Relationships Between Tree Height and Carbon Isotope Discrimination

Abstract

Understanding how tree size impacts leaf-and crown-level gas exchange is essential to predicting forest yields and carbon and water budgets. The stable carbon isotope ratio (d 13 C) of organic matter has been used to examine the relationship of gas exchange to tree size for a host of species because it carries a temporally integrated signature of foliar photosynthesis and stomatal conductance. The carbon isotope composition of leaves reflects discrimination against 13 C relative to 12 C during photosynthesis and is the net result of the balance of change in CO 2 supply and demand at the sites of photosynthesis within the leaf mesophyll. Interpreting the patterns of changes in d 13 C with tree size are not always clear, however, because multiple factors that regulate gas exchange and carbon isotope discrimination (D) co-vary with height, such as solar irradiance and hydraulic conductance. Here we review 36 carbon isotope datasets from 38 tree species and conclude that there is a consistent, linear decline of D with height. The most parsimonious explanation of this result is that gravitational constraints on maximum leaf water potential set an ultimate boundary on the shape and sign of the relationship. These hydraulic constraints are manifest both over the long term through impacts on leaf structure, and over diel periods via impacts on stomatal conductance, photosynthesis and leaf hydraulic conductance. Shading induces a positive offset to the linear decline, consistent with light limitations reducing carbon fixation and increasing partial pressures of CO 2 inside of the leaf, p c at a given height. Biome differences between tropical and temperate forests were more important in predicting D and its relationship to height than wood type associated with being an angiosperm or gymnosperm. It is not yet clear how leaf internal conductance varies with leaf mass area, but some data in particularly tall, temperate conifers suggest that photosynthetic capacity may not vary dramatically with height when compared between tree-tops, while stomatal and leaf internal conductances do decline in unison with height within canopy gradients. It is also clear that light is a critical variable low in the canopy, whereas hydrostatic constraints dominate the relationship between D and height in the upper canopy. The trend of increasing maximum height with decreasing minimum D suggests that trees that become particularly tall may be adapted to tolerate particularly low values of p c .

Determinants of maximum tree height inEucalyptusspecies along a rainfall gradient in Victoria, Australia

“…4). Globally, D increases with moisture supply in many contexts (Schulze et al 1998, Miller et al 2001, Turner et al 2008, Diefendorf et al 2010, Kohn 2010, Prentice et al 2010, Mencuccini et al 2011) and decreases with tree height locally (Koch et al 2004, Burgess and Dawson 2007, McDowell et al 2011. Our data suggest that, when relative moisture supply and tree height co-vary along a gradient, the effects of the former on maximum tree height will dominate.…”

“…This astonishing growth rate is accompanied by exceptionally slender stems, the production of notably light wood (ca. 425 kg/m 3 ), with numerous thin-walled, large-diameter, long vessels (;280 lm 3 1.8 m) that frequently collapse when sawn wood is dried, and by strong xylem tapering that results in much of the axial resistance to water flow being confined to the last 1 m below the tree's apex, suggesting exceptionally shallow declines in w leaf with height (Petit et al 2010), as also seen in many conifers from temperate rain forests in the Pacific Northwest (McDowell et al 2011). …”

“…Third, while plant species can differ in their sensitivity to water potential, w leaf at the top of the tallest Sequoia is much less negative than the minimum observed in many desert plants that can survive and photosynthesize under much harsher conditions (e.g., less than À6.0 MPa for Larrea tridentata in North American deserts [Naumberg et al 2004], and less than À6.5 to possibly À10 MPa in Acacia aneura and A. ayersiana in Australian mulga [O'Grady et al 2009, Page et al 2011, Cleverly et al 2013), making it unlikely that low water potentials in tall trees per se halt carbon gain. Increases in d 13 C (and, thus, decreases in the carbon isotope discrimination D) with tree height appear to be quite general, both within tree crowns and among well-illuminated crown tops, with the rate of decline being least in gymnosperms and greatest in temperate angiosperms (McDowell et al 2011). Optimal stomatal conductance and w leaf should decline together under a variety of models: (1) to maximize photosynthesis for a given total amount of transpiration, assuming that A does not depend on w leaf (Cowan andFarquhar 1976, Cowan 1986), given that less water is available in the stem and the ground when predawn w leaf is more negative; (2) to maximize photosynthesis when A does depend on w leaf , given that optimal g should drop not with decreasing w leaf but with increasing values of ]A/]w leaf and decreasing values of plant hydraulic conductance (Fig.…”

“…We predict that tree height should increase with relative moisture supply, reflecting the effects of both hydraulic limitation and resource allocation; that stomatal limitation should decrease with increasing rainfall at maximum tree height; and thus that moisture supply should have a stronger effect than tree height on apparent stomatal limitation. Recent global metaanalyses show a roughly 4% increase in D with rainfall from 300 to 1600 mm (Diefendorf et al 2010, Kohn 2010, but roughly the same decrease in D with tree height from 3 to 90 m across a wide range of different habitats (McDowell et al 2011), so that the null expectation might in some sense be zero change in D along the Victoria gradient.…”

“…Variation in maximum tree height has profound implications for the photosynthesis, growth, crown and bole dimensions, tissue allocation, hydraulic conductance, and population density of individual trees (Givnish 1995, Ryan and Yoder 1997, McDowell et al 2011, Mencuccini et al 2011, Larjavaara 2013, stand biomass, carbon storage, and productivity (Feldpausch et al 2012, Hui et al 2012, Ruiz-Benito et al 2014, leaf area index and transpiration (Roberts et al 2001, Falster et al 2010, net input to ground and surface waters (Bleby et al 2012), and the availability of suitable habitat for certain animals (Wormington et al 2002, Garcı´a-Ferced et al 2011. Over the last three decades, several models have been advanced to account for maximum tree height and its variation with environmental conditions and plant traits.…”