FIELD VALIDATION OF ISOTOPIC PROCEDURES FOR DETERMINING SOURCES OF WATER USED BY PLANTS IN A SEMIARID ENVIRONMENT

Methods for using stable isotopes of water (H-2, O-18) for determining the sources of water transpired by plants in a semi-arid field situation are validated. A comparison of the isotopic composition of the soil water in zones of high soil water potential (and hence high plant water availability) with that in plants shows that the overall approach is subject to a randomly distributed error of 5% in H-2 isotopic composition with no indication of any bias. This includes errors involved in the extraction of water from plants and soil, in spatial heterogeneity in soil profiles, in heterogeneity within plant canopies, in the time of travel of water through the plant, and in assumptions about fractionation of isotopes during water uptake by plants. These errors are generally smaller than natural variations of isotope, thus allowing the isotopic method to be used. Similarly, an estimate for the error in O-18 is 1%. A compartmental model was used to quantify errors associated with an inversion model using isotopic data of water coming from two soil layers. The error in the estimation of the fraction of water taken from each compartment was 20%. For many hydrological purposes, this error is acceptable. This is the first study estimating the errors associated with the method.