1.1 Physical description of Hunga volcano and events leading up to the eruption
1.2 The 2022 Hunga eruption
1.3 Background on historic eruptions as context for the 2022 Hunga eruption
Science questions:
2.1 Volcanic Eruption and the First 20 Minutes
The Hunga volcano became active again on 20 December 2021. The main eruption started on 13 January 2022 when a large eruption sent a plume up to 20 km. The climax was reached on 15 January 2022 at 4:14 UTC with a very large and explosive eruption producing a plume that reached 58 km within 20 minutes [check duration on hima8]. The atmospheric plume was estimated to contain 1 Tg of solid ejecta (from deposits, Cairn), a much larger mass 8-9 Tg remained under water. The altitude reached suggests that the plume had a mass proportion of 60% steam water (Mastin, to be published in Bulletin of Volcanology, using a 1D model, that is about 0.8 km3 of water, and the rest being very thin ash. Injection of cold water would have instead reduced the top altitude.3D calculations in progress (Bruckert, KIT, Suzuki, U. Tokyo) might unveil more information and better estimates since the 1D model has limitations (steady state, no fallout, simplified entertainment). Deposits show heavy particles (rho > 2) with an unseen proportion of fractures.
The submarine nature of the eruption implied that the majority of the emissions (ash, gases) remained under the ocean surface and that water was massively entrained, which makes the composition of the plume very unusual with respect of other recent documented eruptions.
2.2 Plume Initial Stage
The plume generated from the eruption reached a maximum altitude of 58 km as reported from geostationary imaging (Carr, Proud).
The convective plume as seen from Hima 8 rapidly collapsed (Carr, Legras). Radiative calculations (Thompson, to be published in Bulletin of Volcanology), MISR retrieval of aerosol properties (Kahn, to be pub) and model (Bruckert) suggest formation of ice compatible with a large injection of water (see below). The initial convective plume led to several stratified aerosol layers.
Above 35 km, a 35-40 km layer with low AOD was detected on 15 Jan by CALIOP with low AOD, depolarizing and moving westward at high speed (40 m/s) in a layer of large shear (Khaykin). However CALIOP is blind above 40 km and OMPS-LP detected aerosols up to 50 km in the first days after the eruption. These layers all disappear after one week (Khaykin, Taha, Baron)
A low altitude plume by 18-20 km moving slowly westward (10 m/s) containing products from the 13 and 15 January eruption, mixed together (Taha). More SO2 in this plume than aerosol (presumed to be composed of H2SO4; Asher). Depolarizing and weakly absorbing (Boichu, Kloss). Need a characterization of evolution during the first month.
Two clouds of non depolarizing particles, presumably spherical droplets, moving westward at 20 m/s initially at 30-32 km and 26-28 km (Sellitto, Legras, Taha). These two clouds, first detected on 15 Jan at 19 UTC by Hima 8 and 16 Jan at 15 UTC by CALIOP are collocated with SO2 (Legras) and appear (on Hima 8 images) as emerging from the background within a few hours between 15 Jan 19 UTC and 16 Jan.4 UTC due to the fast conversion from SO2. See below in section vi. These two clouds were the source of the persisting aerosols in the stratosphere. By January 22, in this layer, more aerosol (also presumed to be composed of H2SO4) in the plume than SO2; Asher).
2.3 Gravity Waves
The explosivity has generated a train of GW detected in the early stage by several instruments GEOs, IASI, ...which have been followed for X days with several round the globe journeys. Balloon measurements show at least two successive explosions linked with the two main aerosol plumes [confirmed?].
2.4 Water Vapor Injection
The distinguished character of this eruption is the amount of water that was injected saturating the atmosphere up to 35 km at least (Sellitto, Khaykin, Vomel, Randel). The amount that remained in the stratosphere is estimated to be approximately 150 Tg, resulting in an instantaneous increase of 11% in stratospheric water vapor. There is good agreement between remote sensing and in situ measurements. The initial injection of water vapor was certainly much larger but the excess (above saturation) condensed as ice and fell out [estimates might come from models, Bruckert, Mastin?] entraining ash in the fall and leaving a plume mostly made of water vapor. This might be discussed as a function of altitude (e.g. condensed water in the mesosphere might have sublimated in the warmer stratosphere and ash might remain in the lower plume).
A question to be addressed; Why does water vapor exhibits a layer structure after a few days if it was initially saturated all the way?
This has a number of consequences for the impact of the eruption. In the first days, water vapor was mixed with the two main clouds initially of SO2, with the upper more western cloud being moister than the eastern and lower cloud. As the presence of water accelerates the production of sulfate (ref to the chemistry chapter) it explains the early detection by CALIOP (compare with Raikoke with higher SO2 emission) and the evolution over the first few days. See section vi An other important effect of water was to produce intense cooling by radiative longwave emission that led to a rapid descent of the two clouds of mixed aerosols and water (Sellitto, Legras, Wallis), up to 400-500 m per day initially that dropped to 150 m per day after one week.The two clouds had different water content and behaved differently, the highest one being moister descended faster and the conversion of SO2 also occurred faster (see vi below).
2.5 Other Gas Phase Chemicall Coimposition and Sea Salt
HCl and ClO have been detected by MLS (Millan, Evan) in small but significant quantities. The large amount of lightnings (20000/min) produced short-lived NOx, which, however, does not have a big effect on stratospheric chemistry (Zhu). At later stage (Wilmouth), NOx is depleted and ClO enhanced by OH produced from water, with effects on ozone (see chapter 5). The presence of sea salt has been found as coating the deposited tephras (Colombier) and detected during the BRAVO campaign at 20 km (Vernier). However, no direct evidence of solid non depolarizing aerosols or of significant short wave absorption in the two main clouds.
