Home / Our Science / Natural Hazards and Risks / Tsunami / 2022 Hunga Tonga Tsunami

Hunga Tonga - Hunga Ha’apai volcano-induced sea level oscillations and tsunami simulations

Aditya Gusman and Jean Roger

How to cite : Gusman, A.R. & Roger, J. (2022). Hunga Tonga - Hunga Ha’apai volcano-induced sea level oscillations and tsunami simulations. GNS Science webpage, Accessed at https://doi.org/10.21420/DYKJ-RK41 on [insert date].

The eruption of Hunga Tonga – Hunga Ha’apai volcano on 15 January 2022 resulted in widely observed sea level changes at coastal gauges and DART buoy stations around the world, and particularly in the Pacific. The volcano is a part of the Tonga-Kermadec arc and is located ~67 km north-northwest of Nuku’alofa at 20.57° S and 175.38° W, the capital city of the Kingdom of Tonga (Colombier et al., 2018; Garvin et al., 2018). Based on the Himawari-8 satellite images the largest eruption occurred sometime between 4:00 – 4:10 UTC, and a tsunami may have been generated within this time (volcano.ssec.wisc.edu). For the tsunami arrival time comparison below, we assumed that the tsunami was generated at 4:00 UTC. Tides and short period (<2 min) waves were removed from the original record to reveal tsunami-like waves, but the filtered waves may not consist purely of tsunami waves that have propagated from the source.

The arrival time, and peak amplitude of the first wave recorded at the New Zealand DART stations (Fry et al., 2020) are shown in Table 1 and Figure 2.  The maximum amplitude of the waves recorded by coastal gauges are shown in Table 2, Figure 3, and Figure 4. At DART NZG, the closest New Zealand DART from the volcano, the maximum amplitude was ~20 cm, and the largest crest-to-trough height was ~40 cm (Figure 2). The largest observed amplitude was typically not in the first wave and often occurred several hours after the initial wave (Figure 3 and Figure 4). Some components of these waveforms may be a consequence of pressure changes caused directly by the volcanic shockwave rather than changes in water surface level, though this is still being evaluated (see below).

Generally, the records show a set of long-period waves either followed or superimposed with higher frequency waves. In some locations (e.g. PKEM-Port Kembla and TBWC-Twofold Bay, Australia, SPRG-Spring Bay, Tasmania, and DART NZB, New Zealand), these first waves have a recorded period of ~1 hour. In other places, these first waves show periods of several minutes to a few tens of minutes (e.g. HUAHI-Huahine, French Polynesia, LIFO-Lifou, New Caledonia). This may be an indication that the waves propagated in a non-isotropic pattern (and perhaps also that some locations are sensitive to different components of the propagating waves). At the time of tsunami propagation a cyclone (Cody) was located (https://zoom.earth/#view=-39.5,161.4,5z/date=2022-01-15,pm/layers=daily) between Tonga and New Zealand, consequently the coastal gauges in New Zealand recorded the combined influences of both the tsunami and the cyclone. 

Tsunami Simulation

There are at least eight known volcanic source mechanisms which include 1) underwater explosion, 2) pyroclastic flow, 3) earthquake, 4) flank failure, 5) caldera subsidence, 6) shock wave, 7) lahar, and 8) collapse of lava bench (Paris et al., 2014, 2015). The tsunami from Hunga Tonga – Hunga Ha’apai volcano eruption could have been generated by any or a combination of these mechanisms. We simulated the tsunami using a simple circular sea surface deformation model with a characteristic diameter of 10 km. This kind of circular model can be used to roughly model an underwater explosion or a caldera collapse mechanism. A tsunami numerical model that solves the linear shallow water equations (Satake, 1995) was used to simulate the tsunami. The simulated long waves were then corrected using the Watada et al. (2014) method to include dispersion effects.

