Influence of Saline Irrigation on Growth, Ion Accumulation and Partitioning, and Leaf Gas Exchange of Carrot (Daucus carota L.)

Abstract

Like those of many horticultural crop species, the growth and leaf gas exchange responses of carrot (Daucus carota L.) to salinity are poorly understood. In this study ion accumulation in root tissues (periderm, xylem and phloem tissues) and in leaves of different ages was assessed for carrot plants grown in the field with a low level of salinity (5·8 mm Na⁺ and 7·5 mm Cl^–) and in a glasshouse with salinity ranging from 1–80 mm. At low levels of salinity (1–7·5 mm), in both the field and glasshouse, carrot leaves accumulated high concentrations of Cl^– (140–200 mm); these appear to be the result of a high affinity for Cl^– uptake and a low retention of Cl^– in the root system. However, Cl^– uptake is under tight control, with an 80‐fold increase in external salinity resulting in only a 1·5‐fold change in the Cl^– concentration of the shoot and no increase in the Cl^– concentration of the root xylem tissue. In contrast to Cl^–, shoot Na⁺ concentrations were comparatively low (30–40 mm) but increased by seven‐fold when salinity was increased by 80‐fold. Growth over the 56‐d treatment period in the glasshouse was insensitive to salinity less than 20 mm, but at higher concentrations the yield of carrot tap roots declined by 7 % for each 10 mm increase in salinity. At low levels of salinity the accumulation of high concentrations of Cl^– (150 mm) in carrot laminae did not appear to limit leaf gas exchange. However, photosynthesis and stomatal conductance were reduced by 38 and 53 %, respectively, for plants grown at a salinity of 80 mm compared with those grown at 1 mm. Salinity‐induced reductions in both p_i and carbon isotope discrimination (Δ) were small (2·5 Pa and 1·4 ‰, respectively, at 80 mm) indicating that the reduction in photosynthesis was only marginally influenced by CO₂ supply. At a salinity of 80 mm the photosynthetic capacity was reduced, with a 30 % reduction in the CO₂‐saturated rate of photosynthesis (A_max) and a 40 % reduction in both the apparent rate of RuBP‐carboxylase‐limited CO₂ fixation (V_cmax) and the electron transport rate limiting RuBP regeneration (J_max). This study has shown that carrot growth and leaf gas exchange are insensitive to the high leaf Cl^– concentrations that occur at low levels (1–7 mm) of salinity. However, growth is limited at salinity levels above 20 mm and leaf gas exchange is limited at salinity levels above 8 mm.

INTRODUCTION

The salinization of major waterways, horticultural land and underground water reserves is an important issue for the sustainability of irrigated horticulture. An improved understanding of the responses of horticultural crop species to salinity may aid the development of more tolerant cultivars and improved management practices. Yet, even for some major crops, there is conjecture in the literature regarding levels of tolerance. Bernstein and Ayers (1953) found that carrot was sensitive to salinity, with a tolerance lower than that of other horticultural crops such as potato, lettuce and beans. However, in later studies large differences in salinity tolerance have been reported for carrot (Maas and Hoffmann, 1977; Matsubara and Tasaka, 1988; Mangal et al., 1989). This issue may be confounded by intervarietal differences in ion uptake reported for carrot. Bernstein and Ayers (1953) reported that the Na⁺ and Cl^– concentrations of tissue water in shoots of a number of varieties of carrot grown under ‘control’ conditions ranged from 13 to 33 mm Na⁺ and from 107 to 135 mm Cl^– at maturity. In particular, these high Cl^– levels may indicate that carrot has a high affinity for Cl^– uptake at low salinity and that a high affinity for Cl^– uptake may have ramifications for carrot growth and leaf gas exchange under saline conditions.

There are no published reports on the influence of salinity on carrot leaf gas exchange. In other species, salinity decreases assimilation through reductions in leaf area (Papp et al., 1983; Munns et al., 2000), stomatal conductance (Brugnoli and Lauteri, 1991; Ouerghi et al., 2000), mesophyll conductance (Delfine et al., 1998), inhibited carbon metabolism (Seemann and Sharkey, 1986; Brugnoli and Bjorkman, 1992) and/or the concentration or efficiency of photosynthetic enzymes (Seemann and Critchley, 1985; Yeo et al., 1985; Brugnoli and Bjorkman, 1992). A number of authors have shown that leaf gas exchange can be correlated with laminae Na⁺ (Yeo et al., 1985) and Cl^– (Downton, 1977; Walker et al., 1981; Seemann and Critchley, 1985) concentrations, and there remains considerable conjecture with respect to the relative importance of the toxic effects of ions on photosynthesis compared with the effects mediated by water relations or impaired nutrition under saline conditions. In most instances it is likely that a number of processes operate in concert and vary with the severity and duration of the stress and the genotype (for a review, see Munns, 1993).

An initial study was conducted to confirm the high levels of Cl^– accumulation in laminae of plants grown in the field and irrigated with slightly saline water. There were then three major aims of this paper: first to resolve some of the conjecture in the literature over the salt tolerance of carrot; secondly to further describe the uptake and partitioning (between organs) of Na⁺, K⁺ and Cl^– in plants grown with a range of salinity levels; and finally to determine the influence of salinity on carrot leaf gas exchange and photosynthetic capacity.

