Change search
ReferencesLink to record
Permanent link

Direct link
Dispersion modelling of volcanic emissions
Uppsala University, Disciplinary Domain of Science and Technology, Earth Sciences, Department of Earth Sciences, LUVAL. Centre for Natural Disaster Science.ORCID iD: 0000-0001-7909-0640
2016 (English)Doctoral thesis, comprehensive summary (Other academic)Alternative title
Spridningsmodellering av utsläpp från vulkaner (Swedish)
Abstract [en]

Gases and particles released by volcanoes pose a serious hazard to humans and society. Emissions can be transported over long distances before being reduced to harmless concentrations. Knowing which areas are, or will be, exposed to volcanic emissions is an important part inreducing the impact on human health and society. In this thesis, the dispersion of volcanic emissions is studied using a set of atmospheric models.

The work includes contribution to the development of the Lagrangian Particle Dispersion Model FLEXPART-WRF. Three case studies have been performed, one studying potential ash emissions from potential future eruptions on Iceland, a second covering SO2 emissions from Mt. Nyiragongo in D.R. Congo, and a third studying the SO2 emission rate of the Holuhraun eruption (Iceland) in 2014–2015.

The first study covers volcanic ash hazard for air traffic over Europe. Three years of meteorological data are used to repeatedly simulate dispersion from different eruption scenarios. The simulations are used to study the probability of hazardous concentrations in ash in European airspace. The ash hazard shows a seasonal variation with a higher probability of efficient eastward transport in winter, while summer eruptions pose a more persistent hazard.

In the second study, regional gas exposure around Mt. Nyiragongo is modelled using flux measurements to improve the description of the emission source. Gases are generally transported to the north-west in June–August and to the south-west in December–January. A diurnal variation due to land breeze around lake Kivu contributes to high concentrations of SO2 along the northern shore during the night. Potentially hazardous concentrations are occasionally reached in populated areas in the region, but mainly during the nights.

The third study uses inverse dispersion modelling to determine the height and emission rates based on traverse measurements of the plume at 80–240 km from the source. The calculated source term yields better agreement with satellite observations compared to commonly used column sources.

The work in this thesis presents improvements in dispersion modelling of volcanic emissions through improved models, more accurate representation of the source terms, and through incorporating new types of measurements into the modelling systems.

Abstract [sv]

Gas- och partikelutsläpp från vulkaner utgör en fara för människor och för vårt samhälle. Utsläppen kan transporteras över långa avstånd innan de reduceras till oskadliga halter. Att känna till vilka områden som utsätts för, eller kommer utsättas för, utsläppen är ett viktigt verktyg föratt minska påverkan på folkhälsa och samhälle. I avhandlingen studeras spridningen av utsläpp från vulkanutbrott med hjälp av en uppsättning numeriska atmosfärsmodeller.

Den Lagrangiska Partikelspridningsmodellen FLEXPART-WRF har förbättrats och applicerats för spridningsmodellering av vulkanutbrott. Tre studier har utförts, en fokuserar på vulkanaska från potentiella framtida utbrott på Island, den andra studerar SO2-ustläpp från vulkanen Nyiragongo i Demokratiska Republiken Kongo, och den tredje studerar SO2-ustläpp från utbrottet i Holuhraun (Island) 2014–2015.

Den första studien uppskattar sannolikheten för att vulkanaska från framtida vulkanutbrott på Island ska överskrida de gränsvärden som tillämpas för flygtrafik. Tre år av meteorologisk data används för att simulera spridningen från olika utbrottsscenarier. Sannolikheten för skadliga halter aska varierar med årstid, med en högre sannolikhet för effektiv transport österut under vintermånaderna, sommarutbrott är istället mer benägna att orsaka långvariga problem överspecifika områden.

