GROWTH AND YIELD CHARACTERS OF POTATO GENOTYPES GROWN IN DROUGHT AND IRRIGATED CONDITIONS OF NEPAL

This experiment was conducted to assess the genotypic variation for the growth, yield and yield components of potato grown under drought and irrigated conditions at Hattiban Research Farm, Khumaltar during the summer cropping season (Feb.-May) of the two consecutive years (2013 and 2014). The results revealed that canopy cover, stem height and number of leavers were more sensitive to drought and the effect was more pronounced in early cultivar Desiree. Stem height and leaf number of plant were more (36% and 45%, respectively) reduced in Desiree than other genotypes. Mean tuber number decreased by 55% under drought condition. Drought stress reduced marketable tuber number by 79% as compared to irrigated treatment. Drought stress reduced the marketable tuber yield from 70% to 87%. The clones CIP 392242.25 and LBr-40 had relatively lower yield loss, and less drought susceptibility index under drought indicating their tolerance to drought in field condition. Further experiment is recommended to study the physiological parameters of these genotypes under different water and soil conditions.


Introduction
Potato is an important non-cereal food crop of Nepal. In mid hills, it is mainly used as vegetables and grown in two times of the year i.e. summer and august, respectively. Summer is considered to be normal season for potato production in the hills. In mid-hills, the normal season potato faces the drought during tuberization period (NPRP, 2012). The planting of potato soon after harvesting the rice in ricepotato cropping system utilizes the residual moisture for tuber emergence. However, the drought condition exists thereafter during the tuberization and bulking period (NPRP, 2012). Drought stress occurs even in irrigated potato (Van Loon, 1981;Jefferies, 1993).
In the sub-tropical region, where potato production is only possible through irrigation but short periods of drought often arise because of inadequate irrigation techniques or shortages of water. Water stress may occur even in good irrigated condition because of high transpiration rate particularly during mid-day when root system cannot completely meet the water requirements of the plant (Minhas and Sukumaran, 1988). Drought stress affects the development and the growth of shoots, root and tubers (Lahlou et al., 2003). Reduced leaf growth and accelerated leaf senescence are common responses to water deficits condition (Monneveux, 1991). The effect of drought on tuber number has been studied by several researchers (MacKerron and Jefferies, 1986;Haverkort et al., 1991, Deblonde andLedent, 2001;Lahlou et al., 2003). Potato crop may suffer significant yield reduction even under short periods of water depletion because of a relatively shallow root system and low capacity of recuperation after water stress (Iwama, 2008).
The drought effect on potato foliar characters affects tuber yield which had been studied by several researchers. Jefferies (1993) showed that the final size of individual leaves was reduced by drought but the magnitude of the effect differed significantly between cultivars. Drought reduced the number of green leaves (Deblonde and Ledent, 2001). Screening drought tolerance genotypes for breeding purposes need reliable parameters to assess the drought tolerance traits. In addition, the potato plant can follow different growth strategies to adapt drought condition. The potato genotypes developed so far may differ on their morphological characters, precocity and their variation for drought tolerance. The main objective of this study was to investigate the genotypic differences for their agro-Research Article B.P. Luitel et al. (2015) Five CIP clones (391598.75, 392242.25, 391011.47, 378711.7, and LBr-40), one German variety NPI-106, and a standard check represented by the Dutch cultivar Desiree, a very popular released variety of Nepal were used in the experiment. The experiment was conducted at Hattiban Research Farm, Khumaltar (27°40' N, 85° 20'E, 1340 m asl) during 2013-2014. The type of soil was sandy loam with medium organic matter content and soil pH was 5.5. The average, minimum and maximum temperature and rainfall were collected from Agrometeorological Station of Agronomy Research Division (ARD), Khumaltar within the distance of 1 km from the experimental field. Total rainfall during the cropping period was 314.7 mm and 199.4 mm in 2013 and 2014, respectively. Maximal temperature was higher in 2014 than 2013 (Table 1).

Materials and Methods
The uniform sizes of well-sprouted tubers ranging from 25 to 50 g size from each variety were chosen for the trial. Before planting, the plot was well-leveled, rows were made and NP2O5K2O @ 100:100:60 kgha -1 was placed, and the compost was mixed homogenously in the row. Then, sprouted tubers were placed manually in rows in upward direction. Tubers were planted at the spacing of 60 x 25 cm in two rows maintaining the plot sized 3 m 2 . A split plot design with two factors (water status and cultivar) and three replicates were used. The drought treatments (T1) in main plot and potato genotypes were placed in subplot. In drought plot, irrigation was applied once for emergence just after planting; then, drought treatment was applied after the emergence of tuber by withholding irrigation totally. Additionally, the water from rainfall was excluded by covering a strong plastic sheet just above the drought plot. The irrigated (T2) plot was received water from rainfall and irrigation and it was applied by furrow system. Soil moisture content was determined using TRIME-PICo64/32 device at 10 cm depth in upper soil layer. When volumetric soil moisture content (%) dropped below 8%, irrigation was given to create ideal moisture condition (8-16%) but in drought plot, soil moisture content was maintained below 8% after tuber emergence to maintain dry or drought condition according to soil moisture scale mentioned by Zotarelli et al. (2010). In the both years, the trial was planted in February 1 and 10 and harvested in May 20 and 28, respectively.

