The influence of drip irrigation or subirrigation on tomato grown in closed-loop substrate culture with saline water

Scientia Horticulturae 107 (2006) 365–372 www.elsevier.com/locate/scihorti The influence of drip irrigation or subirrigation on tomato grown in closed-loop substrate culture with saline water L. Incrocci *, F. Malorgio, A. Della Bartola, A. Pardossi Dipartimento di Biologia delle Piante Agrarie, Viale delle Piagge, 23 I-56124 Pisa, Italy Received 21 January 2005; received in revised form 13 September 2005; accepted 7 December 2005 Abstract In closed-loop soilless culture, one of the most relevant problems may be the accumulation in the recirculating nutrient solution of ions contained in irrigation water that are not or scarcely absorbed by the plants (e.g. Na, Cl) In order to verify the possibility to reduce the rate of recirculating water salinisation by means of subirrigation, an experiment was carried out in the spring of 2002 and 2004, with tomato plants (cv. Jama) grown in glasshouse and watered by conventional drip irrigation (D) or by subirrigation (trough bench system; S). The plants were cultivated in pots filled with a peat-perlite substrate for approximately 3 months and fed with complete nutrient solution, which was prepared with fresh water containing 10 mol m
3 NaCl; the nutrient solution in the collecting tank was replaced when the value of electrical conductivity (EC) exceeded 6.0 dS m
1. Water and nutrient crop use, salt accumulation in the substrate, and fruit yield were monitored. In S culture, the composition and EC of the recirculating nutrient solution changed slightly, while in D treatment there was a fast water salinisation that made it necessary to flushed out the nutrient solution in six different occasions, with consequent loss of water and fertilisers. In S culture, the upward water movement in the substrate, coupled with selective mineral uptake by the roots, caused salinity build-up in the upper region of the substrate, which was associated with Na+ accumulation. No significant influence of irrigation methods on fruit yield and quality was observed. These findings suggest that subirrigation can be a tool to reduce the water consumption and nutrient runoff in closed-loop substrate culture of tomato conducted with saline water. # 2005 Elsevier B.V. All rights reserved. Keywords: Hydroponics; Lycopersicon esculentum mill; NaCl; Salinity; Soilless culture; Subirrigation; Substrate 1. Introduction In protected horticulture closed-loop aggregate (substrate) cultures with drip (top) or bottom irrigation (trough or ebb-andflow bench systems) are increasingly used since they can minimise water use and limit the environmental pollution resulting from nutrient runoff (Voogt and Sonneveld, 1997). In these systems, the nutrient solution drained from the growing medium is collected and recirculated after proper adjustment of pH and nutrient concentration, the latter being generally assessed by measuring electrical conductivity (EC). In terms of water relations, four main components can be identified in closed substrate culture: (1) the water added to the mixing tank to compensate for plant transpiration and growth; (2) the recirculating nutrient solution; (3) the substrate; (4) the plants. Inevitably, any ion contained in the irrigation water that is not fully absorbed by the plants (such as Na and Cl) would * Corresponding author. Tel.: +39 050 2216529; fax: +39 050 2216524. E-mail address: incrocci@agr.unipi.it (L. Incrocci). 0304-4238/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.scienta.2005.12.001 accumulate in the recirculating water (Silverbush and BenAsher, 2001; Carmassi et al., 2005) and in the growing medium, in a proportion that depends on how the water moves in the substrate, then on the irrigation method. In drip-irrigated culture, water is generally distributed in excess. Therefore, the salts not taken up by the crop are removed from the substrate by drainage water and tend to accumulate in the recirculating nutrient solution, which has to be flushed out more or less frequently (semi-closed system), with consequent waste of water and fertilisers. Compared to drip irrigation, bottom irrigation (subirrigation) may reduce salt accumulation in the recirculating nutrient solution, since the upward water movement (by capillary force and evapotranspiration-driven mass flow) makes soluble salts to accumulate in the growing medium (Reed, 1996). The use of poor-quality water accelerates the salinity build-up in the substrate and may produce