In den andra studien undersöks spridningen av SO2 från Nyiragongo över en ettårsperiod. Flödesmätningar av plymen används för att förbättra källtermen i modellen. Gaserna transporteras i regel mot nordväst i juni–augusti och mot sydväst i december–februari En dygnsvariation, kopplad till mesoskaliga processer runt Kivusjön, bidrar till förhöjda halter av SO2 nattetid längs Kivusjöns norra kust. Potentiellt skadliga halter av SO2 uppnås av och till i befolkade områden men huvudsakligen nattetid.

Den tredje studien utnyttjar inversmodellering för att avgöra plymhöjd och gasutsläpp baserat på traversmätningar av plymen runt 80–240 km från utsläppskällan. Den beräknade källtermen resulterar i bättre överensstämmelse mellan modell- och satellitdata jämfört med enklare källtermer.

Arbetet i den här avhandlingen presenterar flertalet förbättringar för spridningsmodellering av vulkanutbrott genom bättre modeller, nogrannare beskrivning av källtermer, och genom nya metoder för tillämpning av olika typer av mätdata.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 2016. , 53 p.
Series
Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1651-6214 ; 1433
Keyword [en]
dispersion modelling, atmospheric, volcano, gas emissions, volcanic ash, FLEXPART, FLEXPART-WRF
Keyword [sv]
Spridningsmodellering, atmosfär, vulkan, gasutsläpp, vulkanaska, FLEXPART, FLEXPART-WRF
National Category
Meteorology and Atmospheric Sciences
Research subject
Meteorology
Identifiers
URN: urn:nbn:se:uu:diva-303959ISBN: 978-91-554-9704-0OAI: oai:DiVA.org:uu-303959DiVA: diva2:974854
Public defence
2016-11-17, Axel Hambergssalen, Villavägen 16, Uppsala, 10:00 (English)
Opponent
Supervisors
Available from: 2016-10-27 Created: 2016-09-27 Last updated: 2016-11-02
List of papers
1. The Lagrangian particle dispersion model FLEXPART-WRF version 3.1
Open this publication in new window or tab >>The Lagrangian particle dispersion model FLEXPART-WRF version 3.1
Show others...
2013 (English)In: Geoscientific Model Development, ISSN 1991-959X, E-ISSN 1991-9603, Vol. 6, no 6, 1889-1904 p.Article in journal (Refereed) Published
Abstract [en]

The Lagrangian particle dispersion model FLEXPART was originally designed for calculating long-range and mesoscale dispersion of air pollutants from point sources, such that occurring after an accident in a nuclear power plant. In the meantime, FLEXPART has evolved into a comprehensive tool for atmospheric transport modeling and analysis at different scales. A need for further multiscale modeling and analysis has encouraged new developments in FLEXPART. In this paper, we present a FLEXPART version that works with the Weather Research and Forecasting (WRF) mesoscale meteorological model. We explain how to run this new model and present special options and features that differ from those of the preceding versions. For instance, a novel turbulence scheme for the convective boundary layer has been included that considers both the skewness of turbulence in the vertical velocity as well as the vertical gradient in the air density. To our knowledge, FLEXPART is the first model for which such a scheme has been developed. On a more technical level, FLEXPART-WRF now offers effective parallelization, and details on computational performance are presented here. FLEXPART-WRF output can either be in binary or Network Common Data Form (NetCDF) format, both of which have efficient data compression. In addition, test case data and the source code are provided to the reader as a Supplement. This material and future developments will be accessible at http://www.flexpart.eu.

National Category
Meteorology and Atmospheric Sciences
Research subject
Meteorology
Identifiers
urn:nbn:se:uu:diva-233381 (URN)10.5194/gmd-6-1889-2013 (DOI)000329050500003 ()
Available from: 2014-10-02 Created: 2014-10-02 Last updated: 2016-09-27Bibliographically approved
2. Estimating volcanic ash hazard in European airspace
Open this publication in new window or tab >>Estimating volcanic ash hazard in European airspace
2014 (English)In: Journal of Volcanology and Geothermal Research, ISSN 0377-0273, Vol. 286, 55-66 p.Article in journal (Refereed) Published
Abstract [en]

The widespread disruption of European air traffic in late April 2010, during the eruption of Eyjafjallajökull,showed the importance of early assessment of volcanic hazard from explosive eruptions. In this study, wefocus on the short-term hazard of airborne ash from a climatological perspective, focusing on eruptions onIceland. By studying eruptions of different intensity and frequency, we estimate the overall probability that ashconcentration levels considered hazardous to aviation are exceeded over different parts of Europe.

