Introduction

Potato (Solanum tuberosum L.) is grown and eaten in more countries than any other crop (Jackson 1999). The crop is mainly grown in climates with cool temperate with full sunlight, moderate day temperatures and cool nights. Short days generally induce tuber formation in potato, although many modern cultivars can initiate tubers under the long days of northern temperate regions as well (Tarn et al. 1992). Potato is among the most important crops in the world (Fernie and Willmitzer 2001) and in Iran (FAO 2008). In Iran, potato is ranked fourth in annual production after the cereals rice, wheat and barley. Iran is the world’s 12th potato producer and the third biggest producer in Asia, after China and India (FAO 2008).

Water stress is a common stress in potato production areas. Water stress leads to yield and tuber quality losses. Potato is susceptible to drought (Foti et al. 1995). Supplying sufficient water is very important to increase potato quality and quantity. It is also very important to study the tolerance of different potato cultivars towards water deficit stress and to assess how much water is consumed by the potato crop.

In the potato growing area of Ardabil, Iran, water deficit is a serious problem. Moreover, the climate in this region has changed recently, resulting in changes in the distribution of precipation over time and changes in river flows and well water availability. Therefore, it is necessary to assess the agronomic characteristics of new potato cultivars in relation to their response to the water availability conditions in Ardabil and to analyse how much water these new cultivars require to obtain a good yield of high quality.

In comparison with other species, potato is very sensitive to water stress because of its shallow root system (Iwama and Yamaguchi 2006). Water deficit strongly decreases the number of leaves, plant water potential (Frensch 1997), leaf area, stem height, ground coverage, canopy radiation interception, number of tubers, growth and yield. In contrast, effects of water stress on radiation use efficiency, harvest index and tuber dry matter concentration (Schittenhelma et al. 2005) and on nitrate reductase (Foyer et al. 1998; Chandra et al. 2004; Das et al. 2005; Kar et al. 2005; Xu and Guang 2006) are relatively small. Water deficit also increases reducing sugar concentration in the tuber, tuber cracking and malformation, surface abrasions, hollow heart, brown centre, internal brown spot, vascular discolouration or bruising, starch degradation in the tuber stem end and total glycoalkaloid concentration (Papathanasiou et al. 1999). Reflectance indices were used to measure biomass and drought stress, changes in leaf water content (Francois and De Proft 2005) and water stress (Bahrun et al. 2003). A set of drought tolerance genes previously found to be up-regulated in tolerant potato under drought (Schafleitner et al. 2007) was assayed for expression changes in potato under drought.

The objective of this study was to evaluate different potato cultivars for their response to water deficit stress under in vivo conditions in order to release cultivars suitable for regions with water deficit in Iran.

Materials and Methods

Experiments were laid out in a split-plot design in three replications. Factor A included three irrigation regimes (normal, mild stress and severe stress, i.e. irrigation after 25%, 35% and 50% discharge of the available water, respectively) and factor B included seven potato cultivars (Agria, Savalan, Satina, Caesar, Kennebec, Marfona and Santé). Experiments were carried out in the growing seasons of 2005 and 2006. Every plot consisted of five rows of 6 m length with a row distance of 0.75 m. Planting arrangement was 75 × 25 cm. Distances between plots were 1.5 m.

The water amount used was regularly calculated according to the collected evaporation of a Class A Basin using the equation:

$$ {\hbox{IW}}/{\hbox{CPE}} = 0.{8} $$

where IW = the amount of irrigation water (mm) and CPE = the collected evaporation calculated from evaporation pan (mm).

The amount of irrigation supplied was measured by a water metre. The start of irrigation was on the basis of 30 mm evaporation from a Class A evaporation pan. Amount of precipitation was measured by an udometer and daily evaporation by a Class A evaporation pan.

During crop growth and at harvesting, many characteristics were measured, including the number of main stems, plant height, number of tubers and tuber weight per plant, total and marketable tuber yield, and dry matter concentration. Analysis of variance was carried out, and means were separated using Duncan’s range test.

Water use efficiency (WUE; kg m−3) was calculated for potato cultivars under different irrigation regimes, as follows:

$$ {\hbox{WUE}} = \frac{\text{TY}}{{\hbox{TWU}}} $$

where, TY = tuber yield (in kg ha−1) and TWU = total water used (in m3 ha−1). Indices used for evaluating potato cultivars included Fischer and Maurer stress index (SSI), Fernandez tolerance index (STI), Rosielle and Hamblin tolerance index (TOL), Baron geometric index (GMP) and modified stress tolerance index (MSTI) as indicated below (Fischer and Maurer 1978; Fernandez 1992; Rosielle and Hamblin 1981; Naderi et al. 1999):

  1. 1.