negative effects on salt-sensitive crops. Several papers (e.g. Molitor, 1990; Dole et al., 1994; Reed, 1996; Morvant et al., 1997; Todd and Reed, 1998; Uva et al., 1998; Treder et al., 1999; Lenzi et al., 2000; Cox, 2001; James and van Iersel, 2001; Guarino et al., 2002) have documented the 366 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 advantages of zero-runoff subirrigation (e.g. reduced labour costs and lower risk for root diseases) for greenhouse cultivation of pot ornamentals. However, less attention has been paid to the application of this technique to vegetable production (Santamaria et al., 2003; Serio et al., 2004). In this study, drip irrigation and subirrigation (trough bench system) were compared to cultivate tomato plants in a closedloop system using fresh water with 10 mol m
3 NaCl for preparing the nutrient solution. The aim was to verify the possibility to reduce, by means of subirrigation, the rate of salt accumulation in the recirculating nutrient solution with a consequent decrease in water use and nutrient leaching. Santamaria et al. (2003) published a similar paper, in which open-loop (without recirculating nutrient solution) drip irrigation and a trough bench system were evaluated to grow cherry tomato, which is considered relatively tolerant to salinity (Cough and Hobson, 1990), for 4 months under the typical greenhouse conditions of Mediterranean autumn. With respect to that work, in our study: (i) water was recycled in both irrigation systems; (ii) a more salt-sensitive tomato genotype and saline (NaCl-enriched) water were employed; (iii) the plants were grown in spring, that is under conditions enhancing evapotranspiration and rendering the plants more susceptible to salinity stress; (iv) an expedient was adopted to mitigate the possible effect of salinity build-up in the root zone of subirrigated plants (specifically, the volume of substrate was doubled with respect to drip irrigation treatment). system (roughly 20 L m
2) was greater by 25% than the one in D culture (roughly 16 L m
2), notwithstanding a slightly lower substrate moisture in S pots (1.9 L pot
1) with respect to D ones (2.5 L pot
1). Drip irrigation was automatically controlled on the basis of GR, which was measured with a piranometer. Depending on the growing stage, a watering dose of 0.6–1.2 L m
2 was supplied, whenever cumulated GR reached 1.0 MJ m
2; one emitter was placed in each pot. Leaching fraction (i.e. the ratio between drainage and irrigation volume) was around 0.60. In S treatment the pots were flooded, at a flow rate of approximately 3.0 L min
1 and to a height of 20–22 mm, for 15 min, six times during the day and once during the night, since nocturnal plant transpiration was not negligible (it accounted for up to 10–15% of daily water uptake). Every day, the collecting tank of each growing unit was automatically refilled with newly prepared nutrient solution to compensate for crop water uptake. The EC and the ion composition (mol m
3) of the refill water was the following: EC 3.0 dS m
1; 11.0 N-NO3
; 1.0 H2PO4
, 7.5 K+, 2.5 Mg2+, 4.0 Ca2+, plus Hoagland’s microelement concentration. The nutrient solution was prepared using NaCl-enriched water (10 mol m
3). The nutrient solution was checked daily for EC and pH; the latter was eventually adjusted to 5.5–6.0 with sulphuric acid. The nutrient solution of the collecting tank was renewed whenever the EC value exceeded 6.0 dS m
1. 2. Materials and methods 2.2. Determinations 2.1. Plant material and growing conditions In addition to the weight and number of fruits harvested in due course at turning-red stage, the incidence of nonmarketable (too small, cracked, affected by blossom-end rot or by Botritys) berries was recorded. Several quality attributes were also determined on the purée of marketable fruits: dry matter content, acidity (determined by titration with 0.1 mM NaOH and expressed as citric acid) and total soluble solids (determined with a bench refractometer). The volume and ion concentration of the refill water and of exhausted solutions (EC > 6.0 dS m
