Analysis of variance for normal condition yield (YP) and stress condition yield (YS) indicated a significant difference in the level of 1% probability between oilseeed rape genotypes (Table 1). Accordingly, significant differences were observed for oilseeed rape genotypes regarding all drought tolerance indices, including TOL, MP, GMP, HM, STI, YSI, SSI, YR, K1 STI, K2STI, YI and RDI (Table 2). These results indicate genetic diversity between genotypes to drought stress tolerance. Breeding programs depend on the knowledge of genetic systems controlling their and genetic and environmental factors that influence their expression (Kahrizi et al., 2010).
Table 1
Analysis of variance for oilseeed rape genotypes in drought stress tolerance indices
SOV | df | YP | YS | TOL | MP | GMP | HM | STI |
Replication | 2 | 0.136ns | 0.004 ns | 0.187 ns | 0.023 ns | 0.016 ns | 0.011 ns | 0.003 ns |
Genotype | 37 | 0.261** | 0.351** | 0.577** | 0.161** | 0.175** | 0.195** | 0.024** |
Error | 74 | 0.098 | 0.051 | 0.149 | 0.037 | 0.036 | 0.037 | 0.005 |
CV% | - | 6.95 | 6.95 | 30.51 | 4.98 | 5 | 5.15 | 10.12 |
SOV | df | YSI | SSI | YR | K1STI | K2STI | YI | RDI |
Replication | 2 | 0.006 ns | 0.084 ns | 66.79 ns | 0.821 ns | 0.041 ns | 0.009 ns | 0.012 ns |
Genotype | 37 | 0.022** | 0.291** | 229.1** | 1.62** | 2** | 0.033** | 0.044** |
Error | 74 | 0.005 | 0.068 | 53.6 | 0.58 | 0.294 | 0.005 | 0.01 |
CV | - | 10.1 | 26.59 | 26.59 | 22.91 | 22.24 | 7.0 | 10.1 |
Based on the yield of the genotypes under different conditions and also drought tolerant indices, the genotypes G13 (5.924 t ha-1) showed the highest yield under normal condition (YP), while the genotypes G10 (4.2 t ha-1) was found as a genotype with the highest yield under drought stress condition (YS). Additionally, the G10 genotype showed the MP (4.488 t ha-1), HM (4.468 t ha-1), GMP (4.478 t ha-1) and STI (0.984 ). On the other hand, the lowest MP (3.325 t ha-1), HM (3.240 t ha-1), GMP (3.282 t ha-1) and STI (0.528) were obtained for G11 (Table 3). Jahangiri and Kahrizi (2015), reported a oilseeed rape genotype (Karaj3) with the highest YS obtained highest MP, GMP and STI. Fernandez (1992) classified plants based on their yield in stress and normal condition to four groups: genotypes with high yield in both conditions (group A), genotypes with high yield only in normal condition (group B) and stress condition (group C), and genotypes with low yield in both conditions (group D). It has been reported that the MP index is affected by grain yield under normal conditions and is incapable of distinguishing A and B group genotypes (Fernandez, 1992). Low values of TOL and SSI indicate higher tolerance to drought stress. Based on TOL (0.455) and SSI (0.379) the G33 was identified as a drought tolerance genotype. The G10 and G18 were in same group with G33 based TOL and SSI, while the G8 has been showed the highest mean TOL (2.011), SSI (1.43). It should be noted that the values of indices TOL and SSI are not suitable to select a genotype for drought tolerance because these two indices do not distinguish between genotypes with low and high yields (Rosille and Hambilin, 1981). The G8, G13, G3, G30, G1 and G25 founded for the highest relative yield reduction (YR) and lowest yield stability index (YSI) and Relative drought index (RDI) while G33 showed highest YSI and RDI and lowest YR. According to the K1STI (4.073) and K2STI (5.368) index, the genotypes G10, was identified as a drought tolerance genotype. Based on the YI index, the genotypes G10 (1.292) was identified as drought tolerant genotypes, while the genotypes G1 (0.808) showed that the lowest amount of YI was drought susceptible genotype.
