A mos. Chem. Phys., 25, 7343–7368, 2025
h ps://doi.o g/10.5194/acp-25-7343-2025
© Au ho (s) 2025. This wo k is dis ibu ed unde
he C ea i e Commons A ibu ion 4.0 License.
Resea ch a icle
Volcanic emission es ima es om he in e sion o
ACTRIS lida obse a ions and hei use o
quan i a i e dispe sion modeling
Anna Kampou i1,2, Vassilis Ami idis1, Thanasis Geo giou1,3, S a os Solomos4, Anna Giali aki1,5,
Ma ia Tsichla1,6, Michael Rennie7, Simona Scollo8, and P od omos Zanis2
1Ins i u e o As onomy, As ophysics, Space Applica ions and Remo e Sensing (IAASARS),
Na ional Obse a o y o A hens, 10560 A hens, G eece
2Depa men o Me eo ology and Clima ology, School o Geology,
A is o le Uni e si y o Thessaloniki, 54124 Thessaloniki, G eece
3Labo a o y o A mosphe ic Physics, School o Physics, A is o le Uni e si y o Thessaloniki,
54124 Thessaloniki, G eece
4Resea ch Cen e o A mosphe ic Physics and Clima ology, Academy o A hens, 10680 A hens, G eece
5Depa men o Physics and As onomy, Ea h Obse a ion Science G oup,
Uni e si y o Leices e , Leices e LE1 7RH, UK
6En i onmen al Chemical P ocesses Labo a o y, Depa men o Chemis y,
Uni e si y o C e e, He aklion, G eece
7Eu opean Cen e o Medium-Range Wea he Fo ecas s, Reading RG2 9AX, UK
8Is i u o Nazionale di Geo isica e Vulcanologia, Osse a o io E neo, 95125 Ca ania, I aly
Co espondence: Anna Kampou i ([email p o ec ed])
Recei ed: 14 Oc obe 2024 – Discussion s a ed: 3 Janua y 2025
Re ised: 28 Ma ch 2025 – Accep ed: 28 Ma ch 2025 – Published: 14 July 2025
Abs ac . Modeling he dispe sion o olcanic pa icles ollowing explosi e e up ions is c i ical o a ia ion
sa e y. To cons ain he dispe sion o olcanic plumes and assess haza ds, calcula ions ely on he accu a e
cha ac e iza ion o he e up ion’s sou ce e m, e.g., a ia ion in emission a e and column heigh wi h ime and
he p e ailing wind ields. This s udy in oduces an in e se modeling amewo k ha in eg a es a Lag angian
dispe sion model wi h lida obse a ions o es ima e emission a es o olcanic pa icles eleased du ing an E na
e up ion. The me hodology consis s o using he FLEXPART model o gene a e sou ce– ecep o ela ionships
(SRRs) be ween he olcano and he lida sys em ha obse ed he olcanic plume. These SRRs a e hen used
o de i e he emission a es based on obse a ional da a, including olcanic ash plume heigh s om he INGV-
EO obse a o y and PollyXT lida e ie als. We le e age da a om he ACTRIS PollyXT lida ha ope a es a
he PANhellenic GEophysical obse a o y o An iky he a o he Na ional Obse a o y o A hens (PANGEA-
NOA). The in e sion algo i hm u ilizes lida obse a ions and an empi ical a p io i emission p o ile o es ima e
he olcanic pa icle sou ce s eng h, accoun ing o al i ude and ime o he plume’s e olu ion. Addi ionally, o
s udy he impac ha he wind ields ha e on olcanic ash o ecas ing, he expe imen is epea ed using ields
ha assimila e Aeolus wind lida da a. Ou app oach applied o he 12 Ma ch 2021 E na e up ion and accu a ely
cap u es a dense ae osol laye be ween 8 and 12km abo e he PANGEA-NOA s a ion. Resul s show a minimal
di e ence o he o de o 2% be ween he obse ed and he simula ed ash concen a ions. Fu he mo e, he
s uc u e o he a pos e io i ash plume closely esembles he ash cloud image cap u ed by he SEVIRI sa elli e
abo e An iky he a island, highligh ing he no el y o he in e sion esul s. The p esen ed in e sion algo i hm,
coupled wi h Aeolus da a, op imizes bo h he e ical emission dis ibu ion and E na emission a es, ad ancing
ou unde s anding and p epa edness o olcanic e en s.
Published by Cope nicus Publica ions on behal o he Eu opean Geosciences Union.
7344 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
1 In oduc ion
Volcanic ash cons i u es a signi ican haza d o a ia ion when
i is emi ed a ai c a c uising al i udes (9–11km), wi h po-
en ial consequences including ai c a engine ailu e (Gu -
an i e al., 2005), inaccu a e eadings o c i ical na iga ional
ins umen s, and educed isibili y due o ex e nal ai c a
co osion (Cla kson and Simpson, 2017; ICAO, 2016).
In he case o a olcanic e up ion, u gen decisions a e
necessa y o de e mine sa e ligh ou es and o ensu e ha
ai bo ne ai c a land sa ely. While sa e y emains he op
p io i y, he g ounding and e ou ing o ligh s lead o la ge
inancial losses, e.g., he 2010 e up ion o Eyja jallajökull in
Iceland epo edly cos he ai line indus y o e USD 1 bil-
lion (Mazzocchi e al., 2010; Ox o d Economics, 2012).
In o ma ion on olcanic ash dispe sion a e an e up ion
is p o ided o ope a o s by specialized ea ly wa ning sys-
ems (EWSs) ope a ed by he Volcanic Ash Ad iso y Cen-
e s (VAACs) (Fea nley e al., 2018). These sys ems ypically
ely on de e minis ic olcanic ash anspo and dispe sion
models (VATDM) o o e sho - e m o ecas s o he ol-
canic ash cloud. Al hough VAACs speci y he expec ed loca-
ion o he ash cloud, usually hey do no p o ide quan i a i e
in o ma ion abou ash concen a ion. In he spo ligh o he
expec ed ise in he numbe o ligh s o e olcanically ac i e
egions in he nea u u e (as indica ed by EUROCONTROL,
2022), he p obabili y o encoun e ing olcanic ash a ai c a
c uising al i udes will p opo ionally inc ease. Consequen ly,
he challenge is o minimize unce ain ies in sho - e m o e-
cas s o olcanic ash dispe sion.
