- Email: [email protected]
Plant water uptake and water use efficiency of greenhouse tomato cultivars irrigated with saline water
Agricultural Water Management 78 (2005) 54–66 www.elsevier.com/locate/agwat
Plant water uptake and water use efficiency of greenhouse tomato cultivars irrigated with saline water A. Reina-Sa´nchez, R. Romero-Aranda, J. Cuartero * Estacion Experimental La Mayora, CSIC, 29750 Algarrobo-Costa, Malaga, Spain Accepted 1 April 2005 Available online 6 July 2005
Abstract Effects of salinity on tomato (Lycopersicon esculentum Mill.) fruit yield, plant water uptake and water use efficiency (WUE) have been quantified in experiments carried out under greenhouse and soil-less cultivation with four cultivars (Floradade, L1, L5 and L9) and four salinity levels (0, 25, 50, and 75 mM NaCl). Fruit represented 70% of plant fresh weight while leaves and stems represented 22 and 8%, respectively. Fruit were the most sensitive part of the plant, with the four cultivars showing similar significant fruit yield reduction, namely about 28 g/mM NaCl or 290 g/dS m1. Yield threshold varied from 0 to 3.4 dS m1, values lower than or close to the electrical conductivity (EC) of nutrient solutions used in commercial greenhouses. Yield reduction from threshold to upper salinities was about 8% of maximum yield per dS m1 increase. Blossom end rot increased with salinity although the pattern of increase depended on the cultivar. Tomato fruit grown under saline conditions had higher soluble solids and acid content than those from the control (0 mM) plants. Plants grown under the most saline conditions consumed, on average, 40% less water than control plants. The relationship between total plant water uptake and salinity was linear (negligible threshold) and salinity of the nutrient solution almost entirely explained the variations in plant water uptake (R2 from 0.94 to 1); therefore, salinity of the irrigation water has to be taken into account when calculating tomato water requirements. However, significant differences in the negative slopes of the correlation lines indicate that decreases in plant water uptake, from 3.5 to 5% per dS m1, are cultivar-specific and cannot be generalised. Vegetative (stem and leaves) dry weight was a better indicator of tomato plant water uptake in saline conditions irrespective of cultivars than fruit yield or plant and fruit dry weight. Tomato plants in the control averaged a higher WUE than the most * Corresponding author. Tel.: +34 952552656; fax: +34 952552677. E-mail address: [email protected] (J. Cuartero). 0378-3774/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.agwat.2005.04.021
A. Reina-Sa´nchez et al. / Agricultural Water Management 78 (2005) 54–66
55
salinized plants when WUE was expressed as a function of fruit yield (25 and 13 g fruit L1); however WUE was independent of salinity if expressed as a function of plant and fruit dry matter (approximately 3.0 g dry matter L1). Tomato plants absorbed only a small proportion of the Na+ present in the nutrient solution (from 2.3 to 3.2%) but there were significant differences among the four cultivars which suggest that plant ability to select ions is a trait to be taken into account when selecting tomato genotypes for salt tolerance. # 2005 Elsevier B.V. All rights reserved. Keywords: Lycopersicon esculentum; Yield; Fruit and plant dry weight; Fruit quality; Blossom end rot; Na+ uptake
1. Introduction The tomato crop is adapted to a wide variety of climates ranging from the tropics to within a few degrees of the Arctic Circle. However, in spite of its broad adaptation, horticultural production is concentrated in a few warm and rather dry areas: about 34% of world production comes from countries around the Mediterranean sea and about 14% from California and Mexico (FAO, 2002). These areas are also those where the highest yields are reached but also where salinity is a serious restraint not only for planting new lands to this crop but also for maintaining in high productivity those currently under irrigation (Ghassemi et al., 1995). The increasing demand for domestic, industrial, environmental and recreational water will force agriculturists to manage irrigation water carefully, contributing to environmental preservation. In parallel, brackish and saline water resources not used nowadays could be employed for irrigation if greater knowledge of salt tolerance and proper technology are developed (Shannon and Grieve, 1999). In applying saline/brackish water for irrigation, an integrated approach, which should account for soil, crop and water management at the same time, should be adopted. This approach needs calculation of crop water requirements which are essential for water saving, controlling water table level and drainage volume, and of course the final yield (Ragab, 2002). Tomato could act as a model crop for brackish/ saline water use because it is already grown in large areas with saline conditions, and because there is a wealth of important knowledge of the physiology and genetics of this species. It is known that tomato plants irrigated with a saline solution transpire less water than when fresh water is used (Soria and Cuartero, 1997; Romero-Aranda et al., 2000), but it is necessary to quantify the decrease in plant water uptake in relation to salinity and investigate possible differences in water uptake between cultivars, as a part of the integrated approach described by Ragab (2002) to use saline irrigation water. In this paper, we report on our investigations of those aspects on four tomato cultivars and the relations between plant water uptake and yield and biomass. Water use efficiency (WUE) in agronomical and biological terms (gram fruit and gram dry matter per litre transpired water, respectively) have been also determined as higher efficiency in plant dry matter and fruit formation will lead to relatively less uptake of toxic ions (Na+) (Cuartero and Fernandez-Mun˜ oz, 1999). Finally, we have determined Na+ concentrations in leaves, stem
56
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
and fruit in order to calculate a measure of root Na+ selectivity and transport to the shoot by relating plant water and Na+ uptake.
2. Material and methods Four tomato cultivars (Floradade, L1, L5, and L9) were chosen because of their demonstrated salt tolerance in Syria and Egypt (Floradade) and in Spain (L1, L5, and L9) in previous years. Plants were grown in a commercial polyethylene greenhouse in Malaga, Spain (368450 N, 48020 W) from 14 February to 14 June 2001. Seeds were sown in vermiculite on 12 January and, when seedlings had developed 5 true leaves (14 February), 40 plants/cultivar were transplanted to 40 pots of 20 L filled up with 3 mm diameter sand. Pots were arranged in rows 1 m apart, with 2 plants m1 in the rows. Two extra plants at the beginning and end of every row and two rows before and after the first and final experimental rows were planted to avoid border effects. A standard nutrient solution (10 mM N, 7 mM K, 0.9 mM P, 5 mM Ca, 2 mM Mg, and microelements), prepared with rain water, was supplied to the plants from transplanting to the end of the experiment by an automatic drip irrigation system, with one dripper per plant that discharged 2.3 L h1. To ensure that enough water was always present in the sand around root system, we set 21 irrigation periods/day (at 7, 9, 10, 10.30, 11, 11.30, 12, 12.30, 13, 13.30, 14, 14.30, 15, 15.30, 16, 16.30, 17, 17.30, 18, 19 and 21 h) of 2 min from 19 February to 3 April and of 3 min from 4 April to 14 June. Plants were grown as a single stem by removing side shoots weekly (the side shoots were removed when they were less than 5 cm in length to reduce the effects of wounding). Five days after transplanting, salt treatment began with 0 (control), 25, 50, and 75 mM NaCl added to the nutrient solution, resulting in electrical conductivities (EC) of 1.9, 4.7, 7.1 and 9.1 dS m1, respectively. Those treatments with added salt are referred in the text as saline treatments, with the salinity being determined either in terms of salt concentration or the conductivity of the solution. Ten plants/cultivar and salt treatment were grown although records were only taken from the six central plants; the two at the beginning and the two at the end were grown to reduce border effects. At each salt concentration, four pots without plants were irrigated in order to measure evaporation from the sand. Plant water uptake (transpiration) was calculated weekly for six plants per cultivar and salt treatment. Drainage was collected in trays located below the pots (with and without a plant) during a whole day. Evaporation from the trays was measured in four trays filled with nutrient solution—as weight difference between the beginning and the end of the day (at the times when leacheate was recorded). Plant water uptake was calculated by subtracting from the nutrient solution supplied the nutrient solution drained from planted pots, the water evaporated from unplanted pots and the water evaporated from the trays where drainage was collected. Fruit were harvested six times from 8 May to 14 June. In each harvest, commercial and non-commercial fruit (fruit with blossom end rot) were counted separately and weighed for the same six plants per cultivar and salt treatment used to determine water uptake. Three samples of fruit were taken and dried along the harvest period to calculate total fruit dry weight and to determine Na+ concentration. Vegetative dry weight was obtained by drying
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
57
and weighing stems and leaves of the six plants per cultivar and salt treatment at the end of the experiment; plant dry weight was calculated by adding fruit dry weight to vegetative dry weight. All the leaves produced were on the plant at the end of the experiment because old leaves were not removed during cultivation. Three samples of the dry leaves and three of the dry stem were taken from each plant, to determine Na+ concentration. The fruit dry weight and fruit Na+ concentration were used together with Na+ of dry leaves and stems to calculate total Na+ absorbed by the plant and stored in the shoot. To calculate the relative Na+ exclusion, the amount of Na+ absorbed by each plant was expressed as a percentage of the amount of Na+ that might have been absorbed had the nutrient solution passed through the plant and all the Na remained in the leaves. This parameter was taken as a measure of plant root Na+ exclusion, was calculated as 100 (total Na+ stored in the shoot/total Na+ content in a volume of nutrient solution equivalent to the amount of water transpired). Fruit soluble solids and organic acids content were determined in the three fruit samples following Zerbini et al. (1991); soluble solids (8Brix) were estimated at 20 8C by a refractometer illuminated with sodium light and acidity by titrating tomato juice with 0.1 N NaOH to a pH of 8.1 (NaOH meq/100 ml juice). One-way analysis of variance was used to compare the effects of salinity on each parameter that was measured for each cultivar (six data points per cultivar from the six plants at each salt concentration). A coefficient of correlation was used to quantify the relation between characters (6 plants and 4 salinities, making 24 data points per cultivar). The model proposed by van Genuchten and Hoffman (1984) for evaluating crop salt tolerance and to predict yield in different saline conditions was used to determine cultivar salt response curves for yield and plant water uptake. The program described by van Genuchten (1983) was used to calculate Ym (theoretical maximum yield), slope (yield decrease per unit salinity increase) and EC-threshold (maximum salinity without yield reduction) by nonlinear regression. Results are referred to salinity in mM, except where the van Genuchten and Hoffman (1984) model was used, when EC has been used to facilitate comparison of our results with results described in the literature.
