11:00 AM - 1:00 PM
[U09-P25] Eruption Cloud from Hunga Tonga–Hunga Ha′apai on 15 January 2022 Observed by Meteorological Satellite Himawari-8
Keywords:geostationary meteorological satellites, Himawari-8, HTHH, undercooling, atmospheric gravity wave, blast cloud
Umbrella-shaped eruption cloud and subsequent wind-blown volcanic ash cloud from submarine volcano Hunga Tonga–Hunga Ha′apai (HTHH) on 15 January 2022 were observed by the geostationary meteorological satellite Himawari-8 (e.g. Hayashi et al., 2016, 2018; Smart, 2022). The latter volcanic ash cloud was analyzed in the previous report (Shimbori et al., 2022). In this presentation, we report on the features of the former eruption cloud as well as Smart (2022).
The main results of the Himawari-8 imagery analysis of the HTHH eruption cloud are as follows (indicated time is the one when each satellite image was taken around the volcano):
(i) The origin of the eruption cloud is confirmed at 04:07 UTC in the visible (VIS) image (Fig. 1 (a)).
(ii) The first cloud during the 10-min period from 04:07 to 04:17 UTC had a directivity, and its horizontal velocity averaged more than 64 m/s to the east-southeast from the volcano (Figs. 1 (a), (b)).
(iii) The beginning of the umbrella cloud formation was confirmed at 04:27 UTC in the VIS and infrared (IR) imagery. In the IR (10.4 μm) image, the minimum brightness temperature (BT) of −96.9℃ was observed at 47 km east-southeast of the volcano (Fig. 1 (c)).
(iv) In the satellite imagery from 04:27 to 04:37 UTC, propagating waves circularly were confirmed from the minimum BT observation point (Figs. 1 (c), (d)).
(v) In the satellite imagery at 04:47 UTC, the center of the umbrella cloud approached above the volcano, and the maximum BT of +2.0℃ was observed near the center (12 km southeast of the volcano, Fig. 1 (e)).
(vi) The contour of the umbrella cloud was detected clearly in the split window (SP, 10.4−12.4 μm, e.g. Fig. 1 (e′)). The equivalent radius r, calculated by measuring the umbrella-shaped area, changed with elapsed time t since the eruption (assumed) as shown in the left panel of Fig. 2.
The results of the above analyses (i) – (vi) are discussed below:
(i) Based on the Himawari-8 observation time interval, the eruption is estimated to have started between 03:57 and 04:07 UTC.
(ii) During the period, the first cloud may have been accompanied by upward velocity as its BT temperature fell, and the instantaneous velocity is estimated to be much faster. The Japan Meteorological Agency (JMA) global analysis shows that no such strong westerlies were blowing (the right panels of Fig. 2 of Shimbori et al. (2022)), thus it cannot be considered primarily a wind-driven cloud.
(iii) The observed minimum BT is more than 16℃ cooler than the air temperature near the tropopause based on the JMA analysis (ibid., the left panel of Fig. 2), which is considered to be cloud-top undercooling (Woods and Self, 1992; Sparks et al., 1997) through adiabatic expansion.
(iv) Propagating waves in the umbrella cloud has been observed in eruptions such as Mount Pinatubo on 15 June 1991 (e.g. Tokuno, 1991), and are considered to be atmospheric gravity waves.
(v) The large change of BT was detected due to increased temporal and spatial resolutions of satellite observations of the cloud-top surface where the spatial resolution of Himawari-8 IR imagery has increased to 2 km (at the sub-satellite point).
(vi) The exponent of the elapsed time t depends on where the starting point is taken. If the starting point of t is set around the beginning of the umbrella cloud formation (Fig. 1 (c)), the t-dependence of the radius r is smaller than the exponent shown in the left panel of Fig. 2, after which the axisymmetric umbrella cloud may follow a power law predicted by an incompressible and inviscid gravity current model (e.g. Woods and Kienle, 1994; Sparks et al., 1997; Koyaguchi, 2008). The subsequent decrease in the tendency (the right panel of Fig. 2) estimates that the duration of intrusion into the umbrella region is shorter than 40-60 mins.
The features observed by Himawari-8 in the formation of the umbrella cloud caused by the submarine eruption at HTHH are well analogous to the case of Mount St. Helens eruption on 18 May 1980 (Sparks et al., 1986; Holasek and Self, 1995), and what was observed in the immediate aftermath of this eruption (Fig. 1 (b)) may be a blast cloud (Lipman and Mullineaux, eds., 1981; Woods and Wohletz, 1991). Note that an immediate calculation of undercooling cloud-top heights at any time of day or night is required in the JMA′s tephra transport forecasting operations, for which a combination of Himawari IR imagery and one-dimensional eruption column models (e.g. Ishii et al., 2021) is useful.