The p ima y sou ces o unce ain ies in de e minis ic
anspo models o igina e om he e up ion sou ce pa am-
e e s, he a ious model pa ame e iza ions (such as we de-
posi ion), and he d i ing me eo ological condi ions (Dac e
e al., 2011; P a a and Lynch, 2019; S ohl e al., 2011). Typ-
ically, olcanic ash anspo and dispe sion models (VAT-
DMs) equi e he speci ica ion o pa ame e s abou he ol-
canic e en , including a e ical p o ile o ash emission a es,
pa icle size dis ibu ion, and ash densi y (Ha ey e al.,
2020). The e up ion s a ime can be es ima ed h ough sa el-
li e obse a ions o by local olcano obse a o ies. Va ious
emo e sensing echniques exis o es ima e he heigh o he
ash plume (Pe e sen e al., 2012). I should be men ioned ha
in o ma ion ha elies on obse a ions om passi e senso s
has limi ed sensi i i y o he ash laye heigh . Mass e up ion
a es a e ypically e alua ed using empi ical ela ionships
based on obse ed plume heigh s (Mas in e al., 2009). How-
e e , hese empi ical ela ionships o en ail o conside sec-
onda y ac o s in luencing plume heigh , such as me eo olog-
ical condi ions. The long- ange anspo o olcanic pa i-
cles is in luenced by oposphe ic and/o s a osphe ic winds,
pa icula ly he e ical wind shea , which is equen ly inac-
cu a ely ep esen ed in nume ous nume ical wea he p edic-
ion (NWP) models (Ha ey e al., 2020; Houchi e al., 2010;
S o elen e al., 2020).
Mo eo e , olcanic pa icles can in luence he plane a y
adia i e balance h ough bo h di ec and indi ec e ec s, in-
oducing signi ican unce ain ies in plume dispe sion and
li e ime. The di ec e ec in ol es he sca e ing and abso p-
ion o sola and e es ial adia ion, whe e ine ash and sul-
a e ae osols con ibu e o su ace cooling o a mosphe ic
wa ming depending on pa icle composi ion, size dis ibu-
ion, and injec ion plume heigh (Robock, 2000; Sica d e al.,
2025). The indi ec e ec ela es o he ole o olcanic pa i-
cles in cloud mic o- and mac ophysical p ope ies. Ash pa i-
cles can ac as cloud condensa ion nuclei (CCN), acili a ing
wa e d ople o ma ion and, unde speci ic p essu e and em-
pe a u e condi ions, as ice nuclei (IN) (Gue ie i e al., 2023).
These p ocesses can al e cloud op ical and mic ophysical
p ope ies, enhance cloud e lec i i y, a ec cloud li e imes,
and inc ease he unce ain ies in adia i e ans e . Addi ion-
ally, olcanic ice clouds can hide possible ash laye s and pose
a se e e h ea o a ia ion sa e y. A mosphe ic anspo mod-
els o en s uggle o accoun o hese complex in e ac ions,
leading o unce ain ies in plume e olu ion, ajec o y o e-
cas s, and deposi ion es ima es. Fu he mo e, he absence o
signi ican physical p ocesses, dependence on empi ical ela-
ions, and da a om p e ious e up ions u he con ibu e o
subs an ial unce ain ies in es ima es o he e up ed mass.
O e he pas wo decades, signi ican p og ess has been
made in in eg a ing emo e sensing da a in o a mosphe ic
anspo models o enhance he o ecas ing o olcanic emis-
sions and hei dispe sion. Sa elli e obse a ions om bo h
pola -o bi ing and geosynch onous he mal in a ed ins u-
men s ha e been used o e ie e ash mass loadings (Cla isse
e al., 2010; Pa olonis e al., 2013; P a a and P a a, 2012).
Addi ional senso s, including he Mode a e Resolu ion Imag-
ing Spec ome e (MODIS), Second Gene a ion Spinning
Enhanced Visible and In a-Red Image (SEVIRI), A mo-
sphe ic In a-Red Sounde (AIRS), Ozone Moni o ing In-
s umen (OMI), Mul i-angle Imaging Spec oRadiome e
(MISR), and CALIOP lida on boa d he CALIPSO ha e p o-
ided aluable da a on olcanic ash de ec ion and e ie als
(Eckha d e al., 2008; F ancis e al., 2012). A comp ehen-
si e discussion on he applica ion o sa elli e emo e sensing
o olcanic ash moni o ing in a ia ion haza d mi iga ion is
p o ided by P a a (2009).
In addi ion, g ound-based lida ne wo ks, such as he Eu-
opean Ae osol Resea ch Lida Ne wo k (EARLINET), ha e
played a c ucial ole in alida ing he accu acy o anspo
model ou pu s and imp o ing dispe sion simula ions by p o-
iding high- esolu ion e ical p o iles o olcanic ae osols
(Pappala do e al., 2004).
Ad ancemen s in a mosphe ic anspo and dispe -
sion modeling ha e u he acili a ed he in eg a ion o
hese obse a ional da ase s. Models like he Nume ical
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7345
A mosphe ic-dispe sion Modelling En i onmen (NAME;
Jones e al., 2007) (which is used ope a ionally by he Lon-
don Volcanic Ash Ad iso y Cen e, o LVAAC), he Hy-
b id Single-Pa icle Lag angian In eg a ed T ajec o y (HYS-
PLIT) model (S ein e al., 2015), and he FLEXible PARTi-
cle (FLEXPART) dispe sion model (Eckha d e al., 2008;
K is iansen e al., 2010, 2012, 2014; S ohl e al., 2011) ha e
been ex ensi ely used o olcanic ash o ecas ing, o en in-
co po a ing sa elli e and lida da a o e ine model inpu s and
imp o e p edic i e accu acy.