Table 2
Mean comparison for yield oilseeed rape genotypes for drought stress tolerance indices
Genotype | YP | YS | TOL | MP | GMP | HM | STI | YSI | SSI | YR | K1STI | K2STI | YI | RDI |
G1 | 4.291 | 2.62 | 1.665 | 3.459 | 3.356 | 3.257 | 0.552 | 0.612 | 1.382 | 38.784 | 2.254 | 1.178 | 0.808 | 0.850 |
G2 | 4.479 | 2.87 | 1.609 | 3.674 | 3.584 | 3.496 | 0.630 | 0.642 | 1.273 | 35.712 | 2.841 | 1.602 | 0.883 | 0.893 |
G3 | 4.776 | 2.886 | 1.889 | 3.831 | 3.711 | 3.595 | 0.675 | 0.605 | 1.408 | 39.490 | 3.412 | 1.740 | 0.888 | 0.841 |
G4 | 4.301 | 3.443 | 0.858 | 3.872 | 3.841 | 3.810 | 0.723 | 0.807 | 0.687 | 19.275 | 2.993 | 2.653 | 1.059 | 1.121 |
G5 | 4.579 | 3.333 | 1.246 | 3.956 | 3.894 | 3.833 | 0.743 | 0.737 | 0.937 | 26.301 | 3.466 | 2.568 | 1.025 | 1.024 |
G6 | 3.951 | 3.373 | 0.578 | 3.662 | 3.650 | 3.638 | 0.653 | 0.853 | 0.521 | 14.635 | 2.259 | 2.296 | 1.037 | 1.186 |
G7 | 4.306 | 2.77 | 1.536 | 3.538 | 3.448 | 3.361 | 0.584 | 0.646 | 1.259 | 35.324 | 2.425 | 1.400 | 0.852 | 0.898 |
G8 | 4.994 | 2.983 | 2.011 | 3.989 | 3.858 | 3.732 | 0.730 | 0.597 | 1.434 | 40.235 | 4.031 | 2.01 | 0.917 | 0.830 |
G9 | 4.640 | 3.576 | 1.064 | 4.108 | 4.068 | 4.028 | 0.815 | 0.781 | 0.780 | 21.895 | 4.073 | 3.223 | 1.100 | 1.085 |
G10 | 4.776 | 4.2 | 0.576 | 4.488 | 4.478 | 4.468 | 0.984 | 0.880 | 0.425 | 11.934 | 5.027 | 5.368 | 1.292 | 1.223 |
G11 | 3.850 | 2.8 | 1.050 | 3.325 | 3.282 | 3.240 | 0.528 | 0.727 | 0.972 | 27.271 | 1.734 | 1.283 | 0.861 | 1.010 |
G12 | 4.848 | 3.263 | 1.585 | 4.056 | 3.972 | 3.891 | 0.776 | 0.674 | 1.160 | 32.540 | 4.084 | 2.616 | 1.004 | 0.937 |
G13 | 5.024 | 3.03 | 1.994 | 4.027 | 3.901 | 3.779 | 0.747 | 0.603 | 1.415 | 39.693 | 4.196 | 2.127 | 0.932 | 0.838 |
G14 | 4.655 | 3.75 | 0.905 | 4.202 | 4.173 | 4.144 | 0.857 | 0.813 | 0.663 | 18.620 | 4.261 | 3.733 | 1.153 | 1.131 |
G15 | 4.189 | 2.986 | 1.202 | 3.588 | 3.530 | 3.474 | 0.611 | 0.717 | 1.006 | 28.223 | 2.380 | 1.690 | 0.919 | 0.997 |
G16 | 4.819 | 3.54 | 1.279 | 4.179 | 4.129 | 4.079 | 0.837 | 0.737 | 0.937 | 26.288 | 4.366 | 3.235 | 1.089 | 1.024 |
G17 | 4.822 | 3.46 | 1.362 | 4.141 | 4.082 | 4.025 | 0.817 | 0.719 | 0.999 | 28.023 | 4.236 | 3.009 | 1.064 | 1.0 |
G18 | 4.004 | 3.486 | 0.518 | 3.745 | 3.735 | 3.725 | 0.684 | 0.872 | 0.455 | 12.782 | 2.442 | 2.568 | 1.072 | 1.212 |
G19 | 4.406 | 3.37 | 1.036 | 3.888 | 3.853 | 3.818 | 0.730 | 0.764 | 0.838 | 23.520 | 3.201 | 2.605 | 1.036 | 1.062 |
G20 | 4.472 | 3.133 | 1.338 | 3.802 | 3.743 | 3.684 | 0.692 | 0.701 | 1.064 | 29.849 | 3.194 | 2.166 | 0.964 | 0.975 |
