Drought-tolerance indices: picking tolerant high-yielders across stress and non-stress
Twelve genotypes yielded under irrigated and terminal-drought conditions. SSI, STI, MP and GMP read together to separate true drought tolerance from low overall yield.
Imran screens twelve chickpea genotypes (G1 to G12) for terminal drought tolerance. Each genotype was grown in two regimes in the same season: a fully irrigated block (potential yield Yp) and a terminal-drought block where irrigation was withdrawn at flowering (stress yield Ys). He wants the genotypes that yield well and hold that yield under stress, not just the ones with a small percentage drop.
Question
Which genotypes combine high yield potential with true drought tolerance, judged by SSI, STI, MP and GMP together rather than any single index?
Data, in StatVeda format
One genotype per line, Yp then Ys, grain yield in g per plant. Yp is yield under non-stress, Ys is yield under terminal drought.
G1: 52.0, 38.0 G2: 48.0, 24.0 G3: 60.0, 30.0 G4: 35.0, 28.0 G5: 55.0, 41.0 G6: 50.0, 22.0 G7: 58.0, 44.0 G8: 41.0, 19.0 G9: 62.0, 33.0 G10: 46.0, 35.0 G11: 39.0, 30.0 G12: 57.0, 26.0
What he does in StatVeda
Open Plant Breeding. Pick Drought-Tolerance Indices. Paste the Yp, Ys list. Run. The tool computes SSI, TOL, MP, GMP, STI, YI, YSI and DRI for every genotype using the trial mean Yp and Ys.
Mean Yp = 50.25, mean Ys = 30.83, overall stress intensity SI = 0.387. SSI is scaled by this SI; STI and GMP reward genotypes that are high in both regimes.
Selected genotypes, SSI, STI, MP, GMP:
G7: SSI 0.62, STI 1.65, MP 51.0, GMP 50.5 (top STI and GMP)
G5: SSI 0.66, STI 1.46, MP 48.0, GMP 47.5
G1: SSI 0.70, STI 1.28, MP 45.0, GMP 44.4
G3: SSI 1.30, STI 1.16, MP 45.0, GMP 42.4
G9: SSI 1.21, STI 1.32, MP 47.5, GMP 45.2
G4: SSI 0.52, STI 0.54, MP 31.5, GMP 31.3 (lowest SSI but low yield)
Low SSI means tolerant (small yield loss relative to the population); high STI and GMP mean high yield in both conditions.
What it means
No single index is enough on its own. G4 has the lowest SSI, which in isolation looks like the most drought-tolerant genotype, but its STI and GMP are the lowest of the set: it loses little because it yields little to begin with. That is escape by low productivity, not useful tolerance. G7 is the genotype to select: high yield in both regimes (STI 1.65, GMP 50.5) with an SSI well below 1, meaning it holds yield under stress better than the population average. G5 is the close second. G3 and G9 have high potential yield but SSI above 1, so they are productive but drought-susceptible, suited to assured-moisture conditions only.
Select G7 and G5 as drought-tolerant high-yielders: top STI and GMP with SSI below 1. These combine productivity with stress resilience.
Reject G4 despite its low SSI. A low susceptibility index on a low-yielding genotype is yield escape, not tolerance, and STI and GMP expose it.
Channel G3 and G9 (high Yp, SSI above 1) to irrigated recommendation only; do not promote them as drought varieties.
Imran writes the screening section ranking genotypes by STI with SSI alongside, and states explicitly why the lowest-SSI genotype was not selected. That single sentence pre-empts the most common reviewer objection to drought-index selection.
Why several indices, not just SSI
SSI and YSI measure relative loss, so they reward genotypes that had little to lose. STI, MP and GMP are yield-based, so they reward absolute productivity in both regimes. Using them together is what separates a genuinely tolerant high-yielder from a low-yielding escaper. Reporting only the relative indices is the classic mistake that promotes unproductive genotypes.
What he will do next season
Carry G7 and G5 into a multi-location managed-drought trial across sites and years, then read GxE for the stress trait with a stability analysis. The index screening narrowed twelve genotypes to two; the multi-environment stage confirms whether the tolerance of G7 holds across drought environments rather than one season.