Assimilation of lightning data through comprehensively nudging water contents at the cloud-resolving scale

Based on the two methods of assimilating lightning data that directly increase water vapor by Fierro et al. (denoted F12) and ice-phase particles by Qie et al. (denoted Q14), a combined method that comprehensively increases both water vapor and ice-phase particles (denoted C17) was presented. A squall line case with accurate total lightning observations was chosen to compare the effects of the three lightning assimilation methods mentioned above. Compared with the control experiment, the experiments that assimilated lightning data showed improved simulation for convection. However, the simulations differed in experiments with different assimilation methods. During the assimilation period, the convections were weak in the F12 experiment, while large areas of stratiform clouds formed and the convective core was located downstream of the observation. The intensity and location of convections simulated in the Q14 experiment agreed well with observations, but the simulation of statiform clouds was not improved. The convections in the C17 simulation strengthened with more accurate positions compared with that simulated in the F12, while a larger area of stratiform clouds formed compared to that in the Q14. After the assimilation, the cold pool strengthened while light precipitation extended northeastward in the F12 experiment, suggesting that the positive impacts of lightning assimilation was sustainable. The relatively dry lower-atmosphere in the Q14 resulted in a stronger cold pool and a fast-moving squall line that weakened quickly, while precipitation shifted southward compared to observations, indicating that the positive impacts of lightning assimilation faded soon. The C17 performed best in simulating the coverage and magnitude of the cold pool and the morphology of the squall line with long-lasting positive impacts of lightning assimilation, and large precipitation was successfully simulated.