3. Results Fruit represented 70% of the plant fresh weight while leaves and stems represented 22 and 8%, respectively. The yield of the four cultivars was similar in the absence of salt (control) but decreased with increasing salinity (Fig. 1) and to a greater extent than the other parts of the plant: all four cultivars showed a significant reduction in fruit yield. Even the lowest salt concentration (25 mM, 4.7 dS m1) produced a substantial reduction (from 12 to 38%) in yield, and at the highest salt concentration yield reduction was almost 70% of the control. The average yield reduction for the four cultivars was 28 g/mM or 295 g/ dS m1. Our data fitted well the model proposed by van Genuchten and Hoffman (1984) as R2 values were between 0.91 and 1. Threshold EC varied from 0 (L9) to 3.4 dS m1 (L5); Floradade and L5, showed threshold values higher than EC in control conditions (1.9 dS m1) while L1 and L9 had lower values. Yield reduction at EC values higher than the threshold ranged from 214 to 361 g/dS m1, which represented from 7.7 to 10.5% of the estimated maximum yield (the yield at salnities less than the threshold salinity). Yield
58
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Fig. 1. Relationship between yield (g plant1) and salinity (dS m1) in four tomato cultivars (six plants per cultivar and salinity).
was similar for the four cultivars at control and maximum tested salinity conditions; however, between those extremes L5 yielded more fruit than the others cultivars, which behaved similarly. Tomato yield is determined by fruit weight and fruit number. Expressing yield of the four cultivars in terms of the number of fruit and the weight per fruit, we found for Floradade that the highest salinity (75 mM) significantly reduced the number of fruit (65%) while fruit weight remained almost stable (11%). Salinity reduced fruit number and fruit weight of L1 in similar proportions, but had its major effect on fruit weight for L5 and L9 (54 and 62%, respectively). The average non-commercial fruit (fruit with blossom end rot) for the four cultivars increased with salinity from 2% in control (0 mM Na) conditions to 6, 12 and 16% in 25, 50 and 75 mM, respectively (Table 1). However, this apparently strong correlation between salinity and blossom end rot almost disappeared when the analysis was conducted cultivar by cultivar. Floradade tended to increase blossom end rot in the upper part of the tested salinity range (higher than 25 mM NaCl) while L9 increased blossom end rot in the lower range (from 0 to 50 mM NaCl); L1 showed strong positive correlation between salinity and blossom end rot while L5 showed independence. Fruit dry matter significantly increased with increasing salinity and fruit grown at 75 mM had more than double the dry matter of control fruit (Table 1). There were also differences between cultivars in fruit dry matter content in control conditions, although those differences were reduced by increasing salinity. Due to the higher fruit dry matter in saline conditions no significant differences appeared between salinity treatments when yield was expressed as fruit dry weight.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
59
Table 1 Commercial (C) and non-commercial (NC) yield (g plant1 and percentageof total yield, respectively), fruit dry matter (DM) (%), vegetative dry weight (DW) in grams (stem and leaves), plant dry weight in grams (stem, leaves, and fruit), water uptake (L plant1), and WUE (in grams of fruit and plant dry weight per litre of water) for four tomato cultivars grown at 0, 25, 50, and 75 mM NaCl in the nutrient solution Cultivar
NC Fruit Vegetative Plant Water uptake Fruit Plant NaCl C yield DW (g) (L plant1) (g)/L DW (g)/L (mM) (g plant1) yield (%) DM (%) DW (g) water water
Floradade
0 25 50 75
2789 2057 1800 882
a ab bc c
4.6 4.2 8.3 13.8
a a a a
3.7 5.8 7.6 10.4
d c b a
139 96 88 82
a b b b
249 238 230 220
a a a a
103 83 77 67
a b b b
31 26 23 15
a a a a
2.5 2.6 3.2 3.1
a a a a
L1
0 25 50 75
3310 2363 1438 1043
a ab bc c
3.0 10.8 21.5 26.2
b b a a
5.6 7.7 10.7 11.6
c b a a
215 152 152 143
a ab ab b
402 330 321 296
a ab ab ab
141 118 108 86
a b b c
25 18 14 13
a ab b b
3.5 3.1 3.4 3.4
a a a a
L5
0 25 50 75
3432 2924 2103 1343
a ab bc c
0.5 3.8 1.0 4.2
a a a a
3.6 5.6 6.2 11.3
c b b a
243 179 166 156
a b b b
372 364 299 304
a a a a
151 a 115 b 104bc 95 c
24 24 21 14
a a ab b
2.5 3.2 2.9 3.2
a a a a
L9
0 25 50 75
3170 1821 1362 815
a b b c
0.0 7.7 18.6 19.8
b b a a
5.4 6.8 9.5 11.5
c c b a
289 183 158 130
a b b b
457 308 317 242
a b b b
155 113 103 83
20 15 13 9
a ab ab b
2.9 2.8 3.1 2.9
a a a a
a b b c
Figures followed by different letter within a column and cultivar are significantly different ( p 0.95).
Fruit acidity and 8Brix increased significantly with increasing salinity (Fig. 2). The increase of 8Brix was similar in the four cultivars, however, for acidity, L9 showed a significantly higher increase than Floradade and L5. Fruit acid concentration measured as meq NaOH was about 2.3 times higher than soluble solids expressed as 8Brix in control fruit. Comparing the increase of 8Brix and acidity, the slope of the correlation lines was 2.5–4.5 higher for acidity than for 8Brix, indicating greater accumulation of acids than of soluble solids in fruit grown under saline conditions as well as a greater ratio of acidity/ soluble-solids (meq NaOH/8Brix). Vegetative dry weight was reduced dramatically at low salt concentration (25 mM) with a smaller reduction at higher levels of salinity (Table 1). Shoot plant dry weight (including the number of fruit) at the end of the experiment also tended to decrease when salinity increased but was significantly different from the control only for L5 and L9. Daily plant water uptake increased from 0.1 L plant1 at transplant time (19 February) to 1.70 L plant1 about 70 days after transplant (beginning of May) in the control treatment. From the beginning of May to the end of the experiment (14 June), daily plant water uptake was maintained at the same level. Maximum daily water uptake was lower (0.95 L plant1) at the highest salinity than under control conditions and the maximum was reached earlier, 60 days after transplant. Total plant water uptake was reduced by salinity for the four cultivars with the reduction at the highest salt concentration being almost 40% of the water uptake under control
60
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Fig. 2. Variation of tomato fruit soluble solids (8Brix) and acidity (meq NaOH) with salinity (mM NaCl) in four cultivars (three samples per plant and six plants per cultivar and salinity).
conditions (Fig. 3). Relationship between total plant water uptake and salinity in the nutrient solution fitted well van Genuchten and Hoffman (1984) model, for the four cultivars tested, with coefficients of determination (R2) higher than 0.93 which indicates that the salinity of the nutrient solution explained almost all the variation in water uptake. The intensity of the reduction in total plant water uptake varied between 4.8 and 9.5 L plant1 dS m1 increment in the root zone and was cultivar-specific for there were significant differences between the slopes of the cultivars: Floradade and L1 had statistically similar reductions, lower than for both L9 and L5, which had similar slopes.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
61
Fig. 3. Relationship between total plant water uptake (L plant1) and salinity (dS m1) in the nutrient solution for four tomato cultivars (six plants per cultivar and salinity).