Measurements
Canopy cover was estimated at 70 days after tuber planting. The proportion of the ground covered with green leaves was estimated using a 60 cm x 50 cm grid divided into 6 cm x 5 cm cells held over the central row (covering two plants) and percent cover was calculated as the number of cells at least half-filled by vegetation divided by the total number of cells. Stem height (cm) was measured from soil surface to the apex of the main stem. The number of green leaves was counted on five plants at 80 days after planting and it was averaged. The number of aboveground stems (AGS) was counted on five plants at the maximum growth of the plants when the plants touch each other's between the rows (NPRP, 2014). Leaf length was measured from 14 th to 19 th leaf on the main axis according to Deblonde and Ledent (2000). The internode length was measured between two internodes on the main axis. The number of harvested plants was counted before harvesting and all plants were harvested to assess the total tuber number plant -1 , marketable tuber plant -1 (no.), marketable weight plant -1 (g), and marketable yield (tha -1 ). Total tuber plant -1 was determined using the count of total number of tuber produced divided by the number of plants harvested in a specific plot. The tubers were graded into two categories; >25 g and < 25 g. Tubers <25 g category, and diseased, insect damaged, cracked, green, second growth, and early sprouted tubers were considered as non-marketable, and tubers of > 25 g with well-shaped tubers were considered as marketable tuber (NPRP, 2014). Drought susceptibility index was calculated as (1-yielddrought/yieldirrigated) / (1-meandroughted/meanirrigated) (Schafleitner et al., 2007).

Statistical analysis
The data were analyzed using statistical software GenStat Discovery Edition 4 for Windows. All the measured data derived from the experiment were combined across the years and analyzed. In ANOVA, genotype and water status were considered as fixed effects, while year, and replication were considered as random effects. The genotype, treatment, year and their interaction effects were analyzed by combined ANOVA and pooled data over the years are presented.

Growth parameters
Canopy cover Highly significant genotype and treatment effects were observed in canopy cover ( Table 2). The clone CIP 391011.47 had the highest (80%) canopy cover followed by NPI-106 (77%) and the lowest (49%) in Desiree. Mean canopy cover decreased 39.0% in drought treatments compared to irrigated treatment. The reduction of canopy cover in drought stress was also reported by other researchers (Ojala et al. 1990;Carli et al. 2014). The clone CIP 391011.47 had the highest canopy cover among the studied genotypes under drought and irrigated conditions. The genotype by treatment interaction was significant while year by treatment and year by genotype were not significant.

Stem height
Statistical analysis for stem height within years showed highly significant differences between genotype and treatment ( Table 2). The clone LBr-40 had the tallest (58 cm) height and the lowest (37 cm) in Desiree. The mean stem height reduced by 25.9% in drought than irrigated treatment. Desiree, an early maturing variety was mainly affected by drought stress on stem height. The stem height reduction was 36% in Desiree but in the study of Deblonde and Ledent (2001), they found 25% reduction in stem height due to drought. The shorter vegetative cycle of the early genotype might be the cause of larger reduction of stem height in early bulking variety like Desiree. But Deblonde and Ledent (2001) reported that stem height strongly reduced in late cultivars than early cultivars which contradicts this findings. The stem height difference in irrigated condition among the cultivars mainly is due to genetic differences. The drought stress had less affected on stem height in clone LBr-40, a medium-late maturing variety. Ojala et al. (1990) reported that long term drought stress affects the stem height. The genotype by treatment interaction was non-significant while year by treatment was significant. The year by genotype and year by genotype by treatment interaction were non-significant in stem height.

Number of green leaves
The number of green leaves present on the plant was highly significant in genotype and treatment ( Table 2). The highest number of green leaves was counted on NPI-106 (65) and the lowest in Desiree (43). The number of leaves decreased by 27.9% in drought as compared to irrigated treatment. The genotype by treatment interaction was non-significant while year by treatment and year by genotype were significant. In this study, the number of green leaves in Desiree reduced 45% in drought condition but in the study of Deblonde and Ledent (2001), they reported 25% reduction of green leaves number in Desiree. Fewer numbers of green leaves per plant in irrigated condition showed the lowest leaf number in drought condition in a medium-late maturing clone LBr-40.

Stem number
The genotype had highly significant effect on the main stem number (Table 3). The clone CIP 391011.47 had the highest stem number and LBr-40 had the lowest. The drought stress had highly significant differences on stem number. Drought stress reduced the total stem number by 5% than irrigated treatment. Lahlou et al. (2003) reported 28% reduction of total stem number caused by drought stress. Iritani (1968) reported that total stem number was influenced by drought stress. However, total stem number is affected by other factors like temperature (Haverkort and Harris, 1987) and growth regulators (Marinus and Bodlaander, 1978). Besides, the number of main stem produced plant -1 is a genotypic character (Nielson et al. 1989) and it is also affected by other factors such as length of the pre-sprouting period (Allen, 1978), size of seed tuber (Haverkort et al. 1991;Wurr, 1974) and physiological age (Iritani, 1968). Genotype by treatment interaction was not significant while year by treatment and year by genotype were significant.

