Quantifying gas emissions associated with the 2018 rift eruption of Kīlauea Volcano using ground-based DOAS measurements

Abstract

Starting on 3 May 2018, a series of eruptive fissures opened in Kīlauea Volcano’s lower East Rift Zone (LERZ). Over the course of the next 3 months, intense degassing accompanied lava effusion from these fissures. Here, we report on ground-based observations of the gas emissions associated with Kīlauea’s 2018 eruption. Visual observations combined with radiative transfer modeling show that ultraviolet light could not efficiently penetrate the gas and aerosol plume in the LERZ, complicating SO₂ measurements by differential optical absorption spectroscopy (DOAS). By applying a statistical method that integrates a radiative transfer model with the DOAS retrievals, we were able to calculate sulfur dioxide (SO₂) emission rates along with estimates of their uncertainty. We find that sustained SO₂ emissions were highest in June and early July, when approximately 200 kt SO₂ were emitted daily. At the 68% confidence interval, we estimate that 7.1–13.6 Mt SO₂ were released from the LERZ during the entire May to September eruptive episode. Scaling our results with in situ measurements of plume composition, we calculate that 11–21 Mt H₂O and 1.5–2.8 Mt CO₂ were also emitted. The gas and aerosol emissions caused hazardous conditions in areas proximal to the active vents, but plume dispersion modeling shows that the eruption also significantly impacted air quality hundreds of kilometers downwind. Combined with petrologic studies of the erupted lavas, our measurements indicate that 1.1–2.3 km³ dense-rock equivalent of lava were erupted from the LERZ, which is approximately twice the concomitant collapse volume of the volcano’s summit.

Electronic supplemental material

All spectral data analyzed and presented in this article are available for download from the USGS ScienceBase Catalog: https://doi.org/10.5066/P9LXBJF3

Additional electronic supplemental material is published alongside this article:

The file ESM1.zip contains the lookup table used for the MCMC processing. The zip repository contains 1,188 text files, each of which adhere to the naming convention ‘H_SSA_AOD.txt’, where H is the plume height above the instrument in meters, SSA is the plume aerosol single scatter albedo, and AOD is the plume aerosol optical depth used for the simulation. Additional information on the assumed plume geometry is given in the text. Each file contains two numerical values separated by a tab. The first number gives the air mass factor (AMF) of the simulated upward-looking measurement at 367 nm. Consistent with remote sensing literature, the AMF is defined as SCD/VCD, so values larger than 1 indicate extension of the effective light path in the plume. At this wavelength, the AMF can be considered independent of the SO₂ VCD because SO₂ absorption is weak. The second number gives the simulated darkening ratio I_p/I₀, also simulated at 367 nm. These tables can be used to interpret upward-looking DOAS data of SO₂ plumes in the 345-390 nm wavelength window but cannot be used in other wavelength regions, particularly those where SO₂ absorption is strong and influences the light path distribution.

The tables provided in ESM2.csv, ESM3.csv, ESM4.csv and ESM5.csv contain the SO₂ emission rates and their uncertainties retrieved for all successful mobile DOAS traverses of gas plumes emitted from Kīlauea between April and August 2018. ESM2.csv gives the individual SO₂ emission rates from the active fissures in the LERZ, while ESM3.csv gives the sustained emission rates, which are averaged over three successive measurement days. ESM4.csv gives emissions from the volcano’s summit, while ESM5.csv provides emission rates from the middle East Rift Zone (MERZ). Following our statistical analyses of the LERZ emissions, the most likely emission rate is reported along with both the 68% and the 95% confidence intervals of that value for the LERZ data. As standard DOAS methods were used to retrieve the emission rates from the other source locations, only the standard symmetrical errors are given for these. The wind speeds and wind directions used for derivation of the emission rates are also listed, along with their assumed uncertainties. In cases where it could be determined by dual-beam mobile DOAS measurements, the plume height above sea level is also given. The last column gives the DOAS spectral fit range and analysis method. The identifier ‘MCMC’ indicates that the emission rate was retrieved using the Markov Chain Monte Carlo (MCMC) method.

The file ESM6.pdf provides details of the analysis methods used to retrieve BrO column densities and BrO/SO₂ ratios from the Mobile DOAS data. Examples of the spectral fit and BrO/SO₂ ratio data are also given.

ESM 1

(ZIP 169 kb)

ESM 2

(CSV 12 kb)

ESM 3

(CSV 2 kb)

ESM 4

(CSV 3 kb)

ESM 5

(CSV 5 kb)

ESM 6

(PDF 385 kb)