Correction of cryogenic vacuum extraction biases and potential effects on soil water isotopes application

Abstract

Cryogenic vacuum extraction (CVE) is the most widely used method for obtaining soil water for water stable isotope (2H and 18O) analysis; however, recent studies have shown that it may result in biased isotopic compositions. Here, we presented a method to correct the CVE biases and to assess the influence of the corrections on the conclusions from three different soil water isotope applications in hydrology. Four soil samples with different textures were oven-dried at 205 °C, wetted to varying water contents (0.05–0.30 g g−1) with reference water of known isotope composition, and then subjected to CVE for water extraction. The isotopic differences (biases) between the reference and extracted water were related to the soil textures and water contents. Then, the relationship was used to assess how the cryogenic extraction biases potentially influence soil water isotope applications based on data from three previous studies. The results showed that the biases for 84% CVE samples were more than twice the standard errors of currently available isotope analyzers, and biases increased significantly with increasing clay and decreasing water contents. Correction equations for CVE-obtained δ2H and δ18O were established: R2 = 0.84, P < 0.01, and R2 = 0.60, P < 0.01, respectively. When these corrections were applied to three different scenarios, they led to conclusions significantly different from those established with uncorrected values. This study provides a correction method for soil water vacuum extraction bias to improve the effectiveness of 2H and 18O applications in soil hydrology. We suggest that cryogenic extraction biases should be considered seriously for quantitative analysis under low water contents and with a narrow spread of isotopic compositions among the potential end-members.