The in eg a ion o emo e sensing da a in o a mosphe ic
anspo models has been signi ican ly ad anced h ough in-
e sion algo i hms. In p e ious s udies (Eckha d e al., 2008;
K is iansen e al., 2010), in e sion algo i hms we e de el-
oped using sa elli e column e ie als and es ed o es ima e
he e ical dis ibu ion o sul u dioxide (SO2) emission
a es o quasi-ins an aneous olcanic e up ions such as he
2007 Jebel a Tai and he 2008 Kasa ochi e up ions. Seibe
e al. (2011) examined he unce ain ies o he a ious con-
igu a ions o he 2008 Kasa ochi case s udy and expanded
he me hod o es ima e he unce ain y o he e ie ed sou ce
emissions (a pos e io i unce ain ies).
The in e sion algo i hm was u he used by S ohl e al.
(2011) o olcanic ash emission a es as a unc ion o al i-
ude and ime, while K is iansen e al. (2012) imp o ed he
olcanic ash in e sion echniques using a ious inpu s o be -
e cons ain he 2010 Eyja jallajökull e up ion.
Ami idis e al. (2023) demons a ed ha olcanic ash ea ly
wa ning sys ems can be signi ican ly enhanced by he assim-
ila ion o Aeolus wind ields. No ably, hese imp o emen s
a e mos p onounced o e unde -sampled geog aphical e-
gions, such as he Medi e anean Sea, as olcanoes a e o -
en si ua ed in emo e a eas lacking su ace-based obse a-
ion ne wo ks. Mo eo e , he s udy indica es ha he pos-
i i e e ec o Aeolus wind da a assimila ion is mo e p o-
nounced in he middle and uppe oposphe e (mos ly be-
ween 7 and 15km) compa ed o he lowe oposphe e. This
may highligh unde -sampling issues, since he in si u ob-
se a ions (like adiosondes) adi ionally used o da a as-
simila ion exhibi lowe e ical esolu ion in he uppe o-
posphe e (Rennie e al., 2021). Conside ing ha olcanic
ash plumes a e ypically injec ed in o uppe - oposphe ic and
lowe -s a osphe ic heigh s, hei anspo is la gely in lu-
enced by uppe - oposphe ic winds, hence accu acy in dis-
pe sion modeling is ad anced om high-accu acy wind ield
assimila ion.
Building on hese ad ancemen s, ou s udy u he in es-
iga es imp o emen s in ash emission es ima ions by de el-
oping an in e sion me hod ha in eg a es Aeolus wind da a,
g ound-based lida obse a ions, and anspo model simu-
la ions. This app oach aims o enhance he accu acy o ol-
canic emission sou ce e ms and o educe unce ain ies in
dispe sion o ecas ing.
We speci ically ocus on he E na e up ion ha occu ed on
12 Ma ch 2021, coinciding wi h he in es iga ions p o ided
by Ami idis e al. (2023) and Kampou i e al. (2023). Du ing
his e en , Aeolus had a close o e pass o E na, p o iding
aluable obse a ions a ound he olcano. Addi ionally, he
anspo ed olcanic plume was cap u ed in he eas e n e-
gion o he Medi e anean by he g ound-based PollyXT lida
sys em o he PANhellenic GEophysical obse a o y o An-
iky he a o he Na ional Obse a o y o A hens (PANGEA-
NOA) in G eece, downwind o he E na olcano. This allows
o di ec compa isons o obse a ions agains o ecas s, wi h
and wi hou he assimila ion o Aeolus da a, deno ed as “w”
and “w/o” Aeolus, espec i ely (as indica ed in he s udies by
Ami idis e al., 2023; Kampou i e al., 2023).
2 The case o he E na olcanic e up ion o
12–14 Ma ch 2021
2.1 Volcanic ac i i y
M . E na in I aly, ecognized as one o he mos ac i e ol-
canoes on Ea h, has unde gone signi ican olcanic ac i i y,
pa icula ly since Feb ua y 2021. Du ing his pe iod, he s a-
o olcano expe ienced nume ous pa oxysmal episodes, lead-
ing o equen eph a and SO2emissions. A no able e en
occu ed on 12 Ma ch 2021, ma king one o he mos powe -
ul la a oun ain episodes obse ed a he sou heas e n c a e
since 2020 (Cal a i e al., 2021). The olcanic ac i i y s a ed
wi h S ombolian- ype e up ions a a ound 02:35UTC, esca-
la ing in bo h equency and in ensi y un il 07:35UTC, when
su eillance came as om he Is i u o Nazionale di Geo isica
e Vulcanologia, Osse a o io E neo (INGV-OE) (Co adini
e al., 2018; Scollo e al., 2019) cap u ed he o ma ion o a
sus ained la a oun ain.
Th oughou he pa oxysmal phase, he e up i e column
g adually eached a heigh o 9kma.s.l.(Fig. 1). The a i-
a ion in he e up ion column was de ec ed by he isual
su eillance came a in he CUAD in Ca ania (ECV), cal-
ib a ed by INGV-OE (Fig. 1). The olcanic plume d i ed
eas wa ds unde he in luence o he p e ailing wes e ly
winds dominan in he eas e n Medi e anean egion a he
ime. Acco ding o he Volcano Obse a o y No ice o A i-
a ion (VONA) messages, he INGV-EO obse a o y (INGV,
2025; Co adini e al., 2018; Scollo e al., 2019) issued a ed
wa ning ale , om 06:18 o 08:44UTC, on 12 Ma ch 2021,
when he s onges ash emission was obse ed, while an o -
ange ale was issued a 12:30UTC when he la a oun ain
ceased, and he olcanic ash plume was dispe sed in o he
a mosphe e (Cal a i e al., 2021). Addi ionally, he e up i e
ac i i y esul ed in abundan eph a allou , co e ing se e al
owns on he eas e n lank o he olcano c a e , wi h a la a
low ield expanding on he eas e n and no heas e n lank. In
his s udy, he cloud heigh s epo ed by VONA a e used as
a p io i in o ma ion o ini ialize he olcanic ash dispe sion
simula ions, conduc ed wi h he FLEXPART ( lexible pa i-
cle dispe sion) Lag angian model (B ioude e al., 2013; Pisso
e al., 2019; S ohl e al., 2005). The FLEXPART ash ans-
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7346 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Figu e 1. E na ac i i y on 12 Ma ch 2021 as seen om INGV-OE. Ash plume images om an ECV calib a ed came a moni o ed he explosi e
olcanic ac i i y be ween 5 and 9 kma.s.l.(a) Weak ash plume a 06:30UTC, wi h an uppe pa aligning mo e e ically. (b) S ong e ical
plume a 08:00UTC, which shi ed eas wa d. (c) S ong ash plume a 09:00UTC, wi h a lowe and mo e dilu ed cloud caused by he la a
low expanding eas wa d. (d) Dec ease in he explosi e ac i i y a e 10:00UTC. Figu es a e aken om he INGV-OE au oma ic sys em
desc ibed in Co adini e al. (2018) and Scollo e al. (2019).