There is a question mark on other halogens like bromine.
2.6 Chemical Impacts of Water Vapor Injection
The SO2 emission was modest (0.5 Tg, Carn) for an eruption of this magnitude. According to Carn, most of the SO2 emissions remained under water. The rapid conversion of SO2 to sulfate as observed from satellite (IASI: Sellitto, Legras) and from in situ data (Asher) can be explained by the large amount of water vapor (Zhu 2022). Since the two main layers differ in water vapor content, their behavior differed too. The higher one (26-29 km) being more moist, exhibiting faster initial descent and faster conversion to sulfate (almost complete by 20 Jan) while the lowest one descended less and got fully converted ten days later (Asher, Legras). The conversion of the two aerosol layers can be seen as they travel across the Indian ocean while getting elongated by the zonal shear but still remain distinguished.
The size of the aerosols grow also unexpectedly fast and large during the first weeks (Khaykin, Kloss, Asher). The mass of aerosols estimated from in situ measurements (Asher) or later from SAGEIII (Duchamp) is in good agreement with the initial estimate of 0,5 Tg of injected SO2.
Rapid loss of O3 in the plume (Evan, Zhu 2023) due to the combined effect of injected chlorine and water vapor. Compare perhaps with the ozone loss due to forest fire. The large amount of water contains potential for further depletion on a longer time scale (refer to chapter 5).
2.7 Transport & Dispersion
Initial dispersion in latitude towards equator occurred rapidly within the first days but basically stopped at the northern tropical barrier and most of the dispersion is zonal in the following weeks (Khaykin, Legras, Taha). Vertical transport is also important in the first days and in particular the fast descent of the plume due to water cooling (Sellitto, Legras). The intense cooling was perhaps also responsible for the grouping of aerosols in compact patches with induced anticyclonic rotation that persisted until early March (Legras) [other possibility: KH instability but Rayleigh criterion not satisfied]. During the first month, the part of the aerosol plume between 30°S and 10°S was submitted to a weak vertical shear but a strong horizontal shear that governed the dispersion. In contrast, vertical shear dominated the dispersion in the 10°S-10°N band. Nevertheless the vertical shear gave the appearance of a descending plume in La Reunion since (Baron).
During the initial phase the vertical motion of the plume was governed by the cooling effect of water vapor. Water vapour and aerosols (and remaining SO2) were well mixed. This ended with the dilution due to dispersion and the growth of aerosol size. Then the gravitational sedimentation of aerosols induced a slower but persisting descent of the aerosols while the water vapor stalled and began an ascent due to the large-scale motion consistent with Brewer Dobson Circulation (Legras, Schoeberl; Wilmouth). The two components of the plume then start to separate by mid-February. The cooling effect of water persisted on the large scale (Schoeberl). See chapter 3 for sequel.
2.8 Sources of Uncertainty
The amount of ash, haline or whatever non sulfated aerosol in the plume. The fact that satellite observations are well fitted by non absorbing sulfate aerosol does not rule out entirely a second hidden component (e.g., the optical properties are very compatible with pure sulfates but one cannot rule out the presence of some hidden particles inside droplets). A discussion of the results from the BRAVO campaign (H. Vernier et al.,) will be addressed here because they relate to composition of the aerosol layer plume [a limitation of this campaign is that the balloon flights measured up to 22 km (mostly below 20 km during the extended flight) while the main plume was then between 22 and 26 km and they also did not measure sulfates].
2.9 Radiative Effects
The early work of Sellitto et al. found a positive feedback TOA feedback due to water vapor (e.g., for the first time we have evidence of a volcano warming the atmosphere). This result was however challenged by other studies (Schoeberl, Gupta) who found that aerosol cooling dominates during the first months.
2.10 The Value of Observations
The description of this exceptional eruption benefited a lot from the numerous satellite instruments available. MLS, CALIOP, OMPS-LP, IASI and GEO [also usage of GPS-RO, TROPOMI, EPIC/DSCOVR, others?] were very useful for the early developments and covered a wide range of needed observations on the spatial distribution and composition of the plume. No eruption in the past has concentrated so much observations from space. In situ observations from the rapid response campaign were also very helpful in improving our understanding of the early evolution of the plume, in particular for the characterization of microphysical and optical properties of the aerosols, not all accessible from satellite, and have been used to understand this unusual perturbation. Mention also SAGE III although it has not been useful at the early stage due to its limited sampling but was very useful afterwards.
3.1 Meridional dispersion of water vapour and aerosol
3.2 Evolution of sulfate aerosols
3.3 Transport of volcanic material into Antarctic vortex
3.4 Uncertainty of observations
3.5 Modeling of the HTHH plumes
4.1 Temperature impacts
4.2 Zonal winds, the Brewer-Dobson circulation, and atmospheric wave
4.3 QBO and the Hunga impact on the QBO
5.1 Observations and models used to assess impacts on stratospheric composition
5.2 Initial chemical processing and ozone loss in the fresh plume
5.3 Longer-term impacts on stratospheric composition
5.4 Predictions of the future impacts on the stratospheric ozone layer
6.1. Gravity waves, Lamb waves, vertical coupling (AIM satellite, OMPS-LP, ...)
6.2 HTHH H2O vapor in the deep BDC branch into the upper stratosphere and mesosphere, MLS comparison with model
6.3 Meteor radar observations
6.4 Mesospheric modeling
6.5 ionosphere impact: TIDs, plasma bubbles
7.1 Introduction
7.2 Data and Methods
7.3 HTHH's Radiative Forcing
7.4 Climate Responses
7.5 Summary
Last modified: Monday, 12-Feb-2024 15:56:19 MST