Hydraulic trait variation and relationships with the leaf economics spectrum in wine grapes.

Abstract

Understanding how functional traits and trait spectra vary and covary with one another is a primary goal in plant ecology, however, there remains debate surrounding whether functional trait spectra vary independently of one another. Relationships among trait spectra-referred to as "trait dimensionality"-have been tested in multiple studies, investigating the dimensionality among the most common trait spectra, such as those between the Leaf Economics Spectrum (LES) and other aspects of plant form and function. However, we know considerably less about plant hydraulic traits, and whether or not they form distinct axes of trait variation within plant species, and if a hydraulic trait dimensions covaries with, or is independent of, the LES. Using 12 varieties of grapevine (Vitis vinifera), we evaluated trait dimensionality among eight hydraulic traits derived from pressure-volume curves, to test if these traits form distinct a distinct trait axis, and if this axis covaries within an intraspecific LES in wine grapes. Our results indicate that wine grape cultivars and clones differ significantly in their hydraulic trait expression, with a single principal axis of variation explaining 58.3% of hydraulic trait variation in wine grapes. Along this axis, capacitance at full turgor (C_ft) is especially important in defining hydraulic traits of wine grapes, with this trait trading off against the other hydraulic traits including turgor loss point, relative leaf water content at turgor loss point, modulus of elasticity, and the apoplastic water fraction. Multiple factor analysis further revealed orthogonal relationships between hydraulic traits and a LES in wine grapes, suggesting that these trait spectra vary independently of one another. Maintaining an understanding of hydraulic trait variation is increasingly important in economically significant crops such as V. vinifera, as drought becomes more prevalent, and our results here suggest that crop trait spectra represent independent axes of ecological variation among plants, varieties, and clones.