po model is d i en by wind ields simula ed by he WRF
egional me eo ological model ( e sion 4) (Skama ock e al.,
2019), which, in u n, de i es ini ial and bounda y condi-
ions om he Eu opean Cen e o Medium-Range Wea he
Fo ecas s (ECMWF) In eg a ed Fo ecas ing Sys em (IFS)
(ECMWF, 2021a) global model ( o addi ional in o ma ion,
see Sec . 3.3).
3 Me hods and da a
The in e se me hod employed in his s udy o es ima e ol-
canic ash emissions in eg a es a p io i in o ma ion on ash
emissions, g ound-based lida obse a ions, and simula ions
wi h a dispe sion model, esul ing in imp o ed ash emission
es ima es. Figu e A1 p esen s a schema ic wo k low ou lin-
ing he me hodology ollowed in his s udy, p o iding a clea
o e iew o he s eps in ol ed in ou app oach. In his sec-
ion, we desc ibe he da ase s and me hods employed in he
in e se modeling p ocess.
3.1 PANGEA-NOA g ound-based da a (lida PollyXT)
The PANGEA-NOA obse a o y es ablished i s i s ope -
a ions in June 2018 on he emo e island o An iky he a,
G eece. The a mosphe ic ci cula ion pa e n a PANGEA-
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7347
NOA a o s he anspo o ai masses ca ying an abun-
dance o di e en ae osol ypes such as windblown Saha a
dus , E na olcanic ae osols, smoke om wild i es, and an-
h opogenic pollu ion om majo ci ies. Hence, his coas al
si e cons i u es an ideal place o s udy na u al ae osols unde
he p e ailing backg ound condi ions o he eas e n Medi e -
anean.
The Medi e anean egion, pa icula ly i s eas e n basin,
se es as a con luence o ai masses o igina ing om Eu-
ope, Asia, and A ica. In his egion, an h opogenic emis-
sions om la ge u ban cen e s in e ac wi h na u al emis-
sions om he Saha an and Middle Eas e n dese s, smoke
om equen wild i es, and olcanic pa icles om e up-
ions, no ably om M . E na and Icelandic olcanoes. Addi-
ionally, he a mosphe e o e he eas e n Medi e anean con-
ains backg ound ma ine ae osols and pollen pa icles om
oceanic and ege a i e sou ces. Ae osols exe a a ie y o
e ec s on he egional wea he and clima e, impac ing sola
adia ion, isibili y, and human heal h, and hey pose signi i-
can conce ns o a ia ion sa e y (WMO, 2024).
The eas e n Medi e anean is cha ac e ized by a Medi e -
anean clima e, wi h ho , d y summe s and mild, we win-
e s. This seasonal a iabili y is d i en p ima ily by he
in e ac ion be ween mid-la i ude wes e lies and sub opical
high-p essu e sys ems (Lensky e al., 2018). Du ing win e ,
he egion expe iences he equen passage o ex a opical
cyclones o igina ing om he No h A lan ic and Medi e -
anean s o m acks, b inging p ecipi a ion and colde em-
pe a u es. In con as , summe condi ions a e domina ed by
he expansion o he sub opical heigh , leading o s able a -
mosphe ic condi ions and minimal ain all (ECMWF, 2010).
Synop ic-scale ci cula ion in he eas e n Medi e anean
plays a c ucial ole in shaping wea he pa e ns and a mo-
sphe ic dynamics. The a mosphe ic ci cula ion o e he eas -
e n Medi e anean is domina ed by pe sis en no he ly and
wes e ly winds, a o ing he ad ec ion o olcanic p oduc s
om E na o G eece (Kampou i e al., 2020; Scollo e al.,
2013). Resea ch has iden i ied se e al dominan synop ic
ypes ha in luence he egion, including cyclonic sys ems,
an icyclonic pa e ns, and blocking heigh s (Rousi e al.,
2014). These ci cula ion pa e ns signi ican ly impac he
anspo o ae osols, mois u e, and pollu an s, a ec ing e-
gional ai quali y and clima e a iabili y. Fu he mo e, he
egion’s p oximi y o la ge-scale ci cula ion ea u es such as
he sub opical je s eam and he A ican monsoon sys em
con ibu es o complex seasonal in e ac ions (Lensky e al.,
2018).
Cu en ly, a PollyXT lida sys em (Baa s e al., 2017; En-
gelmann e al., 2016) and a sun–sky pho ome e o CIMEL
Elec onique (Giles e al., 2019; Goloub e al., 2007) ope -
a e con inuously a PANGEA-NOA o p o ide p o iles and
columna ae osol p ope ies wi h high accu acy and esolu-
ion.
PollyXT is a mul i-wa eleng h Raman pola iza ion lida
wi h 24/7 emo e ope a ion capabili y. The sys em ope a es
in 355, 532, and 1064nm and is equipped wi h 12 de ec o s
o measu e ligh elas ically and inelas ically (a 387, 407, and
607nm), backsca e ed om a mosphe ic cons i uen s. Po-
la iza ion capabili y also enables he de ec ion and e ical
sepa a ion o non-sphe ical (e.g., olcanic ash, dus ) om
sphe ical ae osols (e.g., smoke, pollu ion, ma ine pa icles).
The CIMEL sun–sky pho ome e measu es di ec sola
and sky adiance a se e al wa eleng hs (340, 380, 440, 500,
675, 870, 1020, and 1640nm) o de i e column-in eg a ed
ae osol op ical and mic ophysical p ope ies (Dubo ik e al.,
2006).
