preparing for a sustained volcanic degassing episode in

GNS Science Report 2019/58December 2019

Preparing for a sustained volcanic degassing episode in Auckland

C Stewart GS Leonard N Talbot

E Smid I Tomašek

D Charlton TM Wilson

© Institute of Geological and Nuclear Sciences Limited, 2019 www.gns.cri.nz

ISSN 2350-3424 (online) ISBN 978-1-98-856982-6 (online) http://dx.doi.org/10.21420/4NKW-TR68

C Stewart, Massey University, PO Box 756, Wellington 6140, New Zealand E Smid, University of Auckland, xx, Auckland, New Zealand D Charlton, University of Auckland, xx, Auckland, New Zealand GS Leonard, GNS Science, PO Box 30368, Lower Hutt, New Zealand I Tomašek , Vriji Universiteit Brussel, Pleinlaan 2, B1050 Brussel, Belgium TM Wilson, University of Canterbury, Private Bag 4800, Christchurch, New Zealand N Talbot, Auckland Council, Private Bag 92300, Auckland 1142, New Zealand

DISCLAIMER

The Institute of Geological and Nuclear Sciences Limited (GNS Science) and its funders give no warranties of any kind concerning the accuracy, completeness, timeliness or fitness for purpose of the contents of this report. GNS Science accepts no responsibility for any actions taken based on, or reliance placed on the contents of this report and GNS Science and its funders exclude to the full extent permitted by law liability for any loss, damage or expense, direct or indirect, and however caused, whether through negligence or otherwise, resulting from any person’s or organisation’s use of, or reliance on, the contents of this report.

BIBLIOGRAPHIC REFERENCE

Stewart C, Smid E., Charlton D, Leonard GS, Tomašek I, Wilson TM, Talbot N. 2019. Preparing for a sustained volcanic degassing episode in Auckland. Lower Hutt (NZ): GNS Science. 41 p. (GNS Science report; 2019/58). doi:10.21420/4NKW-TR68.

http://www.gns.cri.nz/

GNS Science Report 2019/58 i

CONTENTS

ABSTRACT .......................................................................................................................... III

KEYWORDS ......................................................................................................................... III

1.0 INTRODUCTION ........................................................................................................ 1

2.0 VOLCANIC GAS HAZARDS ...................................................................................... 3

2.1 Introduction to Volcanic Gases ........................................................................ 3 2.2 Health Hazards ................................................................................................ 3 2.3 Infrastructure and Built-Environment Impacts .................................................. 6

2.3.1 Field and Laboratory Studies .............................................................................6 2.3.2 Observational Studies ........................................................................................7

2.4 Aviation Hazards ............................................................................................. 8 2.5 Summary ......................................................................................................... 8

3.0 RECENT EXAMPLES AND KEY LESSONS .............................................................. 9

3.1 2014–2015 Holuhraun Eruption, Iceland .......................................................... 9 3.2 1983–2018 Kīlauea eruption, Hawai’i, USA ................................................... 10 3.3 Ambae 2017–2018 Eruption, Vanuatu ........................................................... 12 3.4 Summary ....................................................................................................... 15

4.0 MONITORING OF SO2 AND PARTICULATE MATTER ........................................... 16

4.1 Regional Monitoring by Auckland Council ...................................................... 16 4.1.1 SO2 .................................................................................................................. 17 4.1.2 Particulate Matter (PM) ................................................................................... 17

4.2 Geonet Monitoring of Gas Fluxes .................................................................. 18 4.3 Summary ....................................................................................................... 19

5.0 ESTIMATING SO2 FLUXES FOR THE AUCKLAND VOLCANIC FIELD.................. 20

5.1 AVF Eruption Scenarios ................................................................................ 20 5.2 SO2 Fluxes from Analogous Eruptions ........................................................... 20

5.2.1 Estimated SO2 Fluxes from AVF Eruption Scenarios in Context of Total Anthropogenic Emissions for New Zealand .................................................... 22

5.2.2 Estimated SO2 Fluxes from AVF Eruption Scenarios in Context of Regional and Local Anthropogenic Emissions for Auckland.......................................... 23

5.3 Summary ....................................................................................................... 25

6.0 KEY ISSUES AND RECOMMENDATIONS .............................................................. 26

6.1 Improved Capability for SO2 Dispersion Modelling and Forecasting............... 26 6.2 Improved Capacity to Forecast Impacts of Volcanic Gases on Infrastructure

and the Built Environment .............................................................................. 27 6.3 Managing Public Health Impacts.................................................................... 28

6.3.1 Exposure Assessment ..................................................................................... 28 6.3.2 Assessing Health Impacts (Syndromic Surveillance) ...................................... 28 6.3.3 Communicating with the Public ....................................................................... 29 6.3.4 Considering Combined Exposure to Volcanic and Urban Air Pollution........... 29

ii GNS Science Report 2019/58

7.0 ACKNOWLEDGEMENTS ......................................................................................... 31

8.0 REFERENCES ......................................................................................................... 31

FIGURES

Figure 2.1 Warning signage and fencing around Mammoth Mountain ski area fumarole .............................. 4 Figure 2.2 Corroded guardrail, bolt and road sign downwind of Kīlauea volcano, June 2015. ...................... 7 Figure 3.1 Variable warning signage near entrance to Hawai’i Volcanoes National Park. ........................... 12 Figure 4.1 Locations of PM10, PM2.5 and SO2 monitoring stations in Auckland region................................. 16 Figure 4.2 Mini-DOAS data for SO2 fluxes at White Island (t/d) since 2012 ................................................ 19 Figure 4.3 10-day means for mini-DOAS SO2 data from White Island (t/d) ................................................. 19 Figure 5.1 Relationship between erupted magma volume and total SO2 emissions for the six eruptions

shown in Table 5.2. .................................................................................................................... 22 Figure 5.2 Average and peak SO2 fluxes from the 2007 Stromboli, 1973 Eldfell and 2014–2015 Holuhraun

eruptions, in comparison to estimated SO2 fluxes for all anthropogenic sources in New Zealand. ................................................................................................................................................... 23

Figure 5.3 Average and peak SO2 fluxes from the 2007 Stromboli, 1973 Eldfell and 2014–2015 Holuhraun eruptions, in comparison to estimated 2016 regional and urban SO2 fluxes in the Auckland region. ........................................................................................................................................ 24

TABLES

Table 1.1 Approaches to including volcanic gas emissions in AVF scenarios. ............................................. 2 Table 2.1 Health effects of exposure to SO2 at different concentrations and air quality guideline values and

standards. ..................................................................................................................................... 5 Table 2.2 Field and laboratory studies of corrosion damage to infrastructure systems and buildings

exposed to volcanic emissions. .................................................................................................... 6 Table 3.1 SO2 observations throughout the 2017-ongoing Ambae eruption .............................................. 13 Table 3.2 West Ambae residents’ perceptions of the impacts of acid rain and volcanic gas when

interviewed in early March 20181................................................................................................ 14 Table 4.1 Standards and guidelines for SO2. ............................................................................................. 17 Table 5.1 Summary details for AVF eruption scenarios ............................................................................. 20 Table 5.2 Erupted volumes, eruption durations, total SO2 emissions, average and peak SO2 fluxes from

recent basaltic eruptions............................................................................................................. 21 Table 5.4 Predicted trends in annual anthropogenic SO2 emissions to Auckland urban area from 2016 to

2040 ........................................................................................................................................... 25 Table 6.1 Comparative features of AERMOD and CALPUFF air dispersion models1. ............................... 26

APPENDICES

APPENDIX 1 HAWAI’I STATE DEPARTMENT OF HEALTH SO2 ADVISORY LEVELS . 41

GNS Science Report 2019/58 iii

ABSTRACT

The Auckland metropolitan area is built upon the intraplate Auckland Volcanic Field (AVF), which poses a considerable threat to the highly exposed people and infrastructure of Auckland. An important component of evaluating and quantifying volcanic risks to Auckland has been the development of AVF eruption scenarios. These are useful for purposes such as assessing impacts on buildings and critical infrastructure and providing decision support for emergency managers and civil authorities. However, while several scenario-based studies have noted the potential for volcanic gas emissions to cause impacts on downwind populations, infrastructure and the built environment, this important volcanic hazard has so far received little systematic attention and there have been no attempts at quantitation.

The purpose of this report is to consider issues arising from volcanic gas emissions in the event of a future AVF eruption. Here we introduce volcanic gases and discuss their hazards to human health, infrastructure, the built environment and aviation; we describe in detail three recent eruption case studies where gas impacts have been important; we describe the current arrangements for monitoring the relevant air pollutants in Auckland; we estimate fluxes of sulfur dioxide (SO2) for selected AVF eruption scenarios and put them in context alongside known anthropogenic sources of SO2 nationally, regionally and locally; and finally we identify key issues and knowledge gaps and provide recommendations on how these might be addressed.

Estimates of gas emission rates for AVF eruption scenarios B and E suggest average and peak SO2 fluxes of ~650 t/d and ~1200 t/d. These fluxes are similar to the range of SO2 fluxes recorded for White Island volcano during 2019 and exceed total daily SO2 fluxes from all known anthropogenic sources in New Zealand by factors of 5-10. In relation to regional and local sources of SO2, new volcanic sources are expected to be even more dominant. A new AVF eruption will therefore be a major new air pollution point source in metropolitan Auckland. Impacts will depend on the following factors: vent location, meteorological conditions, populations and assets exposed, plume height and timing. For effective management of a future AVF eruption, we recommend developing improved capability for gas dispersion modelling and forecasting, improved capacity to forecast impacts of acidic volcanic gases on infrastructure and the built environment, and a multipronged approach to managing impacts on public health.

KEYWORDS

Basaltic volcanism, Auckland Volcanic Field, volcanic gas emissions, sulfur dioxide, monitoring, impacts

iv GNS Science Report 2019/58

This page left intentionally blank.

GNS Science Report 2019/58 1

1.0 INTRODUCTION

Auckland is New Zealand’s largest city, with approximately 1.6 million residents. The Auckland metropolitan area is built upon the intraplate Auckland Volcanic Field (AVF), which poses a considerable threat to the highly exposed people and infrastructure of Auckland. The field is likely to erupt again as the most recent eruption, Rangitoto, was only 550 years ago (Lindsay et al. 2011) and there has been a relatively high eruption rate in the late Quaternary (Leonard et al. 2017). Most AVF volcanic centres are monogenetic (only erupt once) thus it is probable that the next vent will be in a new location. However, there is no clear trend in vent locations over time (Leonard et al. 2017) thus it is completely unknown where or when the next eruption will be. There is also no clear trend in eruption volume over time.

To improve volcanic hazard and risk management in Auckland, the Determining Volcanic Risk in Auckland (DEVORA) research programme was established in 2008. An important component of evaluating and quantifying volcanic risks to Auckland has been the development of AVF eruption scenarios by DEVORA partners; these are useful for purposes such as assessing impacts on buildings and critical infrastructure and providing decision support for emergency managers and civil authorities (Hayes et al. 2018). A brief history of eruption scenarios developed for the AVF is provided in Table 1.1. Most hazardous eruption phenomena (edifice building, lava flows, tephra fall, pyroclastic surges, and ballistic projectiles) are included in these scenarios, with a trend towards quantitation in more recent scenarios. However, while several of these studies (Johnston et al., 1997; Fitzgerald et al. 2015; Hayes et al. 2018; Table 1.1) have noted the potential for volcanic gas emissions to cause impacts on downwind populations, infrastructure and the built environment, this important volcanic hazard has so far received little systematic attention.

The purpose of this report is to consider issues arising from volcanic gas emissions in the event of a future AVF eruption. In this report we introduce volcanic gases and discuss their hazards to human health, infrastructure, the built environment and aviation; we describe in detail three recent eruption case studies where gas impacts have been important; we describe the current arrangements for monitoring the relevant air pollutants in Auckland; we estimate fluxes of SO2 for the eight DEVORA eruption scenarios and put them in context alongside known anthropogenic sources of SO2 nationally, regionally and locally; and finally we identify key issues and knowledge gaps and provide recommendations on how these might be addressed.

2 GNS Science Report 2019/58

Table 1.1 Approaches to including volcanic gas emissions in AVF scenarios.

Study/year Name of scenario/s Comments

Johnston et al. 1997

ARC scenarios 1–5 Five AVF scenarios were commissioned by Auckland Regional Council to explore likely processes, hazards and impacts. Approach to hazard modelling was largely qualitative and narrative-based. The authors noted that continuous, protracted discharges of SO2 gas would be likely to cause health problems for residents downwind, and lead to acid rains which could in turn cause corrosion problems.

MCDEM 2008 Exercise Rūaumoko This was a volcanic unrest scenario developed for an all-of-nation civil defence exercise run from November 2007 to March 2008. Soil gas emission measurements, particularly CO2, were a key component of the simulated GeoNet response as CO2 is one of the first gases to exsolve from rising magma (Mazot et al. 2013). SO2 emissions were also included as a parameter in BET_EF modelling as SO2 is a characteristic magmatic gas (Lindsay et al. 2010).

Dohaney et al. 2015

Fitzgerald et al. 2015

Mt Rūaumoko educational scenario

A volcanology team at the University of Canterbury developed an extended geophysical scenario for educational purposes, as a simulation exercise for tertiary students. The scenario included seismicity and deformation data in addition to eruptive hazards. It was based loosely on analogous eruptions from around the world. Emission data from a VEI2 eruption of Stromboli in 2007 was used to supply indicative SO2 fluxes for the exercise.

Deligne et al. 2015

ERI Mt Rūaumoko scenario

The ‘Mt Rūaumoko educational scenario’ was adapted for use as the input for a model (Modelling the Economics of Resilient Infrastructure Tool, or MERIT) to quantify the economic consequences of infrastructure failure following this hypothetical eruption. Modifications included the inclusion of Volcanic Alert Levels along with their implications for civil defence; the addition of a base surge as a hazard; and the inclusion of rainfall data. Gas emissions were specifically excluded from this scenario on the grounds of posing a negligible threat to infrastructure.

