Impact of soil moisture variability on convective rainfall activity over the Indian sub-continent

Soil moisture is an important geophysical parameter affecting land atmosphere processes, and hence free convection, by controlling the partitioning of the surface heat flux into latent and sensible heat flux. Interaction between these fluxes and the atmosphere gives rise to different types of soil moisture-precipitation feedback, namely “wet advantage” where rain is favoured over a wet (high latent heat flux) surface and “dry advantage” where rain is favoured over a dry (high sensible heat flux) surface. Previous studies over different parts of the world have shown that these feedback processes can take different pathways, according to one-dimensional and three-dimensional models. According to the one-dimensional model there is probability of rain initiation when the boundary layer top meets the level of free convection either by heating (increase in sensible heat flux over a dry surface) or by moistening (increase of latent heat flux over wet soil) of the boundary layer. On the other hand three-dimensional models explain convective triggering due to wind convergence near gradients in soil moisture. This is a first study to compare and evaluate the existing soil moisture-precipitation feedback theories presented in the literature, over the Indian sub-continent under a single environment, by using high resolution convection-permitting (non-parameterized, or “explicit” convection) EMBRACE model simulation. Initially, a brief synoptic observational study shows evidence of surface-atmosphere coupling. More detailed case studies from the model output show further evidence for the land-atmosphere interaction in this region. The model indicates that all the processes defined by different theoretical models do exist under different surface, and atmospheric conditions. The relative contribution of different processes under different soil moisture conditions prevailing over different climatic zones of the Indian sub-continent during the 20-day wet monsoon period from mid-July to early August is statistically studied. Dry-to-wet downwind soil moisture gradient is found to be the statistically significant pattern for initiation of the majority of afternoon convective initiation in the East, Centre and South study domains of India. It is also found that the so-called “CTP-HIlow” predictive framework is not sufficient to address the observed behaviour of convective initiation under the full three-dimensional modelling environment. The use of the parameter HIlow, which is defined as the sum of humidity within and just above the inversion, as a predictive parameter is not physically understandable. This framework also lacks generality and solutions are empirically derived based on one-dimensional modelling and observations, which vary from place to place. To offer a solution to these theoretical difficulties, this study provides a new quantitative model, using the basic idea behind the CTP-HIlow framework to find new predictive parameters depending on sound physical relationships instead of empirical solutions. The system is governed by two non-dimensional parameters, namely inversion Bowen ratio and a “stiffness ratio”, and a third, dimensional parameter ΔR. Analysis of the EMBRACE simulations shows occurrence of both the dry and wet advantage, but the majority of the morning profiles favour prediction of dry advantage. Thus, the equations derived from the new quantitative model offer a quantitative prediction of wet and dry advantage occurring systematically, which is a question of great importance to weather and climate prediction, especially over moisture-limited areas.