Hydrogen isotope ratios of leaf wax n-alkanes in grasses are insensitive to transpiration

We analyzed hydrogen isotope ratios of high-molecular weight n-alkanes (δD_l) and oxygen isotope ratios of α-cellulose (δ¹⁸O_C) for C₃ and C₄ grasses grown in the field and in controlled-environment growth chambers. The relatively firm understanding of ¹⁸O-enrichment in leaf water and α-cellulose was used to elucidate fractionation patterns of δD_l signatures. In the different relative humidity environments of the growth chambers, we observed clear and predictable effects of leaf-water enrichment on δ¹⁸O_C values. Using a Craig-Gordon model, we demonstrate that leaf water in the growth chamber grasses should have experienced significant D-enriched due to transpiration. Nonetheless, we found no effect of transpirational D-enrichment on the δD_l values. In field samples, we saw clear evidence of enrichment (correlating with relative humidity of the field sites) in both δ¹⁸O_C and δD_l. These seemingly contrasting results could be explained if leaf waxes are synthesized in an environment that is isotopically similar to water entering plant roots due to either temporal or spatial isolation from evaporatively enriched leaf waters. For grasses in the controlled environment, there was no enrichment of source water, whereas enrichment of grass source water via evaporation from soils and/or stems was likely for grass samples grown in the field. Based on these results, evaporation from soils and/or stems appears to affect δD_l, but transpiration from leaves does not. Further evidence for this conclusion is found in modeling expected net evapotranspirational enrichment. A Craig-Gordon model applied to each of the field sites yields leaf water oxygen isotope ratios that can be used to accurately predict the observed δ¹⁸O_C values. In contrast, the calculated leaf water hydrogen isotope ratios are more enriched than what is required to predict observed δD_l values. These calculations lend support to the conclusion that while δ¹⁸O_C reflects both soil evaporation and transpiration, δD_l appears to only record evaporation from soils and/or stems. Therefore, the δD of n-alkanes can likely be used to reconstruct the δD of water entering a leaf, supporting the soil-enrichment model of Smith and Freeman (2006). In both the field and controlled studies, we found significant photosynthetic pathway effects on n-alkane δD suggesting that biochemical pathways or plant phylogeny have a greater effect on leaf wax δD than leaf-water enrichment in grasses.