MATERIALS AND METHODS

Experiment 1: Ion accumulation and tissue partitioning in field‐grown plants

Site details.

Carrot plants (Daucus carota L. ‘Ivor’ F₁ syn. Nantes; Clause Seeds, France) were sampled from a large commercial property 15 km east of the coastal town of Guilderton (31·21°S, 115·30°E), approx. 100 km north of Perth in the south‐west of Australia. Irrigation was by centre pivot irrigator (total area = 66 ha) with irrigations of 3 to 9 mm applied up to four times daily. The crop was sown during December 1997 and harvested in late March 1998. The soil was well flushed throughout the growing period: the average rainfall plus irrigation during the growth of the crop exceeded pan evaporation by 1·4‐fold. The soil at the site is a vertically and horizontally homogeneous deep sand, classified as a Karakatta sand (McArthur and Bettenay, 1974) with a grey phase from 0 to 0·5 m over a yellow phase to depth. Irrigation water was drawn from an aquifer at 60 to 75 m depth. The composition of the irrigation water was assessed on four occasions during the crop cycle: it contained (mean, n = 4 ± s.e.) 5·8 ± 0·13 mm Na⁺ and 7·5 ± 0·06 mm Cl^–. Further details of cultural conditions for a similar crop and site can be found in Gibberd et al. (2000).

Sampling.

Plants were sampled just prior to harvesting (90 d after sowing, DAS) at five points, 20 m apart on a transect radially orientated from the origin of the centre pivot irrigator. At each sample point nine plants were sampled; the three largest and three smallest were discarded and three median‐sized plants were retained.

Shoots were washed twice in tap water and three times in de‐ionized water before being blotted dry and divided into one of three categories: (1) emergent leaf, emerging from the shoot base and yet to expand; (2) youngest fully expanded (hereafter called mature) leaf; and (3) the remainder of leaves (hereafter referred to as ‘bulk shoots’). All samples included both laminae and petioles. Fresh weights were determined. Samples were then oven dried at 80 °C for 3 d before being reweighed to determine dry weights.

Storage roots were gently sponged to remove soil before being rinsed briefly, twice in tap water and three times in de‐ionized water. A 4‐cm sub‐sample was sectioned from 40 to 80 mm below the shoulder of the storage root. This ensured that the sub‐sample was derived from root rather than from hypocotyl tissue, which can extend approx. 25 mm below the base of the shoot (Esau, 1940). The sub‐sample was then separated into three sections by first using a vegetable peeler to cut an approx. 3‐mm‐thick layer from the surface of the tap root to form a sample composed predominantly of periderm and pericycle tissues. The remainder of the sub‐sample was then cut longitudinally into quarters, and the outer cylinder of predominately secondary phloem and parenchyma tissue was separated from the inner cylinder of predominately secondary xylem and parenchyma tissues along the inner cambium (tissue identification and nomenclature as per Esau, 1940). Root tissue samples are hereafter referred to as periderm, phloem and xylem. Fresh weights were determined before tissues were dried at 80 °C for 7 d and reweighed to determine dry weights.

Ion analysis: cations.

Dried material was ground in a ball mill to a particle size of <0·5 mm. Ions were extracted according to Hunt (1982). In brief, 0·1 g of dried material was extracted in 10 ml of 0·5 m HCl constantly rotating at 23 °C for 24 h. The extract was filtered through Whatman No. 42 paper and the filtrate was analysed for cation (Na⁺ and K⁺) concentrations by flame photometry (Spectra AA 600 Spectrophotometer; Spectra, Varian, CA, USA).

Ion analysis: chloride.

Samples of dried, ground material weighing 0·1 g were extracted in 10 ml of H₂O at 60 °C for 4 h. The extract was filtered through Whatman No. 42 paper; 0·2 ml of extract was diluted with 0·1 n HNO₃, 10 % v/v CH₃COOH to a volume of 5 ml and the Cl^– concentration determined by coulombic titration against Ag⁺ (Digital chloridometer; Buchler instruments, Kansas, KS, USA).

Experiment 2: Growth, ion accumulation and ion partitioning of container‐grown plants irrigated with saline nutrient solution

Plant culture.

Carrot plants were grown in a naturally‐lit glasshouse at 25/15 °C (day/night temperatures), with midday photosynthetically active radiation (PAR) averaging 1425 µmol m^–2 s^–1 at shoot height, and a day length of approx. 16 h. ‘Pots’ were free‐draining black polyethylene bags (15 l) containing 22 kg of grey‐phase Karakatta sand sourced from uncultivated soil adjacent to the field site used in expt 1. Seven seeds were sown in each pot at 12 mm depth. Prior to emergence the pots were irrigated three times daily with a solution containing (mm): Ca²⁺, 1·0; SO₄^2–, 1·2; K⁺, 0·4; H₃BO₃, 0·5 × 10^–3; and pH = 5·8. After emergence, plants were irrigated with 0·2‐strength nutrient solution which was increased to full‐strength in 0·2‐strength increments imposed every second day. At each irrigation a volume 1·6‐fold greater than that required to bring the soil to field capacity was applied. At full‐strength the nutrient solution contained (mm): K⁺, 1·1; Mg²⁺, 0·2; Ca²⁺, 0·9; NH₄⁺, 0·1; NO₃^–, 0·7; SO₄^2–, 0·7; Cl^–, 1·2; H₂PO₄^–, 0·1; H₃BO₃, 5 × 10^–3; Zn, 7·5 × 10^–4; Cu, 2·0 × 10^–4; Mn, 1 × 10^–2; Mo, 0·02 × 10^–3; and Fe‐EDDHA, 2·5 × 10^–2. The pH of the solution was adjusted to 5·8 using KOH. At 24 DAS seedlings were thinned to two per pot.

