Increasing the stability of crop production is one of the most important challenges in this century (Powell et al., 2012). Yield stability in agriculture is defined as consistency in crop performance when grown at different locations with varying biotic or abiotic stresses over time (Döring et al., 2018). Most arable crop production is dominated by monocultures of modern elite cultivars that have been bred and subsequently grown under high inputs of artificial fertilizers and pesticides. In barley (Hordeum vulgare), which is the fourth most planted crop in the world by area grown (FAO, 2004), the most common genetic structure is the pedigree-bred pure line (elite cultivar). However, barley is also grown in different parts of Europe and world-wide as landraces and other traditional material. Genetically diverse lines, such as traditional landraces and composite cross populations (CCPs) have wider variation in agronomic traits compared to elite cultivars (Nandety et al., 2014). Furthermore, some traditional barley material have increased yield complementation and compensation under varying growing conditions (Ceccarelli, 1994). Most traditional or diverse barley landraces have been locally adapted, often under low input farming, and with high consistency in yield.
In high input systems, the expectation is that yield of elite cultivars will outperform traditional landraces. Although landraces and CCPs may have inferior yield under high agronomic inputs, they demonstrate increased resilience under reduced inputs, or more stressful conditions (Ceccarelli, 1994). However, some traditional landraces and CCPs confer increased yield stability compared to elite cultivars, when grown under challenging environmental or climatic conditions, including high disease burden, extremes in precipitation and variation in soil nitrogen (N) supply (Dwivedi et al., 2017). Explanations for this enhanced stability in crop performance has been attributed to improved resource capture in relation to niche variation (Zuppinger-Dingley et al., 2014), specific genetic variation underlying stability (Mickelbart et al., 2015) or local adaptation (Dwivedi et al., 2017) in the more diverse barley lines.
An approach to exploiting wider genetic resources, including landraces, is to quantify variation and stability in agronomic traits under a common field protocol. Such approaches can be used to compare landraces from different geographic and climatic backgrounds for agronomic and pre-breeding value. This would include a practical guide to phenotypic variance in agronomic traits such as leaf morphology, shoot production (tillering), grain number per ear and mean grain weight (thousand grain weight) that are being associated with improved yield performance. Theoretical studies (Zhai et al, 2014) suggest a trade-off (or yield penalty) between phenotypic plasticity (robustness) and other traits (yield); however, such a practical approach has rarely been tested in agricultural field crop systems (but see Sadras et al., 2016; Fletcher at al., 2015).
Traditional landraces tend to originate from low input conditions under local environmental conditions where genetic by environmental (GxE) effects are best exploited under local conditions (Gage et al, 2019). This contrasts with high inputs systems, or elite cultivars, in which high GxE is controlled using high agronomic inputs. There are very few studies that have compared genetically diverse landraces from different geographic backgrounds in relation to phenotypic expression and stability against elite cultivars when grown under common protocol for achieving high yield. Using pre-adapted genetic resources such as landraces and CCPs provides an opportunity to discover untapped agronomic value, as a large proportion of traditional barley material has remained stored in gene banks with very little information on its agronomic performance.
In order to exploit the potential benefits of wider genetic variation for future crop breeding, we tested the hypothesis that traditional landraces a CCP (Harlan & Martini, 1929), express wider phenotypic variation in agronomic traits, but with greater yield stability than modern elite cultivars. We tested this hypothesis by comparing five geo-climatic landrace groups comprised of two or six rowed ear habit along with a group of Harlan CCPs and a group of elite cultivars grown under a common high-yielding field evaluation protocol over several years and sites. We used European geographic region of origin (North, East, South and West; Metzger, 2005; Peel et al., 2007) based on gene bank information. Agronomic traits of interest were plant height, shoot production (tillering), final ear number (population), number of seeds per ear, ear length thousand grain weight (TGW), and flag leaf dimensions. Breadth of phenotypic expression and stability across sites were examined among the nine groups, with main comparisons being between landraces, CCPs and cultivars, and between the and 2 or 6 rowed ear habit.