The set of data obtained in this experiment allowed calculation of the correlation between water uptake and yield, plant dry weight (including fruit) and dry weight of the vegetative part. The correlation coefficient between water uptake and yield was lower than the coefficients between water uptake and plant dry weight and dry weight of the vegetative part (Table 2). The correlations between plant water uptake and yield calculated independently for each salt concentration were low (from 0.21 to 0.39) and statistically not significant, whereas the overall correlation between plant water uptake and yield appeared high (0.60). Plant dry weight and dry weight of the vegetative part were well correlated with water uptake under all of the tested salt concentrations (Table 2). In order to discover which one out of the three parameters—yield, plant dry weight, or vegetative dry weight—was a better indicator of plant water uptake, linear regressions Table 2 Coefficients of correlation between plant water uptake and plant yield, plant dry weight (fruit, stem and leaves) and vegetative dry weight (stem and leaves) at 0, 25, 50, and 75 mM NaCl (24 plants/NaCl concentration) and in all salt concentrations together (general, 96 plants)
General 0 mM NaCl 25 mM NaCl 50 mM NaCl 75 mM NaCl
Yield
Plant DW
Vegetative DW
0.60** 0.29 ns 0.39 ns 0.27 ns 0.21 ns
0.79** 0.65** 0.87** 0.78** 0.65**
0.83** 0.49* 0.86** 0.87** 0.77**
ns: Not significant. * Coefficients of correlation significant at p 0.95. ** Coefficients of correlation significant at p 0.99.
62
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Table 3 Slopes of the correlation lines between plant water uptake (PWU) and plant yield, plant dry weight (fruit, stem and leaves; PDW), and vegetative dry weight (stem and leaves; VDW) at 0, 25, 50, and 75 mM NaCl, for Floradade, L1, L5, and L9 tomato cultivars (24 plants/cultivar) Cultivar
PWU-yield
PWU-PDW
PWU-VDW
Floradade L1 L5 L9
0.0229 0.0229 0.0309 0.0333
1.4749 0.4991 0.5436 0.3568
0.6445 0.6043 0.6963 0.4685
a a b b
a b b b
a a a a
Figures followed by different letter in each column are significantly different ( p 0.95).
were calculated separately for the four cultivars. The regression lines between water uptake and yield showed that the slopes of the correlation lines varied according to the cultivar: Floradade and L1 had similar slopes and different from L5 and L9, which had also similar slopes (Table 3). The slopes of the regression lines for plant water uptake versus plant dry weight were also dependant on the cultivar. However, in the case of plant water uptake versus dry weight of the vegetative part, the slope was the same for all four cultivars. WUE was calculated in terms of yield (gram of fresh fruit produced per litre of water transpired) and in terms of plant dry weight (gram of dry weight produced per litre of water). The four cultivars needed more water to produce 1 g of fruit in saline than in control conditions, although a similar WUE in plant dry weight was observed in the four salinities tested (Table 1). Floradade, L1, and L5 transpired different amount of water (Table 1) but the average of Na+ absorbed in 25, 50 and 75 mM as percentage of the Na+ present in the nutrient solution was similar (2.34 0.22%, 2.64 0.16%, and 2.50 0.10%, respectively): L9 absorbed a higher percentage (3.24 0.22%). The percentage of Na+ absorbed (average for the 4 cultivars) was similar at 25 mM NaCl (2.63 0.17%), 50 mM NaCl (2.52 0.08%) and 75 mM NaCl (2.87 0.18%) in the nutrient solution
4. Discussion Floradade is a cultivar recommended for outdoor cultivation with saline water in Syria and Egypt (Gaibeh and N. Malash, personal communication, 2000) while L1, L5, and L9 were bred for use in greenhouses and selected because of their performance under saline conditions in a greenhouse in Malaga (Spain). However, fruit yield in this experiment demonstrated that the four cultivars had similar yield potential in a greenhouse with a short cultivation cycle (5 weeks of harvest). The estimated threshold values for the four cultivars ranged from 0 to 3.4 dS m1, broader values than the 2.5 dS m1 reported by Maas (1986) and the 2.0–2.5 reported by Saranga et al. (1991). Those authors carried out their experiments in soil (soil is known to act as buffer, reducing the range of plant responses, Cuartero and Fernandez-Mun˜ oz, 1999). However, our threshold values are comparable to the 0–3.2 dS m1 obtained by Caro et al. (1991) for Lycopersicon esculentum accessions on soil-less culture. According to the estimated thresholds, L5 should be the more salt tolerant cultivar, followed by Floradade and L1.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
63
In our experiment, rain water collected from the roof of the greenhouses (0.1 dS m1) was used to prepare the nutrient solutions: fertilisers added a little less than 2 dS m1 to the water EC. When ground or river water is used, the EC of the nutrient solution can be around 2.5 dS m1 or higher, a value greater than the threshold of 3 of the 4 cultivars tested here and, probably, higher than the threshold for most tomato cultivars. Consequently, thresholds would be of little value in soil-less tomato cultivation. Yield reduction from threshold to upper salinities varied from 7.7 (Floradade) to 10.5% (L5), close to the 9–10% found by Maas (1986) and Saranga et al. (1991). Reduction between 8 to 10% seems common in tomato because we grew different tomato cultivars than Maas (1986) and Saranga et al. (1991) and our plants had different environmental conditions than theirs. L5 showed the largest decrease and, theoretically, it was the most saline sensitive cultivar. However, as L5 had the highest yield in control conditions and the highest threshold, it was the cultivar with the highest yield between 0 and 9.1 dS m1. Yield reduction in saline conditions was due to reduction in fruit size in the case of L1, L5, and L9 which agrees with literature indicating that, at relatively low salinity, yield reduction is mainly caused by reduction in fruit weight whilst fruit number remains almost unchanged (Caro et al., 1991; van Ieperen, 1996). Floradade’s reduced yield through reduced fruit number was possibly due to the lack of adaptation to the greenhouse environment. The increase of fruit showing blossom end rot under salinity also contributed to yield reduction. The relatively low percentages of blossom end rot fruit found for the four tested cultivar (2–16%) suggest the cultivars are close to the ‘‘low sensitive cultivars to blossom end rot’’ according to Cuartero and Fernandez-Mun˜ oz (1999). Salinity increased blossom end rot regularly when the four cultivars were considered together, although each cultivar showed a different salinity-blossom end rot pattern. The different behaviour of the four cultivars supports the idea of Adams and Ho (1992) that sensitivity of cultivars to blossom end rot rather than salinity per se is the cause of blossom end rot in saline conditions. Shoot vegetative dry weight (leaves and stem) decreased with increasing salinity in the root zone in the four tested cultivars, although the decrease was lower than for fruit yield. A lower sensitivity to salinity of shoot development in comparison to fruit had been also recorded by Bolarin et al. (1991) and Romero-Aranda et al. (2002). Tomato fruit soluble solids and fruit dry weight did not decrease proportionally to fresh weight because under saline conditions tomato fruit have higher soluble solids content than in non-saline conditions (Li et al., 2001). Tomato fruit soluble solids (closely related to sugar concentration), acid content and their interrelation is important to sweetness, sourness and flavour (Stevens et al., 1977). It is well known that soluble solids increase with salinity (Mizrahi et al., 1988; Saranga et al., 1991) although there is much less information about acid content. For the four tested cultivars, soluble solids and acid content increased linearly with increasing the salinity of the root substrate although acid increased to a higher degree than soluble solids: the same trends were described by Cuartero and FernandezMun˜ oz (1999). High sugar concentrations together with relatively high acids are required for best flavour (Grierson and Kader, 1986), therefore, tomato fruit grown under saline conditions will have enhanced fruit flavour and taste although fruit would be more acid than when irrigated with fresh water. Daily plant water uptake increased from transplanting (February) to the beginning of May because of plant development and increasing temperature. At the beginning of May,
64
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
