Mobile Water Vapor Raman Lidar for Localized Heavy Rainfall Forecasting in Japan

In recent years, the occurrence frequency of localized heavy rainfall that can produce flash floods and landslides, has been increasing in Japan. To improve forecast accuracy and lead time (period of time between the issuance of a forecast and the occurrence of the event) of heavy rainfall, we developed a mobile Raman lidar (MRL) system for measuring the vertical distribution of the water vapor mixing ratio in the lower troposphere. The MRL was installed in a small trailer for easy deployment to the upwind side of potential rainfall areas to monitor the inflow of humid air that is expected to precede rainfall event.

In this presentation, we describe the MRL system and present validation results obtained by comparing the RL-measured data with collocated radiosonde and Global Navigation Satellite System (GNSS) data. The comparison was also performed by using a high-resolution objective analysis data of Japan Meteorological Agency (JMA).

Furthermore, we show a preliminary result of the MRL data assimilation experiment on the heavy rainfall occurred on the northern part of Kanto Plain, Japan, on 17 August 2016, to investigate the impact of the lidar data on the heavy rainfall forecast. In the experiment, the MRL-derived water vapor mixing ratio profiles were assimilated to the JMA nonhydrostatic mesoscale model with a local ensemble transform Kalman filter (NHM-LETKF). The results of the experiment showed that, even though the timing of the simulated heavy rainfall preceded the observation by 3 hours, the accuracies of the simulated humidity field and rainfall intensity improved significantly.