Treatments.

At 34 DAS six treatments were imposed: a control (no added NaCl with a background Cl^– concentration of 1·2 mm, hereafter referred to as 1 mm) and five ‘salinity’ levels of 5, 10, 20, 40 and 80 mm. The salinity treatments incorporated a combination of NaCl and CaCl₂ to maintain a Na⁺ : Ca²⁺ ratio of 15 : 1; salinity levels are hereafter referred to on the basis of their Cl^– concentrations. At the onset of the treatments, salinity levels of the irrigation solution were increased at a rate of 20 mm per day. During the treatment period pots were flushed with solution twice daily. The irrigation solution did not come into contact with the laminae, as it did in the field experiment.

Harvest and ion analysis.

Plants were harvested at 90 DAS and were divided in the same manner as in expt 1. Ion analysis was performed on oven‐dried, ground material as described above.

Statistical analyses.

Experiment 2 consisted of a control and five salinity levels. There were four replicates arranged in a randomized block design with four blocks. One‐way ANOVAs were used to determine the effect of salinity level on the measured variables, and l s.d. values were calculated at the 5 % confidence level.

Experiment 3: Influence of saline irrigation on leaf gas exchange.

Plant culture.

Culture methods and environmental conditions were similar to those used in expt 2, with the exception that the salinity treatments did not commence until 50 DAS when the carrot storage roots had started to turn orange and expand. Three treatments were imposed (1, 8 and 80 mm) with six replicate pots each containing two plants.

Leaf gas exchange measurements.

Leaves were tagged at emergence and leaf gas exchange was measured on five occasions at 7‐d intervals using a CIRAS‐1 (PP Systems, Hitchin, UK) portable infrared gas analysis system. On each occasion, measurements were performed on terminal leaflets using a different 25‐ to 27‐d‐old (fully expanded) leaf of one plant in each pot. Measurements were performed at 36·5 Pa CO₂, 40 % relative humidity (RH), 20 °C and a PAR of 1200 µmol m^–2 s^–1 (saturating light). Each terminal leaflet used for gas exchange was removed and its area was determined after digitizing with a flat‐bed scanner (ScanJet IIcx; Hewlett Packard, Palo Alto, CA, USA). Calculations of gas exchange and internal CO₂ partial pressure (p_i) were performed as described by von Caemmerer and Farquhar (1981). Immediately following gas exchange measurements two additional leaflets adjacent to the terminal leaflet were removed for Cl^– and Na⁺ analysis (as described previously). A preliminary study showed that there was no significant (P ≥ 0·05) difference in the ion content of the terminal leaflets (used for gas exchange) and the adjacent leaflets (used for ion analysis). Responses of CO₂ assimilation (P_n) to p_i (P_n/p_i curves) were assessed on the terminal leaflets of 25‐ to 27‐d‐old leaves that had developed during the treatment period. After an initial equilibration period (20 min) at 36·5 Pa CO₂, 40 % RH, 20 °C and 1200 µmol m^–2 s^–1 PAR (saturating light), the external CO₂ concentration (p_a) was varied step‐wise from 5·0 to 150·0 Pa CO₂. At each step gas exchange parameters were recorded at steady state. Maximum P_n at saturating CO₂ concentrations (P_nmax) was determined and the RuBP‐carboxylase‐limited CO₂ fixation capacity (V_cmax) was calculated from the slope of the initial linear phase of the relationship between net photosynthetic CO₂ fixation and p_i. The photosynthetic electron transport rate limiting RuBP‐regeneration (J_max) was calculated from the P_n/p_i relationship where P_n attained a maximum and was independent from any further increase in p_i [for details of calculations see von Caemmerer and Farquhar (1981) and Harley et al. (1992)]. Leaf area and laminae Cl^– and Na⁺ concentrations were recorded as described previously.

Carbon isotope composition and calculation of the ratio of intercellular to atmospheric CO₂ partial pressures (p_i/p_a).

