The hydrogen and oxygen stable isotopic compositions (δ2H and δ18O) of terrestrial waters are usually considered equal to the amount-weighted annual mean δ values of precipitation, but this is not always the case. This isotopic shift relative to precipitation is introduced by the fractionation and/or selection effect. However, these two effects have never been evaluated separately, and insufficient data might lead to incorrect results. Here, we evaluated the isotopic shift over averaging periods of different lengths using 13 years of isotope data. The results showed that the isotopic shift based on one or a few years of data is significantly overestimated/underestimated since the interannual to decadal isotopic variability is greater in precipitation than in streamflow. An isotope-incorporated 1D vertical model of soil–vegetation–atmosphere transfer (IsoRHEA2) could reproduce the observed isotopic shift when the altitude effect is appropriately corrected. The positive isotopic shift due to the fractionation effect depends mainly on the vegetation conditions. In contrast, winter-precipitation-enhanced recharge introduces a negative shift via the selection effect. While these two opposite effects cancel each other out, the fractionation effect is superior to the selection effect in δ18O. It may introduce significant errors in estimating recharge elevations or other applications.
Adachi et al. (Thu,) studied this question.