harvesting started and tomato plants reached an equilibrium between new developing fruit and fruit harvested which was reflected in a daily plant water uptake that was approximately constant from the beginning of May to the end of the experiment (14 June), a similar situation to that recorded by Romero-Aranda et al. (2002) from the beginning of harvest onwards. Plants irrigated with saline water reached maximum daily water uptake earlier than control plants because salinity enhanced plant senescence. Salinity reduced total plant water uptake in the four cultivars and seemed to be a very important variable affecting total plant water uptake since R2 values were very close to 1 (0.93–0.97). Salinity of the irrigation water should be taken into account when calculating tomato water requirements. Salinity and water uptake were only independent for L1, and in this case, only from 0 to 1.9 dS m1. As the EC of nutrient solutions is usually around or higher than 2.5 dS m1 in commercial soil-less tomato cultivation irrigated with good quality water (from 0.5 to 1 dS m1), plant water uptake will, in practice, be proportional to the EC of irrigation water. Substantial savings in irrigation water can be made if the EC of irrigation water is taken into account: from 4.5 to 6.1% per dS m1 of irrigation water according to our results. However, the amount of irrigation water that could be saved in a tomato crop is cultivar-specific and cannot be generalised because significant differences in the slopes of the correlation lines have been found. To understand the cause, some cultivarspecific variables such as yield and biomass have been compared to water uptake. Yield and biomass have been correlated with plant water uptake in several crops (Hanks, 1974) and under several stress conditions including salinity (Letey and Dinar, 1986; Shani and Dudley, 2001). We have obtained significant positive correlation between yield and water uptake (0.60), but this correlation can be mainly explained by the averages in the four salinities tested and to a much lesser extent by correlation of yield with water uptake in every saline condition (Table 2). Ben-Gal and Shani (2003) provided data also indicating higher correlation for five water stress conditions as a whole than for each one separately. According to our results, plant dry weight, including fruit and dry weight of the vegetative part, are better correlated with plant water uptake than yield for the four cultivars. However, when the data were separated by cultivar, dry weight of the vegetative part gave similar slopes for the four cultivars and plant dry weight including fruit did not. Consequently, dry weight of the vegetative part is a better correlated variable with plant water uptake in saline conditions than yield and plant dry weight, irrespective of cultivars WUE measured as gram dry matter per litre of water transpired was constant in the range of salinities tested here. Similar results have been reported by Romero-Aranda et al. (2000), although Al-Karaki (2000) found a decrease in tomato dry matter produced per litre of water when salinity increased. This discrepancy could be attributed to the range of salinity tested by Al-Karaki (2000), which was much higher than ours and that of RomeroAranda et al. (2000). Moreover, the cultivars used were also different. In addition, we obtained constant WUE for the four cultivars and Romero-Aranda et al. (2000) also reported independence between WUE and salinity for two different cultivars grown in similar salinity range than ours. WUE in terms of commercial fruit (gram) per litre of water transpired is an important trait for the grower. The four cultivars produced about 25 g of marketable fruit per litre of water transpired in control conditions, and about 26 g of total fruit (marketable and nonmarketable). WUE depends on the environmental conditions under which plants have been
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
65
growing and the length of the harvest period in relation to the whole cropping period. However, our figures of WUE are similar to the 25 g L1 reported by van Os (2001) for tomato grown in soil in Spanish greenhouses although lower than the 33 (van Os, 2001) or the 34 g L1 (Ben-Gal and Shani, 2003) reported for tomatoes in Israel. We have obtained significant WUE reduction (gram of fruit per litre of water transpired) as salinity in the nutrient solution increased. Romero-Aranda et al. (2002) also reported a lower WUE in saline than control conditions. Water transpired by plants and WUE (g fruit L1) decreased as salinity increased, while water used to produce dry matter remained constant. Water uptake by plants under stressful conditions was probably primarily used to meet transpiration and vegetative growing demands and secondarily allocated to fruit growth, as suggested by Ehret and Ho (1986) and Johnson et al. (1992). Plants growing under saline conditions have to take up water and nutrients from a soil solution with a high Na+ concentration. Water and nutrients are essential to growth, while the high concentration of Na+ is toxic. Cuartero and Fernandez-Mun˜ oz (1999) hypothesized that a salt-tolerance related trait would have the capacity of absorbing water and nutrients while rejecting Na+. The relationship between total Na+ content in the shoot and total Na+ content in the nutrient solution absorbed by the plant is a measure of the root Na+ selectivity and Na+ transport to the shoot, as a balance between the root capacity to discriminate against Na+ entrance and the capacity of the root to extrude Na+ to the medium. Tomato plants absorbed only a small proportion of the Na+ from the nutrient solution (from 2.3 to 3.2%) but there were significant differences among the four cultivars suggesting that this would be a trait to be taken into account when selecting tomato genotypes for salt tolerance. However, percentage of Na+ uptake seems to be independent of the Na+ concentration in nutrient solution.
Acknowledgments To Drs. Anthony Yeo, James Oster and Timothy Flowers for their revision of the manuscript. The experiments described here have been 50% funded by EU contract IC18CT98-0301, ‘A systems approach to a sustainable increase in irrigated vegetable crop production in salinity-prone areas of the Mediterranean region’ and by project IFAPA2002890, ‘Control inteligente de sensores para regular soluciones nutritivas recirculadas y minimizar vertidos de agua y nutrientes’.
References Adams, P., Ho, L.C., 1992. The susceptibility of modern tomato cultivars to blossom end rot in relation to salinity. J. Hort. Sci. 67, 827–839. Al-Karaki, G.N., 2000. Growth, water use efficiency, and sodium and potassium acquisition by tomato cultivars grown under salt stress. J. Plant Nutr. 23, 1–8. Ben-Gal, A., Shani, U., 2003. Water use and yield of tomatoes under limited water and excess boron. Plant Soil 256, 179–186. Bolarin, M.C., Fernandez, F.G., Cruz, V., Cuartero, J., 1991. Salinity tolerance in four wild tomato species using vegetative yield salinity responses curves. J. Am. Soc. Hort. Sci. 116, 286–290.
66
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Caro, M., Cruz, V., Cuartero, J., Estan, M.T., Bolarin, M.C., 1991. Salinity tolerance of normal and cherry tomato cultivars. Plant Soil 136, 249–255. Cuartero, J., Fernandez-Mun˜ oz, R., 1999. Tomato and salinity. Sci. Hort. 78, 83–125. Ehret, D.L., Ho, L.C., 1986. Effects of osmotic potential in nutrient solution on diurnal growth of tomato fruit. J. Exp. Bot. 37, 1294–1302. FAO, 2002. Statistics. Available from: http://www.fao.org/. Ghassemi, F., Jakeman, A.J., Nix, H.A. (Eds.), 1995. Salinisation of Land and Water Resources. Human Causes, Extent Management and Case Studies. CAB International, Wallingford, Oxon, UK, p. 256. Grierson, D., Kader, A.A., 1986. Fruit ripening and quality. In: Atherton, J.G., Rudich, J. (Eds.), The Tomato Crop. A Scientific Base for Improvement. Chapman and Hall, London, pp. 241–280. Hanks, R.J., 1974. Model for predicting plant yield as influenced by water use. Agron. J. 66, 660–664. Johnson, R.W., Dixon, M.A., Lee, D.R., 1992. Water relations of the tomato during fruit growth. Plant Cell Environ. 15, 947–953. Letey, J., Dinar, A., 1986. Simulated crop-water production function for several crops when irrigated with saline water. Hilgardia 54, 1–32. Li, Y.L., Stanghellini, C., Challa, H., 2001. Effect of electrical conductivity and transpiration on production of greenhouse tomato. Sci. Hort. 88, 11–29. Maas, E.V., 1986. Salt tolerance of plants. Appl. Agric. Res. 1, 12–26. Mizrahi, Y., Taleisnik, E., Kagan-Zur, V., Zohas, Y., Offenbach, R., Matan, E., Golan, R., 1988. A saline irrigation regime for improving tomato quality without reducing yield. J. Am. Soc. Hort. Sci. 113, 202–205. Ragab, R., 2002. A holistic generic integrated approach for irrigation, crop and field management: the SALTMED model. Environ. Modell. Softw. 17, 345–361. Romero-Aranda, R., Soria, T., Cuartero, J., 2000. Tomato plant-water uptake and plant-water relationships under saline growth conditions. Plant Sci. 160, 265–272. Romero-Aranda, R., Soria, T., Cuartero, J., 2002. Greenhouse mist improves yield of tomato plants grown under saline conditions. J. Am. Soc. Hort. Sci. 127, 644–648. Saranga, Y., Zamir, D., Marani, A., Rudich, J., 1991. Breeding tomatoes for salt tolerance: field evaluation of Lycopersicon germplasm for yield and dry-matter production. J. Am. Soc. Hort. Sci. 116, 1067–1071. Shani, U., Dudley, L.M., 2001. Field studies of crop response to drought and salt stress. Soil Sci. Soc. Am. J. 65, 1522–1528. Shannon, M.C., Grieve, C.M., 1999. Tolerance of vegetable crops to salinity. Sci. Hort. 78, 5–38. Soria, T., Cuartero, J., 1997. Tomato fruit yield and water consumption with saline water irrigation. Acta Hort. 458, 215–219. Stevens, M.A., Kader, A.A., Albright-Holton, M., Algazi, M., 1977. Genotypic variation for flavour and composition in fresh market tomatoes. J. Am. Soc. Hort. Sci. 102, 680–689. van Genuchten, M.T., 1983. Analyzing Crop Salt Tolerance Data: Model Description and Users’s Manual. Research Report 120 USDA Agricultural Research Service. U.S. Salinity Laboratory, Riverside, CA, 50 pp. van Genuchten, M.T., Hoffman, G.J., 1984. Analysis of crop salt tolerance data. In: Shainberg, I., Shalhevet, J. (Eds.), Soil Salinity Under Irrigation. Ecological Studies, vol. 8. Spring Verlag, Berlin, pp. 58–271. van Ieperen, W., 1996. Effect of different day nd night salinity levels on vegetative growth, yield and quality of tomato. J. Hort. Sci. 71, 99–111. van Os, E., 2001. Diffussion and environmental aspects of soilless growing systems. Italus Hortus 8, 9–15. Zerbini, P.E., Gorini, F., Polesello, A. 1991. Measurement of the Quality of the Tomatoes: Recommendations of an EEC Working Group. IVTPA, Milano, 36 pp.