At the end of the experiment, laminae of the three youngest fully expanded leaves (all formed during the treatment period) of each plant were harvested, dried at 80 °C for 2 d and then finely ground (<100 µm). The ¹³C : ¹²C ratio of the laminae dry matter was determined by ratio mass spectrometry (20–20 isotope ratio mass spectrometer; PDZ Europa, Northwich, UK) using a working standard of known carbon isotope composition relative to Pee Dee Belemnite. Carbon isotope discrimination (Δ) was calculated as described by Farquhar and Richards (1984) assuming a ¹³C : ¹²C ratio of CO₂ in air equal to 7·6 ‰. The mean ratios of intercellular to atmospheric CO₂ partial pressures (p_i/p_a) over the duration of the treatment period were calculated from the Δ values according to Farquhar and Richards (1984):

p_i/p_a = (Δ – a)/(b – a)(1)

where a is the discrimination associated with diffusion of CO₂ in air (= 4·4 ‰) and b is discrimination associated with carboxylation (= 30 ‰). The discrimination of CO₂ during dissolution, liquid phase diffusion, respiration and photorespiration was assumed to be similar for the leaves of plants in the various salinity treatments and is incorporated into the estimated value of b.

Statistical analyses.

Experiment 3 consisted of three salinity treatments, with six replicates arranged in a randomized block design with two blocks. One‐way ANOVAs were used to determine the effect of salinity level on the measured variables and l s.d. values were calculated at the 5 % confidence level.

RESULTS

Experiment 1: Ion accumulation and tissue partitioning in field‐grown plants

Carrot plants grown under field conditions and irrigated with water containing low levels of Na⁺ (5·8 mm) and Cl^– (7·5 mm) accumulated up to 160 mm Cl^– and 90 mm Na⁺ in their leaves (Fig. 1A). There was no significant difference in Cl^– and Na⁺ concentrations between the youngest mature leaves and the bulk shoot sample. However, concentrations of Cl^– and Na⁺ in the emergent leaves were 50 % lower than those in older leaves. The K⁺ concentration was highest (102 mm) in emergent leaves and decreased with leaf age. The Cl^– concentration did not differ significantly among the root tissues. The average Cl^– concentration of the root tissues was 35 mm, 78 % lower than in the youngest mature leaf and bulk shoot samples (Fig. 1B). Sodium concentrations exceeded Cl^– concentrations in the phloem tissue by 1·2‐fold and in the xylem tissue by 1·8‐fold, whereas Na⁺ and Cl^– concentrations were similar in the periderm. Potassium concentrations in the periderm were 2·9‐ and 1·6‐fold higher than those in the phloem and xylem tissues, respectively.

Fig. 1. Concentration of K⁺, Na⁺ and Cl^– in the tissue water of emergent leaves (EL), youngest mature leaves (YML) and bulk shoot sample (A), and root tissues (B) of carrot grown in the field for 90 d on a sandy soil irrigated with water containing 5·8 mm Na⁺ and 7·5 mm Cl^–. Data are means ± s.e.m. (n = 4).

Experiment 2: Influence of salinity on growth and ion accumulation of container‐grown plants

Growth.

Biomass of 90‐d‐old plants was not affected by irrigation with nutrient solutions (over a 54‐d period) containing salinity levels up to 20 mm (Fig. 2). However, plant biomass was reduced by 15 and 37 % at salinity levels of 40 and 80 mm, respectively. At salinities up to 40 mm, the proportion of biomass partitioned to roots was seven‐fold greater than that partitioned to shoots. Biomass partitioning to roots was reduced to 5·9 g root per g shoot at 80 mm, indicating a greater effect of high salinity on root growth than on shoot growth.

Fig. 2. Fresh weight of storage roots (A) and shoots (B) of carrot grown for 90 d in sandy soil irrigated with nutrient solution containing salinity levels from 1 to 80 mm applied as NaCl and CaCl₂ with an Na⁺ : Ca²⁺ ratio of 15 : 1. Data are means ± s.e.m. (n = 4).

Leaf ion concentration.

At the lowest salinity level (1 mm) carrot plants accumulated very high (140–210 mm) concentrations of Cl^– in their shoots, even in young, emergent leaves (Fig. 3A). Over the treatment range of 5 to 40 mm, leaf Cl^– concentration for all leaf tissues was constant (no significant treatment effect) and was 1·4‐fold greater than at the 1 mm salinity level. At the highest soil salinity level of 80 mm, the leaf Cl^– concentrations of emergent leaves and the bulk leaf sample increased to 290 and 330 mm Cl^–, respectively.

Fig. 3. Tissue Cl^– (triangles), K⁺ (circles) and Na⁺ (squares) concentrations of leaves (A) and root tissues (B) of carrot grown for 90 d in sandy soil irrigated with nutrient solution containing salinity levels from 1 to 80 mm applied as NaCl and CaCl₂ with an Na⁺ : Ca²⁺ ratio of 15 : 1. Bars are s.e.m. (n = 4) where larger than the symbol.

In contrast to Cl^–, leaf Na⁺ concentration was low (30–40 mm) at the 1 mm treatment, and in all leaf types it increased almost linearly with the increase in salinity of the irrigation solution to reach an average value of 220 mm at a salinity level of 80 mm. Leaf K⁺ concentration was lower in the bulk leaves than in the youngest mature leaves and emergent leaves but remained constant across all salinity treatments (Fig. 3A). For the three leaf tissues, and across the salinity treatments, leaf Na⁺ concentration was approximately equal to the leaf Cl^– content minus the leaf K⁺ concentration. Hence the shoot ratio of [Na⁺ + K⁺] : Cl^– remained comparatively constant across the salinity levels with a ratio of 0·65 at 1 mm increasing to 1·0 at 80 mm (Table 1). At the 1 mm salinity level the K⁺ : Na⁺ ratio of the shoots was between 1·8 and 2·3. However, the ratio decreased with increasing salinity and at the treatment levels of 40 and 80 mm it was less than 1.