Deligne et al. 2017

Māngere Bridge scenario

The ERI Mt Rūaumoko scenario was renamed the Māngere Bridge scenario. This article described the scenario development process and explored the impacts of the hypothetical eruption on electricity service provision. A further article (Blake et al. 2017) explored consequences for surface transport in Auckland.

Hayes et al. 2018

DEVORA AVF scenarios A–H

Hayes et al. (2018) developed a suite of seven new geophysical scenarios for the AVF to encompass the credible range of future AVF eruptions. The eight scenarios (including Māngere Bridge) are geographically spread across Auckland (although are not located on known volcanic centres); allow for the exploration of different eruption styles and hazards; and allow for the exploration of impacts to different assets. The authors noted that areas downwind of each hypothetical eruption could plausibly be affected by volcanic gases but did not attempt to model gas dispersion because of the lack of input parameters.


2.0 VOLCANIC GAS HAZARDS

2.1 Introduction to Volcanic Gases

Magma contains dissolved gases. As magma rises towards the surface and pressure decreases, gases separate from the magma as bubbles, continue to rise and are eventually released into the atmosphere. Large eruptions can release large quantities of gas. However, even if magma does not reach the Earth’s surface, gases can escape into the atmosphere through soil, volcanic vents, fumaroles and hydrothermal systems. The most abundant volcanic gas is water vapour, with significant quantities of carbon dioxide (CO2), sulfur dioxide (SO2), hydrogen sulfide (H2S), hydrogen chloride (HCl) and hydrogen fluoride (HF) also emitted.

Gas emissions can cause impacts on different spatial and temporal scales (Edmonds et al. 2018). Diffuse degassing of CO2 and H2S through fractures and faults may persist for years, but typically only present hazards to people on a local scale when they accumulate in basements or topographic lows. Acidic tropospheric plumes, primarily comprised of SO2 and derived sulfate aerosol, HCl and HF, are typically dispersed over tens of kilometres and can persist for many years, potentially inducing respiratory problems in downwind populations. Finally, large, explosive eruptions inject SO2 directly into the stratosphere where it converts to sulfate aerosol, forming an ‘aerosol veil’ (Mather 2015). Consequences such as climate cooling and crop failure can affect large areas, on regional to global scales, lasting for years (Oppenheimer et al. 2011).

The focus of this report is on SO2 hazards and impacts, although we do pay attention to the hazard of CO2 accumulation. Basaltic magmas, such as those which underlie the AVF (Leonard et al. 2017), have a high sulfur content (Thordarson et al. 2003; Shinohara 2008), typically 2-4 times higher than silicic magmas, and basaltic eruptions typically release large quantities of sulfur, in the form of SO2, into the atmosphere. Adverse health and environmental effects from SO2 emissions are reported for a range of basaltic volcanoes worldwide, including Kīlauea volcano, Hawaii (Section 3.2); Masaya volcano, Nicaragua (Delmelle et al. 2001; 2002), and Miyakejima volcano, Japan (Fujita et al. 2003).

2.2 Health Hazards

Adverse health effects from exposure to volcanic gases range from mild to serious and occasionally lethal. Volcanic gases have claimed the lives of over 2000 people in the past 600 years (Auker et al. 2013; Edmonds et al. 2018) with 1700 lives lost in a single event: the 1986 Lake Nyos gas disaster, Cameroon (Hansell and Oppenheimer 2004). A dense cloud of CO2 gas was released suddenly during a lake overturn event and flowed down surrounding valleys, suffocating local residents as they slept. The invisible and insidious nature of this hazard led Edmonds et al. (2018) to characterise volcanic gases as ‘silent killers’ as there may be no warning signs until concentrations accumulate to lethal levels, or sudden inundations may allow no time to escape. These authors further describe strategies developed to mitigate risks of CO2 flows and accumulations including gas sensor networks linked to sirens; artificially degassing deep lake waters to prevent sudden gas releases; and warning signage advising the public of the potential for gas accumulation in topographic lows.

Ground emissions of CO2 can be particularly hazardous as there may be little warning of high concentrations. Unrest beneath Mammoth Mountain lava dome complex, California, has led to extensive emissions of CO2 from soil and fumaroles since approximately 1990. Several cases of near-asphyxia have been reported for people entering snowed-in back country cabins in the


area (Farrar et al. 1995; Sorey et al, 1998). A cross-country skier who fell into a tree well in the area is thought to have died of asphyxia (Hill 2000). In 2006, two ski patrollers at Mammoth Mountain ski area fell through snow into a deep hole created by a fumarole, then a third patroller carrying oxygen descended into the hole, but rapidly lost consciousness (Cantrell and Young 2009). All three were in cardiopulmonary arrest after being extracted and could not be resuscitated. Autopsies suggested their deaths were consistent with asphyxiation. The concentration of CO2 in the Mammoth Mountain fumarole was measured at 98.7% (Cantrell and Young 2009). The fumarole is now fenced off and warning signage installed (Figure 2.1).

Figure 2.1 Warning signage and fencing around Mammoth Mountain ski area fumarole (photo taken in 2013).

Eruptive fluxes of CO2 may also be hazardous. The only fatality caused by the 1973 eruption of Eldfell on the island of Heimaey, Iceland, was a person breathing gas inside a building, with several other people partially overcome. Gases accumulated in topographic lows in the eastern part of Vestmannaeyjar town closest to the lava flows, and also in houses partially buried by tephra. This gas was comprised of 98% CO2 with minor quantities of carbon monoxide and methane (Williams and Moore 1983).

A further class of volcanic gas hazard is associated with low-altitude persistent gas plumes which can affect downwind populations (Edmonds et al. 2018). These plumes are typically rich in acidic gas species such as SO2, HCl and HF. Sulfur dioxide oxidises in the atmosphere to sulfuric acid, which forms sulfate aerosol particles typically comprised of 75 wt.% H2SO4 and 25 wt.% H2O when hydrated. Plumes rich in sulfate aerosol can form a thick, choking haze known as ‘vog’. Areas closer to eruptive vents are exposed to both SO2 and aerosol, whereas areas further downwind are primarily affected by aerosol. In urban populations, many studies have shown that exposure to fine particulate matter (PM2.5, or particles smaller than 2.5 µm in diameter) is associated with increasing or inducing respiratory and cardiovascular diseases, as well as substantial premature mortality and excess morbidity (summarised in REVIHAAP report, WHO 2013). However, the health impacts of sulfate aerosol exposure in air pollution are much less well understood. Epidemiological studies report associations between sulfate and adverse impacts on human health, but it is unclear whether sulfate is simply a marker for


causal agents in PM2.5, or whether sulfate reacts with trace metals in PM to form toxic species (Ilyanskaya et al. 2017).

Health effects of volcanic gases depend on the type of gas and its concentration in the air we breathe, exposure duration and individual sensitivity. The short-term health effects of exposure to different concentrations of SO2 are shown in Table 2.1, alongside New Zealand and international air quality guideline values for different averaging periods. Short term exposure to SO2 causes inflammation and irritation of the eyes and airways, resulting in burning eyes, coughing, difficulty breathing and/or chest tightness. Inflammation of the respiratory tract also makes people more susceptible to respiratory tract infections. Higher concentrations of airborne SO2 are associated with increased hospital admissions for cardiac disease and cardiac mortality. Longo et al. (2008) note that long-term residency in degassing volcanic areas may have an adverse effect on cardiorespiratory health in adults.

While people vary in their sensitivity, asthmatics are known to be particularly susceptible to SO2, and exposure may aggravate their asthma symptoms. Other groups who may be more sensitive to experiencing ill effects include: people with other respiratory or heart conditions, older adults (due to declining heart and lung function) and children, because they commonly breathe faster and have larger lungs relative to their body size than adults.

While airborne SO2 concentrations in air can be compared against air quality standards and guideline values to assess health risks to exposed populations, it is important to note that in volcanic plumes, SO2 will occur in combination with other volcanic gases and fine PM, which may act synergistically to increase adverse health effects. Health impacts on downwind populations exposed to such ‘combined’ plumes are discussed further in Section 3.2.

Table 2.1 Health effects of exposure to SO2 at different concentrations and air quality guideline values and standards.

SO2 concentration in air (µg/m3)1 Short term health effects1

520 Threshold for respiratory response in asthmatic individuals

2600-13,000 Threshold for respiratory response in healthy individuals

13,000 Increased airway resistance in healthy individuals,

26,000 Worsening irritation of the eyes, nose and throat

52,000 Paralysis or death after extended exposure

>260,000 Rapid unconsciousness, respiratory paralysis and death

Air quality guideline values and standards2

500 WHO2 guideline value (10-minute average)

570 New Zealand NES-AQ upper standard (1-hour average)3

350 New Zealand NES-AQ lower standard (1-hour average)3

120 New Zealand national ambient air quality guideline (24-hour average)4

20 WHO2 guideline value for (24-hour average) 1. http://www.ivhhn.org/information/information-different-volcanic-gases/sulphur-dioxide

Note that units for SO2 (µg/m3) presented in Table 1.1 are converted and rounded from units used in the IVHHN summary table (ppm) where 1 ppm = 2620 µg/m3.

2. See Table 4.1 for data sources.

http://www.ivhhn.org/information/information-different-volcanic-gases/sulphur-dioxide


2.3 Infrastructure and Built-Environment Impacts

2.3.1 Field and Laboratory Studies

Field studies suggest that corrosion is a commonly-observed phenomenon in areas exposed to volcanic emissions (Table 2.2).

Table 2.2 Field and laboratory studies of corrosion damage to infrastructure systems and buildings exposed to volcanic emissions.

Study Type of study and exposure setting Findings

Izumo et al. 19901

Laboratory study of combined impacts of SO2 gas (150-200 ppm) and volcanic ash on corrosion of carbon steel, zinc-plated steel, copper and aluminium

Copper corrodes more quickly in the absence of volcanic ash. For zinc-plated steel, carbon steel and aluminium, corrosion is greatly accelerated in the presence of ash.

Lichti et al 1996

Kurata et al. 1995.

Field study of different corrosion environments on White Island including fumaroles, acidic hot pools, volcanic soils and the ambient atmosphere, using test coupons.

The corrosion rate of carbon steel in the atmosphere (0.9 mm/year) was slightly higher than its corrosion rate in an acidic pool, pH 3.2-5.2. Atmospheric concentrations of SO2 and H2S in µg/m3 were not accurately determined.

Watanabe et al. 2006

Field study of corrosion of copper and silver test plates on Miyaki Island downwind from Oyaki volcano which was emitting SO2 and H2S, for eight-month period.

Copper plates were covered in corrosion products after eight months. This study did not quantify SO2 and H2S concentrations and did not quantify corrosion in terms of mass loss. The authors noted that sea-salt aerosol also contributed to observed corrosion.

Hawthorn et al. 2007

Field study using outdoor corrosion test site at Kīlauea volcano in 2006 for one year. Test coupons exposed to combined volcanic emissions.

Corrosion rate of exposed aluminium test coupons at Kīlauea site over seven times higher than industrial site on Oahu.

Oze et al. 2014 Laboratory study of synthetic volcanic ash corrosion to metal roofing materials in a weathering chamber.

Corrosion had not been initiated at the end of a one-month exposure period.

Li et al. 2018 Field study using test panels deployed across Rotorua geothermal field. H2S and SO2 quantified using diffusion tubes.

Extremely high corrosion rates of mild steel, zinc and copper were recorded at sites exposed to highest concentrations of H2S and SO2 combined emissions. Corrosion rates correlate to H2S concentration (SO2 correlation not tested for).

1 Only the abstract for this paper is available in English.

While field studies indicate that corrosion is accelerated in areas affected by volcanic emissions, they are beset by problems that limit the development of quantitative relationships between components of volcanic emissions and corrosion rates. Perhaps the major limitation is that few studies have measured ambient concentrations of gases to assess exposure. A further challenge associated with field studies in volcanic environments is that test materials are typically exposed to complex, unknown combinations of acidic gases, acidic aerosol, volcanic ash and acid rain and it remains a difficult challenge to tease apart these components.


Nonetheless, this remains an important goal. Lichti et al. (1996) note that in general, findings from studies of SO2 corrosion in urban industrial environments cannot be applied to volcanic environments because of the complex mixture of conditions and components, and that in situ studies are required. Some progress has been made by isolating different components in laboratory studies (Izumo et al. 1990; Oze et al. 2014).

2.3.2 Observational Studies

Findings from field and laboratory studies (Table 2.2) are supported by observations of corrosion damage to building materials and infrastructure components following volcanic eruptions. Blong (2003) documented damage to buildings in Rabaul township, Papua New Guinea, following the September 1994 eruption of Rabaul volcano, and noted that exposed sheet metal deteriorated rapidly following the eruption. Johnston et al. (2000) reported severe corrosion damage to ski field equipment on Mt Ruapehu, New Zealand, following the 1995-1996 eruptions of Ruapehu volcano. Galvanised nuts and bolts attaching chairlift hangers to cables were heavily rusted, and chairlift towers were heavily rusted on the uphill sides facing the crater. A small number of insurance claims to the Earthquake Commission for corrosion damage to roofing materials were paid out on.

Observations by utility company staff about corrosion damage can be particularly useful as they may be able to comment on the relative vulnerability and any reduction in lifetime of different components and equipment exposed to volcanic emissions. Staff from several utilities on the Big Island of Hawai’i were interviewed in June 2015 about impacts of the ongoing eruption of Kīlauea volcano (see Section 3.2) by a joint New Zealand/US impact assessment field team. The State Department of Transportation Highways representatives reported that roadside signs and guard rails corrode prematurely, and the lifetime of a guardrail is reduced from 25 years, in other parts of Hawai’i, to around five years (Figure 2.2). Staff from electricity lines company Helco reported accelerated corrosion of cotter pins, nuts and bolts in areas downwind of Kīlauea. The manager of a hardware store in Pāhoa noted that nuts, bolts and exposed hinges are highly corrosion-prone, that areas of standing water or splash zones around buildings are particularly corrosion-prone, and that the store began to sell a higher proportion of stainless-steel products after the escalation in SO2 emissions in 2008.