Obse a ions om bo h senso s a e o s ong in e -
es o Pan-Eu opean and global ne wo ks such as he
Ae osol, Clouds and T ace Gases Resea ch In as uc u e
(ACTRIS-RI), he Eu opean Ae osol Resea ch Lida Ne -
wo k (EARLINET), and he global AE osol RObo ic NET-
wo k (AERONET: h ps://ae one .gs c.nasa.go /, las access:
27 June 2025). In all o hese ne wo ks, measu emen s aken
a PANGEA-NOA a e submi ed on a egula basis.
3.1.1 Ash mass calcula ions using emo e sensing da a
Volcanic ash mass es ima es we e de i ed om a combi-
na ion o PollyXT lida measu emen s and sun pho ome e
obse a ions. Fi s , he lida measu emen s we e a e aged
o e he 3 h pe iod when he olcanic laye was obse ed
abo e An iky he a, and he s anda dized EARLINET algo-
i hm Single Calculus Chain (SCC) (D’Amico e al., 2015)
was used o de i e he pa icle backsca e coe icien (βp)
and pa icle linea depola iza ion a io (δp) p o iles.
These p o iles we e hen used o disen angle he con ibu-
ion o la ge, non-sphe ical ash pa icles om he obse ed
olcanic plume and o calcula e he ash mass concen a-
ion wi h he POla iza ion-LIda PHO ome e Ne wo king
(POLIPHON) me hod (Ansmann e al., 2012; Mamou i and
Ansmann, 2017), ailo ed o E na ash, as desc ibed in Kam-
pou i e al. (2020).
Mo e speci ically, he ollowing equa ion was used:
ma=ρa×c ,a(λ)×βp,a(h,λ)×Sp,a(h,λ),(1)
whe e mis he mass concen a ion, aindica es an ae osol
ype, ρ ep esen s he pa icle mass densi y ( o olcanic ash
pa icles, his is 2.6±0.6 gm−3 ollowing he s udy o Ans-
mann e al., 2011a), λis he wa eleng h, c (λ) is he so-
called olume- o-ex inc ion con e sion ac o (de i ed om
sun pho ome e measu emen s), his he heigh abo e g ound,
and Sp(λ,h) is he a io o he pa icle ex inc ion o pa icle
backsca e coe icien (lida a io).
As he macalcula ion is sensi i e o he ae osol ype, unde
he simul aneous p esence o mul iple ae osol componen s
in he a mosphe ic column, a decomposi ion o he o al pa -
icle backsca e coe icien βpis needed p io o he mass
concen a ion calcula ion. In POLIPHON, his decomposi-
ion is suppo ed o up o wo ae osol ypes, one exhibi -
ing la ge pa icle depola iza ion a io alues (usually dus
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7348 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Table 1. Pa ame e s used o lida p o ile decomposi ion and mass concen a ion calcula ion.
ρα[µmcm−3]cν,α,532nm δp,α,532nm(h)Sp,α,532nm(h) [s ]
Ash pa icles 2.6±0.6 0.6±0.1 0.36±0.02 50±10
Sul a es 1.5±0.3 0.18±0.02 0.05±0.01 60±20
o olcanic ash) and one ha does no (ma ine, con inen-
al, o oposphe ic smoke and hei mix u es). To sepa a e
he con ibu ion o he depola izing (βp,d(h,λ)) and he non-
depola izing (βp,nd(h,λ)) ae osol componen om he o al
pa icle backsca e coe icien , we apply he ollowing equa-
ions:
βp,d(h,λ)=βp(h,λ)
×(δp(h,λ)−δp,nd(h,λ))(1+δp,d(h,λ))
(δp,d(h,λ)−δp,nd(h,λ))(1+δp(h,λ)) (2)
βp,nd(h,λ)=βp(h,λ)−βp,d(h,λ).(3)
PollyXT lida signals a e sensi i e o ae osol pa icles in
he adius ange om abou 50nm o a ew mic ome e s
(Wei kamp, 2005). Fo FLEXPART, he size ange consid-
e ed o olcanic ash pa icles is be ween 5 and 21 µm in
diame e and hus wi hin he ange ha is de ec able by
PollyXT. Unce ain ies in he ash mass concen a ion calcu-
la ion using he POLIPHON me hod a ise om he inpu pa-
ame e e o s ha p opaga e in o Eq. (1) and a e expec ed o
be in he o de o ∼40% (Ansmann e al., 2011b). The ech-
nique has been alida ed agains syne gis ic e ie als ha
combine mul i-wa eleng h lida and sun–sky adiome e ob-
se a ions (sensi i e up o 15µm in pa icle adius (Lopa in
e al., 2013, 2021)) o dus and olcanic ash pa icles and
has been ound o pe o m well (Kons a e al., 2021; Wagne
e al., 2013).
In Table 1, we summa ize he alues and unce ain ies o
he pa ame e s used as inpu o he abo e. The lida a io
o coa se-mode olcanic ash a 532nm is epo ed o ange
be ween 40 and 60s in he li e a u e (see o example G oß
e al., 2012; Table 3 o pa icle ex inc ion and backsca e
alues in Flou si e al., 2023; and Gas eige e al., 2011).
Fo he ine-mode ae osols, we use a mean alue o 60 s ,
ollowing he alues epo ed in he li e a u e o pa icles
o a sul u ing na u e (see o example Flou si e al., 2023;
Mülle e al., 2007). We also accoun o a lida a io e-
ie al unce ain y o ∼30% o cap u e he measu emen
ange (Ansmann e al., 2012; Giannakaki e al., 2015; G oß
e al., 2013). The pa icle densi y alues ρ ollow om he
OPAC model o he coa se-mode mine al componen and
he wa e -soluble componen o ash and sul a e pa icles, e-
spec i ely (Hess e al., 1998; Koepke e al., 2015). Fo he
wa e -soluble componen , we assume alues a a ela i e hu-
midi y o 0%, which is conside ed ep esen a i e o he al-
i udes o he olcanic laye s. The coa se-mode componen
is no conside ed o be hyd ophilic. Finally, he ex inc ion-
o-mass con e sion ac o s c a e aken om Ansmann e al.