Table 1.

Ratios of [Na⁺ + K⁺] : Cl^– and K⁺ : Na⁺ for tap roots and shoots of carrot grown for 90 d in sand culture irrigated for 54 d with nutrient solution containing salinity levels from 1 to 80 mm applied as NaCl and CaCl₂ with an Na⁺ : Ca²⁺ ratio of 15 : 1

	[Na+ + K+] : Cl–		K+ : Na+
Salinity level (mm)	Tap root	Shoot	Tap root	Shoot
1 (control)	5·25a	0·65a	2·33a	1·78a
5	4·15b	0·57a	1·75b	1·19b
10	4·33a,b	0·66a,b	1·84b	1·24b
20	4·39a,b	0·80b	1·35c	0·89c
40	3·07c	0·82b	0·82d	0·56d
80	3·37c	1·0c	0·60d	0·51d
LSD(P= 0·05)	1·04	0·15	0·27	0·27

Data are means (n = 4). Means followed by different superscripts are significantly different (P < 0·05).

Root ion concentration.

Unlike those in leaves, Cl^– concentrations in root tissues of plants irrigated with a 1 mm saline solution were low (18, 17 and 25 mm Cl^– for the periderm, phloem and xylem tissues, respectively; Fig. 3B). The Cl^– concentration of the periderm and phloem tissues increased as the external Cl^– concentration increased. However, the Cl^– concentration of the phloem tissue was consistently about 50 % less than that of the periderm. The Cl^– concentration of the xylem tissue remained low (25 mm) and was insensitive to external Cl^– over a salinity range of 1 to 80 mm (Fig. 3B).

The K⁺ concentration was high (130 mm) in the periderm at low salinity and decreased with increasing external salinity in all root tissues (Fig. 3B). Sodium concentrations in root tissues increased with external salinity and, unlike in the shoot tissues, concentrations of Na⁺ in the root tissues exceeded those of Cl^– by up to five‐fold. The total number of Na⁺ and K⁺ ions in the tap root was three‐ to five‐fold higher than that of Cl^– ions (Table 1) and, as the tap root comprised the bulk of total biomass, the [Na⁺ + K⁺] : Cl^– ratio of the whole plant was approximately equal to two and was constant over all the salinity treatments.

Experiment 3: Influence of salinity on carrot leaf gas exchange

Prior to the onset of treatments, stomatal conductance (g_s) and the rate of photosynthesis (P_n) of laminae were 255 mmol H₂O m^–2 s^–1 and 16 µmol CO₂ m^–2 s^–1, respectively. As in expt 2 (Fig. 3), leaf Cl^– concentrations at the start of the treatments, and in the 1 mm treatment thereafter, were comparatively high (100 mm), whereas leaf Na⁺ concentrations were low (30 mm).

Negative linear relationships were evident between g_s, P_n and leaf Cl^– (Fig. 4A) for data above an apparent threshold leaf Cl^– concentration of 150 mm. Although Na⁺ concentrations remained low throughout the experiment compared with Cl^– concentrations, negative linear relationships were also evident for g_s, P_n and leaf Na⁺ (Fig. 4B).

Fig. 4. Relationships between leaf Cl^– concentration and stomatal conductance (g_s) and photosynthesis (P_n) (A) and between leaf Na⁺ and stomatal conductance and photosynthesis (B) of carrot grown with nutrient solution containing salinities of 1 (circles), 8 (squares) or 80 mm (triangles) applied as NaCl and CaCl₂ with an Na⁺ : Ca²⁺ ratio of 15 : 1. Gas exchange was measured under ambient conditions, and ion concentrations in the terminal leaflets of 25‐ to 27‐d‐old leaves. Data are means (n = 6). Measurements were performed on one occasion prior to the onset of treatments (open symbols) and then on five subsequent occasions 7 d apart (closed symbols). Equations are linear regressions for observations with leaf Cl^– concentrations above 150 mm with coefficients of determination.

For leaves that developed during the salinity treatment (between days 28 and 35 of treatments), the 8 and 80 mm treatments reduced P_n by 13 and 38 %, respectively (Table 2) compared with that of leaves of plants grown at 1 mm. The influence of salinity on g_s was greater, with an 18 and 53 % reduction at the 8 and 80 mm salinity treatments, respectively. This was reflected in a 1·12‐ and a 1·30‐fold increase in the P_n : g_s ratio of leaves of plants subjected to 8 and 80 mm salinity treatments compared with that of the 1 mm controls (data not shown).

Table 2.