Figure 2.2 Corroded guardrail, bolt and road sign downwind of Kīlauea volcano, June 2015. In all cases

corrosion was most severe on the side facing the volcano.


2.4 Aviation Hazards

While airborne volcanic ash is a very well-known hazard to aviation (e.g. Prata and Tupper 2009), hazards from volcanic plume gases have received far less attention. In general, SO2 gas is only considered hazardous to aircraft bodies and engines once it has oxidised to sulphuric acid which may then form sulphuric acid aerosol particles which have the potential to cause corrosion damage (Schmidt et al. 2014 and references therein). While there are no documented cases of damage to aircraft due to encounters with SO2 plumes to date, there have been several documented encounters of aircraft with SO2 and/or sulfuric acid aerosol that have caused distress to pilots and passengers (Casadevall et al. 1996). There are currently no criteria to define when airspace is hazardous because of volcanic SO2 plumes (Schmidt et al. 2014).

Schmidt et al. (2014) used a modelling approach to simulate SO2 concentrations along North Atlantic flight paths for a range of hypothetical explosive Icelandic eruptions. These concentrations were then compared to available air quality standards for SO2, especially the World Health Organisation (WHO) outdoor health protection guideline of 500 µg/m3 which is set for an averaging period of 10 minutes. For the long duration 2010 Eyjafjallajökull eruption, SO2 data was retrieved from OMI (NASA’s Aura satellite Ozone Monitoring Instrument) and the authors found that at no time were air quality standards exceeded at flight altitudes. For a hypothetical short duration eruption similar to the 2000 Hekla eruption, the probability of a 15 minute or longer exposure of aircraft and passengers to concentrations >500 µg/m3 was calculated to be about 0.1%.

2.5 Summary

• Health hazards of SO2 are very well characterised and a colour-coded advisory system has been developed linking airborne concentrations, health impacts and recommended actions.

• People vary markedly in their susceptibility to the effects of SO2, thus it is important to ensure that advice clearly targets at-risk groups.

• Field and observational studies suggest accelerated corrosion of metal roofing materials and infrastructure components exposed to volcanic emissions. Nuts and bolts appear to be particularly susceptible.

• There has been limited development of quantitative relationships between components of volcanic emissions and corrosion rates, thus there is currently very little basis for assessing likely damage to building materials and infrastructure in the event of major/sustained SO2 emissions.

• Hazards of gas plumes to aviation appear to be minor.


3.0 RECENT EXAMPLES AND KEY LESSONS

3.1 2014–2015 Holuhraun Eruption, Iceland

The 2014–2015 Bárđarbunga-Veidivötn fissure eruption at Holuhraun was the largest eruption in Iceland for over 200 years, since the 1783-1784 Laki eruption. It produced approximately 1.6 km3 of lava over a six-month period and emitted enormous quantities of SO2 into the lower troposphere (Schmidt et al. 2015). The total mass of SO2 emitted over the course of the eruption was 11±5 Mt (Gíslason et al. 2015). For comparison, the total combined emissions from the USA, Canada, Western and Central Europe for the year 2011 were 14.7 Mt/year (Gíslason et al. 2015). The average SO2 flux over the course of the eruption was approximately 60 kt/d (Gíslason et al. 2015; Galeczka et al. 2017), with peak fluxes as high as 112 kt/day (BGVN 39:10, Global Volcanism Program 2014). These fluxes dwarf other long-term volcanic SO2 sources such as Ambrym, Vanuatu (7.4 t/d) or Kīlauea, Hawai’i (5.0 kt/d) (Carn et al. 2016).

This eruption has been described as “one of the most intense, large-scale volcanogenic air pollution events in centuries” (Ilyinskaya et al. 2017). Ilyinskaya et al. (2017) carried out air quality sampling in January 2015 using hand-held instruments. A 15-minute average concentration of 43,000 µg/m3 in the airborne plume near the vent was recorded; and 3100 µg/m3 in the grounding plume 4-20 km from the vent. The near-vent concentration was immediately hazardous to human health (Table 1.1). The ground-level concentration of SO2 exceeded the European Commission air quality standards over much of Iceland for days to weeks (Gíslason et al. 2015). For the nearest population centre (Reykjahlíđ town, ~100 km from the vent, population ~300), the hourly limit of 350 µg/m3 was exceeded 88 times and the daily limit of 125 µg/m3 exceeded 10 times. For the capital city, Reykjavik, ~250 km away and with a population of ~120,000, exceedances of the air quality standards were less frequent (34 hours/7 days). Prior to the Holuhraun eruption, ground-level SO2 concentrations had never exceeded 350 µg/m3 since monitoring started in 1968 (Gíslason et al. 2015). It is likely that the eruption’s impacts on air quality were mitigated somewhat due to its remote location.

Despite the eruption plumes being just 1-3 km in height, volcanic SO2 was transported over long distances, and was detected by air quality monitoring stations almost 3000 km away in Austria (247 µg/m3 as a one-hour average on 22 Sept 2014, Schmidt et al. 2015). For context, background SO2 concentrations in Europe are typically <6 µg/m3, and the SO2 concentrations recorded across Europe during two air pollution episodes (4-8 Sept and 20-25 Sept 2014) were the highest since 1990. In terms of air quality standards, exceedances were minor and transient, and Schmidt et al. (2015) concluded that the risk of detrimental health impacts on European populations would be low.

Scavenging of volatiles from the plume by snow was assessed by Galeczka et al. (2017) to be of minor importance, based on measurements of acidity, Cl, F and S in the snowpack of the Vatnajökull glacier immediately downwind of the eruption site. The lowest pH value recorded in the snowpack was 4.45. Gíslason et al. (2015) report that the lowest pH of fresh snowmelt at the eruption site was 3.3, and 3.2 in precipitation collected 105 km downwind. Thus, while volatile scavenging by precipitation may have been low, impacts on the composition of snowmelt and precipitation were clearly discernible, and these authors note that the primary environmental pressure on the Icelandic environment is likely to be on surface waters, soils and vegetation. They further noted that the timing of the eruption (in autumn/winter) may have mitigated its environmental impacts due to the following factors. Vegetation growth was at its minimum; grazing animals were at lower altitudes; windspeeds were relatively high, dispersing


the plume and mobilising dust, which interacts with precipitation to neutralise added acidity; and the oxidation of SO2 to sulfuric acid was minimised due to limited sunlight.

During the eruption, the Icelandic Meteorological Office (IMO) provided a range of information to the public. Citizen science reporting of volcanic gas smells was encouraged, using an online form, which asked whether the respondent could detect a sulfur smell; the date and time of the observation; physical symptoms such as nausea and effects on the eyes and throat; whether pollution was visible as a haze; windspeed and direction and any rainfall; and respondents’ location. All observations were made available on an interactive map. The IMO website hosted a link to health advice on the effects of SO2 at different concentrations and appropriate preventive actions, using the same colour-coded advisory system as used in Hawai’i (see Appendix 1).

The IMO also provided probabilistic hazard maps of surface SO2 concentrations using the CALPUFF air dispersion model (Barsotti 2014). A preliminary set of two maps were produced using meteorological data for one month. One map was intended for the public, and showed simple concentric zones depicting a 50% probability of exceeding a series of health-based threshold concentrations. The other, less-conservative, map was for authorised response personnel, and based the zones on a 90% probability of exceeding the same thresholds. The second set of maps had higher spatial resolution and was based on ten years of meteorological data. One map showed the probability that at a specific location across Iceland the hourly concentration of SO2 would exceed the hourly air quality standard of 350 µg/m3; the second map showed the probability of exceedance of a higher threshold of 2600 µg/m3 in the vicinity of Holuhraun. This higher threshold is where everyone will experience respiratory symptoms; the public are advised to remain indoors, close all doors and windows and shut down air conditioning; and work is prohibited unless gas masks are used.

To date, we are unaware of any published studies on health impacts of the 2014-2015 Holuhraun eruption on the Icelandic population.

3.2 1983–2018 Kīlauea eruption, Hawai’i, USA

Kīlauea Volcano, Hawai’i, was in continuous eruption from 1983 until 2018. Magma outgassing occurred at both the summit area, since 2008, and at the East Rift Zone (Longo et al 2010). Regular monitoring of SO2 began in 1979 (Sutton and Elias 2014). Emissions from Pu’u Ō’ō were initially around 2000 t/d, although they decreased in mid-2008 in response to pre-eruptive degassing at the summit caldera. The mid-2008 summit eruption caused SO2 emissions from the summit caldera to increase sharply from 100-200 t/d to around 2000 t/d, from where they have gradually declined (Sutton and Elias 2014). The average SO2 flux between 2005 and 2015 for Kīlauea was calculated to be 5019 t/d, from OMI measurements (Carn et al. 2016), the second-largest volcanic SO2 source globally. Eruptive fluxes during the 2018 Lower East Rift Zone (LERZ) eruption were considerably higher at over 30,000 t/d.

SO2 reacts in the atmosphere with oxygen, sunlight, moisture, and other gases and particles and, within hours to days, converts to fine aerosol particles (PM) which scatter sunlight, causing the visible haze (‘vog’, see Section 1.4.1) that is observed downwind of Kīlauea. Vog contains both SO2 and fine PM in varying proportions. Areas far downwind, such as the communities on the western side of the Big Island and other islands in the state, are mostly affected by the fine PM, but areas immediately downwind of eruptive vents can be exposed to both SO2 and fine PM during vog episodes.


Different air quality issues were experienced during the 2014 Pāhoa lava flow crisis, and the 2018 LERZ fissure eruption. The vent for the 2014 lava flows was located many kilometres distant from any residential areas, and the active lava flows had lost most of their gas content to the atmosphere by the time they neared the Pāhoa community. However, air quality was compromised by lava combustion of roading materials, structures and vegetation, which generated a complex, unknown mixture of smoke and volatile combustion products. State Department of Health staff interviewed in June 2015 by a joint NZ/US impact assessment team noted that deterioration of air quality was a major concern for the public. The DOH also adapted an air quality colour code alert system used for wildfires in California for use in local schools.

The eruptive vents for the 2018 LERZ eruption were located in or very close to residential areas, and the primary air quality problem was the highly hazardous concentrations of SO2. Near to the erupting fissures, SO2 concentrations spiking at >260,000 µg/m3 were recorded; these are immediate life safety hazards (Table 1.1). In residential areas of Lower Puna, concentrations of approximately 13,000 µg/m3 were recorded. These concentrations are sufficient to cause respiratory symptoms in healthy adults from short-term exposures and are well above concentrations that will affect asthmatic individuals. Health advice to residents included measures to reduce exposure such as staying indoors, closing all doors and windows and using an air cleaner, and also for sensitive individuals to consider voluntary evacuation.

Kīlauea has been the focus of a number of recent epidemiological studies, particularly in relation to the increased SO2 emissions in 2008. Prior to this escalation in emissions, nearby residents self-reported increased lung, eye and nasal symptoms compared to residents in areas unaffected by vog (Longo et al. 2008; Longo, 2009). Longo et al. (2010) used a within-clinic retrospective cohort design to compare cases of medically-diagnosed acute illnesses before and after the 2008 escalation in emissions and reported statistically-significant positive associations between high vog exposure and visits for cough, headache, acute pharyngitis, and acute airway problems. Both self-reported ailments and physical measurements (blood pressure and oxygen saturation) were used to document strong statistical correlations with vog exposure (Longo 2013). In this study, half the participants perceived that the post-2008 increased emissions had negatively affected their health. Most recently, Tam et al. (2016) reported an association between vog exposure and increased cough, but not with physician-diagnosed asthma, persistent cough or wheeze among children resident on the Big Island.

In response to the need for monitoring and warning information to the public, concentration data from sensors within the national park and around the island are combined with SO2

emission rates and a HYSPLIT model for plume dispersion to produce an air quality vog forecast (VMAP) for the Hawaiian Islands (Reikard 2012). Warnings are disseminated to the public via the web, radio, field units and variable road signs (Figure 3.1). The IVHHN website has a ‘vog dashboard’ (https://vog.ivhhn.org/) which includes links to vog forecasts, real-time air quality data which includes links to a colour-coded SO2 advisory system (shown in Appendix 1.1), health protection advice, vog fact sheets and information on the 2018 LERZ eruption. Edmonds et al. (2018) note that this comprehensive style of monitoring, modelling, forecasting, warning and communication might profitably be applied to other volcanic centres facing tropospheric volcanic air pollution in the future.

https://vog.ivhhn.org/


Figure 3.1 Variable warning signage near entrance to Hawai’i Volcanoes National Park.

A final air pollution hazard that has emerged during the 1983-present Kīlauea eruption is ‘laze’ (lava plus haze). Laze is produced when lava flows into the ocean and reacts vigorously to generate acidic steam plumes laden with HCl gas and volcanic glass particles (Kullman et al. 1994). Laze was implicated in the deaths of two hikers in November 2000, found near the lava ocean entry point. The final cause of death determined by the medical examiner was death as a result of pulmonary oedema caused by inhalation of volcanic laze, sustained when the victims were exposed to the plume near the ocean entry (Heggie et al. 2009).

3.3 Ambae 2017–2018 Eruption, Vanuatu

The island of Ambae is a large basaltic shield volcano in the Vanuatu archipelago. Its population in 2016 was 11,670 when a national mini-census was carried out (VNSO 2016).