G21 | 4.510 | 3.243 | 1.267 | 3.876 | 3.822 | 3.769 | 0.717 | 0.718 | 1.002 | 28.107 | 3.232 | 2.356 | 0.997 | 0.999 |
G22 | 4.253 | 2.856 | 1.396 | 3.555 | 3.485 | 3.417 | 0.596 | 0.671 | 1.170 | 32.821 | 2.399 | 1.505 | 0.879 | 0.933 |
G23 | 4.229 | 3.56 | 0.669 | 3.894 | 3.877 | 3.860 | 0.737 | 0.844 | 0.553 | 15.517 | 2.947 | 2.890 | 1.095 | 1.174 |
G24 | 4.879 | 3.006 | 1.872 | 3.942 | 3.824 | 3.709 | 0.716 | 0.620 | 1.354 | 37.991 | 3.787 | 2.002 | 0.925 | 0.861 |
G25 | 4.675 | 2.86 | 1.815 | 3.767 | 3.653 | 3.542 | 0.654 | 0.612 | 1.381 | 38.743 | 3.169 | 1.674 | 0.88 | 0.851 |
G26 | 4.371 | 3.17 | 1.201 | 3.770 | 3.720 | 3.670 | 0.678 | 0.725 | 0.977 | 27.422 | 2.870 | 2.117 | 0.975 | 1.008 |
G27 | 4.785 | 3.156 | 1.628 | 3.971 | 3.884 | 3.800 | 0.740 | 0.66 | 1.212 | 34.004 | 3.758 | 2.292 | 0.971 | 0.917 |
G28 | 4.405 | 3.523 | 0.881 | 3.964 | 3.931 | 3.9 | 0.760 | 0.80 | 0.681 | 19.099 | 3.362 | 2.937 | 1.084 | 1.124 |
G29 | 4.828 | 3.276 | 1.551 | 4.052 | 3.977 | 3.903 | 0.779 | 0.677 | 1.149 | 32.230 | 4.106 | 2.654 | 1.008 | 0.941 |
G30 | 4.951 | 2.986 | 1.964 | 3.969 | 3.841 | 3.718 | 0.724 | 0.605 | 1.407 | 39.47 | 3.943 | 2.002 | 0.919 | 0.841 |
G31 | 4.434 | 3.516 | 0.917 | 3.975 | 3.947 | 3.919 | 0.764 | 0.795 | 0.728 | 20.424 | 3.371 | 2.915 | 1.082 | 1.105 |
G32 | 4.755 | 2.956 | 1.798 | 3.855 | 3.744 | 3.637 | 0.687 | 0.624 | 1.339 | 37.557 | 3.455 | 1.865 | 0.909 | 0.867 |
G33 | 4.255 | 3.8 | 0.455 | 4.027 | 4.020 | 4.013 | 0.792 | 0.893 | 0.379 | 10.640 | 3.178 | 3.522 | 1.169 | 1.241 |
G34 | 4.618 | 3.73 | 0.888 | 4.174 | 4.140 | 4.108 | 0.840 | 0.816 | 0.655 | 18.375 | 4.006 | 3.610 | 1.147 | 1.134 |
G35 | 4.437 | 3.286 | 1.150 | 3.862 | 3.818 | 3.774 | 0.715 | 0.740 | 0.925 | 25.947 | 3.121 | 2.397 | 1.011 | 1.029 |
G36 | 4.377 | 3.346 | 1.031 | 3.862 | 3.819 | 3.777 | 0.719 | 0.767 | 0.828 | 23.223 | 3.110 | 2.585 | 1.029 | 1.067 |
G37 | 4.182 | 2.87 | 1.312 | 3.526 | 3.463 | 3.401 | 0.588 | 0.687 | 1.113 | 31.239 | 2.284 | 1.490 | 0.883 | 0.955 |
G38 | 4.489 | 3.463 | 1.026 | 3.976 | 3.942 | 3.909 | 0.762 | 0.771 | 0.815 | 22.858 | 3.401 | 2.821 | 1.065 | 1.072 |
LSD 5% | 0.502 | 0.361 | 0.618 | 0.309 | 0.306 | 0.310 | 0.116 | 0.117 | 0.417 | 11.716 | 1.219 | 0.868 | 0.113 | 0.162 |
LSD 1% | 0.661 | 0.476 | 0.814 | 0.407 | 0.402 | 0.408 | 0.153 | 0.154 | 0.549 | 15.423 | 1.605 | 1.143 | 0.180 | 0.148 |