The uplift of initial sea surface deformation model was adjusted to fit the tsunami amplitude recorded at DART NZG, the closest DART buoy station from the volcano. However, the arrival time and amplitude of the simulated waveforms cannot match the observations at the other DART buoys. We found that the arrival time difference between the observed and simulated wave becomes larger at DART stations farther away from the source. As an example, the tsunami at DART NZE arrived about 30 min earlier than the simulation, and the observed peak amplitude is much larger than the simulated one (Figure 6). This result suggests that the tsunami source mechanism was very complex. The earlier observed tsunami arrival time may suggest that a component of the tsunami source could be from the shock wave of the eruption. This is also suggested by the far-field observations of sea-level oscillations in regions like the Mediterranean and the Caribbean (e.g. Palermo in Sicily and Carloforte in Sardinia for the Mediterranean, Fort de France in Martinica for the Caribbean; Figure 7) that are well-protected from any tsunami from Tonga propagating through the ocean. Other mechanisms that could have been contributed to the tsunami generation in this event include underwater explosion, pyroclastic flow, caldera collapse and flank failure. The investigation of the tsunami source mechanism is currently still underway.

Figure 1. DART stations (green triangles) and theoretical tsunami (long wave) travel time (TTT) map.  The contour intervals for the TTT is 1 hour. Blue star represents the location of Hunga Tonga - Hunga Ha'apai volcano.

Figure 1. DART stations (green triangles) and theoretical tsunami (long wave) travel time (TTT) map. The contour intervals for the TTT is 1 hour. Blue star represents the location of Hunga Tonga - Hunga Ha'apai volcano.

Figure 2 Waveforms recorded at New Zealand DART buoy stations

Figure 2 Waveforms recorded at New Zealand DART buoy stations

Figure 3 Waveforms recorded at New Zealand coastal gauges

Figure 3 Waveforms recorded at New Zealand coastal gauges

Figure 4. Waveforms recorded at stations around the South West Pacific.

Figure 4. Waveforms recorded at stations around the South West Pacific.

Figure 5. Maximum wave amplitude recorded at coastal gauges around the South West Pacific on 15 January 2022.

Figure 5. Maximum wave amplitude recorded at coastal gauges around the South West Pacific on 15 January 2022.

Figure 6. Comparison between simulated (blue lines) and observed (orange lines) tsunami waveforms at DART stations. The time is in hours after 15 January 2022 4:00 UTC.

Figure 6. Comparison between simulated (blue lines) and observed (orange lines) tsunami waveforms at DART stations. The time is in hours after 15 January 2022 4:00 UTC.

Figure 7. Sea-level variations recorded at Palermo (Sicily, IT), Carloforte (Sardinia, IT) and Fort-de-France (Martinique, FWI) coastal gauges following the Tonga volcano eruption. These three gauges are respectively located ~18,000 (for the first two) and ~13,000 km away from the volcano.

Figure 7. Sea-level variations recorded at Palermo (Sicily, IT), Carloforte (Sardinia, IT) and Fort-de-France (Martinique, FWI) coastal gauges following the Tonga volcano eruption. These three gauges are respectively located ~18,000 (for the first two) and ~13,000 km away from the volcano.

Table 1. Peak and trough amplitudes of the first wave observed at DART stations

Station Lat Lon First peak (cm) First trough (cm) Distance from HTHH volcano (km) Arrival time of the first peak (UTC)
NZG -23.3516 186.5988 15.74 -25.81 370 4:42
NZF -29.6843 184.9874 11.21 -8.511 1013 5:25
NZE -36.0493 182.292 7.902 -4.713 1735 6:00
NZC -38.2001 180.2022 5.414 -3.961 2004 6:15
NZD -36.0998 178.6037 3.748 -2.511 1822 7:13
NZB -40.6003 179.0996 2.856 -1.342 2286 6:30
NZA -42.3707 176.9109 3.271 -2.53 2528 6:43
NZK -24.3093 169.4988 5.168 -5.696 1607 5:53
NZL -19.3096 166.782 4.977 -4.377 1868 6:8
NZJ -26.6672 163.9549 4.129 -4.175 2207 6:24

Table 2. Maximum wave amplitudes observed at coastal gauges.