The 2017-2018 Ambae eruption had four main phases. The eruption started in early September, with the first phase extending from September to November 2017. Initially, between 1 and 28 September, evacuations remained within-island, but due to fears of the eruption escalating, all residents were evacuated off-island between 29 September and 21 October. Residents moved back to Ambae between 22 October and 1 November. Strong degassing events caused acid rain damage to crops during November 2017.

The second phase of the eruption extended from December 2017 to early February 2018; during this period there were moderate SO2 emissions but relatively little ashfall. The third phase, between late February and April 2018, produced acid rain in the west of the island. Three substantial ash-producing explosive eruptions caused damage to buildings and crops and prompted within-island evacuations from the areas worst affected by thick ashfalls. The fourth phase was between July and November 2018, with two large ashfalls that affected the previously unaffected north-eastern tip of the island. The combination of structural damage to buildings (McSporran et al. 2019, in prep), ash ingress into buildings, heavy crop damage and

https://www.sciencedirect.com/topics/nursing-and-health-professions/lung-edema


contamination of water supplies prompted another mandatory off-island evacuation of the whole population, which was completed by 14 August. The State of Emergency was lifted on 26 November 2018. Since then some residents have returned to the island, with government support.

Ambae is the fifth-strongest volcanic source of passive SO2 emissions globally (2005-2015 average 2.9 kt/d, Carn et al. 2016). While the Vanuatu Meteorology and Geohazards Department (VMGD) has limited capacity for instrumental measurements of SO2 fluxes, information is available from NASA’s Aura satellite Ozone Monitoring Instrument for plumes that are sufficiently large to be detected by OMI. For the 2017-ongoing eruption of Ambae, this has been summarised in Bulletins of the Global Volcanism Program (Global Volcanism Program 2018a, 2018b and 2019) and a brief summary is presented in Table 3.1, together with other observational data. It is important to note that this is not a complete catalogue of SO2 emissions for the eruption. A further limitation is that on occasions OMI is unable to differentiate emissions from Ambae and Ambrym as spatial resolution is relatively poor, and acquisitions are everyone to two days allowing for substantial plume drift.

Table 3.1 SO2 observations throughout the 2017-ongoing Ambae eruption (adapted from BGVN 43:02, 43:07 and 44:02).

Date SO2 and related observations

October–December 2017 Significant quantities of SO2 were released with variable dispersion. Emissions were substantial enough to be recorded by OMI on October 23 and 28 and November 8, 13 and 17.

The dispersion of the 17 November plume approximately 1000 km to the SE was captured on 21 November by both OMI and TROPOMI (Theys et al. 2019).

Local residents reported rapid and widespread acid rain damage to crops on November 11 and 20 (Food Security and Agriculture Cluster 2017)

January–February 2018 Emissions were substantial enough to be recorded by OMI on January 2 and 11 and February 17 and 19.

March 2018 Reports of acid rain damage to crops.

April 2018 A large plume associated with an explosive eruption on April 5–6 was captured by OMI and drifted west towards Australia. It was described as one of the strongest SO2 plumes measured globally since 2015. Further eruptions continued on April 9–10 with a SO2 plume extending SE across most of the South Pacific.

July 2018–January 2019 On July 13, an observation flight reported ongoing steam-and-gas and ash emissions. Residents on Ambae and neighbouring islands reported ‘gas smells’ on July 13 and August 22.

Large SO2 plumes were recorded by OMI on July 22, 24, 26 and 31. These generally drifted eastwards.

Intermittent gas and steam, and ash, plumes have continued through to January 2019.

From Table 3.1, Ambae residents reported acid rain impacts on crops, particularly in November 2017 and March 2018. Further details about crop damage are documented in a needs-assessment report carried out on Ambae between 21-27 November 2017 by a range of crop, livestock and fisheries officers and food security and nutrition experts from Vanuatu government agencies (FSAC 2017). Acid damage was recorded to food, cash and forestry crops and also to farmed fish, which were an important source of protein to the island diet. Fish (tilapia) are reportedly very sensitive to water acidity and died following the November 2017 acid rains. Reported crop damage included yellowing of foliage and, in severe cases, ‘burning’,


blackening and stripping of foliage. Similar effects on vegetation have been noted downwind of Masaya volcano, Nicaragua, during degassing crises (Delmelle et al. 2001; 2002). As a note of caution, it is unclear from this report whether the observed damage was due primarily to SO2 gas (or other volcanic gases), sulfate aerosol, acid rain or volcanic ash deposition as all these constituents of volcanic plumes occur together (‘volgara’).

Nineteen residents of West Ambae were interviewed between 6-9 March 2018 about their experiences of volcanic impacts and damage during the Ambae eruption, by a joint VMGD/GNS Science field impact assessment team. Residents were asked which of a range of volcanic impacts they had experienced at their location, and whether impacts were minor, major or no impact, for two time periods: September through to December 2017, and then January through to early March 2018 (Table 3.2). Perceived levels of impact reduced markedly for both hazards between the earlier and the later time period. In the earlier time period, acid rains were perceived as having greater impact than gas, with 62% of respondents describing impacts as major compared to 44% of respondents describing gas impacts as major, but the reverse was the case for the later time period. Reports from further afield often indicated this crop damage was from ashfall. It was not until this survey was conducted and the nature of the impacts were discussed in detail that the balance of acid rain over gas or ashfall was clarified as the main cause of damage. We acknowledge the limitations inherent in a study involving such a small number of participants.

Table 3.2 West Ambae residents’ perceptions of the impacts of acid rain and volcanic gas when interviewed in early March 20181.

Acid rain Volcanic gas No impact

Minor impact

Major impact

No impact

Minor impact

Major impact

% of valid responses2 % of valid responses1

Sept–Dec 2017 0% 38% 62% 13% 44% 44%

Jan–early March 2018 83% 17% 0% 67% 33% 0% 1 This project was assessed as low-risk by the Massey University Human Ethics Committee (Ethics Notification

Number: 4000019031) 2 There were 18 valid responses for the Sept-Dec 2017 question and 16 valid responses for the Jan 2017 to early

March 2018 question

While residents expressed concern about volcanic gas smells, no surface gas monitoring data was available for Ambae thus it was not possible to provide specific advice. Generic advice on coping with ash, gas and acid rain was prepared by the field team and is hosted on the VMGD website as ‘Ambae Volcano Advice Key Messages’. The key features of this advice were that:

1. it was self-contained and included advice for ash, gas and acid rain as well as related advice such as awareness of lahar hazards

2. it was appropriate for the local context and

3. it was multi-agency.

A wide range of agencies were consulted in its preparation, including international (World Health Organisation, the International Volcanic Health Hazard Network and World Animal Protection), New Zealand and Vanuatu government agencies, and the inclusion of these organisations’ logos on the advice added to its authority and credibility.

https://www.vmgd.gov.vu/vmgd/index.php/geohazards/volcano/volcano-info/resources

https://www.vmgd.gov.vu/vmgd/index.php/geohazards/volcano/volcano-info/resources


3.4 Summary

The 2014-2015 Holuhraun eruption was the largest in Iceland since the 1783–1784 Laki eruption and has been described as “one of the most intense, large-scale volcanogenic air pollution events in centuries” (Ilyinskaya et al. 2017). While its local impacts may have been mitigated somewhat by both its remote location and timing (in autumn/winter), SO2 emissions nonetheless exceeded air quality standards in populated areas of Iceland for days to weeks, and caused minor, transient deteriorations in air quality thousands of kilometres downwind in Europe. The Icelandic Meteorological Office responded with effective air quality monitoring (using pre-existing networks) and modelling, a citizen science initiative to report volcanic gas smells and health impacts, and the dissemination of effective public health advice.

The ongoing eruption of Kīlauea has affected downwind populations since 1983, and especially since 2008. Over this extended timeframe, comprehensive and effective systems have been developed for monitoring, modelling, forecasting, warning and communication to the public about volcanic emissions. These systems provide a model for other volcanic centres facing tropospheric volcanic air pollution in the future (Edmonds et al. 2018).

Ambae volcano, Vanuatu, is one of the strongest volcanic SO2 sources globally. While eruptive fluxes during the 2017-ongoing eruption were not measured in-country, OMI data showed substantial SO2 emissions during this period, and wide dispersal of the plume across the South Pacific. The eruption vent is at ~1500 metres altitude, thus the plume may not have grounded locally with emissions dispersed away from the island. Consequently, the main impact on local communities may have been acid rain rather than SO2 exposure although it is difficult to disentangle the effects of volcanic ash, gas and acid rain. The value of self-contained, multi-hazard, locally appropriate, multi-agency advice to local populations was also clear.

The following key lessons can be taken from these case studies:

• For eruptive vents close to residential areas, the potential for hazardous concentrations of SO2 is high.

• Pre-existing air quality monitoring networks (for SO2 and fine particles) are of great value in the event of a SO2-rich eruption or a sustained non-eruptive degassing episode.

• The nature of volcanic air pollution will vary depending on proximity to the vent. In the immediate vicinity of the vent, the primary hazard is gas-phase SO2. At distances of a few km to tens of km, populations are likely to be exposed to SO2 and sulfate aerosol particles. At greater distances downwind, particles (assessed as PM10 and PM2.5) are the primary air pollution hazard.

• Impacts on local populations and the environment depend on vent location, timing of the eruption, and plume dispersion height.

• Acid rain may be a significant associated hazard and may cause impacts on surface waters, crops and vegetation.

• In practice, it may be difficult to disentangle impacts from gas, ash and acid rain.

• The comprehensive system of SO2/vog monitoring, modelling, forecasting, warning and communication developed in Hawai’i serves as a useful example for emergency managers for future volcanic air pollution episodes.

• Public messaging should be multi-hazard, locally appropriate and multi-agency.


4.0 MONITORING OF SO2 AND PARTICULATE MATTER

4.1 Regional Monitoring by Auckland Council

Regional councils and unitary authorities measure outdoor air quality in their regions as part of their responsibilities for managing air quality under the Resource Management Act 1991. New Zealand has National Environmental Standards for Air Quality (NES-AQ) that set limits in outdoor air for five air pollutants, including SO2 and PM10.

In Auckland, PM10 is currently monitored by Auckland Council at nine sites across the region, and PM2.5 at seven of these nine sites (Figure 4.1). At the time of writing, SO2 is only monitored at the Penrose site, but another SO2 monitoring site at Customs St, adjacent to the port area, is expected to go live shortly. A temporary monitoring station for SO2 was set up in the Ports of Auckland area during the period 2011 to 2014, primarily to monitor emissions from shipping (Talbot and Reid 2017).

Figure 4.1 Locations of PM10, PM2.5 and SO2 monitoring stations in Auckland region (NOx monitoring not

shown).


4.1.1 SO2

There are two NES-AQ for SO2 in outdoor air in New Zealand. The lower threshold is 350 µg/m3 as a one-hour average, with nine exceedances permitted per year in each airshed. The upper threshold is 570 µg/m3, for which no exceedances are permitted. These standards are listed alongside other New Zealand and international guideline values for different averaging periods (Table 4.1).

Table 4.1 Standards and guidelines for SO2.

Averaging period 10 minutes 1 hour 24 hours

World Health Organisation guideline value1 500 µg/m3

NES-AQ lower threshold2 350 µg/m3

NES-AQ upper threshold2 570 µg/m3

NZ outdoor guideline value2 120 µg/m3

World Health Organisation outdoor guideline value3 20 µg/m3

1 https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health 2 https://www.stats.govt.nz/indicators/sulphur-dioxide-concentrations

3 https://www.mfe.govt.nz/air/air-guidance-and-wood-burners/ambient-air-quality-guidelines

Concentrations of SO2 are generally very low at the Penrose monitoring site. During the period 2008–2017, there were no exceedances of the lower NES-AQ threshold of 350 µg/m3 and just two exceedances of the far more stringent WHO 24-hour average guideline value of 20 µg/m3. In contrast, the Ports of Auckland site recorded 9 exceedances of the lower NES-AQ threshold and 50 exceedances of the WHO 24-hour standard during 2011-2014, indicating the importance of contributions from shipping. Based on a survey using passive diffusion tubes, which allow for greater spatial resolution, Talbot and Reid (2017) note that concentrations of SO2 at the waterfront are approximately four times higher than at any other site in Auckland. In downtown Auckland, concentrations are highest when the wind blows from the direction of the shipping sources.

4.1.2 Particulate Matter (PM)

At present, the only New Zealand NES-AQ for particulate matter in outdoor air is a 24-hour average value of 50 µg/m3 for PM10, for which one exceedance is permitted per year in each airshed and monitoring of PM10 is mandatory. Internationally, the trend has been towards monitoring PM2.5 as a more relevant metric for assessing health hazards of particulate air pollution (WHO 2013), but so far it is not mandatory in New Zealand, and monitoring is limited to sites in Auckland, Wellington and Canterbury. Auckland Council has drawn up its own Auckland Ambient Air Quality Targets (AAAQTs, Table 4.2) based on international best practice (WHO 2013).

https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health

https://www.stats.govt.nz/indicators/sulphur-dioxide-concentrations

https://www.mfe.govt.nz/air/air-guidance-and-wood-burners/ambient-air-quality-guidelines


Table 4.2 PM standards, guidelines and targets for Auckland (from Talbot et al. 2017)

Particle size PM10 PM2.5

24 hours Annual 24 hours Annual

AAAQTs 20 µg/m3 25 µg/m3 10 µg/m3

NES-AQ 50 µg/m3

Allowable exceedances per year

1

A recent report on outdoor air quality trends in Auckland over the period 2006-2015 (Talbot et al. 2017) describes generally declining trends in PM10 and PM2.5 across the region. These reductions have been attributed to reduced contributions from vehicles as fuel quality has improved, with the most marked declines at peak traffic sites. From elemental analysis of PM (Davy et al. 2016), secondary sulfate (formed from oxidation of SO2) makes a minor but appreciable contribution to both PM10 (9-14%) and PM2.5 (12-17%) at all five sites.