(2011a) o ash and ine-mode pa icles, espec i ely.
3.2 Aeolus high-spec al- esolu ion lida (HSRL) da a
Aeolus, he wind mission o he Eu opean Space Agency
(ESA), ca ied he wo ld’s i s high-spec al- esolu ion
Dopple wind lida in space (S o elen e al., 2006; S aume-
Lindne e al., 2021). Launched in Augus 2018, Aeolus’s
aim was o e ie e ho izon al wind p o iles in he opo-
sphe e and lowe s a osphe e. The mission’s p ima y objec-
i e was o showcase his inno a i e echnology in space o
enhance wea he o ecas s and o ad ance ou unde s and-
ing o a mosphe ic dynamics, pa icula ly in he opics.
Addi ionally, Aeolus aimed o con ibu e aluable insigh s
in o he in ica e in e ac ions be ween he a mosphe ic con-
s i uen s, wa e cycles, and b oade clima e sys em (Rennie
e al., 2021; S aume-Lindne e al., 2021). Aeolus wind da a
demons a ed no able quali y and co e age, leading o sub-
s an ial enhancemen s in NWP o ecas s, pa icula ly wi hin
he opics and Sou he n Hemisphe e. The imp o emen in
wind o ecas anges om 0.5% o 2%, in e ms o oo
mean squa e e o , main ains a signi ican impac e en in
medium- ange wea he o ecas ing. The mos subs an ial im-
pac was obse ed a app oxima ely 100hPa in he opics,
pa icula ly o e he eas e n Paci ic Ocean. This is a ibu ed,
in pa , o he opics ha ing a ela i ely limi ed co e age
o high-quali y adiosonde wind p o iles. Addi ionally, he
wind ield in he opics is less cons ained by empe a u e
in o ma ion om o he sa elli es (Rennie e al., 2021). Fu -
he mo e, Aeolus had he capabili y o e ie e ae osol and
cloud p o iles, o e ing aluable da a o assimila ion o e al-
ua ion in olcanic ash dispe sion modeling. I is essen ial o
no e, howe e , ha hese e ie als ace limi a ions due o
he absence o a dedica ed lida channel o de ec ing c oss-
pola ized ligh e u ns (wi h espec o he emi ed adia-
ion). This absence is pa icula ly c ucial o cap u ing he
backsca e ed ligh om non-sphe ical pa icles like olcanic
ash. Consequen ly, cau ion is ad ised when u ilizing Aeolus
obse a ions in such cases. Despi e his limi a ion, he Aeo-
lus mission demons a ed i s e icacy in enhancing wind o e-
cas s, pa icula ly o e unde -sampled egions, such as he
opics (Rennie e al., 2021). Simila ly, Aeolus can be used
o e unde -sampled emo e a eas wi h ac i e olcanoes, con-
ibu ing o imp o ed simula ions o olcanic ash dispe sion
ollowing e up ions.
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7349
Table 2. Con igu a ion o he PP schemes o he WRF-ARW simula ions.
PP Schemes Re e ences
Mic ophysics (MP) Thompson Thompson e al. (2008)
Su ace laye (SFL) Monin–Obukho (Janjic E a) Janjic (2002)
Plane a y bounda y laye (PBL) Mello –Yamada–Janjic (MYJ) Janjic (2003)
Cumulus pa ame e iza ion (CUM) Tied ke Zhang e al. (2011)
Longwa e and sho wa e adia ion (RAD) Rapid adia i e ans e model (RRTM) Iacono e al. (2008)
Land su ace (LSM) NOAH Chen and Dudhia (2001)
3.3 FLEXPART-WRF model se up
To pe o m me eo ological simula ions o e he s udy e-
gion o he eas e n Medi e anean, he Ad anced Resea ch
WRF model ( e sion 4) (Skama ock e al., 2019) is used.
The spa ial esolu ion o he model is 12km×12 km o a
o al o 351×252 g id poin s and 31 e ical le els (up o
50hPa). The simula ion pe iod s a s on 12 Ma ch 2021 a
00:00UTC (6 h ea lie han he FLEXPART uns o accom-
moda e he model’s 12 h spin-up) and ends on 14 Ma ch 2021
a 18:00UTC, wi h hou ly ou pu s. Table 2 summa izes he
physics pa ame e iza ion (PP) schemes o he WRF-ARW
simula ions.
In he con ex o his s udy, wo e sions (ECMWF, 2021a)
o he ini ial and bounda y condi ion ields om he IFS
we e u ilized. These ields, p o ided a a spa ial esolu ion
o 0.125°×0.125°, wi h 137 e ical model le els, se e as
inpu s o he WRF-ARW egional model. One e sion inco -
po a es assimila ed Aeolus Rayleigh-clea and Mie-cloudy
ho izon al line-o -sigh (HLOS) L2B wind p o iles ( e e ed
o as he “w” Aeolus expe imen ), while he o he e sion
is wi hou Aeolus da a ( e e ed o as he “w/o” Aeolus ex-
pe imen ). The ini ial condi ions wi hou Aeolus assimila ion
adhe e o he model se up u ilized in he obse ing sys em
expe imen s (OSEs) conduc ed by S o elen e al. (2006).
The WRF-ARW uns ely on ini ial and bounda y con-
di ions gene a ed om ECMWF-IFS, wi h bounda y condi-
ions upda ed a 6-hou in e als. Sea su ace empe a u e
(SST) analysis da a, ob ained om he Cope nicus Ma ine
En i onmen Moni o ing Se ice (CMEMS) a a spa ial es-
olu ion o 1/12° supplemen hese simula ions. The WRF-
ARW model con igu a ion u ilized in his s udy is consis en
wi h ha employed in he s udy o Ami idis e al. (2023).
