Leaf traits predict performance under varying levels of drought stress in cultivated sunflower (Helianthus annuusL.)

Abstract

ABSTRACTDrought is a major agricultural challenge and is expected to worsen with climate change. Exploring plant traits and how they respond to drought has the potential to improve understanding of drought tolerance and inform breeding efforts to develop more drought tolerant plants. Given their importance in plant-water relations, we explored variation and plasticity in leaf traits in response to water limitation in cultivated sunflower (Helianthus annuusL.). A set of four sunflower genotypes was grown under four different levels of water availability and leaf vein and stomatal traits were measured along with total biomass (as an indicator of performance), leaf mass per area (LMA), chlorophyll content, and various mass fraction traits related to resource allocation (e.g., leaf, root, and stem mass fraction). Traits exhibited numerous bivariate correlations within treatments that generally followed expectations based on the literature. For example, stomatal size and density were negatively correlated while stomatal density and vein length per area (VLA) were positively correlated. Most traits exhibited substantial plasticity, as evidenced by significant shifts in trait values across environments and multivariate analyses revealed differentiation in trait space across treatment levels. This included an overall reduction in growth/productivity in response to stress, accompanied by a shift in traits relating to gas exchange and hydraulics including stomatal and vein density (increased), stomatal size (decreased), and theoretical gsmax (increased). We found that variation in performance across treatments (estimated as total biomass) can be largely explained by a small number of putatively size-independent traits (i.e., VLA, stomatal length and density and LMA;R2= 0.74). Moreover, on average, more extreme changes in VLA were associated with more extreme decreases in performance across environments. A small number of leaf traits can predict plant performance, with plasticity in VLA being the best predictor of changes in productivity.