4.2 Geonet Monitoring of Gas Fluxes

Geonet monitors emissions of the gases CO2, SO2 and H2S at Ruapehu, Tongariro, Ngauruhoe and White Island volcanoes. The most intensive monitoring is carried out at White Island, which is currently the largest source of volcanic gas emissions. Both ground-based and flight-based methods are used. For ground-based methods, an installation of two mini-DOAS (miniature differential optical absorption spectrometer) instruments are used to profile the plume, and fluxes of carbon dioxide and hydrogen sulphide from soil are also measured regularly. Gas flights are scheduled approximately monthly but may be stepped up in response to increased activity. Remote sensing, using a correlation spectrometer (COSPEC) or the lightweight FLYSPEC version, is used to quantify outputs of SO2, and plume contouring (where plume gases are sampled directly and analysed) is used for all three gases.

A visualisation of SO2 mini-DOAS data (Figure 4.2) shows that fluxes are highly variable but are generally consistent with an estimated (from OMI data) average SO2 flux during 2005-2015 of 254 t/d (Carn et al. 2016). Ten-day means are also routinely calculated from mini-DOAS data (Figure 4.3); from this data, a declining trend since 2014 is more apparent.

Data from gas flights is reported in kg/s, as data is collected in four-minute time intervals, and can be visualised on the interactive FITS/GUI tool (https://fits.geonet.org.nz/). This data extends further back, to 2003, and illustrates the same general trend of decreasing emissions since 2014. There has been a recent (since March 2019) surge in emissions, with a flux of ~22 kg/s (~1900 t/d) recorded on 26 June 2019.

https://fits.geonet.org.nz/


Figure 4.2 Mini-DOAS data for SO2 fluxes at White Island (t/d) since 2012 (credit: Steve Sherburn).

Figure 4.3 10-day means for mini-DOAS SO2 data from White Island (t/d)(credit: Steve Sherburn).

4.3 Summary

• Ambient SO2 concentrations are currently monitored at one site in Auckland, in Penrose, with a second site adjacent to the port area expected to go live shortly.

• Concentrations of SO2 in Auckland are generally very low. The highest concentrations occur around the waterfront due to emissions from shipping.

• PM10 is monitored at nine sites across the region, and PM2.5 at seven sites. Both of these parameters have declined in recent years, particularly at high-traffic sites.

• Geonet monitors fluxes to air (in tonnes/day) of SO2, H2S and CO2 at Ruapehu, Tongariro, Ngauruhoe and White Island volcanoes. For SO2, regular monitoring is carried out at White Island and Ruapehu, and on an as-required basis at Tongariro and Ngauruhoe. The average flux from White Island from OMI data was 254 t/d during the period 2005-2015, but there has been a general trend of decline since 2014 with a recent surge in emissions in 2019.


5.0 ESTIMATING SO2 FLUXES FOR THE AUCKLAND VOLCANIC FIELD

5.1 AVF Eruption Scenarios

As noted in Section 1.0 and Table 1.1, eight eruption scenarios for the AVF have been developed by DEVORA partners for the purpose of evaluation and quantifying volcanic risks to Auckland. Vent locations, erupted volumes and eruption durations from Hayes et al. 2018 are shown in Table 5.1.

Table 5.1 Summary details for AVF eruption scenarios (from Hayes et al. 2018).

DEVORA scenario Location

Erupted volume (km3)

Erupted volume (m3)

Eruption duration (days)

A Auckland airport 1 x 10-3 1.0 x 106 4

B Ōtāhuhu 1.95 x 10-2 20 x 106 32

C Māngere Bridge 1 x 10-1 100 x 106 32

D Mt Eden 8.6 x 10-2 86 x 106 320

E Waitematā Port 7.6 x 10-3 7.6 x 106 27

F Birkenhead 1.43 x 10-2 14 x 106 160

G Rangitoto Channel 2.25 x 10-4 0.2 x 106 8

H Rangitoto Island 1.5 x 10-1 150 x 106 104

Work is underway by one of the authors of this report using petrological analysis of the sulfur content in melt inclusions and quenched eruption products (e.g., Thordarson et al. 2011) to estimate gas fluxes, including SO2, for each of the eight AVF scenarios. Estimated average and maximum fluxes will be used as an input to air dispersion models (see Section 6.1) to estimate impacts of future AVF eruptions on ground-level airborne SO2 concentrations, which can then be related to health-based air quality guidelines (Table 2.1).

5.2 SO2 Fluxes from Analogous Eruptions

An alternative, interim approach to estimating SO2 fluxes from future eruptions of the AVF is to refer to analogous basaltic eruptions worldwide for which erupted magma volumes and SO2 emissions and fluxes have been reported (Table 5.2 and Figure 5.1). Several of these eruptions are broadly similar to AVF eruption scenarios in terms of erupted volume and duration.


Table 5.2 Erupted volumes, eruption durations, total SO2 emissions, average and peak SO2 fluxes from recent basaltic eruptions.

Volcano (year of

eruption)

Erupted volume

(m3)

Eruption duration

(d)

Total SO2 emissions

(t)

Average SO2 flux

(t/d)

Peak SO2 flux (t/d)

Similar to DEVORA scenario:

Miyakejima (2000)1 3 x 106 ~1200 18,000,000 42,000 200,000 -

Stromboli (2007)2

3.2-11 × 106 34 22,000 650 ~1,200 B, E

Etna (2015)3 5-12 x 106 <3

21,000 ± 2,730

>7,000 13,900 -

Nyiragongo (2002)4 30±10 x 106 <1

15,000-48,000

~15,000-48,000

73,000–148,000

-

Eldfell (1973)5 110 x 106 21

~300,000 ± 100,000

(est)

~15,000± 5,000 (est)

~30,000± 10,000 (est)

C

Holuhraun (2014–2015)6 1600±300 x 106 180

11,000,000

±5,000,000 60,000 112,000 -

1 Geshi et al. 2002; Kazahaya et al, 2004; Harada et al. 2013. 2 Burton et al. 2009; Calvari et al. 2010. 3 D’Aleo et al. 2019. 4 Carn 2002. 5 Global Volcanism Program 1973a,b. The reported eruptive volume was for the first 21 days of the eruption. In this

case, SO2 emissions and fluxes were not reported, so are estimated here from the relationship proposed by Wallace (2001) between erupted magma volume and minimum SO2 emissions and apply only to the 21-day period.

6 Gíslason et al. 2015; Schmidt et al. 2015; Ilyanskaya et al. 2017.

An approximately linear positive relationship exists between erupted magma volume and total SO2 emissions (Figure 5.1), with the notable exception of the 2000 eruption of Miyakejima volcano, Japan, which had an exceptionally large emission rate relative to its erupted volume. The total SO2 emission of 18 Mt was comparable to the 20 Mt released by the 1991 Pinatubo eruption (Kazahaya et al. 2004). These authors suggested that the creation of a large magma conduit system formed by caldera collapse (Geshi et al. 2002) was probably the cause of these extremely high gas emission rates.

For the following sections, we assume that the 2007 Stromboli eruption is broadly similar to AVF eruption scenarios B and E, and that the 1973 Eldfell eruption is broadly similar to AVF eruption scenario C based on erupted volume and eruption duration, and we apply the average and peak SO2 fluxes listed in Table 5.2. Eruption scenarios B and E are close to the median AVF erupted volume (Hayes et al. 2018; Kereszturi et al. 2013).


––

Figure 5.1 Relationship between erupted magma volume and total SO2 emissions for the six eruptions shown

in Table 5.2. Bars represent reported ranges.

5.2.1 Estimated SO2 Fluxes from AVF Eruption Scenarios in Context of Total Anthropogenic Emissions for New Zealand

A national emissions inventory was commissioned by the Ministry for the Environment in 2017. Emissions for all significant anthropogenic sources of the air pollutants PM10, PM2.5, CO, NOx and SO2 were estimated for the base year of 2015 (Metcalfe and Sridhar 2018). Total SO2

emissions from anthropogenic sources in New Zealand were estimated as 49,900 tonnes per annum in 2015, or approximately 140 t/d. Significant anthropogenic sources of SO2 include industrial processes (primarily fertiliser manufacturing, aluminium smelting and steel-making) and fossil fuel combustion, particularly for marine fuel oil and coal. The largest single point source emitter of SO2 is the Tiwai Point aluminium smelter in Southland, which emits ~20 t/d, followed by the NZ Steel plant in Glenbrook, 40 km south of Auckland (~2.5 t/d).

Transport sector emissions are dominated by shipping, because most marine fuel oils have a high sulfur content (typically 2.7%, or 27,000 mg/kg S) and are not currently regulated in New Zealand (Peeters 2018). In contrast, the current regulated limit of S in petrol and diesel in New Zealand is 10 mg/kg (Metcalfe and Sridhar 2018).

Total daily SO2 emissions from anthropogenic sources in New Zealand (~140 t/d) are smaller than New Zealand’s largest point source of SO2, White Island volcano in the Bay of Plenty. Carn et al. (2016) estimated an average flux for the period 2005–2015 (a period of passive degassing with no eruptions) for White Island of 254 t/d, from Ozone Monitoring Instrument (OMI) data from NASA’s Aura satellite, broadly similar to mini-DOAS data (Figure 4.3). Emissions from White Island have been generally higher in 2019 (Section 4.2).

1

10

100

1,000

10,000

100,000

1 10 100 1000 10000

Tota

l SO

2em

issio

n (k

t)

Erupted volume (x106 m3)

Miyakejima (2000) Holuhraun (2014-2015)

Eldfell (1973)

Nyiragongo (2002)Stromboli (2007)

Etna (2015)


Fluxes for the 2007 Stromboli, 1973 Eldfell and 2014–2015 Holuhraun eruptions are shown on Figure 5.2, together with total daily SO2 emissions from all anthropogenic sources for New Zealand and emissions from White Island.

Figure 5.2 Average and peak SO2 fluxes from the 2007 Stromboli, 1973 Eldfell and 2014–2015 Holuhraun

eruptions, in comparison to estimated SO2 fluxes for all anthropogenic sources in New Zealand. Emission rates from New Zealand’s two largest industrial point sources are also shown. Bars represent reported ranges.

For the 2007 Stromboli eruption (similar to AVF scenarios B and E), both average and peak fluxes exceed total NZ anthropogenic fluxes by factors of 4.6 and 8.6 respectively. The larger 1973 Eldfell eruption, similar to AVF scenario C, exceeds total anthropogenic fluxes by a factor of ~100 for its average flux and ~200 for its peak flux. The dominance of volcanic over anthropogenic sources of SO2 was also observed for the 2014–2015 Bárđabunga fissure eruption, Iceland, where during the eruption, daily SO2 emissions exceeded daily SO2 emissions from all anthropogenic sources in Europe by a factor of at least 3 (Schmidt et al. 2015; Carboni et al. 2019).

Relative to passive degassing at White Island, which is currently New Zealand’s largest point source of SO2, 2007 Stromboli average and peak fluxes are generally similar to emission rates recorded during 2019. The 1973 Eldfell average and peak fluxes are approximately an order of magnitude higher than 2019 White Island emission rates.

5.2.2 Estimated SO2 Fluxes from AVF Eruption Scenarios in Context of Regional and Local Anthropogenic Emissions for Auckland

Specific air pollution emission inventories have also been created for Auckland, for industrial sources (Crimmins 2018), shipping (Peeters 2018), road and air transport (Sridhar and Metcalfe 2019) and home heating (Metcalfe et al. 2018), for the base year 2016 (Table 5.2).

1

10

100

1000

10000

100000

2007 Stromboliaverage

2007 Strombolipeak

1973 Eldfellaverage

1973 Eldfell peak 2014-2015Holuhraun

average

SO2

emis

sion

, t/d

Analogous basaltic eruptions

Total NZ anthropogenicemissions (est. for 2015)

White Island (2005-2015)

Tiwai Point aluminiumsmelter (est. for 2015)

NZ Steel (est. for 2015)

White Island (2019 peak flux)

White Island (31 Oct 2019)


Table 5.3 Estimated fluxes of SO2 to air in the Auckland region by sector for the year 2016.

Sector Study SO2 flux

(t/y) SO2 flux (t/day)

Industry Crimmins et al. (2018) 1043 2.9

Industry, urban area only Crimmins et al. (2018) 114 0.3

Shipping Peeters et al. (2018) 472 1.3

Road and air transport Sridhar and Metcalfe (2019) 196 0.5

Home heating Metcalfe et al. (2018) 38 0.1

Total regional anthropogenic sources Above refs 1749 4.8

Total anthropogenic sources, urban area only Above refs 820 2.2

For the 2007 Stromboli eruption (similar to AVF scenarios B and E), both average and peak fluxes exceed Auckland regional anthropogenic fluxes by factors of ~140 and ~250 respectively. The larger 1973 Eldfell eruption, similar to AVF scenario C, exceeds Auckland regional anthropogenic fluxes by a factor of >3000 for its average flux and >5000 for its peak flux (Figure 5.3). The dominance of a new AVF source of SO2 is likely to be even more pronounced if emissions to Auckland’s urban area only are considered (Figure 5.3). This is because regional-scale SO2 emissions are heavily dominated by the NZ Steel plant at Glenbrook, which is outside the Auckland urban area.

Figure 5.3 Average and peak SO2 fluxes from the 2007 Stromboli, 1973 Eldfell and 2014–2015 Holuhraun

eruptions, in comparison to estimated 2016 regional and urban SO2 fluxes in the Auckland region.