The method involves setting up a range of eruption scenarios based on the eruptive history of Icelandic volcanoes,and repeated simulation of these scenarios for 2 years' worth of meteorological data. Simulations are conducted using meteorological data from the ERA-Interim reanalysis set, which is downscaled using the Weather Researchand Forecasting (WRF) model. The weather data are then used to drive the Lagrangian particle dispersion model FLEXPART-WRF for each of the eruption scenarios. A set of threshold values, commonly used in Volcanic Ash Advisories, are used to analyze concentration data from the dispersion model.

We see that the dispersion of ash is highly dominated by the mid-latitude westerlies and mainly affect northern UK and the Scandinavian peninsula. The occurrence of high ash levels from Icelandic volcanoes is lower over con-tinental Europe but should not be neglected for eruptions when the release rate of fine ash (<16 μm) is in theorder of 107 kg s−1 or higher.

There is a clear seasonal variation in the ash hazard. During the summer months, the dominating dispersiondirection is less distinct with some plumes extending to the northwest and Greenland. In contrast, during thewinter months, the strong westerly winds tend to transport most of the emissions eastwards. The affected area of a winter-time eruption is likely to be larger as high concentrations can be found at a further distance downwind from the volcano, effectively increasing the probability of hazardous levels of ash reaching the European continent.

The concentration thresholds for aviation, which were adopted after the Eyjafjallajökull eruption in 2010, havestrong influence on the hazard estimates for weaker eruptions but is less important for larger eruptions; thusash forecasts for weaker eruptions are likely more uncertain in comparison to larger eruptions.

Keyword
Dispersion modelling, FLEXPART, Aviation safety, Climatology Hazard, Iceland
National Category
Meteorology and Atmospheric Sciences Geosciences, Multidisciplinary
Identifiers
urn:nbn:se:uu:diva-233386 (URN)10.1016/j.jvolgeores.2014.08.022 (DOI)000346551400006 ()
Available from: 2014-10-02 Created: 2014-10-02 Last updated: 2016-09-27Bibliographically approved
3. Seasonal and diurnal patterns in the dispersionof SO2 from Mt. Nyiragongo
Open this publication in new window or tab >>Seasonal and diurnal patterns in the dispersionof SO2 from Mt. Nyiragongo
Show others...
2016 (English)In: Atmospheric Environment, ISSN 1352-2310, E-ISSN 1873-2844, Vol. 132, 19-29 p.Article in journal (Refereed) Published
Abstract [en]

Mt. Nyiragongo is an active volcano located in the Democratic Republic of Congo, close to the border of Rwanda and about 15 km north of the city of Goma (similar to 1,000,000 inhabitants). Gases emitted from Nyiragongo might pose a persistent hazard to local inhabitants and the environment. While both ground- and satellite-based observations of the emissions exist, prior to this study, no detailed analysis of the dispersion of the emissions have been made. We have conducted a dispersion study, using a modelling system to determine the geographical distribution of SO2. A combination of a meteorological model (WRF), a Lagrangian particle dispersion model (FLEXPART-WRF) and flux data based on DOAS measurements from the NOVAC-network is used. Since observations can only be made during the day, we use random sampling of fluxes and ensemble modelling to estimate night-time emissions. Seasonal variations in the dispersion follows the migration of the Inter Tropical Convergence Zone. In June-August, the area with the highest surface concentrations is located to the northwest, and in December-February, to the southwest of the source. Diurnal variations in surface concentrations were determined by the development of the planetary boundary layer and the lake-/land breeze cycle around lake Kivu. Both processes contribute to low surface concentrations during the day and high concentrations during the night. However, the strong northerly trade winds in November-March weakened the lake breeze, contributing to higher daytime surface concentrations along the northern shore of Lake Kivu, including the city of Goma. For further analysis and measurements, it is important to include both seasonal and diurnal cycles in order to safely cover periods of high and potentially hazardous concentrations.

