Ice versus Liquid Water Saturation in Regional Climate Simulations of the Indian Summer Monsoon

At the same temperature, below 0oC, the saturation vapor pressure (SVP) over ice is slightly less than the SVP over liquid water. Numerical models use the Clausius-Clapeyron relation to calculate the SVP and relative humidity, but there is not a consistent method for the treatment of saturation above the freezing level where ice and mixed-phase clouds may be present. In the context of current challenges presented by cloud microphysics in climate models, we argue that a better understanding of the impact that this treatment has on saturation-related processes like cloud formation and precipitation, is needed. This study explores the importance of the SVP calculation through model simulations of the Indian Summer Monsoon (ISM) using atmosphere-only simulations with the Regional Spectral Model (RSM) and RSM coupled to the Regional Ocean Modeling System (RSM-ROMS). Atmosphere-only simulations are conducted with two saturation parameterizations. In one, the SVP over liquid water is prescribed through the entire atmospheric column (woIce), and in another the SVP over ice is used above the freezing level (wIce). When SVP over ice is prescribed, a thermodynamic drying of the middle and upper troposphere above the freezing level occurs due to increased condensation. In the wIce runs, the model responds to the slight decrease in the saturation condition by increasing, relative to the SVP over liquid water only run, grid-scale condensation of water. Changes in the cloud layer amounts in the wIce simulation cause in increase in the net heat flux (NHF) at the surface of 2-3 W/m2 over the Arabian Sea (AS) and a decrease of similar magnitude over the eastern equatorial Indian Ocean (EEIO). Motivated by these NHF changes the wIce and woIce experiments were repeated in the coupled simulations. With coupling added, the ocean is allowed to respond to any NHF changes; however we find that the NHF difference between wIce-woIce over the AS is near zero. It is proposed that with the inclusion of air-sea coupling the atmospheric and oceanic response to changes in the SVP is damped relative to the forced RSM integrations. The importance of air-sea interaction for the northward propagation and evolution of the Indian monsoon intrareasonal oscillation (ISO) is examined through a comparison between the uncoupled and coupled simulations, and the observed ISO. It was found that the observed ISO contains a robust air-sea interaction during its evolution which would suggest that coupling is required to simulate the observed relationship between the ocean and atmosphere during the ISO. However, the uncoupled simulations show the ability to simulate realistic amplitude ISOs without coupling to the ocean, suggesting that there is an internal atmospheric component that is important for simulating the observed ISO period and amplitude.