    Stress susceptility index (SSI):

\( {\hbox{SSI}} = \left( {1 - \frac{{{Y_{\rm{Si}}}}}{{{Y_{\rm{Pi}}}}}} \right)/{\hbox{SI}} \) where Y Si and Y Pi = yield of cultivar under stress and normal conditions, respectively and \( {\hbox{SI}} = 1 - \frac{{{Y_{\rm{S}}}}}{{{Y_{\rm{p}}}}} \) where Y S and Y P = total yield mean under stress and normal condition, respectively

  1. 2.

    Stress tolerance index (STI)

\( {\hbox{STI}} = \frac{{{Y_{\rm{pi}}} \times {Y_{\rm{Si}}}}}{{{{({{\overline Y }_{\rm{p}}})}^2}}} \)

  1. 3.

    Tolerance index (TOL)

\( {\hbox{TOL}} = {Y_{\rm{Pi}}}-{ }{Y_{\rm{Si}}} \)

  1. 4.

    Geometric mean index (GMP)

\( {\hbox{GMP}} = \sqrt {{{Y_{\rm{Pi}}} \times {Y_{\rm{Si}}}}} \)

  1. 5.

    Mean productivity (MP)

\( {\hbox{MP}} = \frac{{{Y_{\rm{Pi}}} + {Y_{\rm{Si}}}}}{2} \)

  1. 6.

    Modified stress tolerance index (MSTI):

\( {\hbox{MSTI}} = K \times \left( {\frac{{{Y_{\rm{pi}}} \times {Y_{\rm{Si}}}}}{{{{({{\overline Y }_{\rm{p}}})}^2}}}} \right) \)where \( K = Y^{2}_{{{\text{Si}}}} /Y^{2}_{{\text{S}}} \)

Analysis of variance and the mean separation using Duncan’s range test was carried out using MSTATC software.

Results and Discussion

Analysis of variance showed significant effects of years for tuber yield, plant height, tuber number and weight per plant, tuber weight less than 35 mm, between 35 and 55 mm and bigger than 55 mm; irrigation treatments had significant effects on total and marketable tuber yield, tuber number and weight per plant, tuber number and weight less than 35 mm and bigger than 55 mm; cultivar effects were significant for total and marketable tuber yield, main stem number, plant height, tuber number and weight per plant, tuber number and weight less than 35 mm, between 35 and 55 mm and bigger than 55 mm; the interactions between irrigation regime and cultivar were significant for total and marketable tuber yield and tuber weight per plant (Tables 1 and 2).

Table 1 Analysis of variance for different tuber traits of potato cultivars grown under different irrigation regimes
Full size table
Table 2 Analysis of variance for different tuber size distribution traits of potato cultivars grown under different irrigation regimes
Full size table

The highest total and marketable tuber yield, tuber number and weight per plant were obtained for cultivars Savalan, Caesar and Kennebec in all irrigation regimes (Table 3). Mean comparisons of attributes among cultivars showed that tuber yield of cultivars Savalan, Caesar and Kennebec was higher than yields of other cultivars under stress and non-stress conditions. The yield loss caused by water stress differed among cultivars. In both mild and severe stress conditions, cultivars Savalan, Caesar and Kennebec showed a relatively small decline in tuber yield and could be classified as cultivars tolerant to water stress. These cultivars can produce a relatively high tuber weight per plant in all conditions. Yield decrease in mild and severe water stress conditions compared with normal conditions was 5.6 and 10.9 Mg ha−1 in cultivar Savalan, 3.7 and 5.9 Mg ha−1 in cultivar Caesar and 1.6 and 6.2 Mg ha−1 in cultivar Kennebec, respectively (Table 3). We observed that the most susceptible traits were tuber weight and number in the size classes 35–55 mm and bigger than 55 mm. It can be concluded that water stress decreased total and marketable tuber yield. Marfona had the largest yield loss, both under mild and severe water stress.

Table 3 Attributes mean in potato different under irrigation regimes
Full size table

There were significant and positive correlations between total tuber yield and marketable tuber yield, tuber weight per plant, tuber number and weight between 35 and 55 mm and bigger than 55 mm; between marketable tuber yield and tuber weight between 35 and 55 mm and bigger than 55 mm, and tuber number bigger than 55 mm; between tuber number per plant and plant height, main stem number, tuber number and weight less than 35 mm and between 35 and 55 mm; between tuber weight per plant and total and marketable tuber yield, tuber number and weight bigger than 55 mm; between main stem number and tuber number per plant, plant height, tuber number and weight less than 35 mm and between 35 and 55 mm; between plant height and tuber number per plant, main stem number, tuber number and weight less than 35 mm, tuber number between 35 and 55 mm and tuber weight bigger than 55 mm (Table 4).