Keyword
Dispersion modelling; Volcanic degassing; Nyiragongo; Sulfur dioxide; FLEXPART-WRF
National Category
Earth and Related Environmental Sciences Meteorology and Atmospheric Sciences
Research subject
Meteorology
Identifiers
urn:nbn:se:uu:diva-264437 (URN)10.1016/j.atmosenv.2016.02.030 (DOI)000374614500003 ()
Funder
EU, European Research Council, 18354Sida - Swedish International Development Cooperation Agency, SWE-2008-064Swedish National Infrastructure for Computing (SNIC), p2011191
Available from: 2015-10-12 Created: 2015-10-12 Last updated: 2016-09-27Bibliographically approved
4. Using DOAS traverses and atmospheric modelling to determine plumeheight and eruption rate of the 2015 Holuhraun eruption
Open this publication in new window or tab >>Using DOAS traverses and atmospheric modelling to determine plumeheight and eruption rate of the 2015 Holuhraun eruption
(English)Manuscript (preprint) (Other academic)
Abstract [en]

The fissure eruption in Holuhraun — part of the Barðarbunga volcanic system — in 2014–2015 was amajor emitter of SO 2 . We present estimates of the SO 2 release rate of the eruption based on inverseatmospheric modelling and mobile-DOAS measurements made 80–240 km downwind of the main fissurevent. The FLEXPART-WRF and FLEXPART models are used to simulate the dispersion, using differentmeteorological data sets as input. Different inversion schemes were used to determine emission rates basedon modelled and measured column densities. The results were compared with OMI satellite observationsfor validation. This is, to our knowledge, the first case where a ground-based mobile-DOAS measurementshave been used in an atmospheric dispersion model for inverse modelling.

Comparisons were made between dispersion simulations based on meteorological data from different con-figurations of the WRF-model (at a resolution of 1.5 km); comparisons were also made using analysis andforecast data from ECMWF (at 0.2 degree resolution). Dispersion simulations based on data from ECMWFshowed better agreement with OMI satellite data than any of the simulations based on data from WRF.The inversion technique produced less variable emission rates compared to previous estimates, except duringperiods with low directional wind shear. Plume heights determined by the inverse modelling were below4 km for all periods.

We estimate the emission rate of the eruption to 500 kg/s in September 2014 and to 150 kg/s in thebeginning of February 2015, with a steady decrease over time.

Keyword
Dispersion modelling, FLEXPART, Volcanic eruption, Sulfur dioxide, Holuhraun
National Category
Meteorology and Atmospheric Sciences
Research subject
Meteorology
Identifiers
urn:nbn:se:uu:diva-303950 (URN)
Available from: 2016-09-27 Created: 2016-09-27 Last updated: 2016-09-27Bibliographically approved

Open Access in DiVA

fulltext(4310 kB)105 downloads
File information
File name FULLTEXT01.pdfFile size 4310 kBChecksum SHA-512
f9bebc74f2c5e33785f9a2068696369fd223f8c7f874734191d303885ed61d2c4cf2df3eb848f8c25802f3c3249377f8692d91ef6c0551634cfeb360134bc732
Type fulltextMimetype application/pdf
Buy this publication >>

Search in DiVA

By author/editor
Dingwell, Adam
By organisation
LUVAL
Meteorology and Atmospheric Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 105 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 425 hits
ReferencesLink to record
Permanent link

Direct link