The olcanic ash plume anspo simula ions we e done
wi h he Lag angian pa icle dispe sion model FLEXPART
(B ioude e al., 2013; Pisso e al., 2019; S ohl e al., 2005) in
a o wa d mode. These simula ions ely on hou ly me eo o-
logical ields om he WRF-ARW model, ini ia ed wi h IFS
da ase s. The use o 1-hou ly WRF me eo ological ields a a
12km ×12 km spa ial esolu ion allows o a mo e de ailed
ep esen a ion o he olcanic plume dispe sion. The ini ial
simula ions, in which we used an a p io i emission p o ile
o he e up ion emissions aken om VONA ale s ( om
now on e e ed o as “a p io i olcanic ash plume ans-
po ”), we e ini ia ed a he epo ed s a ime o he e up-
ion a 07:00UTC on 12 Ma ch 2021 and we e comple ed a
00:00UTC on 14 Ma ch 2021, wi h a o al o 100000 pa i-
cles eleased in each o ecas . The model laye s we e di ided
in o 18 laye s wi h 1km e ical esolu ion in he ange ex-
ending om 1 o 18km abo e g ound le el (a.g.l.). We es i-
ma e he a p io i mass e up ion a e (MER) o ash pa icles
ollowing Mas in e al. (2009) and Scollo e al. (2019) by
in e ing he obse ed plume heigh s o e he E na summi
c a e om he VONA epo s and ield obse a ions, as ob-
se ed by he INGV obse a o y, using he 1D plume model
o Deg uy e and Bonadonna (2012). The ini ial injec ion
heigh in he model is se o he al i ude o he E na summi
c a e s (3.3kma.s.l.) up o 9 kma.s.l., based on he VONA
epo s (Co adini e al., 2018; Scollo e al., 2019) and ield
obse a ions. Also, he g a i a ional pa icle se ling (Näs-
lund and Thaning, 1991) was de e mined assuming sphe ical
pa icles wi h a densi y o 2450kgm−3. The pa icle den-
si y alue used in he FLEXPART model di e s sligh ly om
he densi y used in Table 1 (2.6±0.6gcm−3) due o di e -
ences in shape assump ions, size dis ibu ions, and li e a u e
sou ces e e enced in a ious calcula ions. The size dis ibu-
ion o olcanic ash pa icles was desc ibed using ou size
bins (3, 5, 9, and 21µm in diame e ), as hese co e he size
dis ibu ion ele an o long- ange anspo (≤25µm diam-
e e ) (Becke e al., 2022; Dac e e al., 2011; Du an e al.,
2010).
To de i e he sou ce– ecep o ela ionships (SRRs), he
FLEXPART-WRF model was used, once again in a o wa d
mode (see Appendix A, Fig. A2), conside ing he same ou
ash size bins as hose used in he a p io i olcanic ash plume
anspo . The SRR model da a, which ep esen all po en-
ial dispe sion scena ios o he ash plume, a e compa ed wi h
he lida e ie als a PANGEA-NOA. Fo each g id poin
in he conside ed domain, he FLEXPART ash column load-
ings eleased om one pa icula emission ime and heigh
a e ma ched wi h he co esponding ime and g id poin o
he lida ash mass e ie al.
The FLEXPART SRRs we e d i en by he same hou ly
me eo ological ields om he WRF-ARW model, u ilizing
bo h con ol and assimila ed da ase s (ECMWF, 2021) o
quan i a i ely e alua e he impac o da a assimila ion. Sub-
sequen ly, hese SRRs we e used o ini ialize he in e sion al-
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7350 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
go i hm, cons ained by he PollyXT g ound-based lida mea-
su emen s o olcanic pa icles.
I was assumed ha he ash emissions occu ed be ween
he g ound and 16kma.g.l.o e he E na olcano. The o al
heigh ange was disc e ized in o 79 laye s o 200m hick-
ness. Fo each laye , 150 000 uni mass pa icle aces we e
uni o mly eleased along a e ical line sou ce e e y 2 h
( om 04:00 o 06:00UTC un il 12:00 o 14:00 UTC). Ad-
di ionally, he model laye s we e di ided in o 74: 70 lay-
e s be ween 200m and 14 km, wi h a e ical esolu ion o
200m; 3 laye s be ween 14 and 16 kma.g.l.(pe 1km); and
ano he laye om 22 o 50kma.g.l.These model-de i ed
column alues ep esen sou ce– ecep o ela ionships, since
hey we e ob ained wi h a uni mass as he sou ce. The ac ual
mass eleased a each le el is de e mined h ough he in e -
sion. Following he in e sion, a single, longe “pos e io i”
simula ion o e he pe iod 12 o 14 Ma ch 2021 was made,
eleasing 200000 pa icles acco ding o he es ima ed emis-
sion p o ile. The ou pu om his simula ion was p oduced
a he same e ical and ho izon al esolu ion as he a p io i
FLEXPART simula ion.
3.4 In e sion algo i hm
The in e sion me hod employed he e o ash sou ce es i-
ma ions is based on a cos unc ion minimiza ion app oach.
Simila wo k has been done by Eckha d e al. (2008), K is-
iansen e al. (2010), and S ohl e al. (2011). In hese s udies,
an in e sion algo i hm was de eloped o calcula e he e i-
cal dis ibu ion o sul u dioxide and he ash emission a es
o ins an aneous olcanic e up ions. Sa elli e e ie als, yp-
ically o ash column loading, ha e been combined in hose
analyses wi h VATDM simula ions using in e sion ech-
niques o p o ide ime-e ol ing es ima es o hese signi ican
quan i ies.
In sa elli e e ie al echniques, nume ous ad an ages ex-
is whe e es ima es o ash cloud op heigh and ash column
loading a e ypically a ailable (F ancis e al., 2012; Pa olo-
nis e al., 2013). Addi ionally, MER can be es ima ed h ough
empi ical ela ionships unde speci ic assump ions, which
a e especially use ul when sa elli e images a e una ailable o
limi ed, such as du ing he ea ly s ages o an e up ion (Pouge
e al., 2013; P a a e al., 2022). Howe e , di ec e ie als
o he e ical dis ibu ion wi hin he e up ion column a e
no easible. G ound-based and ai bo ne ada obse a ions,
which a e sensi i e o la ge pa icles and can pene a e op i-
cally hick plumes, p o ide a complemen a y sou ce o in o -
ma ion o e ie e nea -sou ce plume p ope ies such as mass
e up ion a e and column heigh .
