Lidar Observations and Data Assimilation of Low-Level Moist Inflows Causing Severe Local Rainfall Associated with a Mesoscale Convective System

Abstract

Abstract We conducted an observational survey using a ground-based water vapor Raman lidar (RL) during the warm season in Japan to investigate the water vapor structure of low-level inflows that contribute to the formation of a mesoscale convective system (MCS). After the passage of a warm front, low-level moisture convergence contributed to the initiation and development of numerous convective clouds that composed the MCS. The RL observations showed that the vertical profiles of the water vapor mixing ratio (WVMR) associated with low-level inflows into the MCS exceeded 20 g kg−1 below 500 m above sea level, which is comparable to WVMRs in previous reports associated with MCSs in Japan and the United States. We conducted two assimilation experiments using a four-dimensional variational data assimilation system: one is to assimilate operational observational data (CNTL), and the other is to assimilate WVMR vertical profiles and operational observational data (TEST). A comparison between TEST and CNTL showed that data assimilation of the WVMR vertical profiles not only modified the moisture field but also the wind field. It appears that the modifications observed in horizontal wind are related to the modification of the WVMR in the analysis fields. These WVMR and wind modifications improved the reproduction of the frontal surface and forecasting of 6-h precipitation amount slightly. Data assimilation of vertical profiles of the WVMR has positive and negative impacts on the WVMR and horizontal wind, respectively, implying that the vertical profiles of both the horizontal wind and the WVMR might better estimate initial conditions and forecasts. Significance Statement Low-level moisture inflows are one of the key parameters involved in the formation of mesoscale convective systems (MCSs). Therefore, data assimilation of low-level moisture profiles is one of the prospective methods for better forecasting heavy precipitation associated with MCSs. However, few direct observations of the low-level moisture structure associated with MCSs and data assimilation experiments have been undertaken to date. We observed the vertical profiles of moisture associated with an MCS in Japan using a ground-based water vapor Raman lidar and show the existence of a relatively moist low-level inflow into the MCS. The data assimilation of low-level moisture has positive and negative impacts on moisture and horizontal wind, respectively, and improves slightly 6-h precipitation forecasts.