Comment on bg-2022-230

Abstract

<strong class="journal-contentHeaderColor">Abstract.</strong> The triple oxygen isotope composition of phytoliths (¹⁷O-excess_phyto) can provide key information on past atmospheric relative humidity (RH) over land. Here, we examined how leaf-to-air temperature gradients and changes in the silica polymerization rate in response to stomatal conductance influence the interpretation of ¹⁷O-excess_phyto in terms of RH. Further, we assessed the reliability of a theoretical isotope model of leaf water evaporation to predict the triple oxygen isotope composition of leaf water on diurnal and seasonal scale. For this purpose, we monitored a grass plot within a natural Mediterranean woodland for one year. We measured in particular the isotope composition of atmospheric water vapor and plot-scale grass leaf temperatures &ndash; two variables that are often only estimated. Grass leaf blades were collected in different seasons and over a 24-hour period for leaf water and phytolith isotope analysis. We found that the steady state model reliably predicts the triple oxygen isotope composition of leaf water during daytime but remains sensitive to uncertainties on the leaf-to-air temperature difference. Deviations from isotope steady state at night are well represented by the non-steady state model. In our study, the ¹⁷O-excess_phyto best reflects average daytime RH over the growth period, rather than daily RH. Average daytime leaf-to-air temperature gradients of less than 2&thinsp;&deg;C introduce an insignificant bias to the RH estimate. The results also confirm the established triple oxygen isotope fractionation factors between phytoliths and leaf water. The findings of this study help to better understand how to interpret ¹⁷O-excess_phyto of fossil phytolith assemblages in terms of past RH.