Station Lat Lon Maximum amplitude (cm)
UPOL -13.8268 188.2387 31.2
PAGO -14.2766 189.3093 47.2
NKFA -21.1368 184.8193 93.6
RARO -21.2 200.217 61.0
VITI -18.1342 178.4236 29.2
 LEVU -17.6049 177.4383 18.8
 FONG -8.5025 179.1952 13.7
 VANU -17.7553 168.3077 102.1
 LUGA -15.5156 167.1886 58.1
 LITZ -16.1128 167.444 44.9
 OUIN -21.9829 166.6833 116.6
 LIFO -20.9185 167.2787 86.1
 OUVE -20.5498 166.5619 53.3
 THIO -21.6138 166.2415 71.8
 HIEN -20.6928 164.9422 46.0
 KJNI -29.0591 167.9536 131.5
 GCSB -27.9387 153.4326 68.0
 PKEM -34.4738 150.9119 39.3
 TBWC -37.1003 149.9266 74.5
 SPRG -42.5459 147.9327 29.9
 TUBUA -23.3418 210.5245 21.6
 AUCT -36.8314 174.7865 21.0
 PUYT -46.0848 166.5894 21.9
 LOTT -37.5503 178.159 93.6
 CHIT -44.0247 183.6312 91.0
 GIST -38.6754 178.0229 79.4
 TAUT -37.6411 176.1812 26.7
 KAIT -42.4129 173.7028 26.7
 NCPT -34.41 173.05 71.6
 MNKT -37.0466 174.5117 31.7
 GBIT -36.189 175.4889 152.9
 NAPT -39.4757 176.9201 40.4
 CPIT -40.8993 176.2317 27.2
 WLGT -41.2846 174.7791 30.3
 SUMT -43.5696 172.7732 3.9
 CHST -41.903 171.4341 99.6
 OTAT -45.8143 170.6294 8.5

 

Download an animation here:

Download animation

Acknowledgements

The New Zealand coastal gauge and New Zealand DART station data are available from GEONET (https://www.geonet.org.nz) while the coastal gauge data in Tonga, Fiji, New Caledonia, Vanuatu, Australia, Italy and Martinica are available from the UNESCO-IOC sea level monitoring site (http://www.ioc-sealevelmonitoring.org/). We thank David Burbidge, Xiaoming Wang, and William Power for their comments.

The information on how to download the New Zealand DART data (https://doi.org/10.21420/8tcz-tv02) can be found here: https://tilde.geonet.org.nz. The information on how to download the New Zealand coastal tsunami gauge data (https://doi.org/10.21420/EJ6W-RC96) can be found here: https://www.geonet.org.nz/data/tools/FDSN

References

Colombier, M., Scheu, B., Wadsworth, F. B., Cronin, S., Vasseur, J., Dobson, K. J., ... & Dingwell, D. B. (2018). Vesiculation and Quenching During Surtseyan Eruptions at Hunga Tonga‐Hunga Ha'apai Volcano, Tonga. Journal of Geophysical Research: Solid Earth123(5), 3762-3779.

Flanders Marine Institute (VLIZ); Intergovernmental Oceanographic Commission (IOC) (2022): Sea level station monitoring facility. Accessed at http://www.ioc-sealevelmonitoring.org on 2022-01-19 at VLIZ. DOI: 10.14284/482

Fry, B., McCurrach, S.-J., Gledhill, K.R., Power, W.L., Williams, M., Angove, M., Arcas, D., Moore, C. (2020) Sensor network warns of stealth tsunamis. Eos, 101: DOI: 10.1029/2020EO144274

Garvin, J. B., Slayback, D. A., Ferrini, V., Frawley, J., Giguere, C., Asrar, G. R., & Andersen, K. (2018). Monitoring and Modeling the Rapid Evolution of Earth's Newest Volcanic Island: Hunga Tonga Hunga Ha'apai (Tonga) Using High Spatial Resolution Satellite Observations. Geophysical research letters45(8), 3445-3452.

Paris, R., Switzer, A. D., Belousova, M., Belousov, A., Ontowirjo, B., Whelley, P. L., & Ulvrova, M. (2014). Volcanic tsunami: a review of source mechanisms, past events and hazards in Southeast Asia (Indonesia, Philippines, Papua New Guinea). Natural Hazards70(1), 447-470.

Paris, R. (2015). Source mechanisms of volcanic tsunamis. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences373(2053), 20140380.

Satake, K. (1995). Linear and nonlinear computations of the 1992 Nicaragua earthquake tsunami. Pure and Applied Geophysics144(3), 455-470.

Watada, S., Kusumoto, S., & Satake, K. (2014). Traveltime delay and initial phase reversal of distant tsunamis coupled with the self‐gravitating elastic Earth. Journal of Geophysical Research: Solid Earth119(5), 4287-4310

Volcano.ssec.wisc.edu.