Photosynthesis (P_n), stomatal conductance (g_s), the ratio of internal to external CO₂ concentration calculated from leaf gas exchange (p_i/p_a*), carbon isotope discrimination (Δ), and the p_i/p_a^† ratio calculated [eqn (1)] from Δ measurements for 90‐d‐old carrot plants grown with nutrient solution containing salinity levels of 1, 8 or 80 mm

Salinity level (mm)	Pn (µmol CO2 m–2 s–1)	gs (mmol H2O m–2 s–1)	pi/pa*	Δ	pi/pa†
1 (control)	15·3a	259a	0·77a	23·3a	0·74a
8	13·4b	212b	0·75ab	22·5ab	0·71b
80	9·6c	121c	0·72b	21·9b	0·68c
LSD(P= 0·05)	1·8	39	0·03	0·9	0·01

Gas exchange measurements were performed on 25‐ to 27‐d‐old leaves formed during the salinity treatments and were conducted at a leaf temperature of 20 °C, 36·5 Pa CO₂, 40 % RH and a PAR of 1200 µmol m^–2 s^–1. Data are means (n = 6). Means followed by different superscripts are significantly different (P < 0·05).

The salinity‐mediated reduction of P_n was associated (curvilinear relationship) with the decrease in g_s (Fig. 5A), and the ratio of the internal to the external carbon dioxide concentration (p_i/p_a) calculated from leaf gas exchange measurements declined slightly from 0·77 for leaves of plants grown at 1 mm to 0·72 for those grown at 80 mm (Table 2). In addition, carbon isotope discrimination of leaf material that developed during the treatment period decreased from 23·3 ‰ for plants grown at a salinity of 1 mm to 22·5 and 21·0 ‰ for plants grown at 8 and 80 mm, respectively. Based on eqn (1), this corresponds to a reduction in p_i/p_a from 0·74 for plants grown at a salinity of 1 mm to 0·71 and 0·68 for plants grown at salinity levels of 8 and 80 mm, respectively, which is similar in magnitude to the decline in p_i/p_a calculated from leaf gas exchange measurements (Table 2). While the reduction in p_i/p_a is significant and, in the first instance, is indicative of a stomatal limitation to P_n, the magnitude of the reduction in p_iis small by comparison with the reduction in P_n (Fig. 5B).

Fig. 5. Relationships between laminae photosynthesis and stomatal conductance (A), photosynthesis and the internal concentration of CO₂ (p_i) at ambient CO₂ (36·5 Pa) (B), and photosynthesis and the internal concentration of CO₂ (p_i) under variable external CO₂ concentrations (C) for carrot grown with nutrient solution containing salinities of 1 (circles), 8 (squares) or 80 mm (triangles) applied as NaCl and CaCl₂ with an Na⁺ : Ca²⁺ ratio of 15 : 1. Data for A and B are values of individual plants, curves for C are plots of mean P_n and p_i values across a range of p_a levels. Calculated parameters with l s.d. values are given in Table 3. All measurements were performed on the terminal leaflets of 25‐ to 27‐d‐old‐leaves that developed during the salinity treatment.

The response of P_n to p_i over a p_a range from 5 to 150 Pa CO₂ (Fig. 5C) was investigated to determine the effect of salinity on parameters influencing photosynthetic capacity. For plants grown at a salinity level of 1 mm, P_n at CO₂‐saturating concentrations (P_nmax) was equal to 25·4 µmol CO₂ m^–2 s^–1 and declined by 15 and 30 % at salinity levels of 8 and 80 mm (Table 3). For plants grown at a salinity level of 1 mm, the RUBP‐carboxylase limited capacity for CO₂ fixation (V_cmax) was equal to 68·2 µmol CO₂ m^–2 s^–1 and it declined by 22 and 40 % to 53·0 and 41·3 µmol CO₂ m^–2 s^–1 at salinity levels of 8 and 80 mm, respectively. Salinity had a similar effect on J_max, the calculated maximum capacity for photosynthetic electron transport associated with RuBP regeneration, which declined from 129 µmol e^– m^–2 s^–1 at a salinity level of 1 mm to 100 and 79·7 µmol e^– m^–2 s^–1 at salinity levels of 8 and 80 mm, respectively (Table 3).

Table 3.

Rates of CO₂‐saturated photosynthesis (P_nmax), calculated maximum capacities for RuBP‐carboxylation (V_cmax) and photosynthetic electron transport associated with RuBP regeneration (J_max) and tissue‐water Cl^– and Na⁺ concentrations

Salinity level (mm)	Cl– (mm)	Na+ (mm)	Pnmax (µmol CO2 m–2 s–1)	Vcmax (µmol CO2 m–2 s–1)	Jmax (µmol e– m–2 s–1)
1 (control)	112·9a	34·3a	25·4a	68·2a	128·8a
8	144·8b	82·7b	21·6b	53·0b	100·1b
80	239·6c	111·2c	17·8c	41·3c	79·7c
LSD(P= 0·05)	31·0	16·1	3·4	10·8	21·0

Data are means (n = 6). Means followed by different superscripts are significantly different (P < 0·05).