Acknowledgements
We used ″SATAID″ imagery analysis software developed by the MSC (Meteorological Satellite Center, JMA).
References
See Japanese Proceedings.
The main results of the Himawari-8 imagery analysis of the HTHH eruption cloud are as follows (indicated time is the one when each satellite image was taken around the volcano):
(i) The origin of the eruption cloud is confirmed at 04:07 UTC in the visible (VIS) image (Fig. 1 (a)).
(ii) The first cloud during the 10-min period from 04:07 to 04:17 UTC had a directivity, and its horizontal velocity averaged more than 64 m/s to the east-southeast from the volcano (Figs. 1 (a), (b)).
(iii) The beginning of the umbrella cloud formation was confirmed at 04:27 UTC in the VIS and infrared (IR) imagery. In the IR (10.4 μm) image, the minimum brightness temperature (BT) of −96.9℃ was observed at 47 km east-southeast of the volcano (Fig. 1 (c)).
(iv) In the satellite imagery from 04:27 to 04:37 UTC, propagating waves circularly were confirmed from the minimum BT observation point (Figs. 1 (c), (d)).
(v) In the satellite imagery at 04:47 UTC, the center of the umbrella cloud approached above the volcano, and the maximum BT of +2.0℃ was observed near the center (12 km southeast of the volcano, Fig. 1 (e)).
(vi) The contour of the umbrella cloud was detected clearly in the split window (SP, 10.4−12.4 μm, e.g. Fig. 1 (e′)). The equivalent radius r, calculated by measuring the umbrella-shaped area, changed with elapsed time t since the eruption (assumed) as shown in the left panel of Fig. 2.
The results of the above analyses (i) – (vi) are discussed below:
(i) Based on the Himawari-8 observation time interval, the eruption is estimated to have started between 03:57 and 04:07 UTC.
(ii) During the period, the first cloud may have been accompanied by upward velocity as its BT temperature fell, and the instantaneous velocity is estimated to be much faster. The Japan Meteorological Agency (JMA) global analysis shows that no such strong westerlies were blowing (the right panels of Fig. 2 of Shimbori et al. (2022)), thus it cannot be considered primarily a wind-driven cloud.
(iii) The observed minimum BT is more than 16℃ cooler than the air temperature near the tropopause based on the JMA analysis (ibid., the left panel of Fig. 2), which is considered to be cloud-top undercooling (Woods and Self, 1992; Sparks et al., 1997) through adiabatic expansion.
(iv) Propagating waves in the umbrella cloud has been observed in eruptions such as Mount Pinatubo on 15 June 1991 (e.g. Tokuno, 1991), and are considered to be atmospheric gravity waves.
(v) The large change of BT was detected due to increased temporal and spatial resolutions of satellite observations of the cloud-top surface where the spatial resolution of Himawari-8 IR imagery has increased to 2 km (at the sub-satellite point).
(vi) The exponent of the elapsed time t depends on where the starting point is taken. If the starting point of t is set around the beginning of the umbrella cloud formation (Fig. 1 (c)), the t-dependence of the radius r is smaller than the exponent shown in the left panel of Fig. 2, after which the axisymmetric umbrella cloud may follow a power law predicted by an incompressible and inviscid gravity current model (e.g. Woods and Kienle, 1994; Sparks et al., 1997; Koyaguchi, 2008). The subsequent decrease in the tendency (the right panel of Fig. 2) estimates that the duration of intrusion into the umbrella region is shorter than 40-60 mins.
The features observed by Himawari-8 in the formation of the umbrella cloud caused by the submarine eruption at HTHH are well analogous to the case of Mount St. Helens eruption on 18 May 1980 (Sparks et al., 1986; Holasek and Self, 1995), and what was observed in the immediate aftermath of this eruption (Fig. 1 (b)) may be a blast cloud (Lipman and Mullineaux, eds., 1981; Woods and Wohletz, 1991). Note that an immediate calculation of undercooling cloud-top heights at any time of day or night is required in the JMA′s tephra transport forecasting operations, for which a combination of Himawari IR imagery and one-dimensional eruption column models (e.g. Ishii et al., 2021) is useful.
Acknowledgements
We used ″SATAID″ imagery analysis software developed by the MSC (Meteorological Satellite Center, JMA).
References
See Japanese Proceedings.