Technical note: Water vapor sampling for stable isotope analysis: systematic testing of sampling conditions and development of a robust protocol

Ecohydrological water isotope studies often rely on destructive sampling of soils and plants. Recently, techniques for collecting equilibrated water vapor were developed as alternative. Here, we present a systematic evaluation of water vapor sampling methods with the goal of identifying how key parameters influence δ18O and δ2H values. In controlled laboratory experiments we first tested the isotopic stability of three container types: 250 mL infusion glass bottles, 1 L FlexFoil sample bags, and 500 mL Aluminum-zip bags. The most suitable container was subsequently tested for different storage times (6 h to 7 d), storage temperatures (4 to 40 °C) and sampling flow rates (35 to 125 mL min−1). The 250 mL infusion glass bottles showed the best performance, with deviations to reference standards of ±0.5 ‰ for δ18O and ±1 ‰ for δ2H followed by FlexFoil sample bags (deviations of ±1 ‰ for δ18O and −3 ‰ to +2 ‰ for δ2H). Aluminum bags showed the largest deviations (−2.5 ‰ to −3 ‰ for δ18O and −12 ‰ to −25 ‰ for δ2H). Further extensive testing of the most suitable container type – the infusion glass bottles – revealed that the Mean Absolute Error (MAE) for δ18O remained stable under all tested conditions (<0.6 ‰). For δ2H, best results were obtained using flow rates of 100–125 mL min−1, and storage times of ≤1 d under ambient conditions (20–25 °C) (MAE =±0.5 for δ18O and ±1.5 ‰ for δ2H); hence, our recommendation is using these settings. Although vapor sampling cannot match the analytical precision of conventional methods, it offers a practical, inexpensive framework. The protocol enables reliable water vapor isotope measurements and is especially advantageous in remote areas or locations with limited infrastructure. Therefore, it has great potential for investigating ecohydrological processes in a customizable spatiotemporal resolution without the need of repeated destructive sampling or full in situ setups.