Contrasted reaction norms of wheat yield in pure vs mixed stands explained by tillering plasticities and shade avoidance

Abstract

Background and Aims: Cultivar mixture is an agronomic practice of diversification increasingly used in the framework of the agroecological transition. However, even though the yield of such mixtures is higher on average than the mean yield of the monocultures, the variance of mixture yield is large. This variability is likely due to the co-occurence of multiple ecophysiological processes shaping plant-plant interactions, yet it remains poorly understood, notably in crops. With winter wheat (Triticum aestivum L.) as a case study, we designed a field experiment to explore phenotypic plasticity at both genotype and plant levels along a gradient of neighborhood heterogeneity. Methods: Eight wheat commercial cultivars were grown in pure and mixed stands in field plots for two seasons. Two quaternary mixtures were assembled with cultivars contrasted either for height or earliness. Thanks to a precision sowing in mixtures, genotypes were tracked at plant scale from sowing to harvest, and individual plants were phenotyped for above-ground traits throughout growth. Phenotypic plasticity between pure and mixed stands was then analyzed at both within- and between-genotype scales, according to a new conceptual framework distinguishing mean and variability differences. Key results: Some genotypes dominated others in mixed stands, i.e., they produced a significantly higher mean yield, whereas all these genotypes yielded similarly in pure stands. These between-genotype dominance relationships remained stable over the two seasons despite strong contrasts in temperature and light sums. We showed that these dominance relationships in mixed stands were caused by contrasted phenotypic plasticity of yield components and biomass allocation in pure versus mixed stands. Tillering dynamics, determined by light competition between individual plants, was a main causal factor explaining between-genotype plasticity in both pure and mixed stands. Conclusions: Our innovative experimental design enabled us to measure phenotypic plasticity at both within- and between- genotype levels. Plasticity in tillering dynamics and yield components allowed to decipher the genotype strategies in mixtures.