Towards a refined method for estimating 18O autotrophic fractionation during sucrose synthesis in a bulk leaf

A quantitative understanding of ¹⁸O fractionation mechanisms in plants is highly desirable for effective utilization of the δ¹⁸O signatures of plant cellulose (δ¹⁸O_cel) in diverse climatic and ecological applications. According to the isotope theory, biochemical fractionation associated with oxygen isotopic exchange represents a critical control of δ¹⁸O_cel. Biochemical fractionation operates in both autotrophic (i.e., leaf) and heterotrophic (i.e., stem/trunk) organs, with the current δ¹⁸O_cel model assuming that its effect amounts to c. 27‰ in both types of organs. However, with respect to the autotrophic fractionation factor (ε_{bio_A}), calculations of the deviations of the δ¹⁸O enrichment of sucrose (Δ¹⁸O_ls) from that of water (Δ¹⁸O_lw) in a bulk leaf, -- as performed in many previous studies, -- have led to a wide range of ε_{bio_A} estimates (i.e., from 22.4 to 34‰) across different species. Such a bulk-leaf based estimation method, however, does not provide a precise quantification of ε_{bio_A}. This is because the oxygen exchange-determined intrinsic relationship between Δ¹⁸O_ls and Δ¹⁸O_lw, namely Δ¹⁸O_ls = Δ¹⁸O_lw + ε_{bio_A}, is not expected to hold at the bulk leaf level owing to complications arising from within-leaf heterogeneity that is commonly present in Δ¹⁸O_lw and sucrose synthesis rate (r_suc).

Here, based on explicit consideration of potential within-leaf heterogeneity in r_suc, as well as of spatial variation characteristics of Δ¹⁸O_lw as informed by the Farquhar-Gan model, we suggest that the isotopic relationship between bulk Δ¹⁸O_ls and Δ¹⁸O_lw should instead be expressed as the following: Δ¹⁸O_ls = β*Δ¹⁸O_lw + ε_{bio_A}, with β being a composite variable highly relevant to spatial variation of r_suc across the leaf. Further analysis demonstrates that β can be markedly larger or smaller than unity depending on whether r_suc progressively increases or decreases along the leaf length. With the derivation of this new equation delineating bulk leaf isotopic relationships, we propose a regression approach via which ε_{bio_A} can be robustly quantified as the intercept of a linear relationship between Δ¹⁸O_ls and Δ¹⁸O_lw. We subsequently applied such a regression method under highly controlled experimental settings to determine ε_{bio_A} in diverse plant species under different growth temperatures. The application of this new method allowed us to successfully constrain the estimate of ε_{bio_A} at 25°C to a narrow range of 26.4‰ ± 1.5 per mil across a range of plant species, closely aligning with the traditionally assumed value of 27‰. Additionally, a significantly inverse relationship of ε_{bio_A} with temperature was revealed from our experiment. Further comparisons will be made between our revealed temperature dependence of ε_{bio_A} with that of the heterotrophic factor ε_{bio_H} as reported in Sternberg and Ellsworth (2011). Our study represents a step forward in constraining isotope parameters in the δ¹⁸O_cel model, which has important implications for isotope-based paleoclimatic reconstruction and ecophysiological applications.