1

10

100

1000

10000

100000

2007 Stromboliaverage

2007 Strombolipeak

1973 Eldfellaverage

1973 Eldfell peak 2014-2015Holuhraun

average

SO2

emis

sion

, t/d

Analogous basaltic eruptions

Auckland regional emissions est. for 2016Auckland urbanemissions est. for 2016


A final consideration is that the relative dominance of a new SO2 source from a future AVF eruption compared to existing local emissions is expected to increase in the future, because the major source of SO2 to urban Auckland (shipping) is expected to decrease markedly in future due to the implementation of a cap on the allowable sulfur content in marine fuel oil by 2020 (Peeters 2018). As satellite-based SO2 measurements will have low SO2 background concentrations, new volcanic sources of SO2 would be expected to have a strong signal:noise ratio.

Table 5.4 Predicted trends in annual anthropogenic SO2 emissions to Auckland urban area from 2016 to 2040 (from Peeters 2018).

Year SO2 flux (tonnes/year)

2016 2026 2036 2040

Shipping 472 110 19 16

Total SO2 emissions to Auckland urban area 820 458 367 364

5.3 Summary

• For this section, we have identified several recent basaltic eruptions for which erupted magma volumes and SO2 emissions and fluxes have been reported.

• Based on erupted volume and eruption duration, the 2007 Stromboli eruption is broadly similar to AVF scenarios B and E, and the 1973 Eldfell eruption is broadly similar to AVF scenario C.

• We then compared average and peak SO2 emission rates for the 2007 Stromboli and 1973 Eldfell eruptions to total anthropogenic SO2 sources for New Zealand, Auckland regional and urban sources, and emissions from White Island.

• SO2 average and peak emission rates from the 2007 Stromboli eruption are similar to the range of SO2 emission rates measured for White Island during 2019. They exceed total anthropogenic sources of SO2 for New Zealand by factors of ~5 (for average flux) and ~10 (for peak flux). They heavily dominate SO2 emissions for the Auckland region (by factors of ~140 and ~250) and for Auckland’s urban area (by factors of ~300 and ~600).

• For the larger 1973 Eldfell eruption, dominance over White Island SO2 emissions and anthropogenic emissions is at least an order of magnitude larger than for the 2007 Stromboli eruption.

• Work is currently underway by one of the authors of this report on petrological analysis of the sulfur content in melt inclusions and quenched eruption products for AVF erupted magmas. This work may provide more specific estimates of gas emissions for the eight DEVORA AVF eruption scenarios.


6.0 KEY ISSUES AND RECOMMENDATIONS

Estimates of gas emissions for DEVORA AVF scenarios B and E (noting their broad similarity to the 2007 Stromboli eruption) suggest that average and peak SO2 fluxes of ~650 t/d and ~1200 t/d are expected. These fluxes are similar to the range of SO2 fluxes recorded for White Island volcano during 2019 and exceed total daily SO2 fluxes from all known anthropogenic sources in New Zealand by factors of 5–10. In relation to regional and local sources of SO2, new volcanic sources are expected to be even more dominant. A new AVF eruption will therefore be a major new air pollution point source in metropolitan Auckland. Impacts will depend on the following factors: vent location, meteorological conditions, populations and assets exposed, dispersion height and timing.

For effective management of a future AVF eruption we identify the following information needs and recommend approaches to address these:

6.1 Improved Capability for SO2 Dispersion Modelling and Forecasting

Developing the capability to model SO2 dispersion from a new volcanic source will be useful in two main ways. Firstly, it can be used in pre-crisis settings to generate hazard footprints for the eight DEVORA scenarios in order to explore the consequences for downwind populations, buildings and infrastructure, to add to existing efforts for other hazards (Deligne et al. 2015, 2017; Blake et al. 2017). Secondly, given that the warning period for a new AVF eruption may be as short as a few days (Brenna et al. 2018; Sherburn et al. 2007), rapid hazard assessment will be necessary in the event of unrest to provide timely information to emergency managers.

In New Zealand, two air dispersion models are currently approved for regulatory purposes. They are summarised in Table 6.1.

Table 6.1 Comparative features of AERMOD and CALPUFF air dispersion models1.

Feature AERMOD CALPUFF

Description Gaussian plume dispersion model Multi-layer, multi-species puff dispersion model

Complexity Relatively simple Advanced (can account for gas to particle conversions)

Range Short (less than 50 km) Long (up to 100s of km)

Steady state or non-steady state Steady state Non-steady state

Ease of use Moderate Difficult

Accessibility Open source Closed source

Examples of use for SO2 hazard modelling

Jenkins et al. 2015 Gíslason et al. 2015

1 http://tools.envirolink.govt.nz/

Jenkins et al. (2015) listed the following model input parameters for AERMOD: vent elevation, exit temperature, proportion of SO2, gas exit velocity and density, vent diameter, albedo, Bowen ratio and surface roughness. Outputs were in the form of average percentages of days (in a two-year degassing period) that SO2 exceeded air quality guidelines in different population centres. Gíslason et al. (2015) used the CALPUFF model initialised with DOAS estimates of SO2 fluxes to simulate the dispersal of the SO2 plume and its surface concentration across Iceland. The output was a map showing the frequency of hourly concentrations of SO2

http://tools.envirolink.govt.nz/


exceeding the 350 µg/m3 European Union one-hour limit. The Icelandic Meteorological Office also used simulations to produce three-day-long forecast models for SO2, twice daily.

Other models that have been used for SO2 dispersion modelling include HYSPLIT for vog modelling and forecasting in Hawai’i (see Section 3.2), and the UK Met Office’s NAME atmospheric dispersion model (Schmidt et al. 2014, 2015). Schmidt et al. (2015) noted that in the far field, modelled SO2 concentrations were biased low compared to surface measurements by a factor of up to eight. They further suggested that SO2 measurements need to be assimilated for near real time air quality forecasting; and that existing monitoring networks be retained or extended. It may also be worth considering the deployment of mobile monitoring capacity such as portable SO2 meters (Joseph 2004). Joseph et al. (2019) also note that deploying passive diffusion tubes is a simple and cost-effective air quality monitoring tool, although this method provides only time-integrated concentrations over a given exposure period rather than providing real-time information on ambient SO2 concentrations.

To develop local capability for SO2 dispersion modelling and forecasting, we recommend a workshop approach involving key parties such as volcanologists, air quality and environmental health scientists, atmospheric dispersion modelling specialists and meteorologists.

6.2 Improved Capacity to Forecast Impacts of Volcanic Gases on Infrastructure and the Built Environment

Observational and field studies pointing towards accelerated corrosion of metal building materials and infrastructure components exposed to volcanic emissions were summarised in Section 2.2. However, there has been very limited progress towards the development of quantitative relationships between hazard intensity (i.e. ambient concentrations of SO2 and other corrosive gases) and impact (i.e. rates of corrosion of different materials).

Of the field studies listed in Table 2.2, only Li et al. (2018) determined ambient concentrations of SO2 and H2S. This comprehensive study by BRANZ staff measured time-averaged concentrations of SO2 and H2S, recorded meteorological variables, recorded corrosion rates in terms of mass loss, identified corrosion products, and included reference sites in the experimental design. An extremely high corrosion rate for mild steel (mass loss of 3000 g/m2/year) was recorded at a site in the geothermal field with an average SO2 concentration of ~100 µg/m3. This compares to rates of 170 g/m2/year at a rural reference site and 650 g/m2/year at a ‘severe marine’ reference site exposed to sea salt aerosol.

However, even the highest SO2 concentrations measured in the Li et al. (2018) study (~100 µg/m3) are dwarfed by SO2 concentrations measured close to erupting vents (>260,000 µg/m3 recorded close to LERZ fissure in 2018 Kīlauea eruption, Section 3.2) or in grounding plumes (3100 µg/m3 recorded at distances of 4–20 km from the vent of the 2014–2015 Holuhraun fissure eruption). Therefore, the applicability of the findings from the Li et al. (2018) study to SO2-rich eruptions or sustained degassing episodes may be limited.

We suggest that further investigations of corrosion rates at higher concentrations of SO2 are warranted, to develop fragility functions for acid gas damage to infrastructure components and building materials in the event of sustained volcanic gas emissions. To this end we plan to develop a research proposal in collaboration with BRANZ, extending the previous work into more extreme environments, such as White Island.


6.3 Managing Public Health Impacts

6.3.1 Exposure Assessment

Exposure levels to volcanic gas emissions can be assessed using data on ambient concentrations of SO2, PM10 and PM2.5 from existing air quality monitoring stations. These may be supplemented as required by deploying portable meters and/or passive diffusion tubes (Section 6.1) for increased spatial coverage, particularly for SO2 where there is currently only one existing and one planned monitoring site in Auckland.

Volcanic gases HCl and HF are not routinely monitored in New Zealand. Analysis of rainfall for acidity, sulfate, chloride and fluoride should provide insight into plume gas composition (Edmonds et al. 2003; Hardin and Miller 1982).

Monitoring of CO2 and H2S in confined spaces such as basements is also recommended to assess risks of these gases accumulating to hazardous concentrations (Durand and Scott 2005).

In the event of lava entry into seawater, an exclusion zone to protect the public from ‘laze’ (see Section 3.2) is strongly recommended.

6.3.2 Assessing Health Impacts (Syndromic Surveillance)

Assessment of health impacts from volcanic activity is vitally important for the provision of evidence-based public health messaging to vulnerable groups exposed to volcanic emissions, as well as the general public. Data from syndromic surveillance can be used to answer the question: has there been a short-term increase, at population level, in adverse health outcomes following volcanic activity?

The International Volcanic Health Hazard Network has recently developed a set of standardised protocols to assess the health impacts from volcanic eruptions. A basic, standardised epidemiological study protocol to assess short term respiratory and other health impacts (IVHHN 2019) recommends:

• Assessing exposure to volcanic emissions (Section 6.31)

• Identifying all healthcare facilities receiving patients from exposed areas, and collating numbers of visits and reasons for visits, along with demographic information

• Combining data from facilities and assessing the counts and proportions of respiratory and cardiovascular visits in exposed areas separately for pre- and post-eruption periods, separated into age groups and by gender

• Assessing mortality counts over the same period for the same areas

• Comparing counts and proportional visits during and after the event with pre-eruption levels

• Determining if there has been a statistical increase in visits or mortality.

Evidence of adverse impacts on health may be useful to justify deployment of resources and/or more in-depth investigations and forms the basis of public health messaging.


In terms of reasons for visits, New Zealand codes visits according to the International Classification for Diseases scheme ICD-10-AM/ACHI/ACS 11th Edition. The IVHHN (2019) protocol recommends the following ICD categories as particularly relevant to assessing health outcomes from volcanic eruptions:

• Respiratory disease (J00–J99), focusing on:

- Acute lower respiratory infections or diseases (J20–J22)

- Chronic lower respiratory infections or diseases (J40–J47)

- Acute upper respiratory infections or diseases (J00–J06)

- Chronic upper respiratory infections or diseases (J40–J47)

• Cardiovascular endpoints (I20–I52)

• Cerebrovascular endpoints (I60–I69)

• Disorders of the eye (H55–H59)

• Accidents (V01–X59)

• Mental health, such as reaction to severe stress (F43)

Other data that may be useful can include the dispensing of medications such as inhalers and calls to official advice services such as Healthline. Efforts to undertake such a study during or following an eruption need to be coordinated with emergency management responses, and ideally should be planned in advance (including a consideration of whether ethics approval is needed).

6.3.3 Communicating with the Public

Public demand for information in the event of a sustained volcanic degassing episode in Auckland can be anticipated. We intend to produce two new DEVORA fact sheets on the basis of this report: 1) background information on volcanic gases in the context of the AVF, and 2) a guide to the health impacts of volcanic gases and suggested mitigation measures. Evidence-based public messaging should follow, when results of health impacts assessments are available, with particular attention to vulnerable groups.

Ideally, real-time information on air quality will be communicated to the public in the event of a future AVF eruption. A dedicated website, including an interactive map linked with mitigation measures, could be an effective solution. The use of a colour-coded system linking airborne concentrations of SO2 to information on health impacts and recommended actions for sensitive groups and the general public (Appendix 1.1) may also be useful.

6.3.4 Considering Combined Exposure to Volcanic and Urban Air Pollution

Communities resident in urban areas located near active volcanoes can experience volcanic ash and gas exposures during and following an eruption, in addition to sustained exposures to high concentrations of anthropogenic air pollutants such as vehicle exhaust emissions. Few epidemiological studies have considered human health hazards and/or effects associated with the combined inhalation exposures that result from the addition of volcanic PM to the urban environment.

Two recent laboratory-based, in vitro studies assessed the biological impact of a short-term, concomitant exposure to respirable volcanic ash with diesel exhaust particles and complete gasoline exhaust (Tomašek et al. 2016; 2018). When co-exposed, volcanic ash and diesel


exhaust particles were found to induce a heightened pro-inflammatory response in vitro, greater than the response noted for each particle type separately, implying a potentially-greater hazard of simultaneously inhaling both particle types (Tomašek et al. 2016). The observed responses were largely attributed to the effect of the greater combined dose of particles delivered to the lung cell surface, indicating that the potential health impact is likely to be similar to that expected to result from the increased concentration of particulate matter in the air. A lack of such effect was seen, however, during exposure to complete gasoline exhaust in the presence (or absence) of volcanic ash (Tomašek et al. 2018).

Based on the current knowledge, it could be assumed that, when combined in the ambient air, PM of volcanic and/or anthropogenic origin may contribute to the effects of inhaling gaseous volcanic SO2 and acid aerosols during an eruption, but the biological impacts of such co-exposures warrant further studies. In the absence of this evidence, emergency management agencies may wish to take a precautionary approach to public health guidance, especially if combined exposure to volcanic and urban PM potentially poses an increased hazard toward respiratory health. Agencies could advise that citizens should minimise or avoid exposure especially during periods of intense exposure to both urban pollution and volcanic emissions.


7.0 ACKNOWLEDGEMENTS

CS received funding from DEVORA via GNS EQC subcontract GNS-EQC00021. IT was supported by the VUB Strategic Research Program.

