Abstract
Precipitation isotopes are valuable tracers for understanding the hydrologic cycle and climate variations. Distinct from d-excess, 17O-excess has recently emerged as a promising new tracer of precipitation processes because of its insensitivity to moisture source temperature. However, the control mechanisms on precipitation 17O-excess remain poorly understood. In this study, we evaluated the performance of the GISS-E2.1 climate model in simulating the precipitation isotopes, focusing on 17O-excess. Through comprehensive analysis, we explored how variations in seawater isotopes, rain evaporation, kinetic isotope fractionation parameters, and supersaturation factors influence the simulated precipitation d-excess and 17O-excess. Our findings reveal that GISS-E2.1 accurately captures the spatial distribution and temporal variations of precipitation δ18O. Moreover, it reasonably reproduces the spatial patterns of precipitation d-excess, though slightly underestimating the mean value in the low latitudes. Although most simulated 17O-excess values fall within the observed range, evaluating the accuracy of 17O-excess simulations is challenging due to the limited availability of observational data. Notably, in tropical regions, the spatiotemporal distributions of d-excess and 17O-excess are sensitive to convective processes, such as rain evaporation. The model's limitations in 17O-excess simulation suggest that current formulations are inadequate to fully capture the variability of 17O-excess. This underscores the complexity of the processes influencing 17O-excess and highlights the need for additional data and further research to comprehensively understand its controlling factors. Our findings contribute to our understanding of the mechanisms driving the observed variation in precipitation triple oxygen isotopes and to the validation and improvement of climate models.