Differential Responses of Rice Seed Vigour to Accelerated Ageing Among Sub-Populations and Geographic Origins as Revealed via Multi-Spectral Imaging

Seed vigour is a key trait determining field seedling establishment and storage longevity in rice. In this study, we aimed to systematically evaluate the vigour variation patterns of rice germplasm from different sub-populations and geographic origins under artificial ageing and explore the feasibility of multi-spectral imaging for non-destructive vigour assessment. Specifically, we selected 244 rice accessions covering seven sub-populations and originating from six continents. After 0, 10 and 15 days of accelerated ageing, standard germination tests combined with multi-spectral image analysis were conducted to determine the germination rate, germination potential and vigour index. The results showed that accelerated ageing significantly inhibited the germination ability of all seeds, and this effect exhibited significant sub-population specificity. Among the sub-populations, aus and Asian-origin accessions displayed the strongest ageing tolerance, while the temperate japonica sub-population and European-origin accessions were highly sensitive, with vigour almost completely lost after 10 days of ageing. Multi-spectral data analysis revealed that reflectance in the 780–970 nm wavelength band was highly significantly correlated with the vigour index. In conclusion, the genetic background and geographical ecotype may be key determinants of seed storability in rice; the aus sub-population and Asian-origin accessions appear to be promising sources of ageing tolerance; and multi-spectral imaging can serve as an effective high-throughput, non-destructive tool for seed vigour evaluation. It should be noted that, given the limited sample size in this study and the fact that geographic origin and sub-population composition may not be independent in the evaluated germplasm panel, the conclusions are only representative of this particular set of accessions. The generalisability of the results therefore requires further validation with additional and more balanced samples.