Chlorophyll fluorescence as a selection criterion for grain yield in durum wheat under Mediterranean conditions

Abstract

The relationships between different fast (non-modulated) chlorophyll fluorescence parameters measured in the field and yield were studied in durum wheat ( Triticum durum Desf.) grown in a Mediterranean region (Northwest Syria) under three different water regimes. A set of 144 genotypes were cultivated in two rainfed trials (hereafter referred to as environments), 125 of which were also grown under supplementary irrigation. Fluorescence measurements were made on attached flag leaf blades about 3 weeks after anthesis in the three environments, plus a further set of measurements, two weeks after, in the irrigation environment. The photochemical capacity of Photosystem (PS) II was measured by means of the ratio of variable to maximum chlorophyll fluorescence ( F v F m ). In addition, the changes in variable ( F v = F m − F 0) fluorescence, the absolute values F 0, F m, and the half-time of the increase from F 0 to F m (t 1 2 ) were also determined. Growing environment was found to have a strong effect on yield and all the fluorescence parameters. Except for F v F m , genotype effect was also significant ( P ⩽ 0.001) within environments. Interaction between genotype and environment was only significant for grain yield. F 0 showed the highest broad-sense heritability ( h 2 = 0.745), followed very closely by F m ( h 2 = 0.729), t 1 2 (h 2 = 0.721) and F v ( h 2 = 0.679), whereas F v F m showed much lower heritability ( h 2 = 0.143) and grain yield an intermediate value ( h 2 = 0.506). The pattern of changes in chlorophyll fluorescence parameters between the irrigation and the driest environments paralleled those associated with ageing in the former. Thus, mean values of F v F m , F m, and t 1 2 decreased and F 0 increased with ageing or in the driest environment. Across genotypes fluorescence parameters changed in a coordinated manner different from that caused by environment or ontogeny. High-yielding genotypes showed lower t 1 2 , accompanied by higher F 0 and F m. The parameter which showed the best genetic correlation with grain yield was t 1 2 ( r = −0.92), followed by F 0 ( r = 0.88), F m ( r = 0.74) and F v ( r = 0.71), whereas F v F v (r = 0.34) was the parameter least correlated. The coordinated pattern of changes in fluorescence parameters across genotypes suggest that the more productive genotypes are those which can avoid or escape the development of drought stress during grain filling. The observed correlations of t 1 2 and F 0 with phenological parameters (days from planting to heading or maturity) or indicators of water status of the crop (carbon isotope discrimination of kernels) support this conclusion.