Atmospheric processes governing the changes in water isotopologues during ENSO events from model and satellite measurements

ENSO (El Niño-Southern Oscillation) has profound effects on the global water cycle, which can be examined at the process level by investigating the associated water isotopologues. Many isotope-based studies are aimed at understanding ENSO variability in the tropics, however, focusing principally on near-surface processes and isotopologue signals. The goal of the present study is to investigate the atmospheric processes governing the changes in the isotopic composition of water vapor both near the surface and at midtroposphere in the Pacific region during ENSO events, using a combination of remote sensing data and model simulations. For the lower atmosphere (i.e., 1000 hPa), our results show that rainout processes, less rain reevaporation of falling droplets, and increase of convective updrafts and diffusive exchange within the convective systems contribute to "the isotope amount effect" and isotopically deplete the water vapor during wet conditions, in agreement with previous studies. However, we find that the ENSO-associated isotopic signal in the midtroposphere (i.e., 500 hPa) diverges from the near-surface response. Analysis suggests that transport of enriched water vapor from lower atmospheric layers through convective updrafts controls the enrichment of midtropospheric water vapor over the Pacific Ocean. In the observations, a strong positive correlation between the increase of convective precipitation and the isotopic composition of water vapor clearly points to such a mechanism (R of 0.7-0.8 in the Central Pacific and 0.5-0.6 in the West Pacific). Model results confirm this mechanisms though producing slightly lower correlation values, with R values of 0.6 in the Central Pacific and 0.5 in the West Pacific. However, the distinction between convective and stratiform precipitation remains a result of model-dependent parameterization. Our analysis suggests that two issues should be investigated in more detail in further studies: (1) the equilibrium and disequilibrium between rain droplets and surrounding vapor for convective and stratiform precipitation and (2) different convection schemes in the different isotopic general circulation models (GCMs) describing the triggering of convection and uplift of lower layer air to higher layers. Ideally, such a comparison of different isotopic GCMs can provide us with an interesting benchmark test for the performance of the different convection schemes during ENSO and can help to disentangle the importance of the different processes contributing to the amount effect.