Identification of Drought Tolerant Oilseed Rape Genotypes using Multivariate Analysis

 Drought stress as one of the most important abiotic stress is the main limiting factor of oilseed rape cultivation in arid and semi-arid climates. Therefore, the identification of drought tolerant genotypes is the essential programs in these regions. One of the appropriate methods to identify drought tolerant genotypes is the use of stress tolerance indices. To this end, the present study was designed to detect drought tolerant among 38 winter genotypes using drought tolerance indices.

 The experiment was based on randomized complete block design (RCBD) with three replications under both normal and drought conditions during 2016 to 2017 cropping season. In the well-watered experiment, irrigation was performed in five stages, while in drought treatment, irrigation was stopped before flower initiation. The yield of genotypes under normal and drought conditions using drought tolerant indices, including; TOL, MP, GMP, HM, STI, YSI, SSI, YR, K1STI, K2STI and RDI were investigated to identify drought tolerant and susceptible genotypes.

 Analysis of variance showed a significant difference among evaluated genotypes for all drought tolerance indices. The highest coefficient of genotype variability was observed for TOL and SSI, YR and K2ST1 indices. Correlation analysis showed a significant positive correlation between Yp and TOL, SSI and YR, while a negative correlation was observed for Ys. The principal component analysis showed that the two first components covered 99.64% of all data variations. 65.36 and 34.28% of the variation were covered by the first and second components, respectively.  The first component detected tolerant genotypes, but the second component identified drought sensitive genotypes. Cluster analysis, while confirming the results of principal component analysis, classified genotypes into three categories. Cluster analysis confirmed the PCA results, which separated drought tolerant genotypes (Class 3) from sensitive genotypes (Class 1).

 Finally, this research detected five superior genotypes, including; G6, G10, G18, G23 and G33, which were the most drought tolerant genotypes and their yields were not significantly changed under normal and drought stress conditions.