The authors thank Associate Professor Kim Dirks and Dr Simon Thornley of the School of Population Health, University of Auckland, for expert comment on Section 6.3 (managing public health impacts) of this report, and Dr Steve Sherburn of GNS Science for providing information on Geonet monitoring of gas fluxes. We further thank Dr Geoff Kilgour and Dr Perry Davy, of GNS Science, for thorough and constructive peer review of this report.

8.0 REFERENCES Barsotti S. 2014. Hazard zoning. Reykjavik (IS): Icelandic Met Office; [accessed 2019 Aug 13].

https://en.vedur.is/pollution-and-radiation/volcanic-gas/hazard-zoning/

Brenna M, Cronin SJ, Smith IEM, Tollan PME, Scott JM, Prior DJ, Bambery K, Ukstins IA. 2018. Olivine xenocryst diffusion reveals rapid monogenetic basaltic magma ascent following complex storage at Pupuke Maar, Auckland Volcanic Field, New Zealand. Earth and Planetary Science Letters. 499:13–22. doi:10.1016/j.epsl.2018.07.015.

Burton MR, Caltabiano T, Murè F, Salerno G, Randazzo D. 2009. SO2 flux from Stromboli during the 2007 eruption: Results from the FLAME network and traverse measurements. Journal of Volcanology and Geothermal Research. 182(3):214–220. doi:10.1016/j.jvolgeores.2008.11.025.

Cantrell L, Young M. 2009. Fatal fall into a volcanic fumarole. Wilderness & Environmental Medicine. 20(1):77–79. doi:10.1580/08-WEME-CR-199.1.

Carboni E, Mather TA, Schmidt A, Grainger RG, Pfeffer MA, Ialongo I, Theys N. 2019. Satellite–derived sulfur dioxide (SO2) emissions from the 2014–2015 Holuhraun eruption (Iceland). Atmospheric Chemistry and Physics. 19(7):4851–4862. doi:10.5194/acp-19-4851-2019.

Calvari S, Lodato L, Steffke A, Cristaldi A, Harris A, Spampinato L, Boschi E. 2010. The 2007 Stromboli eruption: event chronology and effusion rates using thermal infrared data. Journal of Geophysical Research B: Solid Earth. 115:B04201. doi:10.1029/2009JB006478.

Carn S. 2002. Eruptive and passive degassing of sulfur dioxide at Nyiragongo volcano (D.R. Congo): the 17 January 2002 eruption and its aftermath. Acta Vulcanologica. 14–15:75–86.

Carn S, Clarisse L, Prata F. 2016. Multi–decadal satellite measurements of global volcanic degassing. Journal of Volcanology and Geothermal Research. 311. doi:10.1016/j.jvolgeores.2016.01.002.

Crimmins P. 2018. Auckland air emissions inventory 2016. Industry. [Auckland (NZ)]: Auckland Council. 39 p. Technical Report 2018/019.

Davy PK, Ancelet T, Trompetter WJ, Markwitz A. 2017. Source apportionment and trend analysis of air particulate matter in the Auckland region. 80 p. 2014/194.

D’Aleo R, Bitetto M, Donne DD, Coltelli M, Coppola D, Kilbride BM, Pecora E, Ripepe M, Salem LC, Tamburello G. Aiuppa A. 2019. Understanding the SO2 degassing budget of Mt Etna’s paroxysms: first clues from the December 2015 sequence. Frontiers in Earth Science. 6:article 239. doi:10.3389/feart.2018.00239.

Deligne NI, Blake DM, Davies AJ, Grace ES, Hayes J, Potter SH, Stewart C, Wilson G, Wilson TM. 2015. Economics of Resilient Infrastructure Auckland Volcanic Field Scenario. Lower Hutt (NZ): GNS Science. 151 p. (ERI research report; 2015/03).

https://en.vedur.is/pollution-and-radiation/volcanic-gas/hazard-zoning/


Deligne NI, Fitzgerald RH, Blake DM, Davies AJ, Hayes JL, Stewart C, Wilson G, Wilson TM, Castelino R, Kennedy BM, et al. 2017. Investigating the consequences of urban volcanism using a scenario approach I: Development and application of a hypothetical eruption in the Auckland Volcanic Field, New Zealand. Journal of Volcanology and Geothermal Research. 336:192–208. doi:10.1016/j.jvolgeores.2017.02.023.

Delmelle P, Stix J, Bourque CPA, Baxter PJ, Garcia–Alvarez J, Barquero J. 2001. Dry deposition and heavy acid loading in the vicinity of Masaya Volcano, a major sulfur and chlorine source in Nicaragua. Environmental Science & Technology. 35(7):1289–1293. doi:10.1021/es000153m.

Delmelle P, Stix J, Baxter PJ, Garcia-Alvarez J, Barquero J. 2002. Atmospheric dispersion, environmental effects and potential health hazard associated with the low-altitude gas plume of Masaya Volcano, Nicaragua. Bulletin of Volcanology. 64:423–434. doi:10.1007/s00445-002-0221-6.

Durand M, Scott BJ. 2005. Geothermal ground gas emissions and indoor air pollution in Rotorua, New Zealand. Science of The Total Environment. 345(1):69–80. doi:10.1016/j.scitotenv.2004.10.023.

Edmonds M, Oppenheimer C, Pyle D, Herd R. 2003. Rainwater and ash leachate analysis as proxies for plume chemistry at Soufriere Hills Volcano, Montserrat. In: Oppenheimer C, Pyle DM, Barclay J, editors. Volcanic degassing. London (GB): Geological Society. p. 203–218. (Special publication; 213).

Edmonds M, Grattan J, Michnowicz S. 2018. Volcanic gases: silent killers. In: Fearnley CJ, Bird DK, Haynes K, McGuire WJ, Jolly G, editors. Observing the volcano world: volcano crisis communication. Cham (CZ): Springer. p. 65–83. (Advances in volcanology).

Elias T, Sutton AJ, Stokes JB, Casadevall TJ. 1998. Sulfur dioxide emission rates of Kilauea volcano, Hawaii, 1979–1997. Menlo Park (CA): U.S. Geological Survey. 41 p. Open–File Report 98–462.

Farrar CD, Sorey ML, Evans WC, Howle JF, Kerr BD, Kennedy BM, King CY, Southon JR. 1995. Forest–killing diffuse CO2 emission at Mammoth Mountain as a sign of magmatic unrest. Nature. 376(6542):675–678. doi:10.1038/376675a0.

Fitzgerald RH, Dohaney J, Hill D, Wilson TM, Kennedy B, Lindsay J. 2016. Teaching volcanic hazard management and emergency management concepts through role-play: the science behind the Auckland Volcanic Field Simulation. Lower Hutt (NZ): GNS Science. 61 p. (GNS Science report; 2014/70).

Food Security and Agriculture Cluster. 2017. Ambae volcanic eruption: Food Security and Agricultural Livelihoods (FSAL) Needs assessment, December 2017. [Port Vila (VU)]: [Government of Vanuatu]. 51 p.

Galeczka I, Eiriksdottir ES, Pálsson F, Oelkers E, Lutz S, Benning LG, Stefánsson A, Kjartansdóttir R, Gunnarsson-Robin J, Ono S, et al. 2017. Pollution from the 2014–15 Bárðarbunga eruption monitored by snow cores from the Vatnajökull glacier, Iceland. Journal of Volcanology and Geothermal Research. 347:371–396. doi:10.1016/j.jvolgeores.2017.10.006.

Geshi N, Shimano T, Chiba T, Nakada S. 2002. Caldera collapse during the 2000 eruption of Miyakejima Volcano, Japan. Bulletin of Volcanology. 64:55–68. doi:10.1007/s00445-001-0184-z.

Gislason S, Stefansdottir G, Pfeffer MA, Barsotti S, Jóhannsson T, Galeczka I, Bali E, Sigmarsson O, Stefansson A, Keller N, et al. 2015. Environmental pressure from the 2014–15 eruption of Bárðarbunga volcano, Iceland. Geochemical Perspectives Letters. 1(1):84–93. doi:10.7185/geochemlet.1509.


Global Volcanism Program. 1973a. Vigorous late–January fissure eruption forces evacuations. [Washington (DC)]: Smithsonian Institution. 01/1973 (CSLP 11-73).

Global Volcanism Program. 1973b. Abundant lava flows during mid-February almost close the harbour. [Washington (DC)]: Smithsonian Institution. 02/1973 (CSLP 11-73).

Global Volcanism Program. 2002. Report on Miyakejima (Japan). In: Wunderman R, editor. Bulletin of the Global Volcanism Network, 27:3. [Washington (DC)]: Smithsonian Institution; [accessed 2019 Dec]. https://doi.org/10.5479/si.GVP.BGVN200203-284040.

Global Volcanism Program. 2014. Report on Bardarbunga (Iceland). In: Wunderman R, editor. Bulletin of the Global Volcanism Network, 39(10). [Washington (DC]: Smithsonian Institution; [accessed 2019 Dec]. https://doi.org/10.5479/si.GVP.BGVN201410-373030.

Global Volcanism Program. 2015. Report on Bardarbunga (Iceland). In: Wunderman R, editor. Bulletin of the Global Volcanism Network, 40:1. [Washington (DC]: Smithsonian Institution; [accessed 2019 Dec]. https://doi.org/10.5479/si.GVP.BGVN201501-373030.

Global Volcanism Program. 2018a. Report on Ambae (Vanuatu). In: Crafford AE, Venzke E, editors. Bulletin of the Global Volcanism Network, 43:2. [Washington (DC)]: Smithsonian Institution; [accessed 2019 Dec]. https://doi.org/10.5479/si.GVP.BGVN201802-257030.

Global Volcanism Program. 2018b. Report on Ambae (Vanuatu). In: Crafford ED, Venzke E, editors. Bulletin of the Global Volcanism Network, 43:7. [Washington (DC)]: Smithsonian Institution; [accessed 2019 Dec]. https://doi.org/10.5479/si.GVP.BGVN201807-257030.

Global Volcanism Program. 2019. Report on Ambae (Vanuatu). In: Krippner JB, Venzke E, editors. Bulletin of the Global Volcanism Network, 44:2. [Washington (DC)]: Smithsonian Institution.

Hansell A, Oppenheimer C. 2004. Health hazards from volcanic gases: a systematic literature review. Archives of Environmental Health. 59(12):628–639. doi:10.1080/00039890409602947.

Harada I, Yoshii Y, Kaba Y, Saito H, Goto Y, Alimuddin I, Kuriyama K, Machida I, Kuze H. 2013. Measurement of volcanic SO2 concentration in Miyakejima using differential optical absorption spectroscopy (DOAS). Open Journal of Air Pollution. 2:36–46. doi:10.4236/ojap.2013.22006.

Harding D, Miller JM. 1982. The influence on rain chemistry of the Hawaiian Volcano Kilauea. Journal of Geophysical Research: Oceans. 87(C2):1225–1230. doi:10.1029/JC087iC02p01225.

Hayes JL, Tsang SW, Fitzgerald RH, Blake DM, Deligne NI, Doherty A, Hopkins JL, Hurst AW, N. LC, Leonard GS, et al. 2018. The DEVORA scenarios: multi–hazard eruption scenarios for the Auckland Volcanic Field. Lower Hutt (NZ): GNS Science. 138 p. (GNS Science report; 2018/29).

Heggie TW, Heggie TM, Heggie TJ. 2009. Death by volcanic laze. Wilderness & Environmental Medicine. 20(1):101–103. doi:10.1580/08-WEME-LE-236.1.

Hill PM. 2000. Possible asphyxiation from carbon dioxide of a cross-country skier in eastern California: a deadly volcanic hazard. Wilderness & Environmental Medicine. 11(3):192–195. doi:10.1580/1080-6032(2000)011[0192:PAFCDO]2.3.CO;2.

Ilyinskaya E, Schmidt A, Mather T, Pope F, Witham C, Baxter P, Jóhannsson T, Pfeffer M, Barsotti S, Singh A, et al. 2017. Understanding the environmental impacts of large fissure eruptions: aerosol and gas emissions from the 2014–2015 Holuhraun eruption (Iceland). Earth and Planetary Science Letters. 472:309–322. doi:10.1016/j.epsl.2017.05.025.

[IVHHN] International Volcanic Health Hazard Network. 2019. Standardised epidemiological study protocol to assess short-term respiratory and other health impacts in volcanic eruptions. [place unknown]: International Volcanic Health Hazard Network; [accessed 2019 Dec]. http://www.ivhhn.org/uploads/IVHHN%20Basic%20Protocol.pdf

https://doi.org/10.5479/si.GVP.BGVN200203-284040





http://www.ivhhn.org/uploads/IVHHN%20Basic%20Protocol.pdf


Jenkins S, Barsotti S, Hincks T, Neri A, Phillips J, Sparks R, Sheldrake T, Vougioukalakis G. 2015. Rapid emergency assessment of ash and gas hazard for future eruptions at Santorini Volcano, Greece. Journal of Applied Volcanology. 4:article 16. doi:10.1186/s13617-015-0033-y.

Johnston DM, Nairn I, Thordarson T, Daly M. 1997. Volcanic impact assessment for the Auckland volcanic field. Auckland (NZ): Auckland Regional Council; [accessed 2019 Dec]. http://knowledgeauckland.org.nz/assets/publications/Volcanic_impact_assessment_for_the_Auckland_volcanic_field.pdf.

Johnston DM, Houghton BF, Neall VE, Ronan KR, Paton D. 2000. Impacts of the 1945 and 1995–1996 Ruapehu eruptions, New Zealand: an example of increasing societal vulnerability. GSA Bulletin. 112(5):720–726. doi:10.1130/0016-7606(2000)112<720:Iotare>2.0.Co;2.

Joseph EP, Jackson VB, Beckles DM, Cox L, Edwards S. 2019. A citizen science approach for monitoring volcanic emissions and promoting volcanic hazard awareness at Sulphur Springs, Saint Lucia in the Lesser Antilles arc. Journal of Volcanology and Geothermal Research. 369:50–63. doi:10.1016/j.jvolgeores.2018.11.005.

