Isotope effects accompanying vacuum extraction of soil water for stable isotope analyses

The vacuum distillation method of extracting soil water for stable isotope analysis was tested for three different types of soil characterized by high water content: (1) pure sand, (2) cambisol with high organic matter content, developed on calcareous sandstone under temperate climatic conditions (Austria), and (3) tropical latosol poor in organic matter, developed on sandy clay sediment (Brazil). The method yields accurate and reproducible results for sand, provided that more than 98% of the original soil water is extracted. The time required for complete extraction is a function of sample size and the applied extraction temperature. Column experiments with the clayey soils revealed existence of a weakly bound, easily exchangeable pool of water which is isotopically different from the mobile water. The experiments showed that the extracted soil water is depleted in both deuterium and oxygen-18 by 5–10% and 0.3–0.5%, respectively, when compared with the percolate (mobile water). This depletion depends strongly on the soil type. The reproducibility for replicate extractions of soil water from clayey soils is around ±3% and 0.3% for δD and δ 18 O, respectively.