1) were systematically measured; the ion concentration of recirculating water was also determined every 3–7 days. Flame photometry was used for the analysis of K and Na; Ca and Mg were determined by EDTA titration method, while ion chromatography was used for anions. At the end of the experiment, two plants were sampled from each replicate and subdivided in leaves, stems, fruits and substrate (including some roots). Leaf area was measured with a digital planimeter and leaf area index (LAI) was calculated. Afterwards, plant samples were dried at 80 8C in a ventilated oven and then analysed for N, P, K and Na. Total N was determined by measuring both reduced N (with a mini-Kjeldhal system) and N-NO3; the latter was determined on the aqueous extract of dry matter (1:300, w:w) using Cataldo’s method (Cataldo et al., 1975). After perchloric acid digestion (90 min at 150 8C) of dry matter, K and Na were determined by flame photometry, and P by colorimetric method. The dry matter of fruits that had been previously collected from each plant was Round-fruit tomato plants (Lycopersicon esculentum Mill., cv. Jama F1) were grown in pots (18-cm diameter, 15-cm height, approximately 3.2-L volume) filled with a peat-perlite mixture (1:1, v:v), in a recirculating water system. Plant density was roughly 3.0 plant m
2. The experiment was conducted in a heated glasshouse at the University of Pisa in the spring of 2002 and repeated in 2004. The culture was initiated at the end of March by transplanting 50-day-old seedlings (raised in standard rockwool cubes) and lasted 13 weeks. Plants were grown vertically with single stem and stopped by cutting at the second leaf above the fifth truss. Bumblebees were kept in the glasshouse to improve flower pollination. The minimum (heating) and ventilation air temperature inside the glasshouse were 16 and 27 8C, respectively; maximum temperature reached up to 33–35 8C during sunny hours in late spring. Maximum photosynthetic photon flux density (PPFD) ranged from 500 to 700 mmol m
2 s
1; mean values of daily global radiation (GR) was 9.9 MJ m
2. A complete randomised block experimental design was adopted with three replicates for subirrigation (S) and drip irrigation (D) treatment. Each replicate consisted of 48 plants grown in four gullies (5-m long and 0.22-m wide) connected to an individual collecting tank (140 L; 8.8 L m
2). In S treatment each seedling was planted with the roots split in two adjacent pots. As a consequence, the volume of recirculating water in S 367 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 included in the calculation of total fruit dry weight. The quantity of tested elements accumulated in the root zone during the cultivation was corrected for the substrate mineral content, determined at the transplanting time. At the end of experiment, some determinations were also performed on the aqueous extracts of substrate. The substrate of sampled pots was divided horizontally in three equal layers (5 cm), from which major roots were removed. Afterwards, a sub-sample was extracted with distilled water with a dry substrate:water ratio (w:w) of 1:10 (for that, moisture content was determined in another sub-sample); finally, EC and the concentration of Na and K (as representative of nutrients) were measured in the filtered extracts. The contribution of Na and K to the substrate salinity was calculated on the basis of the equation proposed by Sonneveld et al. (1999): EC ¼ 0:10 C þ (1) This equation allows the calculation of EC from the total equivalent concentration (C+; eq m
3) of cations or anions, assuming that the cationic–anionic balance maintains the electroneutrality of the water solution. The contribution of K and Na was calculated as the percent ratio of their concentration on the total cation concentration calculated with Eq. (1). 2.3. Salt leaching In 2004, a simple experiment was performed to assess the salt leaching from D and S pots during a typical irrigation event. For that, after 10 weeks from planting, eight plants in their pots for each treatment were gently removed from the gullies and placed in separate growing units, which were made by a 1.5-m long gully connected to a tank with 10 L of deionised water. The pots removed from D culture were drip-irrigated for 90 s (about 400 mL of water were supplied to each pot), while S pots were subirrigated for 15 min. By 0, 5, 10 and 15 min, the EC of recirculating water was measured and the following equation was used to convert EC (dS m
1) to salt concentration (SC, g L
1) (Maynard and Hochmuth, 1997): EC ¼ 0:64 SC (2) 2.4. Statistics Since the experiment carried out in 2002 and 2004 gave similar results, only the second experiment has been reported herein. The values are means (S.E.) of three replicates; after ANOVA, LSD test was used for mean separation. 3. Results 3.1. Plant growth, mineral status and fruit yield Neither LAI nor plant dry matter accumulation was affected significantly by watering method (Table 1). On the other hand, compared to D, S decreased the leaf and stem concentration of Table 1 The influence of subirrigation on plant dry weight, leaf area index (LAI), mineral composition and fruit yield of tomato plants grown in closed-loop substrate culture, as determined 91 days after planting Parameter LAI Dry weight (kg m