DISCUSSION

Influence of salinity on yield

Bernstein and Ayers (1953) described carrot as being salt sensitive. Later, Maas and Hoffman (1977) also defined carrot as a salt‐sensitive species, based on four ratings of tolerance, namely sensitive, moderately sensitive, moderately tolerant and tolerant. Maas and Hoffman (1977) suggested that the relationship between salinity and yield might be expressed as a negative linear response function at salinities above a critical threshold, and set the threshold for carrot at 1 mmho cm^–1 (approx. 10 mm NaCl). By using a number of salinity treatments ranging from 1 to 40 mm NaCl, a threshold for yield between 20 and 40 mm NaCl was identified which is significantly higher than that of Maas and Hoffman (1977).

Maas and Hoffman (1977) set the yield decrease for carrot at 14 % per unit increase in salinity (1 mmho cm^–1 or approx. 10 mm NaCl), which implies a 50 % reduction in yield at 45 mm NaCl. If it is assumed that the slope of the yield decrease is linear and that the threshold is 20 mm, then in this experiment the slopes were 7 % between both 20 and 40, and 40 and 80 mm NaCl for every 10 mm increase in Cl^– (data from Fig. 2). A 50 % decrease in yield would occur at about 90 mm NaCl. The yield decrease in this experiment is substantially less than that reported by Maas and Hoffman (1977), but is similar to that in other reports. Mangal and Hooda (1989) reported a 50 % reduction in yield at a salinity level of 7·5 dS m^–1 (78 mm NaCl), and Matsubara and Tasaka (1988) found a 50 % reduction in fresh weight at a salinity level of 4000–6000 mg l^–1 (68–102 mm NaCl). Some of the difference may be explained by the facts that (1) there is potentially large variation in ion uptake among carrot genotypes (Bernstein and Ayers, 1953), and (2) that the earlier studies by Maas and Hoffman (1977) to compare the tolerance of carrot with that of other species often used salt treatments composed of equivalent concentrations of Na⁺ and Ca²⁺ as Cl^– salts. This results in Cl^– concentrations 1·5‐fold higher than a solution of NaCl. Thus, the classification of carrot as salt sensitive may require reassessment, the classification bands may need to be broadened and the tolerance of carrot should be described as in the upper range of the sensitive class.

Ion uptake and partitioning

Using carrot root discs, Cram (1983) demonstrated that Cl^– influx is under tight homeostatic control. Our data would suggest that Cl^– homeostasis also occurs at the whole‐plant level; remarkably high Cl^– concentrations were found in leaf tissue at low and high salinities, but constant low levels of Cl^– were found in root xylem tissue (Fig. 3). For example, an 80‐fold increase in external salinity resulted in only a 1·5‐fold increase in leaf Cl^– concentration and no increase in the Cl^– concentration of the root xylem tissue. Under both field and glasshouse conditions, carrot shoots accumulated high Cl^– concentrations when the irrigation solution contained only low Cl^– concentrations (Figs 1 and 3). High Cl^– concentrations in carrot leaves growing in low‐salt root media have been reported previously (e.g. Bernstein and Ayers, 1953). Chloride accumulation in shoots at low levels of salinity is up to five‐fold greater than that of Na⁺. Thus, carrot roots at low salinity may have a high affinity for Cl^– uptake, a high rate of root‐to‐shoot transport of Cl^– and, possibly, leaf regulation of root Cl^– uptake. Sodium and Cl^– levels in the irrigation water at Guilderton were 5·8 and 7·5 mm, respectively, which suggest that the level of salinity per se would not have reduced crop yields. However, this salinity resulted in high (160 mm) leaf Cl^– concentrations which, on the basis of the second glasshouse experiment, are approaching levels that were associated with a small reduction in leaf gas exchange (Fig. 4).

Influence of salinity on leaf gas exchange

Photosynthesis of control leaves ranged from 15 to 17 µmol CO₂ m^–2 s^–1 (Fig. 4), values similar to those published for carrot grown in the field under well‐watered conditions (Gibberd et al., 2000). The accumulation of bulk‐leaf Cl^– concentrations of up to about 150 mm by plants grown in the 1 mm treatment (and in the 8 mm treatment in the early part of the treatment period) did not influence leaf gas exchange (Fig. 4). Similar leaf Cl^– concentrations have been associated with a 50 % reduction in P_n and g_s of Phaseolus vulgaris (Seemann and Critchley, 1985) and up to a 70 % reduction in P_n of grapevines (Downton, 1977; Walker et al., 1981). The insensitivity of carrot P_n to leaf Cl^– concentrations below 150 mm may be due to effective intra‐ or intercellular partitioning of Cl^–, thereby minimizing the Cl^– concentration in the cytosol of leaf mesophyll cells, as observed for barley (Huang and Steveninck, 1989) and olive (Bongi and Loreto, 1989). This requires further investigation.

Irrigation with nutrient solution containing salinity levels of 8 and 80 mm resulted in large increases in leaf Cl^– and Na⁺ concentrations, and reductions in leaf gas exchange. By spacing gas exchange and tissue ion measurements at weekly intervals it was possible to demonstrate negative linear relationships between P_n (and g_s) and leaf Cl^– concentration above about 150 mm (Fig. 4). The increase in leaf Na⁺ concentration of plants exposed to 8 and 80 mm treatments coincided with the increase in leaf Cl^– concentration, and there was also a negative linear relationship between leaf Na⁺ concentration and P_n and g_s, although Na⁺ concentrations were only 55 to 70 % of those of Cl^–.