A suitable index should have a significant correlation with grain yield under both conditions (Mitra, 2001). The indices that show a high correlation coefficient with yield in normal condition (YP) and stress condition (YS) is considered as a suitable index (Fernandez, 1992). The MP, GMP, STI and K1STI showed high positive correlation with YP and YS, so according to the obtained correlation coefficient, these indices (MP, GMP, STI and K1STI) were founded as suitable indices for yield stability and yield estimate in both condition (Table 3). These results are in accordance with some crops such as wheat (Sio-Se Mardeh et al. 2006), Oilseeed rape (Dehghani et al., 2009), Oilseeed rape (Khalili et al 2012).
Table 3
Correlation coefficients of drought tolerance indices for the oilseeed rape genotypes
| YP | YS | TOL | MP | GMP | HM | STI | YSI | SSI | YR | K1STI | K2STI | YI | RDI |
YP | 1 | | | | | | | | | | | | | |
YS | 0.05 | 1 | | | | | | | | | | | | |
TOL | 0.62* | -0.74* | 1 | | | | | | | | | | | |
MP | 0.67* | 0.77* | -0.14 | 1 | | | | | | | | | | |
GMP | 0.56* | 0.85* | -0.29 | 0.98* | 1 | | | | | | | | | |
HM | 0.44* | 0.91* | -0.41* | 0.96* | 0.99* | 1 | | | | | | | | |
STI | 0.55* | 0.86* | -0.30 | 0.985* | 0.99* | 0.99* | 1 | | | | | | | |
YSI | -0.48* | 0.84* | -0.98* | 0.31 | 0.44* | 0.55* | 0.45* | 1 | | | | | | |
SSI | 0.48* | -0.84* | 0.98* | -0.31 | -0.44* | -0.55* | -0.45* | -0.99* | 1 | | | | | |
YR | 0.48* | -0.84* | 0.98* | -0.31 | -0.44* | -0.55* | -0.45* | -0.99* | 0.99* | 1 | | | | |
K1STI | 0.86* | 0.52* | 0.17 | 0.94* | 0.88* | 0.82* | 0.88* | -0.005 | 0.005 | 0.005 | 1 | | | |
K2STI | 0.23 | 0.96* | -0.6* | 0.86* | 0.92* | 0.95* | 0.93* | 0.71* | -0.71* | -0.71* | 0.675* | 1 | | |
YI | 0.058 | 0.99* | -0.74* | 0.77* | 0.85* | 0.91* | 0.86* | 0.84* | -0.84* | -0.84* | 0.53* | 0.96* | 1 | |
RDI | -0.48* | 0.84* | -0.98* | 0.31* | 0.44* | 0.55* | 0.45* | 0.99* | -0.99* | -0.99* | -0.005 | 0.713* | 0.84* | 1 |
Cluster analysis has been used to describe the variation between genotypes and their grouping based on similarity. The cluster analysis of genotypes based on MP, GMP, STI and K1STI was performed through WARD’s method (Fig. 1). Cluster analysis divided the genotypes into four groups. It was found a group consisting genotypes G10, G33, G34, G14, G17, G16 and G9 which had the highest MP, GMP, STI and K1STI and highest YP and YS, while in another group genotypes G11, G15, G37, G22, G7 and G1 with the lowest values of indices and lowest YP and YS were observed. Genotypes with their average rank for these four indices are available in the Table 4.