Plant water uptake and water use efficiency of greenhouse tomato cultivars irrigated with saline water A. Reina-Sa´nchez, R. Romero-Aranda, J. Cuartero * Estacion Experimental La Mayora, CSIC, 29750 Algarrobo-Costa, Malaga, Spain Accepted 1 April 2005 Available online 6 July 2005
Abstract Effects of salinity on tomato (Lycopersicon esculentum Mill.) fruit yield, plant water uptake and water use efficiency (WUE) have been quantified in experiments carried out under greenhouse and soil-less cultivation with four cultivars (Floradade, L1, L5 and L9) and four salinity levels (0, 25, 50, and 75 mM NaCl). Fruit represented 70% of plant fresh weight while leaves and stems represented 22 and 8%, respectively. Fruit were the most sensitive part of the plant, with the four cultivars showing similar significant fruit yield reduction, namely about 28 g/mM NaCl or 290 g/dS m1. Yield threshold varied from 0 to 3.4 dS m1, values lower than or close to the electrical conductivity (EC) of nutrient solutions used in commercial greenhouses. Yield reduction from threshold to upper salinities was about 8% of maximum yield per dS m1 increase. Blossom end rot increased with salinity although the pattern of increase depended on the cultivar. Tomato fruit grown under saline conditions had higher soluble solids and acid content than those from the control (0 mM) plants. Plants grown under the most saline conditions consumed, on average, 40% less water than control plants. The relationship between total plant water uptake and salinity was linear (negligible threshold) and salinity of the nutrient solution almost entirely explained the variations in plant water uptake (R2 from 0.94 to 1); therefore, salinity of the irrigation water has to be taken into account when calculating tomato water requirements. However, significant differences in the negative slopes of the correlation lines indicate that decreases in plant water uptake, from 3.5 to 5% per dS m1, are cultivar-specific and cannot be generalised. Vegetative (stem and leaves) dry weight was a better indicator of tomato plant water uptake in saline conditions irrespective of cultivars than fruit yield or plant and fruit dry weight. Tomato plants in the control averaged a higher WUE than the most * Corresponding author. Tel.: +34 952552656; fax: +34 952552677. E-mail address: [email protected] (J. Cuartero). 0378-3774/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.agwat.2005.04.021
A. Reina-Sa´nchez et al. / Agricultural Water Management 78 (2005) 54–66
55
salinized plants when WUE was expressed as a function of fruit yield (25 and 13 g fruit L1); however WUE was independent of salinity if expressed as a function of plant and fruit dry matter (approximately 3.0 g dry matter L1). Tomato plants absorbed only a small proportion of the Na+ present in the nutrient solution (from 2.3 to 3.2%) but there were significant differences among the four cultivars which suggest that plant ability to select ions is a trait to be taken into account when selecting tomato genotypes for salt tolerance. # 2005 Elsevier B.V. All rights reserved. Keywords: Lycopersicon esculentum; Yield; Fruit and plant dry weight; Fruit quality; Blossom end rot; Na+ uptake
1. Introduction The tomato crop is adapted to a wide variety of climates ranging from the tropics to within a few degrees of the Arctic Circle. However, in spite of its broad adaptation, horticultural production is concentrated in a few warm and rather dry areas: about 34% of world production comes from countries around the Mediterranean sea and about 14% from California and Mexico (FAO, 2002). These areas are also those where the highest yields are reached but also where salinity is a serious restraint not only for planting new lands to this crop but also for maintaining in high productivity those currently under irrigation (Ghassemi et al., 1995). The increasing demand for domestic, industrial, environmental and recreational water will force agriculturists to manage irrigation water carefully, contributing to environmental preservation. In parallel, brackish and saline water resources not used nowadays could be employed for irrigation if greater knowledge of salt tolerance and proper technology are developed (Shannon and Grieve, 1999). In applying saline/brackish water for irrigation, an integrated approach, which should account for soil, crop and water management at the same time, should be adopted. This approach needs calculation of crop water requirements which are essential for water saving, controlling water table level and drainage volume, and of course the final yield (Ragab, 2002). Tomato could act as a model crop for brackish/ saline water use because it is already grown in large areas with saline conditions, and because there is a wealth of important knowledge of the physiology and genetics of this species. It is known that tomato plants irrigated with a saline solution transpire less water than when fresh water is used (Soria and Cuartero, 1997; Romero-Aranda et al., 2000), but it is necessary to quantify the decrease in plant water uptake in relation to salinity and investigate possible differences in water uptake between cultivars, as a part of the integrated approach described by Ragab (2002) to use saline irrigation water. In this paper, we report on our investigations of those aspects on four tomato cultivars and the relations between plant water uptake and yield and biomass. Water use efficiency (WUE) in agronomical and biological terms (gram fruit and gram dry matter per litre transpired water, respectively) have been also determined as higher efficiency in plant dry matter and fruit formation will lead to relatively less uptake of toxic ions (Na+) (Cuartero and Fernandez-Mun˜ oz, 1999). Finally, we have determined Na+ concentrations in leaves, stem
56
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
and fruit in order to calculate a measure of root Na+ selectivity and transport to the shoot by relating plant water and Na+ uptake.
2. Material and methods Four tomato cultivars (Floradade, L1, L5, and L9) were chosen because of their demonstrated salt tolerance in Syria and Egypt (Floradade) and in Spain (L1, L5, and L9) in previous years. Plants were grown in a commercial polyethylene greenhouse in Malaga, Spain (368450 N, 48020 W) from 14 February to 14 June 2001. Seeds were sown in vermiculite on 12 January and, when seedlings had developed 5 true leaves (14 February), 40 plants/cultivar were transplanted to 40 pots of 20 L filled up with 3 mm diameter sand. Pots were arranged in rows 1 m apart, with 2 plants m1 in the rows. Two extra plants at the beginning and end of every row and two rows before and after the first and final experimental rows were planted to avoid border effects. A standard nutrient solution (10 mM N, 7 mM K, 0.9 mM P, 5 mM Ca, 2 mM Mg, and microelements), prepared with rain water, was supplied to the plants from transplanting to the end of the experiment by an automatic drip irrigation system, with one dripper per plant that discharged 2.3 L h1. To ensure that enough water was always present in the sand around root system, we set 21 irrigation periods/day (at 7, 9, 10, 10.30, 11, 11.30, 12, 12.30, 13, 13.30, 14, 14.30, 15, 15.30, 16, 16.30, 17, 17.30, 18, 19 and 21 h) of 2 min from 19 February to 3 April and of 3 min from 4 April to 14 June. Plants were grown as a single stem by removing side shoots weekly (the side shoots were removed when they were less than 5 cm in length to reduce the effects of wounding). Five days after transplanting, salt treatment began with 0 (control), 25, 50, and 75 mM NaCl added to the nutrient solution, resulting in electrical conductivities (EC) of 1.9, 4.7, 7.1 and 9.1 dS m1, respectively. Those treatments with added salt are referred in the text as saline treatments, with the salinity being determined either in terms of salt concentration or the conductivity of the solution. Ten plants/cultivar and salt treatment were grown although records were only taken from the six central plants; the two at the beginning and the two at the end were grown to reduce border effects. At each salt concentration, four pots without plants were irrigated in order to measure evaporation from the sand. Plant water uptake (transpiration) was calculated weekly for six plants per cultivar and salt treatment. Drainage was collected in trays located below the pots (with and without a plant) during a whole day. Evaporation from the trays was measured in four trays filled with nutrient solution—as weight difference between the beginning and the end of the day (at the times when leacheate was recorded). Plant water uptake was calculated by subtracting from the nutrient solution supplied the nutrient solution drained from planted pots, the water evaporated from unplanted pots and the water evaporated from the trays where drainage was collected. Fruit were harvested six times from 8 May to 14 June. In each harvest, commercial and non-commercial fruit (fruit with blossom end rot) were counted separately and weighed for the same six plants per cultivar and salt treatment used to determine water uptake. Three samples of fruit were taken and dried along the harvest period to calculate total fruit dry weight and to determine Na+ concentration. Vegetative dry weight was obtained by drying
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
57
and weighing stems and leaves of the six plants per cultivar and salt treatment at the end of the experiment; plant dry weight was calculated by adding fruit dry weight to vegetative dry weight. All the leaves produced were on the plant at the end of the experiment because old leaves were not removed during cultivation. Three samples of the dry leaves and three of the dry stem were taken from each plant, to determine Na+ concentration. The fruit dry weight and fruit Na+ concentration were used together with Na+ of dry leaves and stems to calculate total Na+ absorbed by the plant and stored in the shoot. To calculate the relative Na+ exclusion, the amount of Na+ absorbed by each plant was expressed as a percentage of the amount of Na+ that might have been absorbed had the nutrient solution passed through the plant and all the Na remained in the leaves. This parameter was taken as a measure of plant root Na+ exclusion, was calculated as 100 (total Na+ stored in the shoot/total Na+ content in a volume of nutrient solution equivalent to the amount of water transpired). Fruit soluble solids and organic acids content were determined in the three fruit samples following Zerbini et al. (1991); soluble solids (8Brix) were estimated at 20 8C by a refractometer illuminated with sodium light and acidity by titrating tomato juice with 0.1 N NaOH to a pH of 8.1 (NaOH meq/100 ml juice). One-way analysis of variance was used to compare the effects of salinity on each parameter that was measured for each cultivar (six data points per cultivar from the six plants at each salt concentration). A coefficient of correlation was used to quantify the relation between characters (6 plants and 4 salinities, making 24 data points per cultivar). The model proposed by van Genuchten and Hoffman (1984) for evaluating crop salt tolerance and to predict yield in different saline conditions was used to determine cultivar salt response curves for yield and plant water uptake. The program described by van Genuchten (1983) was used to calculate Ym (theoretical maximum yield), slope (yield decrease per unit salinity increase) and EC-threshold (maximum salinity without yield reduction) by nonlinear regression. Results are referred to salinity in mM, except where the van Genuchten and Hoffman (1984) model was used, when EC has been used to facilitate comparison of our results with results described in the literature.