2) Nitrogen (% DM) Leaves and stem Fruits Whole shoot Leaves and stem Fruits Phosphorus (% DM) Leaves and stem Fruits Potassium (% DM) Leaves and stem Fruits Sodium (% DM) Fruit yield Subirrigation Drip irrigation Significance 4.01  0.08 0.62  0.02 4.25  0.12 0.66  0.02 ns ns 0.67  0.02 1.29  0.03 0.70  0.02 1.36  0.02 ns ns 3.06  0.08 3.70  0.15 ** 2.18  0.15 2.01  0.14 ns 0.31  0.01 0.52  0.03 *** 0.25  0.01 0.26  0.01 ns 3.80  0.21 3.96  0.22 ns 3.65  0.29 3.84  0.22 ns Leaves and stem Fruits 2.57  0.29 1.75  0.22 ** 1.50  0.12 1.10  0.08 * Kg m
2 N fruits m
2 10.6  0.5 63  2.0 10.9  0.6 63.9  2.4 ns ns The data represent the mean values of three replicates (S.E.); for each row, the significance of the difference between irrigation systems is shown; ns: not significant. * Significant for P < 0.05. ** Significant for P < 0.01. *** Significant for P < 0.001. both N and P and increased that of Na (Table 1); no differences were recorded between D and S treatments for the concentration of K in green parts. Subirrigation also promoted the accumulation of Na in the fruits, which had a similar concentration of N, P and K in both irrigation systems (Table 1). Regardlessofirrigationmethod,therewas nodifferencein fruit production (Table 1) and quality, as assessed by the determination of a seriesof parameters in marketableberries (data notshown); on average, dry residue, titrable acidity and total soluble solids were 6.4%, 0.80% of citric acid and 5.28 Brix, respectively. Nevertheless, the fruits picked from the last two trusses of S plants had lower fresh weight (
25%) and higher total soluble solids (+10%) with respect to D treatment (data not shown). 3.2. EC and ion concentration of recycling water Due to salinity build-up, the nutrient solution was replaced in six different occasions in D culture (Fig. 1, top; Table 2). On the contrary, S plants were grown with the same nutrient solution during the whole growing period (about 3 months) since EC, albeit increasing, never reached the EC threshold (6.0 dS m
1) for flushing. In D treatment, the first renewal of nutrient solution occurred after 48 days from planting, when the cumulated crop water uptake was around 120 L m
2; at the same date, EC was only 3.95 dS m
1 in S culture. Periodical flushings did not restore the original EC value (3.0 dS m
1) of 368 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 total ion concentration by more than 50% compared to 33% in the fresh nutrient solution (Table 2). A tendency to rise during the last weeks was noticed also for all the macronutrients with the exception of P, as illustrated by Fig. 1 (bottom) where, for the sake of brevity, only K+ concentration has been shown. Presumably, this phenomenon resulted from reduced nutrient uptake by the crop in reason of low growth rate. With respect to the refill nutrient solution, the concentration of macronutrients in the exhausted nutrient solutions of D culture was 14–33% higher, although its contribution to the total salinity was lower (Table 2). 3.3. Water and nutrient balance Fig. 1. The influence of subirrigation (empty circle) or drip irrigation (filled circle) on the electrical conductivity (EC) and the concentration of Na and K of the recirculating nutrient solution in a closed-loop substrate culture of tomato. The value of EC threshold for flushing (as indicated by arrows in the upper graph) and the Na+ or K+ concentration in the refill water are also shown. The data represent the mean values of three replicates (the vertical bars represent the S.E.). recycling water on account of the salt-enriched nutrient solution retained by the substrate. In both irrigation treatments, the progressive rise in the EC of recirculating water was associated with the vast increase in Na concentration (Fig. 1, middle; Table 2), which in D culture rose up to values that were nearly four times greater than in the refill water (10 mol m
3). The change in the Cl concentration of recirculating water paralleled that of the Na. The concentration of these ions was roughly equal and in the exhausted culture solutions of D treatment was 36–38 mol m