Table 4
Average Ranks of oilseeed rape genotypes based on MP, GMP, STI and K1STI
Genotype | Mean Rank | Genotype | Mean Rank | Genotype | Mean Rank | Genotype | Mean Rank |
G10 | 38 | G35 | 14.75 | G25 | 10.75 | G11 | 1 |
G33 | 27.5 | G21 | 14 | G32 | 16.25 | G15 | 5.75 |
G34 | 34 | G5 | 24.25 | G3 | 14 | G37 | 3.75 |
G14 | 28.75 | G38 | 26.25 | G18 | 10.25 | G22 | 5.25 |
G17 | 34.25 | G31 | 23 | G6 | 6.5 | G7 | 4.25 |
G16 | 35.75 | G28 | 24.25 | G26 | 10.75 | G2 | 7.75 |
G9 | 25.25 | G23 | 19.25 | G20 | 14 | G1 | 2 |
| | G36 | 15.75 | | | | |
| | G19 | 20 | | | | |
| | G4 | 16.75 | | | | |
| | G27 | 24.75 | | | | |
| | G29 | 31.5 | | | | |
| | G12 | 31 | | | | |
| | G24 | 20.5 | | | | |
| | G13 | 28.75 | | | | |
| | G30 | 23 | | | | |
| | G8 | 25.25 | | | | |
Average | 31.92 | Average | 22.52 | Average | 11.78 | Average | 4.25 |
Based on two traits of grain yield in normal condition (YP) and drought stress condition (YS), a three-dimensional diagram was drawn for MP, GMP, STI and K1STI indices. Three-dimensional graphs divide the genotypes into four groups and each division represents one combination of the genotypes: high yields under both condition (Group A); high yield in a normal condition (Group B); high yield in a stress condition (Group C); and low yield under both conditions (Group D) (Fernandez, 1992). The results showed that G10 was placed in group A and had the highest yield under normal and stress conditions and was the stable genotype. Also, G11 was placed in group D according to all indices with low yield in both conditions and was the stable genotype. The G13 was placed in group B and it is suitable for cultivation in normal condition and G33 was suitable for cultivation in drought stress condition and was placed in group C (Fig. 2).
The principal component analysis was obtained from 14 drought tolerance indices for 38 oilseeed rape genotypes. The tolerance indices and genotype are graphically displayed in a plot by two first principal components (PC1 and PC2) analysis (Fig. 3). Based on principal components analysis, first and second components explained 99.6% of the variations between index (61.5 and 34.5% for PC1 and PC2, respectively). The biplot diagrams investigate the interrelationship of oilseeed rape genotypes with all drought tolerance indices which is presented in Fig. 3 (A-B). The principal component 1 showed positive correlations with oilseeed rape yield in drought stress conditions and tolerant indices, including HM, STI, GMP and MP, K1STI and K2STI, so the component 1 is related to drought tolerant and yield productivity. The principal component 2 showed positive correlations with YSI and RDI and increasing these components is related to drought stress sensitivity of genotypes. In the polygon diagram, several polygons are obtained based on drought tolerance indices. Genotypes in each polygon have the highest value of indices and the highest correlation coefficient with those indices. Also, in each polyvan, the genotype at the vertex of the polygon has the highest values (Fig. 3A). Genotype 21 was centered on the biplot which has shown moderate yield in drought stress (YS) and normal condition (YP) and ranked as moderate genotype in terms of tolerance and susceptibility indices. The G11 has shown the lowest value of YP and YS and also the lowest rank of indices. The G10 has shown the highest correlation coefficient with YS and drought tolerance indices such as MP, GMP, STI and HM. The G13 has shown the highest value for YP and showed a positive correlation with susceptibility indices which was suitable for normal conditions. Therefore, G10 was chosen as a suitable genotype for drought and normal conditions (Fig. 3A and B).