3. Results Fruit represented 70% of the plant fresh weight while leaves and stems represented 22 and 8%, respectively. The yield of the four cultivars was similar in the absence of salt (control) but decreased with increasing salinity (Fig. 1) and to a greater extent than the other parts of the plant: all four cultivars showed a significant reduction in fruit yield. Even the lowest salt concentration (25 mM, 4.7 dS m1) produced a substantial reduction (from 12 to 38%) in yield, and at the highest salt concentration yield reduction was almost 70% of the control. The average yield reduction for the four cultivars was 28 g/mM or 295 g/ dS m1. Our data fitted well the model proposed by van Genuchten and Hoffman (1984) as R2 values were between 0.91 and 1. Threshold EC varied from 0 (L9) to 3.4 dS m1 (L5); Floradade and L5, showed threshold values higher than EC in control conditions (1.9 dS m1) while L1 and L9 had lower values. Yield reduction at EC values higher than the threshold ranged from 214 to 361 g/dS m1, which represented from 7.7 to 10.5% of the estimated maximum yield (the yield at salnities less than the threshold salinity). Yield
58
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Fig. 1. Relationship between yield (g plant1) and salinity (dS m1) in four tomato cultivars (six plants per cultivar and salinity).
was similar for the four cultivars at control and maximum tested salinity conditions; however, between those extremes L5 yielded more fruit than the others cultivars, which behaved similarly. Tomato yield is determined by fruit weight and fruit number. Expressing yield of the four cultivars in terms of the number of fruit and the weight per fruit, we found for Floradade that the highest salinity (75 mM) significantly reduced the number of fruit (65%) while fruit weight remained almost stable (11%). Salinity reduced fruit number and fruit weight of L1 in similar proportions, but had its major effect on fruit weight for L5 and L9 (54 and 62%, respectively). The average non-commercial fruit (fruit with blossom end rot) for the four cultivars increased with salinity from 2% in control (0 mM Na) conditions to 6, 12 and 16% in 25, 50 and 75 mM, respectively (Table 1). However, this apparently strong correlation between salinity and blossom end rot almost disappeared when the analysis was conducted cultivar by cultivar. Floradade tended to increase blossom end rot in the upper part of the tested salinity range (higher than 25 mM NaCl) while L9 increased blossom end rot in the lower range (from 0 to 50 mM NaCl); L1 showed strong positive correlation between salinity and blossom end rot while L5 showed independence. Fruit dry matter significantly increased with increasing salinity and fruit grown at 75 mM had more than double the dry matter of control fruit (Table 1). There were also differences between cultivars in fruit dry matter content in control conditions, although those differences were reduced by increasing salinity. Due to the higher fruit dry matter in saline conditions no significant differences appeared between salinity treatments when yield was expressed as fruit dry weight.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
59
Table 1 Commercial (C) and non-commercial (NC) yield (g plant1 and percentageof total yield, respectively), fruit dry matter (DM) (%), vegetative dry weight (DW) in grams (stem and leaves), plant dry weight in grams (stem, leaves, and fruit), water uptake (L plant1), and WUE (in grams of fruit and plant dry weight per litre of water) for four tomato cultivars grown at 0, 25, 50, and 75 mM NaCl in the nutrient solution Cultivar
NC Fruit Vegetative Plant Water uptake Fruit Plant NaCl C yield DW (g) (L plant1) (g)/L DW (g)/L (mM) (g plant1) yield (%) DM (%) DW (g) water water
Floradade
0 25 50 75
2789 2057 1800 882
a ab bc c
4.6 4.2 8.3 13.8
a a a a
3.7 5.8 7.6 10.4
d c b a
139 96 88 82
a b b b
249 238 230 220
a a a a
103 83 77 67
a b b b
31 26 23 15
a a a a
2.5 2.6 3.2 3.1
a a a a
L1
0 25 50 75
3310 2363 1438 1043
a ab bc c
3.0 10.8 21.5 26.2
b b a a
5.6 7.7 10.7 11.6
c b a a
215 152 152 143
a ab ab b
402 330 321 296
a ab ab ab
141 118 108 86
a b b c
25 18 14 13
a ab b b
3.5 3.1 3.4 3.4
a a a a
L5
0 25 50 75
3432 2924 2103 1343
a ab bc c
0.5 3.8 1.0 4.2
a a a a
3.6 5.6 6.2 11.3
c b b a
243 179 166 156
a b b b
372 364 299 304
a a a a
151 a 115 b 104bc 95 c
24 24 21 14
a a ab b
2.5 3.2 2.9 3.2
a a a a
L9
0 25 50 75
3170 1821 1362 815
a b b c
0.0 7.7 18.6 19.8
b b a a
5.4 6.8 9.5 11.5
c c b a
289 183 158 130
a b b b
457 308 317 242
a b b b
155 113 103 83
20 15 13 9
a ab ab b
2.9 2.8 3.1 2.9
a a a a
a b b c
Figures followed by different letter within a column and cultivar are significantly different ( p 0.95).
Fruit acidity and 8Brix increased significantly with increasing salinity (Fig. 2). The increase of 8Brix was similar in the four cultivars, however, for acidity, L9 showed a significantly higher increase than Floradade and L5. Fruit acid concentration measured as meq NaOH was about 2.3 times higher than soluble solids expressed as 8Brix in control fruit. Comparing the increase of 8Brix and acidity, the slope of the correlation lines was 2.5–4.5 higher for acidity than for 8Brix, indicating greater accumulation of acids than of soluble solids in fruit grown under saline conditions as well as a greater ratio of acidity/ soluble-solids (meq NaOH/8Brix). Vegetative dry weight was reduced dramatically at low salt concentration (25 mM) with a smaller reduction at higher levels of salinity (Table 1). Shoot plant dry weight (including the number of fruit) at the end of the experiment also tended to decrease when salinity increased but was significantly different from the control only for L5 and L9. Daily plant water uptake increased from 0.1 L plant1 at transplant time (19 February) to 1.70 L plant1 about 70 days after transplant (beginning of May) in the control treatment. From the beginning of May to the end of the experiment (14 June), daily plant water uptake was maintained at the same level. Maximum daily water uptake was lower (0.95 L plant1) at the highest salinity than under control conditions and the maximum was reached earlier, 60 days after transplant. Total plant water uptake was reduced by salinity for the four cultivars with the reduction at the highest salt concentration being almost 40% of the water uptake under control
60
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Fig. 2. Variation of tomato fruit soluble solids (8Brix) and acidity (meq NaOH) with salinity (mM NaCl) in four cultivars (three samples per plant and six plants per cultivar and salinity).
conditions (Fig. 3). Relationship between total plant water uptake and salinity in the nutrient solution fitted well van Genuchten and Hoffman (1984) model, for the four cultivars tested, with coefficients of determination (R2) higher than 0.93 which indicates that the salinity of the nutrient solution explained almost all the variation in water uptake. The intensity of the reduction in total plant water uptake varied between 4.8 and 9.5 L plant1 dS m1 increment in the root zone and was cultivar-specific for there were significant differences between the slopes of the cultivars: Floradade and L1 had statistically similar reductions, lower than for both L9 and L5, which had similar slopes.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
61
Fig. 3. Relationship between total plant water uptake (L plant1) and salinity (dS m1) in the nutrient solution for four tomato cultivars (six plants per cultivar and salinity).
