[M-IS23] Fusion of volcanology and meteorology

[MIS23-03] Weather Radar Detection of Sakurajima Volcanic Eruptions in Precipitation

*Masayuki Maki¹, Haruhisa Nakamichi¹, Masato Iguchi¹, Takaaki Nishi², Yasushi Fujiyoshi³, Eiichi Sato⁴, Takehiko Kobori⁵, Kazuhiko Kaiga⁶, Hidehiko Tokushima⁷ (1.Disaster Prevention Research Institute, Kyoto University, 2.Faculty of Fisheries, Kagoshima University, 3.Emeritus professor, Hokkaido University, 4.Meteorological Research Institute, Japan Meteorological Agency, 5.Mitsubishi Electric Software Corporation, 6.Koden Electronics Co., Ltd., 7.Field Researchers Corporation)

Keywords:volcanic eruption, precipitation, polarimetric radar, echo top height

It has been shown by researches both in the country and overseas that weather radar is one of new monitoring tools for volcanic eruption clouds. One of remaining questions is whether the radar can detect an eruption in precipitation (hereafter we call ‘wet eruption’). To answer this question, we examine the radar data that were collected during volcanic eruptions in precipitation conditions. Radars that we investigated are the MLIT X-band polarimetric radar at Tarumizu (TRM-XMP), the DPRI/Kyoto University X-band polarimetric radar at SVO (SVO-XMP), and the X-band marine radars at SVO (SVO-MARINE), and Kurokami Observatory/Kyoto University (KRK-MARINE). The investigated eruption cases are chosen from the Japan Meteorological Agency's report on Sakurajima eruptions. A typical example of analyzed wet eruptions is the eruption that occurred at 02:59 on June 4, 2020. Figure 1 shows the distribution of time-integrated radar reflectivity factors calculated from PPI scan data of TRM-XMP radar. We can recognize the band-shaped eruption echo that originated from the Minamidake crater and extending eastward in the wide-spreading precipitation echoes. Figure 2 shows the density plots of the size and fall velocity of particles measured by disdrometers installed at different places. It can be seen that the fluctuation in the particle fall speeds observed in the plume area is large compared to that in the precipitation area. The distribution data surrounded by the dashed lines in Fig. 2 (a), (b), and (c) is considered to be mud-rain particles. Figure 3 shows the temporal change of the azimuthally integrated radar reflectivity factor obtained from the sectoral RHI scans of SVO-XMP. The radar captures the plume column immediately after the eruption onset and the temporal changes thereafter. It is turned out that the plume top height, which was supposed to be higher than 1500 m, is actually higher than 7000 m. As a result of investigating all the extracted eruption cases, we found that 1) the radar detectability of a wet eruption depends on the scale of eruption and the intensity of precipitation, 2) RHI observations by the X-band marine radar is the most effective for detecting wet eruptions, 3) the sectoral RHI by the SVO-XMP radar is effective for investigating the vertical structure in the plume column, and 4) PPI observations by the TRM-XMP radar is effective for estimating the ash fall amount, area, and ash fall durations.