A mechanistic model for interpretation of hydrogen and oxygen isotope ratios in tree-ring cellulose

A mechanistic model is presented to quantify both the physical and biochemi cal fractionation events associated with hydrogen and oxygen isotope ratios in tree-ring cellulose. The model predicts the isotope ratios of tree-ring s, incorporating both humidity and source water environmental information. Components;nts of the model include (1) hydrogen and oxygen isotope effects associated with leaf water enrichment; (2) incorporation of leaf water iso tope ratio values into photosynthetic carbohydrates along with the biochemi cal fractionation associated with autotrophic synthesis; (3) transport of e xported carbohydrates (such as sucrose) from leaves to developing xylem in shoots and stems where cellulose is formed; (4) a partial exchange of oxyge n and hydrogen isotopes in carbohydrates with xylem sap water during conver sion into cellulose; and (5) a biochemical fractionation associated with ce llulose synthesis. A modified version of the Craig-Gordon model for evapora tive enrichment adequately described leaf water delta D and delta(18)O valu es. The leaf water model was robust over a wide range of leaf waters for bo th controlled experiments and field studies, far exceeding the range of val ues to be expected under natural conditions. The isotopic composition of ce llulose was modeled using heterotrophic and autotrophic fractionation facto rs from the literature as well as the experimentally derived proportions of H and O that undergo exchange with xylem water during cellulose synthesis in xylem cells of tree-rings. The fraction of H and O from carbohydrates th at exchange with xylem sap water was estimated to be 0.36 and 0.42, respect ively. The proportions were based on controlled, long-term greenhouse exper iments and field studies where the variations in the delta D and delta(18)O of tree-ring cellulose were measured under different source water isotopic compositions. The model prediction that tree-ring cellulose contains infor mation on environmental water source and atmospheric vapor pressure deficit (related to relative humidity) was tested under both field and greenhouse conditions. This model was compared to existing models to explain cellulose isotope ratios under a wide range of source water and humidity conditions. Predictions from our model were consistent with observations, whereas othe r models showed large discrepancies as soon as the isotope ratios of source water and atmospheric water deviated from each other. Our model resolves t he apparently conflicting and disparate interpretations of several previous cellulose stable isotope ratio studies. Copyright (C) 1999 Elsevier Scienc e Ltd.