The set of data obtained in this experiment allowed calculation of the correlation between water uptake and yield, plant dry weight (including fruit) and dry weight of the vegetative part. The correlation coefficient between water uptake and yield was lower than the coefficients between water uptake and plant dry weight and dry weight of the vegetative part (Table 2). The correlations between plant water uptake and yield calculated independently for each salt concentration were low (from 0.21 to 0.39) and statistically not significant, whereas the overall correlation between plant water uptake and yield appeared high (0.60). Plant dry weight and dry weight of the vegetative part were well correlated with water uptake under all of the tested salt concentrations (Table 2). In order to discover which one out of the three parameters—yield, plant dry weight, or vegetative dry weight—was a better indicator of plant water uptake, linear regressions Table 2 Coefficients of correlation between plant water uptake and plant yield, plant dry weight (fruit, stem and leaves) and vegetative dry weight (stem and leaves) at 0, 25, 50, and 75 mM NaCl (24 plants/NaCl concentration) and in all salt concentrations together (general, 96 plants)
General 0 mM NaCl 25 mM NaCl 50 mM NaCl 75 mM NaCl
Yield
Plant DW
Vegetative DW
0.60** 0.29 ns 0.39 ns 0.27 ns 0.21 ns
0.79** 0.65** 0.87** 0.78** 0.65**
0.83** 0.49* 0.86** 0.87** 0.77**
ns: Not significant. * Coefficients of correlation significant at p 0.95. ** Coefficients of correlation significant at p 0.99.
62
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Table 3 Slopes of the correlation lines between plant water uptake (PWU) and plant yield, plant dry weight (fruit, stem and leaves; PDW), and vegetative dry weight (stem and leaves; VDW) at 0, 25, 50, and 75 mM NaCl, for Floradade, L1, L5, and L9 tomato cultivars (24 plants/cultivar) Cultivar
PWU-yield
PWU-PDW
PWU-VDW
Floradade L1 L5 L9
0.0229 0.0229 0.0309 0.0333
1.4749 0.4991 0.5436 0.3568
0.6445 0.6043 0.6963 0.4685
a a b b
a b b b
a a a a
Figures followed by different letter in each column are significantly different ( p 0.95).
were calculated separately for the four cultivars. The regression lines between water uptake and yield showed that the slopes of the correlation lines varied according to the cultivar: Floradade and L1 had similar slopes and different from L5 and L9, which had also similar slopes (Table 3). The slopes of the regression lines for plant water uptake versus plant dry weight were also dependant on the cultivar. However, in the case of plant water uptake versus dry weight of the vegetative part, the slope was the same for all four cultivars. WUE was calculated in terms of yield (gram of fresh fruit produced per litre of water transpired) and in terms of plant dry weight (gram of dry weight produced per litre of water). The four cultivars needed more water to produce 1 g of fruit in saline than in control conditions, although a similar WUE in plant dry weight was observed in the four salinities tested (Table 1). Floradade, L1, and L5 transpired different amount of water (Table 1) but the average of Na+ absorbed in 25, 50 and 75 mM as percentage of the Na+ present in the nutrient solution was similar (2.34 0.22%, 2.64 0.16%, and 2.50 0.10%, respectively): L9 absorbed a higher percentage (3.24 0.22%). The percentage of Na+ absorbed (average for the 4 cultivars) was similar at 25 mM NaCl (2.63 0.17%), 50 mM NaCl (2.52 0.08%) and 75 mM NaCl (2.87 0.18%) in the nutrient solution
4. Discussion Floradade is a cultivar recommended for outdoor cultivation with saline water in Syria and Egypt (Gaibeh and N. Malash, personal communication, 2000) while L1, L5, and L9 were bred for use in greenhouses and selected because of their performance under saline conditions in a greenhouse in Malaga (Spain). However, fruit yield in this experiment demonstrated that the four cultivars had similar yield potential in a greenhouse with a short cultivation cycle (5 weeks of harvest). The estimated threshold values for the four cultivars ranged from 0 to 3.4 dS m1, broader values than the 2.5 dS m1 reported by Maas (1986) and the 2.0–2.5 reported by Saranga et al. (1991). Those authors carried out their experiments in soil (soil is known to act as buffer, reducing the range of plant responses, Cuartero and Fernandez-Mun˜ oz, 1999). However, our threshold values are comparable to the 0–3.2 dS m1 obtained by Caro et al. (1991) for Lycopersicon esculentum accessions on soil-less culture. According to the estimated thresholds, L5 should be the more salt tolerant cultivar, followed by Floradade and L1.
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
63
In our experiment, rain water collected from the roof of the greenhouses (0.1 dS m1) was used to prepare the nutrient solutions: fertilisers added a little less than 2 dS m1 to the water EC. When ground or river water is used, the EC of the nutrient solution can be around 2.5 dS m1 or higher, a value greater than the threshold of 3 of the 4 cultivars tested here and, probably, higher than the threshold for most tomato cultivars. Consequently, thresholds would be of little value in soil-less tomato cultivation. Yield reduction from threshold to upper salinities varied from 7.7 (Floradade) to 10.5% (L5), close to the 9–10% found by Maas (1986) and Saranga et al. (1991). Reduction between 8 to 10% seems common in tomato because we grew different tomato cultivars than Maas (1986) and Saranga et al. (1991) and our plants had different environmental conditions than theirs. L5 showed the largest decrease and, theoretically, it was the most saline sensitive cultivar. However, as L5 had the highest yield in control conditions and the highest threshold, it was the cultivar with the highest yield between 0 and 9.1 dS m1. Yield reduction in saline conditions was due to reduction in fruit size in the case of L1, L5, and L9 which agrees with literature indicating that, at relatively low salinity, yield reduction is mainly caused by reduction in fruit weight whilst fruit number remains almost unchanged (Caro et al., 1991; van Ieperen, 1996). Floradade’s reduced yield through reduced fruit number was possibly due to the lack of adaptation to the greenhouse environment. The increase of fruit showing blossom end rot under salinity also contributed to yield reduction. The relatively low percentages of blossom end rot fruit found for the four tested cultivar (2–16%) suggest the cultivars are close to the ‘‘low sensitive cultivars to blossom end rot’’ according to Cuartero and Fernandez-Mun˜ oz (1999). Salinity increased blossom end rot regularly when the four cultivars were considered together, although each cultivar showed a different salinity-blossom end rot pattern. The different behaviour of the four cultivars supports the idea of Adams and Ho (1992) that sensitivity of cultivars to blossom end rot rather than salinity per se is the cause of blossom end rot in saline conditions. Shoot vegetative dry weight (leaves and stem) decreased with increasing salinity in the root zone in the four tested cultivars, although the decrease was lower than for fruit yield. A lower sensitivity to salinity of shoot development in comparison to fruit had been also recorded by Bolarin et al. (1991) and Romero-Aranda et al. (2002). Tomato fruit soluble solids and fruit dry weight did not decrease proportionally to fresh weight because under saline conditions tomato fruit have higher soluble solids content than in non-saline conditions (Li et al., 2001). Tomato fruit soluble solids (closely related to sugar concentration), acid content and their interrelation is important to sweetness, sourness and flavour (Stevens et al., 1977). It is well known that soluble solids increase with salinity (Mizrahi et al., 1988; Saranga et al., 1991) although there is much less information about acid content. For the four tested cultivars, soluble solids and acid content increased linearly with increasing the salinity of the root substrate although acid increased to a higher degree than soluble solids: the same trends were described by Cuartero and FernandezMun˜ oz (1999). High sugar concentrations together with relatively high acids are required for best flavour (Grierson and Kader, 1986), therefore, tomato fruit grown under saline conditions will have enhanced fruit flavour and taste although fruit would be more acid than when irrigated with fresh water. Daily plant water uptake increased from transplanting (February) to the beginning of May because of plant development and increasing temperature. At the beginning of May,
64
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