Leaf length
Genotype and treatment had highly significant effect on leaf length ( Table 3). The highest length was measured in CIP 378711.7 (19.8 cm) followed by LBr-40 while drought stress reduced the leaf length by 15.5% than irrigated treatment. The clone LBr-40 reduced the leaf length by 8% in drought stress. The reduction of leaf length caused by drought stress was also reported by Deblonde and Ledent (2001) in potato genotypes. Genotype by treatment interaction was not significant while year by treatment and year by genotype were significant.

Internode length
The genotype had significant effect in internode length of the plants. The internode length was reduced in Desiree variety as compared to other genotypes. But the treatment had highly significant effect on in internode length and drought stress decreased the internode length by 14.6% than irrigated treatment (Table 3). The clone CIP 391011.47 had reduced internode length by 7% due to drought stress. Genotype by treatment, year and treatment, and year by genotype were not-significant. The genotype with greatest internode length in irrigation condition had highest reduction in internode length in drought condition.

Total tuber number
The statistical analysis of tuber number measured at final harvest is shown in Table 4. Genotype by treatment interaction for total tuber number was highly significant. NPI-106 had significantly highest and Desiree had lowest number of tubers plant -1 than other genotypes. The drought treatment reduced the total tuber number by 54.8% than irrigated treatment. Desiree was highly influenced by drought stress with regard to total tuber number and the clone LBr-40 had less affected by drought stress. This might due to the different tuberization behavior of the potato genotypes. Lahlou et al. (2003) also reported that tuber number had more affected in early variety. Except these genotypes, drought stress consistently reduced the total tuber number in other genotypes. The reduction of tuber number under drought stress on potato has been described by several authors (MacKerron and Jefferies, 1986;Haverkort et al. 1991;Lahlou et al. 2003;Carli et al. 2014). Genotype by treatment and year by treatment interaction were significant while year by genotype interaction was non-significant.

Marketable tuber number
Genotype had highly significant differences on marketable tuber plant -1 . Except CIP 392242.25, marketable tuber number plant -1 was statistically similar in all the genotypes (Table 4). Similarly, treatment effect was highly significant on marketable tuber number and drought reduced the mean marketable tuber number by 79% as compared to irrigated treatment. The reduction by drought on marketable tuber number was 63.6% in CIP 392242.25, 66.6% in LBr-40, 80% in CIP 378711.7, 82% in CIP 391598.75 and CIP 391011.47, and 83% in Desiree (data not shown) in comparison with their respective irrigated treatment. The reduction of marketable tuber number under drought stress was varied from 64% to 83% and most of tubers under drought plots were observed as malformed, reduced size with defected tubers (data not shown). Schafleitner et al. (2007) had observed the malformed tuber in water deficit condition. Genotype by treatment interaction was significant. Likewise, year by treatment interaction was highly significant but year by genotype interaction was not significant.

Marketable tuber weight
Genotype had highly significant effect on marketable tuber weight (  NS, * and ** indicate non-significant, significant (P≤0.05) and highly significant (P ≤ 0.01), respectively. T1 = Drought, T2 = Irrigated  Cavagnaro et al. (1971) had reported that drought stress decreased the tuber yield. In the study of Lahlou et al. (2003), they reported 11% yield reduction in Desiree, and 38 % in Remarka but in our study, 87% yield reduction was recorded in Desiree. Lahlou et al. (2003) had reported that drought stress reduced yield by 53% in Remarka genotype. The clone CIP 392242.25 had the lowest (0.52) drought susceptibility index followed by the clone CIP LBr-40 (0.60) indicating their drought tolerance and the highest (0.68) DSI for Desiree represented as drought susceptible genotype. But in the study of Schafleitner et al. (2007), they found the lowest (0.50) DSI in cv. Reiche and the highest (1.36) in cv. Achirana.

Conclusion
This study analyzed the agro-morphological characters and yield of potato genotypes under drought and irrigated condition. All the growth parameters including yield and yield components were sensitive to drought stress. The early maturing genotype Desiree was more affected than medium-late maturing genotypes by drought in terms of stem height, number of leaves plant -1 , total tuber number, marketable tuber number and weight, and yield. But drought stress consistently reduced the canopy cover, stem number, leaf length, internode length, yield components and final tuber yield but the magnitude of reduction varied according to the phenology of different genotypes. Drought reduced the marketable yield of all potato genotypes under the investigation in a genotype-dependent manner. The clones CIP 392242.25 and LBr-40 had showed lower yield loss and less drought susceptibility index under drought condition but further experiments are needed to evaluate the physiological response of these genotypes to different water, soil and climatic conditions for confirming this result.