3, accounting for Irrigation method did not influence significantly daily water uptake, which tended to increase during growing period (Fig. 2) because of leaf development and the occurrence of environmental conditions more favourable to leaf transpiration. In D culture, the periodical flushing produced a runoff of nearly 62 L m
2 and, therefore, total water use was significantly larger (+15%) than in S culture, in which water use efficiency (the fruit yield per cubic meter of water applied) was also improved (Table 3). Using the data obtained from the analysis of nutrient solutions, substrates and plant tissues, a balance for N, P, K and Na was compiled for each culture (Table 4). The mineral supply in Table 4 was calculated on the basis of the volume and the ion concentration of refill water. For S culture, the estimated crop mineral uptake incorporated the nutrients contained in the residual nutrient solution at the end of cultivation; diversely, this residue was considered in calculating runoff in D treatment, since EC was higher than 6.0 dS m
1. Irrigation system did not influence considerably the macronutrient consumption, which was estimated as the differences between nutrient supply and leaching, or by determining the nutrients accumulated in leaves, stems, fruits and growing medium. A noteworthy divergence between the two procedures was found only for P. Actually, the value derived from the supply-leaching difference was higher (roughly by 30%) than the one determined on the basis of chemical analysis. This difference was presumably due to the precipitation of P salts in the growing benches and in the collecting tank. On account of leaching requirement, the total use of N, P and K was higher in D treatment than in S (Table 4). In D treatment the runoff of N (115 kg ha
1) and P (18 kg ha
1) was massive, also in consideration of the short growing cycle (3 months). The irrigation method also changed the nutrient allocation. At least for N and P, the amount of nutrients contained in leaves and stems was higher in D than in S (Table 4), as a result of the increased element concentration in dry biomass (Table 1). Due to the larger volume of substrate, in S culture the N, P and K retained by the root zone was two- or three-fold greater than in D (Table 4). Different results were found for Na (Table 4), however. Indeed, the estimated or measured Na consumption in S culture was much larger than in D treatment due to an increase 369 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 Table 2 Electrical conductivity (EC) and ion concentration in the refill water, in the exhausted nutrient solutions periodically discarded from drip-irrigated culture or in the residual nutrient solution in subirrigated culture EC (dS m
1) Na+ (mol m
3) K+ (mol m
3) Mg2+ (mol m
3) Ca2+ (mol m
3) N-NO3
(mol m
3) P-H2PO4
(mol m
3) Cl
(mol m
3) Total concentration (mol m
3) 3.1 10.0 21.7 7.5 16.4 2.5 5.4 4.0 8.7 11.0 23.9 1.0 2.2 10.0 21.7 46.0 100.0 Exhausted nutrient solution (drip irrigation) 6.07 37.1 I (48 dat*) II (59 dat) 6.74 36.3 III (66 dat) 6.26 38.5 IV (72 dat) 6.08 37.0 V (81 dat) 6.81 36.1 VI (86 dat) 6.81 35.7 VII (91 dat; residue) 7.05 35.3 7.9 7.9 8.4 8.7 11.9 12.1 12.9 4.0 3.4 3.8 3.7 3.7 3.7 3.7 5.9 5.5 5.1 5.1 5.2 5.5 5.5 10.0 11.9 12.0 11.6 14.0 14.0 14.5 0.8 0.9 1.0 0.9 0.8 0.9 1.1 35.4 38.4 34.8 36.9 38.9 39.7 39.7 91.2 95.4 94.7 95.1 101.7 102.4 112.7 Refill nutrient solution Contribution (%) Mean value Contribution (%) 6.55 % 36.6 34.3 10.0 9.3 3.6 3.4 5.3 5.0 12.6 11.8 0.9 0.9 37.7 35.3 106.7 100.0 4.20 18.7 25.3 11.9 16.0 5.6 7.5 4.3 5.8 13.0 17.5 1.3 1.7 19.5 26.2 74.3 100.0 Residue (subirrigation) Contribution (%) The contribution of each ion to the total concentration of measured ions is also shown. * Dat: days after transplanting. in plant uptake and, principally, to the massive ion deposit in the root zone (seemingly, in the medium). The apparent ion uptake concentration (CU; i.e. the ratio between ion and water uptake) for Na was 9.5 and 2.7 meq L
1 for S and D plants, respectively. This difference resulted from the Na accumulation in the growing medium, since no difference between S and D treatment was recorded when CU for both Na and macronutrients was calculated on the basis of crop water uptake and genuine element accumulation in plant tissues (data not shown). 