Joseph EP. 2004. Geochemical monitoring of Sulphur Springs, St. Lucia: implication for health and safety. St. Augustine (TT): University of West Indies, Seismic Research Centre. 12 p.

Kazahaya K, Shinohara H, Uto K, Odai M, Nakahori Y, Mori H, Iino H, Miyashita M, Hirabayashi J. 2004. Gigantic SO2 emission from Miyakejima volcano, Japan, caused by caldera collapse. Geology. 32(5):425–428. doi:10.1130/g20399.1.

Kereszturi G, Németh K, Cronin SJ, Agustín-Flores J, Smith IEM, Lindsay J. 2013. A model for calculating eruptive volumes for monogenetic volcanoes — implication for the Quaternary Auckland Volcanic Field, New Zealand. Journal of Volcanology and Geothermal Research. 266:16–33. doi:10.1016/j.jvolgeores.2013.09.003.

Kullman GJ, Jones WG, Cornwell RJ, Parker JE. 1994. Characterization of air contaminants formed by the interaction of lava and sea water. Environmental Health Perspectives. 102(5):478–482. doi:10.1289/ehp.94102478.

Li Z, Marston NJ, Stokes K, Branz. 2018. Materials within geothermal environments. Porirua (NZ): BRANZ. Study Report SR393.

Leonard GS, Calvert AT, Hopkins JL, Wilson CJN, Smid ER, Lindsay JM, Champion DE. 2017. High-precision 40Ar/39Ar dating of Quaternary basalts from Auckland Volcanic Field, New Zealand, with implications for eruption rates and paleomagnetic correlations. Journal of Volcanology and Geothermal Research. 343:60–74. doi:10.1016/j.jvolgeores.2017.05.033.

Lindsay J, Marzocchi W, Jolly G, Constantinescu R, Selva J, Sandri L. 2010. Towards real-time eruption forecasting in the Auckland Volcanic Field: application of BET_EF during the New Zealand National Disaster Exercise ‘Ruaumoko’. Bulletin of Volcanology. 72(2):185–204. doi:10.1007/s00445-009-0311-9.

Longo BM, Rossignol A, Green JB. 2008. Cardiorespiratory health effects associated with sulphurous volcanic air pollution. Public Health. 122(8):809–820. doi:10.1016/j.puhe.2007.09.017.

Longo BM, Yang W, Green JB, Crosby FL, Crosby VL. 2010. Acute health effects associated with exposure to volcanic air pollution (vog) from increased activity at Kilauea volcano in 2008. Journal of Toxicology and Environmental Health, Part A. 73(20):1370–1381. doi:10.1080/15287394.2010.497440.

Mather T. 2015. Volcanoes and the environment: Lessons for understanding Earth's past and future from studies of present-day volcanic emissions. Journal of Volcanology and Geothermal Research. 304:160–179. doi:10.1016/j.jvolgeores.2015.08.016.

http://knowledgeauckland.org.nz/assets/publications/Volcanic_impact_assessment_for_the_Auckland_volcanic_field.pdf

http://knowledgeauckland.org.nz/assets/publications/Volcanic_impact_assessment_for_the_Auckland_volcanic_field.pdf


Mazot A, Smid E, Schwendenmann L, Delgado Granados H, Lindsay J. 2013. Soil CO2 flux baseline in an urban monogenetic volcanic field: the Auckland Volcanic Field, New Zealand. Bulletin of Volcanology. 75:1–9. doi:10.1007/s00445-013-0757-7.

Metcalfe J, Sridhar S. 2018. National Air Emissions Inventory 2015. Auckland (NZ): Emission Impossible. 66 p. Prepared for Ministry for the Environment.

Metcalfe J, Wickham L, Sridhar S. 2018. Auckland air emissions inventory 2016 – home heating. Auckland (NZ): Auckland Council. 48 p. Technical Report TR2018/018.

[MfE] Ministry for the Environment. 2004. Good practice guide for atmospheric dispersion modelling. Wellington (NZ): Ministry for the Environment. 142 p. Report ME 522.

Oppenheimer C, Scaillet B, Martin RS. 2011. Sulfur degassing from volcanoes: source conditions, surveillance, plume chemistry and earth system impacts. Reviews in Mineralogy and Geochemistry. 73(1):363–421. doi:10.2138/rmg.2011.73.13.

Oze C, Cole J, Scott A, Wilson T, Wilson G, Gaw S, Hampton S, Doyle C, Li Z. 2014. Corrosion of metal roof materials related to volcanic ash interactions. Natural Hazards. 71(1):785–802. doi:10.1007/s11069-013-0943-0.

Peeters S. 2018. Auckland Air Emissions Inventory 2016. Sea transport. Auckland (NZ): Auckland Council. Technical Report 2018/017.

Reikard G. 2012. Forecasting volcanic air pollution in Hawaii: Tests of time series models. Atmospheric Environment. 60:593–600. doi:10.1016/j.atmosenv.2012.06.040.

Schmidt A, Witham CS, Theys N, Richards NAD, Thordarson T, Szpek K, Feng W, Hort MC, Woolley AM, Jones AR, et al. 2014. Assessing hazards to aviation from sulfur dioxide emitted by explosive Icelandic eruptions. Journal of Geophysical Research: Atmospheres. 119(24):14,180–114,196. doi:10.1002/2014jd022070.

Schmidt A, Leadbetter S, Theys N, Carboni E, Witham CS, Stevenson JA, Birch CE, Thordarson T, Turnock S, Barsotti S, et al. 2015. Satellite detection, long-range transport, and air quality impacts of volcanic sulfur dioxide from the 2014–2015 flood lava eruption at Bárðarbunga (Iceland). Journal of Geophysical Research: Atmospheres. 120(18):9739–9757. doi:10.1002/2015jd023638.

Sherburn S, Scott BJ, Olsen J, Miller C. 2007. Monitoring seismic precursors to an eruption from the Auckland Volcanic Field, New Zealand. New Zealand Journal of Geology and Geophysics. 50(1):1–11. doi:10.1080/00288300709509814.

Shinohara H. 2008. Excess degassing from volcanoes and its role on eruptive and intrusive activity. Reviews of Geophysics. 46(4):RG4005. doi:10.1029/2007rg000244.

Simmons I, Pfeffer M, Calder E, Galle B, Arellano S, Coppola D, Barsotti S. 2017. Extended SO2 outgassing from the 2014–2015 Holuhraun lava flow field, Iceland. Bulletin of Volcanology. 79:79. doi:10.1007/s00445-017-1160-6.

Smid ER, Lindsay J, Rowe M, Hayes J, Stewart C. 2018. Past gas: using melt inclusion data to estimate volatile emissions for Auckland Volcanic Field Eruption scenarios [abstract]. In: Cities on Volcanoes 10; 2018 Sep 2–7; Naples, Italy. Rome (IT): INGV.

Smid E, Mazot As. 2012. Soil gas CO2 concentrations & CO2 fluxes in the Auckland volcanic field. Auckland (NZ): Institute of Earth Science and Engineering Aotearoa. 26 p. IESE Report 1–2011.02.

Sorey ML, Evans WC, Kennedy BM, Farrar CD, Hainsworth LJ. and Hausback B, 1998. Carbon dioxide and helium emissions from a reservoir of magmatic gas beneath Mammoth Mountain, California. Journal of Geophysical Research. 103(B7):15303–15323. doi:10.1029/98jb01389.


Sridhar S, Metcalfe J. 2019. Auckland air emissions inventory 2016. Transport (revised). Auckland (NZ): Auckland Council. 80 p. Technical Report 2018/016-2.

Sutton AJ, Elias T, Hendley JW and Stauffer PH. 2000. Volcanic air pollution: a hazard in Hawaii. [Reston (VA)]: U.S. Geological Survey. Fact Sheet 169-97. Online Version 1.1, revised June 2000. [accessed 2019 Dec]. https://pubs.usgs.gov/fs/fs169-97/

Sutton AJ, Elias T. 2014. One hundred volatile years of volcanic gas studies at the Hawaiian Volcano Observatory. In: Poland MP, Takahashi TJ, Landowski CM, editors. Characteristics of Hawaiian volcanoes. Reston (VA): U.S. Geological Survey. p. 295–320. (Professional Paper; 1801).

Talbot N, Lehn R. 2018. The impacts of transport emissions on air quality in Auckland’s city centre. Auckland (NZ): Auckland Council. 49 p. Technical Report 2018/028.

Talbot N, Reid N. 2017. A review of research into the effects of shipping on air quality in Auckland: 2006-2016. Auckland (NZ): Auckland Council. 31 p. Technical Report 2017/005.

Talbot N, Reid N, Cummins P. 2017. Auckland ambient air quality trends for PM2.5 and PM10 – 2006–2015. Auckland (NZ): Auckland Council. 36 p. Technical Report 2017/029.

Theys N, Hedelt P, De Smedt I, Lerot C, Yu H, Vlietinck J, Pedergnana M, Arellano S, Galle B, Fernandez D, et al. 2019. Global monitoring of volcanic SO2 degassing with unprecedented resolution from TROPOMI onboard Sentinel-5 Precursor. Scientific Reports. 9:article 2643. doi:10.1038/s41598-019-39279-y.

Thordarson T, Hayward C, Hartley M, Sigmarsson C, Höskuldsson Á, Larsen G. 2011. The 14 April–22 May 2010 summit eruption at Eyjafjallajökull volcano, Iceland: volatile contents and magma degassing [abstract]. In. 8th EGU General Assembly; 2011 Apr 3–8; Vienna (AT). Goettingen (DE): Copernicus Gesellschaft. EGU2011-12046. (Geophysical research abstracts; 13).

Tomašek I. 2018. Environmental and anthropogenic factors affecting the respiratory toxicity of volcanic ash [PhD thesis]. Durham (GB): Durham University. 210 p. + appendices. [accessed 2019 Dec]. http://etheses.dur.ac.uk/12671/

Tomašek I, Horwell CJ, Damby DE, Barošová H, Geers C, Petri-Fink A, Rothen-Rutishauser B, Clift MJD. 2016. Combined exposure of diesel exhaust particles and respirable Soufrière Hills volcanic ash causes a (pro-)inflammatory response in an in vitro multicellular epithelial tissue barrier model. Particle and Fibre Toxicology. 13(1):article 67. doi:10.1186/s12989-016-0178-9.

Tomašek I, Horwell CJ, Bisig C, Damby DE, Comte P, Czerwinski J, Petri-Fink A, Clift MJD, Drasler B, Rothen-Rutishauser B. 2018. Respiratory hazard assessment of combined exposure to complete gasoline exhaust and respirable volcanic ash in a multicellular human lung model at the air-liquid interface. Environmental Pollution. 238:977–987. doi:10.1016/j.envpol.2018.01.115.

[VNSO] Vanuatu National Statistics Office. 2016. Vanuatu Post-TC Pam Mini-Census Report. Vol. 1: Basic tables. Port Vila (VU): Government of Vanuatu, Vanuatu National Statistics Office; [accessed 2019 Dec]. https://vnso.gov.vu/index.php/2-uncategorised/153-mini-census#census-materials

Wallace PJ. 2001. Volcanic SO2 emissions and the abundance and distribution of exsolved gas in magma bodies. Journal of Volcanology and Geothermal Research. 108(1):85–106. doi:10.1016/S0377-0273(00)00279-1.

Werner C, Hurst T, Scott B, Sherburn S, Christenson BW, Britten K, Cole-Baker J, Mullan B. 2008. Variability of passive gas emissions, seismicity, and deformation during crater lake growth at White Island Volcano, New Zealand, 2002–2006. Journal of Geophysical Research: Solid Earth. 113:B01204. doi:10.1029/2007jb005094.

https://pubs.usgs.gov/fs/fs169-97/

http://etheses.dur.ac.uk/12671/

https://vnso.gov.vu/index.php/2-uncategorised/153-mini-census#census-materials

https://vnso.gov.vu/index.php/2-uncategorised/153-mini-census#census-materials


Williams RS, Moore JG. 1983. Man against volcano: the eruption on Heimaey, Vestmannaeyjar, Iceland. [Washington (DC)]: U.S. Geological Survey; [accessed 2019 Dec]. https://pubs.usgs.gov/gip/heimaey/heimaey.pdf.

World Health Organization Regional Office for Europe. 2013. Review of evidence on health aspects of air pollution – REVIHAAP project: technical report. Copenhagen (DK): WHO Regional Office for Europe; [accessed 2019 Dec]. http://www.euro.who.int/__data/assets/pdf_file/0004/193108/REVIHAAP-Final-technical-report.pdf.

https://pubs.usgs.gov/gip/heimaey/heimaey.pdf

http://www.euro.who.int/__data/assets/pdf_file/0004/193108/REVIHAAP-Final-technical-report.pdf

http://www.euro.who.int/__data/assets/pdf_file/0004/193108/REVIHAAP-Final-technical-report.pdf


APPENDICES


APPENDIX 1 HAWAI’I STATE DEPARTMENT OF HEALTH SO2 ADVISORY LEVELS

1 Fairway Drive

Avalon

PO Box 30368

Lower Hutt

New Zealand

T +64-4-570 1444

F +64-4-570 4600

Dunedin Research Centre

764 Cumberland Street

Private Bag 1930

Dunedin

New Zealand

T +64-3-477 4050

F +64-3-477 5232

Wairakei Research Centre

114 Karetoto Road

Wairakei

Private Bag 2000, Taupo

New Zealand

T +64-7-374 8211

F +64-7-374 8199

National Isotope Centre

30 Gracefield Road

PO Box 31312

Lower Hutt

New Zealand

T +64-4-570 1444

F +64-4-570 4657

Principal Location

www.gns.cri.nz

Other Locations

preparing for a sustained volcanic degassing episode in

Documents