Reductions in P_n of plants grown under saline conditions can be associated with the limitation of CO₂ supply through increased stomatal (Brugnoli and Lauteri, 1991; Ouerghi et al., 2000) or mesophyll resistance (Delfine et al., 1998), and/or through direct effects on photosynthetic capacity through reductions in the concentration or efficiency of photosynthetic enzymes (Seemann and Sharkey, 1986; Brugnoli and Bjorkman, 1992). Stomatal limitation of P_n is usually associated with a decline in p_i (Farquhar and Sharkey, 1982). For carrot, the salinity‐induced reduction in P_n was correlated with g_s (Fig. 5A), but the reduction in p_i calculated from leaf gas exchange or from Δ was small (approx. 2 Pa) (Table 2) and, for measurements performed at ambient p_a, variation in P_n was largely independent of variation in p_i (Fig. 5B). Hence, it is likely that stomatal limitation does play a small role in the salinity‐induced reduction of photosynthesis of carrot. However, it appears that there is little, if any, increase in mesophyll resistance in the laminae of carrot grown with saline irrigation. In other species in which salinity results in an increase in mesophyll resistance, the influence of salinity on Δ has been shown to be large and disproportionate to the modelled change in Δ calculated from p_i values [eqn (1)] derived from gas exchange measurements. For example, Brugnoli et al. (1991) and Seemann and Critchley (1985) observed a 2·7 ‰ and 5 ‰ difference in Δ between Phaseolus vulgaris grown at 50 and 150 mm NaCl, respectively, compared with plants grown under control conditions. Likewise, Gibberd et al. (2001) observed a 4·7 ‰ decrease in Δ for grapevine (Vitis vinifera L. ‘Sultana’) grown at 80 mm NaCl compared with plants grown in non‐salinized conditions. The decrease in Δ observed in our experiment was comparatively small with only a 1·4 ‰ difference between the 1 mm and 80 mm treatments (Table 2), and the salinity‐induced change in p_i calculated from Δ was very similar to the change in p_i determined from leaf gas exchange measurements. Thus, while stomatal limitation is likely to be an important component of the salinity‐induced reduction in photosynthesis for carrot, its role appears to be small compared with the reduction in photosynthetic capacity. This is supported by the 30 % reduction in P_nmax for plants grown at a salinity level of 80 mm.

The nitrogen content of carrot leaves was not influenced by salinity (data not shown). However, the CO₂‐saturated rate of photosynthesis (P_nmax), the apparent efficiency of RuBP‐carboxylase (V_cmax), and the rate of photosynthetic electron transport limiting RuBP regeneration (J_max) of plants grown at 80 mm salinity were reduced by approx. 30 to 40 % compared with values for plants grown at 1 mm (Table 3), and were correlated with leaf Cl^– and Na⁺ content (data not shown). Similar reductions in photosynthetic capacity have been observed for other species such as Phaseolus vulgaris (Seemann and Critchley, 1985), spinach (Downton et al., 1985) and olive (Bongi and Loreto, 1989) when grown under saline conditions. It is difficult to determine the extent to which the reduction in photosynthetic capacity of carrot may be attributed to a direct toxic effect of Cl^– and/or Na⁺ on the biochemistry associated with photosynthesis or to other possibilities such as stomatal patchiness leading to the underestimation of P_nmax, V_cmax and J_max as observed for other species (for example, Brugnoli and Lauteri, 1991). The p_i values estimated from Δ measurements on dry matter formed during the experimental period were approx. 1·5 Pa lower than those calculated from leaf gas exchange measurements (Table 2). The overestimation of p_i by leaf gas exchange measurements is small and may reflect inaccuracies in the values assumed for δ¹³C of air (7·6 ‰) and of b (30 ‰). Small errors in either of these values would be sufficient to account for the differences between calculated and measured values of p_i.

CONCLUSIONS

Moderately saline irrigation water (up to 20 mm) did not have a negative influence on carrot growth. However, at salinities above 20 mm, growth was reduced by 7 % for each 10 mmincrement in salinity. At low salinities (e.g. 1 mm), high levels of Cl^– were accumulated by shoots, and shoot‐Cl^– concentrations exceeded shoot‐Na⁺ concentrations by up to five‐fold. However, Cl^– uptake did not increase greatly at high salinity levels (e.g. 80 mm). At low external salinities, leaf gas exchange was not adversely influenced if leaf Cl^– concentrations were less than about 150 mm. However, at higher salinity levels leaf gas exchange was reduced and the reduction in P_n was partly mediated by stomatal limitation of CO₂ supply for photosynthesis and by a reduction in apparent photosynthetic capacity.

ACKNOWLEDGEMENTS

Rana Munns, Richard James and Susanne von Caemmerer are thanked for valuable discussions. Ayalsew Zerihun and Tony Condon are thanked for their comments on the manuscript. The financial assistance of the Food Into Asia initiative is gratefully acknowledged, and the authors thank the Sumich Group Pty Ltd for access to the field site used in expt 1.

Received: 3 July 2002; Returned for revision: 18 July 2002; Accepted: 31 August 2002    Published electronically: 31 October 2002