3.4. Substrate salinity The pattern of salt distribution in the root zone of D and S plants was investigated by measuring, at the end of cultivation, pH, EC and the concentration of K and Na in the aqueous extract of different substrate regions. As indicated by the value of EC (Fig. 3, top), in S treatment the salts tended to concentrate in the upper layer where there were much less roots, which grew preferentially in the lower region (as found by visual assessment). By contrast, a decreasing EC from the bottom to the top third of the substrate was found in D culture, in which roots tended to occupy the whole pot. In D culture, the EC of the upper region was three times lower than in S treatment (Fig. 3, top). At least in S pots, a large vertical gradient was found in Na concentration that showed a tendency to accumulate in the top region (Fig. 3, middle); the reverse phenomenon was observed in D pots. In both treatments, K concentration was less stratified among the substrate regions although, when compared to middle and bottom layer, a significant increase (S) or decrease (D) was found in the top portion (Fig. 3, bottom). Table 3 The influence of subirrigation or drip irrigation on the water use of a closed-loop substrate culture of tomato Fig. 2. The influence of subirrigation (empty columns) or drip irrigation (filled columns) on the weekly averages of daily water uptake of tomato plants grown in closed-loop substrate culture. The data represent the mean values of three replicates (the vertical bars represent the S.E.). Parameter Subirrigation Drip irrigation Significance Water supply (WS; L m
2) Residue or runoff (R; L m
2) Crop water uptake (WS-R; L m
2) Water use efficiency (g L
1) 324 9 315 32.7 373 62 312 29.2 * ** ns * The data represent the mean values of three replicates; for each row, the significance of the difference between irrigation systems is shown; ns: not significant. * Significant for P < 0.05. ** Significant for P < 0.01. 370 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 Table 4 The influence of subirrigation or drip irrigation on the mineral use of a closedloop substrate culture of tomato Parameter Subirrigation Drip irrigation Significance
2 Nitrogen (g m ) Supply Residue or runoff Estimated crop uptakea Measured crop uptakeb Allocated mineral content Leaves and stem Fruits Whole shoot Substrate Phosphorus (g m
2) Supply Residue or runoff Estimated crop uptakea Measured crop uptakeb Allocated mineral content Leaves and stem Fruits Whole shoot Substrate Potassium (g m
2) Supply Residue or runoff Estimated crop uptakea Measured crop uptakeb Allocated mineral content Leaves and stem Fruits Whole shoot Substrate Sodium (g m
2) Supply Residue or runoff Estimated crop uptakea Measured crop uptakeb Allocated mineral content Leaves and stem Fruits Whole shoot Substrate * 48.6 1.5 47.1 46.6 55.8 11.5 44.3 43.5 19.0 14.6 33.6 13.0 24.4 14.1 38.5 5.0 10.0 0.3 9.7 7.0 11.6 1.8 9.8 6.4 1.9 1.7 3.6 3.4 3.4 1.8 5.2 1.2 92.8 3.4 89.4 87.5 106.2 24.1 82.1 79.1 23.6 24.5 48.1 39.4 26.1 26.9 53.0 26.1 ns ns ns 74.5 3.8 70.7 71.0 85.8 51.8 34.0 30.1 * 15.9 10.1 26.0 45.0 11.6 7.7 19.3 10.8 * *** ns ns * ns * *** * *** ns ns *** ns ** *** * *** ns * ** *** *** *** * * *** The data represent the mean values of three replicates; for each row, the significance of the difference between irrigation systems is shown; ns: not significant. a The values were calculated as the difference between supply and residue or runoff. b The values were calculated on the basis of plant dry biomass and its element concentration. * Significant for P < 0.05. ** Significant for P < 0.01. *** Significant for P < 0.001. The contribution of Na to the total cation concentration increased with pot height in S and diminished in D (Fig. 3, middle), while the opposite trends were found for K (Fig. 3, bottom). Irrigation system and sample type did not influence significantly substrate pH, which ranged between 5.8 and 6.4, although a slight alkalisation with increasing pot height Fig. 3. The influence of subirrigation (empty columns) or drip irrigation (filled columns) on the electrical conductivity (EC) and the Na+ or K+ concentration of the aqueous extracts of different substrate layers sampled in a closed-loop culture of tomato. The data represent the mean values of three replicates (the vertical bars represent the S.E.); in each graph, the bars accompanied by the same letter are not significantly different (P < 0.05). The values within brackets represent the Na or K contribution to total cation concentration. was observed in S plants. The initial pH of growing medium was nearly 6.0. 