Table 4 Linear correlation coefficients between different traits in potato cultivars grown at different irrigation regimes
Full size table

Percent water stress intensity was calculated to be 10% (SI = 0.10) under mild stress conditions and 18% (SI = 0.18) under severe stress conditions. This showed that tuber yield under mild and severe stress decreased considerably. Percent tuber yield loss under the conditions of this experiment would be 10% and 18% (Table 5). MP, GMP, STI and MSTI indices indicated that Savalan, Caesar and Kennebec were the better cultivars; the SSI indicated that cultivars Marfona, Kennebec and Satina were best whereas TOL indicated that cultivar Satina was the best cultivar under mild and severe water stress conditions. Phenological investigations of cultivars Savalan and Caesar during susceptible periods can lead to designing strategic approaches in agronomy and breeding. Results for GMP, STI and MSTI were very consistent; Savalan, Caesar and Kennebec had the highest yields compared with other cultivars under mild and severe water stress and under normal conditions. So, these genotypes are tolerant to water stress and suitable for the Ardabil region (Table 5). Analysis of tolerance and sensitivity indices for environmental stress conditions showed that efficiency of these indices depends on the variation between genotypes in yield and on the assessment aimed for. The Fischer and Maurer index classifies genotypes as tolerant or sensitive, regardless of their yield level and therefore is efficient in finding genotypes with resistance genes. TOL’s efficiency is conditional, but after classifying genotypes based on similar TOL values, we can select resistant genotypes with MP. Finding equal TOL in different groups is very hard. With regard to the roles of TOL and MP, genotypes with high MP might not be present in the groups with the least TOL and selecting superior genotypes might be difficult. The Fernandez index (STI) uses yield under stress and non-stress conditions and geometric means but the geometric equation is problematic as it combines data that have different natural background.

Table 5 Estimates of water stress indices for potato cultivars under mild and severe stress
Full size table

In all provinces of Iran, environmental changes are visible. The MSTI index includes the calculation of the KSTI for suitable and unsuitable conditions and is useful for selecting superior genotypes for each region. MSTI results are very notable.

Cultivars Savalan, Caesar and Kennebec had the highest WUE under normal conditions with values of 7.80, 7.17 and 6.96 kg m−3, respectively; under mild stress conditions, values were 7.92, 7.54 and 7.68 kg m−3, and under severe stress conditions values, were 8.88, 9.12 and 8.74 kg m−3, respectively (Table 6).

Table 6 Average water use efficiency (in kilograms per cubic metre) in potato cultivars under mild and severe stress
Full size table

This study shows that water stress increased WUE. Haverkort et al. (1990) also showed that drought increased WUE. Bodlaender (1986) showed that there was no relation between water usage and produced dry matter but that WUE had a significant negative relation with drought resistance. Slight stomata closure decreased transpiration more than photosynthesis and WUE increased. Severe drought led to full closure of stomata and decreased WUE and then yield (Beukema and Van Der Zaag 1990). Wright and Stark (1990) reported that the WUEs for conditions favouring maximum yields range from approximately 50 to 100 kg ha−1 mm−1. Kiziloglu et al. (2006) and Nagaz et al. (2007) reported that the range of WUE was from 44.1 to 63.4 kg ha−1 mm−1 and from 8 to 14 kg m−3, respectively. Nagaz et al. (2007) concluded that WUE varied around 8–9, 6–8 and 11–14 kg m−3 for autumn-, winter- and spring-planted potato, respectively. Nasseri and Baharamloo (2009) showed that the highest increase in WUE was 21.2% relative to control conditions. Based on WUE values, they recommended that potato cultivar Marfona should be irrigated at 59 days after planting to achieve the optimum WUE. Yarnia et al. (2009) indicated that increasing stress intensity decreased WUE in all cultivars. But severe stress had a higher WUE than mild stress. In all stages, increasing stress to a mild level decreased WUE, but beyond that, WUE increased. In all cultivars, irrigation at 50% soil available water increased WUE, and this was associated with a larger decline in tuber yield. Agria cultivar showed the highest WUE.

Conclusion

The cultivars Savalan, Caesar and Kennebec had a higher total and marketable tuber yield, water use efficiency, and more favourable stress indices than the other cultivars, both under mild and severe stress.

Cultivars Caesar and Kennebec were selected as tolerant cultivars; cultivars Savalan and Satina were identified as moderately tolerant cultivars whereas cultivars Agria, Marfona and Santé cultivars were classified as susceptible to water deficit.