A passive method for sampling water in the soil-plant-atmosphere continuum for stable hydrogen and oxygen isotope analyses.

Abstract

Hydrogen and oxygen isotopes in water molecules are powerful tools to constrain the dynamics of water cycling within the soil-plant-atmosphere continuum (SPAC). However, the recovery of water from the SPAC requires logistical arrangements and implementation of different time- and cost-consuming techniques in either the field or the laboratory. We developed a passive method to sample water from the three compartments of the SPAC by using a hygroscopic salt of a high water absorbance capacity (CaCl2 ). This method allows either H2 O(V) -H2 O(L) isotope equilibration in the case of infinite water reservoir (atmospheric water vapor (WV)) or quantitative absorption of water from a finite water reservoir (e.g. soil and plants). The water absorbed by CaCl2 was distilled first and subsequently processed for hydrogen and triple oxygen isotope mass spectrometry analyses. The distillation step can be bypassed when employing isotope analytical techniques that are based on equilibration. Our experiments show that anhydrous CaCl2 absorbs WV of 210 ± 6% and 130 ± 6% of its dry weight from an infinite WV reservoir at relative humidity of 60% and 30%, respectively. Chemical and isotope equilibrations between WV and absorbed water were attained within 3 days at room temperature, enabling the back-calculation of the isotope composition of atmospheric WV. Preliminary experiments to extract water from plant and sand (i.e. finite WV reservoir) demonstrate a quasi-complete recovery of water in these matrices without significant isotope fractionation. The reproducibility of our method is better than 1.6‰, 0.32‰, 0.17‰ and 6‰ per meg for δ2 H, δ18 O, δ17 O and 17 O-excess. The CaCl2 -H2 O absorption (passive) method requires very limited logistics in the field facilitating spatial and temporal water vapor/water sampling from atmosphere and soil at low resolution (i.e. average of 3-5 days). Moreover, it allows high sample throughput for the extraction of plant water in the laboratory. The reproducibility of this method is similar to the analytical uncertainty in mass spectrometry analyses.