3.5. Salt leaching Fig. 4 shows the results of an experiment conducted to quantify the salt leaching that occurred during a typical L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 Fig. 4. The influence of subirrigation (empty circle) or drip irrigation (filled circle) on the salt leakage that occurred from pots with tomato plants during a typical irrigation event. Drip irrigation lasted 90 s and some 400 mL of deionised water were supplied to each pot, while subirrigation lasted 15 min. Salt leakage was assessed on the basis of the measurement of electrical conductivity (EC) of the solution in the mixing tank. The data represent the mean values of three replicates (the vertical bars represent the S.E.). irrigation event in D and S culture. After the end of watering and drainage, it was calculated from EC value that 1.22 g of salts were leached out of each D pot against 0.04 g in S treatment. 4. Discussion This work showed that subirrigation may reduce substantially the salinisation of the recirculating nutrient solution that typically occur in drip-irrigated substrate culture as a consequence of the accumulation of ions contained in the fresh water that are not fully absorbed by the crop. In this growing system, the volume (V) of recirculating nutritive culture, which includes the water retained by the substrate and represents the water buffer, influences the change in salt concentration, according to a model recently published by Carmassi et al. (2005). Using this model, the change in the Na concentration of S culture was simulated using two values for V: the actual one (20 L m
2) and the one used in D culture (16 L m
2), with the aim to estimate how much the difference (in terms of recirculating water salinity) between D and S treatments was due to the irrigation system and to the different water buffer. The rate of Na accumulation was higher with lower V, but the concentration (data not shown) never reached the values (nearly 40 mol m
3; Table 3) that typically led to the EC limit for flushing. In S culture, the salt tended to accumulate in the growing medium, in particular in the upper third, as a result from the ascendant water movement coupled with the selective root mineral uptake. Indeed, in S culture the increase in substrate EC was associated with the large accretion of Na+; in addition, the contribution of this ion to substrate salinity increased with pot height in S and diminished in D, while the opposite trend was found for K. A differential contribution of tested nutritive ions 371 to the EC of different substrate layers was observed in pot culture by other authors (Treder et al., 1999). Apparently the higher EC in the upper portion of substrate did not cause salinity stress to S plants, because their roots grew mostly in the bottom region, where the values of determined parameters were similar in both treatments. In fact, no symptom of nutrient deficiency and/or salt injury (wilting or desiccation of leaves) was shown by S plants and no differences between D and S treatment were found in terms of plant growth, leaf development, crop yield and fruit quality. The absence of any important effect of root zone salinisation on plant growth in S culture was also related to the use of a much larger volume of substrate. In fact, in a preliminary experiment with one pot only per plant, subirrigation decreased remarkably plant growth and fruit yield (data not shown). This result, actually, suggested the expedient of a double volume of substrate for S treatment. The considerable deposit of Na in the root zone of S culture stimulated plant uptake of this ion (as suggested by higher Na concentration of both green parts and fruits of S plants, compared to D treatment) and reduced that of N and P, at least in leaves and stems (Table 1). Other authors reported that in plants Na uptake is proportional to its concentration in the root zone (Sonneveld, 2000; Silverbush and Ben-Asher, 2001; Malorgio et al., 2001), and that a relatively high Na content in the growing medium may reduce the shoot content of N and P (Savvas and Lenz, 2000; Cox, 2001). The accumulation of Na in plant tissues could be responsible for a decrease in fruit production in a cultivation that takes longer than in this study. Indeed, a smaller size of the fruits picked from upper trusses was observed both in our experiment as well as by Santamaria et al. (2003). These authors also found that, compared to open-loop drip irrigation, closed-loop subirrigation reduced significantly total fruit yield, although the quality of marketable berries was slightly improved. To conclude, trough bench system proved to be a method to reduce the water consumption and nutrient runoff of closedloop tomato culture, when saline water is available. Work is in progress to investigate how subirrigation influences plant growth and fruit yield in long-cycle tomato culture and, eventually, to develop measures for limiting crop salinity stress. Acknowledgement This work was supported by Italian Ministry of University and Research (MIUR-PRIN 2003, ‘‘La gestione di sistemi fuori suolo a ciclo chiuso: adattamento, ottimizzazione e controllo in ambienti mediterranei su colture ortofloricole’’: paper n8. 