harvesting started and tomato plants reached an equilibrium between new developing fruit and fruit harvested which was reflected in a daily plant water uptake that was approximately constant from the beginning of May to the end of the experiment (14 June), a similar situation to that recorded by Romero-Aranda et al. (2002) from the beginning of harvest onwards. Plants irrigated with saline water reached maximum daily water uptake earlier than control plants because salinity enhanced plant senescence. Salinity reduced total plant water uptake in the four cultivars and seemed to be a very important variable affecting total plant water uptake since R2 values were very close to 1 (0.93–0.97). Salinity of the irrigation water should be taken into account when calculating tomato water requirements. Salinity and water uptake were only independent for L1, and in this case, only from 0 to 1.9 dS m1. As the EC of nutrient solutions is usually around or higher than 2.5 dS m1 in commercial soil-less tomato cultivation irrigated with good quality water (from 0.5 to 1 dS m1), plant water uptake will, in practice, be proportional to the EC of irrigation water. Substantial savings in irrigation water can be made if the EC of irrigation water is taken into account: from 4.5 to 6.1% per dS m1 of irrigation water according to our results. However, the amount of irrigation water that could be saved in a tomato crop is cultivar-specific and cannot be generalised because significant differences in the slopes of the correlation lines have been found. To understand the cause, some cultivarspecific variables such as yield and biomass have been compared to water uptake. Yield and biomass have been correlated with plant water uptake in several crops (Hanks, 1974) and under several stress conditions including salinity (Letey and Dinar, 1986; Shani and Dudley, 2001). We have obtained significant positive correlation between yield and water uptake (0.60), but this correlation can be mainly explained by the averages in the four salinities tested and to a much lesser extent by correlation of yield with water uptake in every saline condition (Table 2). Ben-Gal and Shani (2003) provided data also indicating higher correlation for five water stress conditions as a whole than for each one separately. According to our results, plant dry weight, including fruit and dry weight of the vegetative part, are better correlated with plant water uptake than yield for the four cultivars. However, when the data were separated by cultivar, dry weight of the vegetative part gave similar slopes for the four cultivars and plant dry weight including fruit did not. Consequently, dry weight of the vegetative part is a better correlated variable with plant water uptake in saline conditions than yield and plant dry weight, irrespective of cultivars WUE measured as gram dry matter per litre of water transpired was constant in the range of salinities tested here. Similar results have been reported by Romero-Aranda et al. (2000), although Al-Karaki (2000) found a decrease in tomato dry matter produced per litre of water when salinity increased. This discrepancy could be attributed to the range of salinity tested by Al-Karaki (2000), which was much higher than ours and that of RomeroAranda et al. (2000). Moreover, the cultivars used were also different. In addition, we obtained constant WUE for the four cultivars and Romero-Aranda et al. (2000) also reported independence between WUE and salinity for two different cultivars grown in similar salinity range than ours. WUE in terms of commercial fruit (gram) per litre of water transpired is an important trait for the grower. The four cultivars produced about 25 g of marketable fruit per litre of water transpired in control conditions, and about 26 g of total fruit (marketable and nonmarketable). WUE depends on the environmental conditions under which plants have been
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
65
growing and the length of the harvest period in relation to the whole cropping period. However, our figures of WUE are similar to the 25 g L1 reported by van Os (2001) for tomato grown in soil in Spanish greenhouses although lower than the 33 (van Os, 2001) or the 34 g L1 (Ben-Gal and Shani, 2003) reported for tomatoes in Israel. We have obtained significant WUE reduction (gram of fruit per litre of water transpired) as salinity in the nutrient solution increased. Romero-Aranda et al. (2002) also reported a lower WUE in saline than control conditions. Water transpired by plants and WUE (g fruit L1) decreased as salinity increased, while water used to produce dry matter remained constant. Water uptake by plants under stressful conditions was probably primarily used to meet transpiration and vegetative growing demands and secondarily allocated to fruit growth, as suggested by Ehret and Ho (1986) and Johnson et al. (1992). Plants growing under saline conditions have to take up water and nutrients from a soil solution with a high Na+ concentration. Water and nutrients are essential to growth, while the high concentration of Na+ is toxic. Cuartero and Fernandez-Mun˜ oz (1999) hypothesized that a salt-tolerance related trait would have the capacity of absorbing water and nutrients while rejecting Na+. The relationship between total Na+ content in the shoot and total Na+ content in the nutrient solution absorbed by the plant is a measure of the root Na+ selectivity and Na+ transport to the shoot, as a balance between the root capacity to discriminate against Na+ entrance and the capacity of the root to extrude Na+ to the medium. Tomato plants absorbed only a small proportion of the Na+ from the nutrient solution (from 2.3 to 3.2%) but there were significant differences among the four cultivars suggesting that this would be a trait to be taken into account when selecting tomato genotypes for salt tolerance. However, percentage of Na+ uptake seems to be independent of the Na+ concentration in nutrient solution.
Acknowledgments To Drs. Anthony Yeo, James Oster and Timothy Flowers for their revision of the manuscript. The experiments described here have been 50% funded by EU contract IC18CT98-0301, ‘A systems approach to a sustainable increase in irrigated vegetable crop production in salinity-prone areas of the Mediterranean region’ and by project IFAPA2002890, ‘Control inteligente de sensores para regular soluciones nutritivas recirculadas y minimizar vertidos de agua y nutrientes’.
References Adams, P., Ho, L.C., 1992. The susceptibility of modern tomato cultivars to blossom end rot in relation to salinity. J. Hort. Sci. 67, 827–839. Al-Karaki, G.N., 2000. Growth, water use efficiency, and sodium and potassium acquisition by tomato cultivars grown under salt stress. J. Plant Nutr. 23, 1–8. Ben-Gal, A., Shani, U., 2003. Water use and yield of tomatoes under limited water and excess boron. Plant Soil 256, 179–186. Bolarin, M.C., Fernandez, F.G., Cruz, V., Cuartero, J., 1991. Salinity tolerance in four wild tomato species using vegetative yield salinity responses curves. J. Am. Soc. Hort. Sci. 116, 286–290.
66
A. Reina-Sa´ nchez et al. / Agricultural Water Management 78 (2005) 54–66
Caro, M., Cruz, V., Cuartero, J., Estan, M.T., Bolarin, M.C., 1991. Salinity tolerance of normal and cherry tomato cultivars. Plant Soil 136, 249–255. Cuartero, J., Fernandez-Mun˜ oz, R., 1999. Tomato and salinity. Sci. Hort. 78, 83–125. Ehret, D.L., Ho, L.C., 1986. Effects of osmotic potential in nutrient solution on diurnal growth of tomato fruit. J. Exp. Bot. 37, 1294–1302. FAO, 2002. Statistics. Available from: http://www.fao.org/. Ghassemi, F., Jakeman, A.J., Nix, H.A. (Eds.), 1995. Salinisation of Land and Water Resources. Human Causes, Extent Management and Case Studies. CAB International, Wallingford, Oxon, UK, p. 256. Grierson, D., Kader, A.A., 1986. Fruit ripening and quality. In: Atherton, J.G., Rudich, J. (Eds.), The Tomato Crop. A Scientific Base for Improvement. Chapman and Hall, London, pp. 241–280. Hanks, R.J., 1974. Model for predicting plant yield as influenced by water use. Agron. J. 66, 660–664. Johnson, R.W., Dixon, M.A., Lee, D.R., 1992. Water relations of the tomato during fruit growth. Plant Cell Environ. 15, 947–953. Letey, J., Dinar, A., 1986. Simulated crop-water production function for several crops when irrigated with saline water. Hilgardia 54, 1–32. Li, Y.L., Stanghellini, C., Challa, H., 2001. Effect of electrical conductivity and transpiration on production of greenhouse tomato. Sci. Hort. 88, 11–29. Maas, E.V., 1986. Salt tolerance of plants. Appl. Agric. Res. 1, 12–26. Mizrahi, Y., Taleisnik, E., Kagan-Zur, V., Zohas, Y., Offenbach, R., Matan, E., Golan, R., 1988. A saline irrigation regime for improving tomato quality without reducing yield. J. Am. Soc. Hort. Sci. 113, 202–205. Ragab, R., 2002. A holistic generic integrated approach for irrigation, crop and field management: the SALTMED model. Environ. Modell. Softw. 17, 345–361. Romero-Aranda, R., Soria, T., Cuartero, J., 2000. Tomato plant-water uptake and plant-water relationships under saline growth conditions. Plant Sci. 160, 265–272. Romero-Aranda, R., Soria, T., Cuartero, J., 2002. Greenhouse mist improves yield of tomato plants grown under saline conditions. J. Am. Soc. Hort. Sci. 127, 644–648. Saranga, Y., Zamir, D., Marani, A., Rudich, J., 1991. Breeding tomatoes for salt tolerance: field evaluation of Lycopersicon germplasm for yield and dry-matter production. J. Am. Soc. Hort. Sci. 116, 1067–1071. Shani, U., Dudley, L.M., 2001. Field studies of crop response to drought and salt stress. Soil Sci. Soc. Am. J. 65, 1522–1528. Shannon, M.C., Grieve, C.M., 1999. Tolerance of vegetable crops to salinity. Sci. Hort. 78, 5–38. Soria, T., Cuartero, J., 1997. Tomato fruit yield and water consumption with saline water irrigation. Acta Hort. 458, 215–219. Stevens, M.A., Kader, A.A., Albright-Holton, M., Algazi, M., 1977. Genotypic variation for flavour and composition in fresh market tomatoes. J. Am. Soc. Hort. Sci. 102, 680–689. van Genuchten, M.T., 1983. Analyzing Crop Salt Tolerance Data: Model Description and Users’s Manual. Research Report 120 USDA Agricultural Research Service. U.S. Salinity Laboratory, Riverside, CA, 50 pp. van Genuchten, M.T., Hoffman, G.J., 1984. Analysis of crop salt tolerance data. In: Shainberg, I., Shalhevet, J. (Eds.), Soil Salinity Under Irrigation. Ecological Studies, vol. 8. Spring Verlag, Berlin, pp. 58–271. van Ieperen, W., 1996. Effect of different day nd night salinity levels on vegetative growth, yield and quality of tomato. J. Hort. Sci. 71, 99–111. van Os, E., 2001. Diffussion and environmental aspects of soilless growing systems. Italus Hortus 8, 9–15. Zerbini, P.E., Gorini, F., Polesello, A. 1991. Measurement of the Quality of the Tomatoes: Recommendations of an EEC Working Group. IVTPA, Milano, 36 pp.