⁺, 2.6 mmolc.L^-1 of K⁺, 4.5 mmolc.L^-1 of NO₃^-, 0.4 mmolc.L^-1 of H₂PO₄^-, 2.6 mmolc.L^-1 of SO₄⁼), according to ^{Steiner (1984)}; 2) saline with two levels, -0.041 Mpa (5.1 mmolc.L^-1 of Ca⁺⁺, 2.3 mmolc.L^-1 of Mg⁺⁺, 4 mmolc.L^-1 of K⁺, 6.8 mmolc.L^-1 of NO₃, 0.6 mmolc.L^-1 of H₂PO₄^-, 4 mmolc.L^-1 of SO₄⁼) and -0.054 Mpa (6.8 mmolc.L^-1 of Ca⁺, 3 mmolc.L^-1 of Mg⁺⁺, 5.3 mmolc.L^-1 of K⁺, 9 mmolc.L^-1 of NO₃^-, 0.8 mmolc.L^-1 of H₂PO₄^-, 5.3 mmolc.L^-1 of SO₄⁼), and 100 % humidity until drainage; also, 2 levels of water stress -66 % and 33 % humidity, in relation to the 100 % of the daily demand- and with an osmotic potential of -0.027 MPa, in both cases. Four repetitions were carried out, in 20 experimental units, and three blueberry plants per unit. The plants of the control treatment were used to evaluate water and salinity stress. Overall, the effect on 60 plants was evaluated. The water used was obtained from the rainwater harvest system of the Colegio de Posgraduados, with an electric conductivity (CE) of 0.01 dS m^-1.

Response variables

The number of fruits per plant was determined on a weekly basis, and then the weight of the fresh matter -expressed in g per plant- was quantified; additionally, the equatorial and longitudinal diameters were also measured using a 6-inch digital caliper, taking into consideration the pedicel and the clycine scars of the fruits.

The leaves were sampled and harvested homogenously between treatments and experimental units in the reproductive and vegetative stages; at the end of the experiment, all the leaves were taken into account to obtain the weight of the fresh and dry matter. To obtain the dry biomass, the leaves and fruits were dried in a forced air circulation oven at 70 °C ±1 for 7 d; subsequently they were weighed and the results were expressed in g of dry matter per plant. The plant material was ground and stored in jars at 5 °C for further analysis. To obtain the dry weight, the roots and stems were analyzed at the end of the cycle, using the same procedure applied to the leaves.

Data analysis

The data were analysed using ANOVA and the means were compared using a Tukey test (p≤0.05) using SAS version 9 (SAS Institute Inc., 1999).

Dry matter of root, stem, leaves and fruits

Statistical differences (p≤0.05) were observed in the dry biomasses of roots, leaves, fruits, and the total dry matter (MST), as a result of the salinity stress of the nutrient solution (Table 1).

The treatments with a PO below -0.027 MPa had a negative impact on the development of roots, leaves, fruits, and total biomass, and were statistically different to the control treatment. ^{Leidi and Pardo (2002)} mention that salinity stress reduces water absorption capability, therefore diminishing foliar expansion and causing loss of turgor. In our experiment, the total dry biomass decreased, as the PO of the nutrient solution increased, since the plant cells exposed to a saline medium balance their hydric potential with the loss of water and this, in turn, reduces osmotic potential and turgor.

Growth reduction was not caused by a reduction in metabolism, but rather by the loss of turgor (physical process). As the water content of the plant decreases, the same occurs inside the cells, decelerating cell volume and the turgor of the cell and increasing the number of solutes and mechanical damages to the cell (Table 2).

Therefore, water stress directly inhibits some cell growth mechanism (Table 2). Total biomass and all its components decreased as the amount of water in the substrate was limited, in relation to the control. The root:stem ratio increased in inverse proportion to the amount of water provided to the plant, indicating that the aerial biomass was more affected by water stress.

Fruit quality

For most of the fruit quality variables, there were no significant statistical differences (p˃0.05) between treatments, as a result of water or salinity stress (Table 3). For water stress, there were only statistical differences (p≤0.05) in the number of fruits per plant variable. No statistical evidence was found that drought has a negative effect on the average fruit weight, or the equatorial and longitudinal diameters evaluated.

In this regard, ^{Mingeau et al. (2001)} show that an increase in water stress reduces fruit size and weight, which has a negative impact on the highbush blueberry plant yields. ^{Ehret et al. (2012)} reported that blueberry plants concentrated more sugars when soil moisture was lower than in control; it also increased the firmness of the fruit and its water loss in postharvest. However, in such treatments, the weight of the fruit, its titratable acidity, and its water content decreased.

Number of leaves, foliar area, specific leaf weight, root length, and plant height

Table 4 shows that only root length has significant statistical differences (p≤0.05) as a consequence of salinity stress, while such differences were found in foliar area, specific leaf weight, and root length in treatments with water stress. According to ^{Wright et al. (1993)}, increasing the salinity of the nutrient solution with 100 mM of Na⁺ in blueberry plants significantly affected these variables. ^{Bryla and Machado (2011)} show that reducing the PO of the nutrient solution with (NH₄)₂SO₄ below -0.054 MPa in highbush blueberry plants reduces the MS content in roots, as well as the AF, which matches the results of this research.

AF and PEH decreased significantly when the blueberry plants recieved 33 % of the daily demand of water (dda), in comparison with the control (100 % of the dda). ^{Mingeau et al. (2001)} point out that blueberry plants are very sensitive to drought, and that increasing water stress reduces transpiration, stem diameter, and foliar area in highbush blueberry plants.

LR was favored when moisture was reduced to 33 % of the dda. This is how plants respond to drought, with a mechanism by means of which their roots tend to elongate faster searching for water further away.

Some variables measured in this research suffer a greater impact from water or salinity stress (foliar area, number of fruits, and dry weight variables) and could serve as a model to be applied in the production of blueberries in areas in which weather conditions pose a challenge regarding water or salinity. Additionally, the negative effect of water stress was greater in blueberry fruits and it was expressed in the dry weight and number of fruits.