Determinants of barley grain yield in a wide range of Mediterranean environments

Barley grain yield in rainfed Mediterranean regions can be largely influenced by terminal drought events. In this study the ecophysiological performance of the ‘Nure’ (winter) × ‘Tremois’ (spring) barley mapping population (118 Doubled Haploids, DHs) was evaluated in a multi-environment trial of eig...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Field crops research 2011, Vol.120 (1), p.169-178
Hauptverfasser: Francia, Enrico, Tondelli, Alessandro, Rizza, Fulvia, Badeck, Franz W., Li Destri Nicosia, Orazio, Akar, Taner, Grando, Stefania, Al-Yassin, Adnan, Benbelkacem, Abdelkader, Thomas, William T.B., van Eeuwijk, Fred, Romagosa, Ignacio, Stanca, A. Michele, Pecchioni, Nicola
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Barley grain yield in rainfed Mediterranean regions can be largely influenced by terminal drought events. In this study the ecophysiological performance of the ‘Nure’ (winter) × ‘Tremois’ (spring) barley mapping population (118 Doubled Haploids, DHs) was evaluated in a multi-environment trial of eighteen site–year combinations across the Mediterranean Basin during two consecutive harvest years (2004 and 2005). Mean grain yield of sites ranged from 0.07 to 5.43 t ha −1, clearly dependent upon both the total water input (rainfall plus irrigation) and the water stress index (WSI) accumulated during the growing season. All DHs were characterized for possessing molecular marker alleles tagging four genes that regulate barley cycle, i.e. Vrn-H1, Vrn-H2, Ppd-H2 and Eam6. Grain yield differences were initially interpreted in terms of mean differences between genotypes (G), environments (E), and for each combination of genotype and environment (GE) through a “full interaction” ANOVA model. Variance components estimates clearly showed the greater importance of GE over G, although both were much lower than E. Alternative linear and bilinear models of increasing complexity were used to describe GE. A linear model fitting allelic variation at the four genes explained genotype main effect and genotype × environment interaction much better than growth habit itself. Adaptation was primarily driven by the allelic constitution at three out of the four segregating major genes, i.e. Vrn-H1, Ppd-H2 and Eam6. In fact, the three genes together explained 47.2% of G and 26.3% of GE sum of squares. Grain yield performance was more determined by the number of grains per unit area than by the grain weight (phenotypic correlation across all genotypic values: r = 0.948 and 0.559, respectively). The inter-relationships among a series of characters defining grain yield and its components were also explored as a function of the length of the different barley developmental phases, i.e. vegetative, reproductive, and grain filling stages. In most environments, the best performing (adapted) genotypes were those with faster development until early occurrence of anthesis. This confirmed the crucial role of the period defining the number of grains per unit area in grain yield determination under Mediterranean environments.
ISSN:0378-4290
1872-6852
DOI:10.1016/j.fcr.2010.09.010