*Rie Nakazawa1, Taro Shinoda1, Tadayasu Ohigashi2, Kosei Yamaguchi3, Haruya Minda1, Moeto Kyushima1, Kazuhisa Tsuboki1, Eiichi Nakakita3
(1.Institute for Space-Earth Environmental Research (ISEE), Nagoya University, 2.National Research Institute for Earth Science and Disaster Resilience (NIED), 3.Disaster Prevention Research Institute (DPRI), Kyoto University)
Keywords:Ka-band cloud radar, early detection, merge
Rapidly developing convective precipitation clouds has short lifespan which is approximately 30 to 60 minutes and sometimes they afford heavy rainfall at surface. This torrential rainfall brought about flash floods and cause terrible damage, for example, five people lost their lives in Toga River in July 2008. This disaster recognized us the importance of early detection and identification of developing convective precipitation clouds. In addition, that is one of the significant goals for the radar meteorology. A Ka-band radar enable us to detect early development of precipitating clouds by utilizing shorter wavelength radio wave (about 8.6 mm) in comparison with typical meteorological radars (e.g., X-band). We installed a Ka-band radar at Kobe International University (Kobe City, Japan) under the research project “Integrated research on state-of-the-art multi-sensors in-situ observation of storm genesis and reduction of serious disaster due to heavy rainfall” in summer season in 2018. The Ka-band radar can detect -17 dBZ as the minimum reflectivity at 20 km distance. Continuous sector PPI (Plan Position Indicator) observations by 90 deg. azimuthal scan with 11 elevation angles were conducted every 2 min. In this study, we chose 8 elevation angles for analysis to focus on detection of initial developing stage and avoid effect of ground clutter. Reflectivity values obtained by the Ka-band radar are projected on the horizontal plane and vertically averaged reflectivity (VAR) is calculated at each grid (horizontally 50 m times 50 m size). We define the developed convective echo as the maximum rainfall intensity exceeds 20 mm/h in an isolated convective precipitation region obtained every 1 min by the X-band radar network. Total 14 convective echoes are analyzed by time series of VAR on August 6 and 16, 2018. Two echoes are evaluated as the developed convective ones. We cannot recognize the significant difference of time series of maximum value of VAR between developed echoes and non-developed ones (Figure 1). However, time series of the area of VAR of developed echoes are clearly larger than those of non-developed ones (Figure 2). By confirming time series of VAR, the developed echoes tend to merge the surrounding small/weak echoes and rapidly increase their area. As the increasing rate of the area of VAR tends to exceed 4 km2 in 2 min, the convective echoes develop as the developing precipitation cell detected by an X-band radar. As a result, using high sensitivity and frequent continuous PPI observations, we have possibility to conduct early detection of rapidly developing convective echoes.