5). References Carmassi, G., Incrocci, L., Maggini, R., Malorgio, F., Tognoni, F., 2005. Modelling salinity build-up in recirculating nutrient solution culture. J. Plant Nutr. 28, 431–445. Cataldo, D.A., Haroon, M., Schrader, L.E., Youngs, V.L., 1975. Rapid colorimetric determination of nitrate in plant tissue by nitration of salicylic acid. Commun. Soil Sci. Plan. 6, 71–80. 372 L. Incrocci et al. / Scientia Horticulturae 107 (2006) 365–372 Cox, D.A., 2001. Growth, nutrient content and growth medium electrical conductivity of poinsettia irrigated by subirrigation or from overhead. J. Plant Nutr. 24, 523–533. Cough, C., Hobson, G.E., 1990. A comparison of the productivity, quality, shelf-life characteristics and consumer’s reaction to the crop from cherry tomato plants grown at different levels of salinity. J. Hortic. Sci. 65, 431– 439. Dole, J.M., Cole, J.C., von Broembsen, S.L., 1994. Growth of poinsettias, nutrient leaching, and water-use efficiency respond to irrigation methods. HortScience 29, 858–864. Guarino, D., Gallina, P.M., Quattrini, E., Pardossi, A., 2002. Poinsettia in vaso, l’influenza della tecnica di irrigazione. Colture Protette 31, 117–123. James, E., van Iersel, M., 2001. Ebb and flow production of petunias and begonias as affected by fertilizers with different phosphorus content. HortScience 36, 282–285. Lenzi, A., Oggiano, N., Rinaldi, M., Maletta, M., 2000. Gerbera a ciclo chiuso: fertirrigazione e substrati. Colture Protette 29, 85–91. Maynard, D.N., Hochmuth, G.J., 1997. Knott’s Handbok for Vegetable Growers, 4th ed. John Wiley & Sons Inc., New York, p. 582. Malorgio, F., Incrocci, L., Carmassi, G., Pardossi, A., Tognoni, F., 2001. Accumulo di sali (NaCl) e consumo minerale in pomodoro coltivato in sistemi idroponici a ciclo chiuso. Italus Hortus 8, 43–48. Molitor, H.D., 1990. The European perspective with emphasis on subirrigation and recirculation of water and nutrients. Acta Hort. 272, 165–170. Morvant, J.K., Dole, J.M., Allen, E., 1997. Irrigation system alter distribution of roots, soluble salts, nitrogen, and pH in the root medium. Hort. Tech. 7, 156– 160. Reed, D.W., 1996. Closed production systems for containerized crops: recirculating subirrigation and zero-leach system. In: Reed, D.W. (Ed.), Water, Media and Nutrition for Greenhouse Crops. Ball Publishing, Batavia, IL, USA, pp. 221–245. Santamaria, P., Campanile, G., Parente, A., Elia, A., 2003. Subirrigation vs. drip-irrigation: effects on yield and quality of soilless grown cherry tomato. J. Hortic. Sci. Biotechnol. 78, 290–296. Savvas, D., Lenz, F., 2000. Effects of NaCl or nutrient-induced salinity on growth, yield and composition of eggplants grown in rockwool. Sci. Hortic. 84, 37–47. Serio, F., Elia, A., Signore, A., Santamaria, P., 2004. Influence of nitrogen form on yield and nitrate content of subirrigated early potato. J. Sci. Food Agric. 84, 1835–1839. Silverbush, M., Ben-Asher, J., 2001. Simulation study of nutrient uptake by plants from soilless cultures as affected by salinity build-up and transpiration. Plant Soil 233, 59–69. Sonneveld, C., Voogt, W., Spaans, L., 1999. A universal algorithm for calculation of nutrient solution. Acta Hort. 481, 331–339. Sonneveld, C., 2000. Effect of salinity on substrate grown vegetables and ornamentals in greenhouse horticulture. PhD Thesis, Wageningen University. Todd, N.M., Reed, D.W., 1998. Characterising salinity limits of new guinea impatiens in recirculating subirrigation. J. Am. Soc. Hortic. Sci. 123, 156– 160. Treder, J., Matysiak, B., Nowak, J.S., Nowak, J., 1999. The effects of potting media and concentration of nutrient solution on growth and nutrient content of three Ficus species cultivated on ebb-and-flow benches. Acta Hort. 481, 433–439. Uva, W.L., Weiler, T.C., Milligan, R.A., 1998. A survey on the planning and adoption of zero runoff subirrigation system in greenhouses operations. HortScience 33, 193–196. Voogt, W., Sonneveld, C., 1997. Nutrient management in closed growing system for greenhouse production. In: Goto, E., Kurata, K., Hayashi, M., Sase, S. (Eds.), Plant Production in Closed Ecosystems. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 83–102.