The p esen s udy b ings oge he (i) he in e se model-
ing by ini ia ing he in e sion simula ions wi h mass concen-
a ions de i ed om g ound-based lida obse a ions down-
wind, combined wi h he sou ce– ecep o ela ionships cal-
cula ed om he FLEXPART-WRF model, and (ii) he in-
eg a ion o Aeolus me eo ological wind ields (ECMWF,
2021a) in o he FLEXPART-WRF model ( o mo e de ail,
see Sec . 3.3). The o e a ching goal is o op imize bo h he
e ical emission dis ibu ion and he ash emission a es nea
he sou ce, ollowing he olcanic e up ion. F om he in e -
sion scheme, a o al ash emission p o ile o he e up ion is
ob ained, which can be u ilized o gene a e obus ash o e-
cas s cons ained by lida obse a ions.
We pe o m he in e sion using a Bayesian app oach o
p o ide he bes es ima e o he emissions p o ile o ine
ash (wi h pa icles 3, 5, 9, and 21µm in diame e ) ha can
be anspo ed o e long dis ances. We ollow he gene al
concep o sou ce– ecep o ela ionships (Seibe and F ank,
2004), whe e he ela ions be ween each measu emen and
a po en ial sou ce o he emission is calcula ed (he e using
FLEXPART-WRF) and s o ed as he sou ce– ecep o ma ix
(SRM) o each e ical le el and o ou ash size bins (as
desc ibed in Sec . 3.3). The n=79 unknowns (sou ce ele-
men s) a e pu in o a s a e ec o xwhile he obse ed alues
ma e pu in o a ec o yo, whe e he subsc ip “o” s ands
o he PollyXT lida obse a ions. Then, he s a e ec o can
be calcula ed om he in e sion o a o wa d model M ha
connec s yoand x, as ollows:
yo=M(x)+ey,(4)
implying a linea ela ionship in which yois a ec o o spa-
io empo al lida measu emen s; Mis he n×mSRM calcu-
la ed by FLEXPART-WRF, desc ibing he sensi i i y o each
obse a ion o a uni elease a e; ey ep esen s lida measu e-
men e o s, which a e no accoun ed o in he algo i hm;
and xis he ash emission ec o o be es ima ed. M(x) is
equi alen o unning a VATDM wi h xas he inpu elease
p o ile. Since Mis calcula ed using such a model, i inhe -
i s he biases ha a e ine i able in VATDMs. As a esul , i
may di e ge om he ue dispe sion and may no necessa ily
align wi h he obse a ions on he le -hand side o Eq. (4),
e en i i is he ue elease p o ile (Fang e al., 2022). Gi en
ha he p oblem is unde de e mined, he solu ion o he lin-
ea in e se p oblem in Eq. (4) is no s aigh o wa d, and u -
he assump ions a e needed.
The mos common a e assump ions imposed on he un-
known emission ec o xsuch as he non-nega i i y o i s
elemen s, smoo hness o he emission (Fang e al., 2022),
o measu emen /emission spa si y (Li e al., 2018) (e.g., he
assump ion ha he emission elemen emains ze o unless
o he e idence is p esen in measu ed and modeled da a). Un-
de hese assump ions, he p oblem in Eq. (4) can be sol ed
by minimizing he dis ance be ween he le and he igh
sides o he equa ion. To enhance he s abili y o he in e -
sion ou come, a p io i emissions a e also used, ep esen ing
ou bes es ima e o xbe o e he obse a ions a e made (see
Sec . 3.4.1). Including an explici a p io i sou ce ec o xa,
we can exp ess he equa ion as ollows:
M(x−xa)≈yo−Mxa,(5)
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7351
and as an abb e ia ion
M˜
x≈˜
y.(6)
The in e sion scheme p esen ed he e is done by minimiz-
ing a cos unc ion C, which comp ises he ollowing sys em
o equa ions:
C1=yo−MTx(7)
C2=x−xa(8)
C3=Dx.(9)
C1quan i ies he di e ence be ween he modeled da a and
he obse a ions, C2is he de ia ion om he a p io i es ima-
ions, and C3imposes a smoo hness egula iza ion e m.
The cos unc ion C i s calcula es he mis i C1be ween
he p o iles a he ecep o poin s, as obse ed by he lida
(yo) and he da a as modeled by FLEXPART (MTx).
The second e m C2(Eq. 8) accoun s o he di e ence be-
ween he a pos e io i es ima es o he emission a es xand
he a p io i es ima es xa. (Fo de ails on he calcula ion o
he a p io i ec o , see Sec . 3.4.1.) To en o ce smoo hness in
he e ical p o ile o emissions, a egula iza ion pa ame e
C3is in oduced, de i ed om a disc e e second-o de di -
e ence ope a o D(Eq. 9). D ep esen s a idiagonal ma ix
whe e he main diagonal elemen s a e equal o −2, elemen s
o he diagonals abo e and below a e equal o 1 (disc e e ep-
esen a ion o he second de i a i e), and is a egula iza ion
pa ame e ha de e mines he weigh o his smoo hness con-
s ain ela i e o he o he wo e ms.
The inal mass emission a es a e ob ained by minimizing
he o al cos unc ion Cusing a s anda d op imiza ion ou-
ine wi h he a p io i emission a es as he ini ial guess. This
app oach ensu es ha he calcula ed ash emission a es a e
consis en wi h bo h he obse ed da a and he a p io i emis-
sion es ima es, while also a o ing a smoo h e ical dis i-
bu ion o emissions.
The in e sion scheme p esen ed in his s udy is no lim-
i ed o M . E na bu can be applied o o he olcanic e up-
ions wo ldwide, p o ided ha sui able obse a ional da a
a e a ailable. The me hodology elies on g ound-based li-
da measu emen s, dispe sion modeling (FLEXPART-WRF),
and an in e sion algo i hm o es ima e olcanic ash emis-
sions. The e o e, i can be adap ed o di e en olcanic se -
ings whe e lida obse a ions o o he emo e sensing da a
like sa elli e-based lida s (CALIPSO, Ea hCARE) and geo-
s a iona y sa elli es (SEVIRI) a e a ailable o cons ain he
sou ce e m. Addi ionally, he app oach can be ex ended o
a egional o global scale by in eg a ing mul iple obse a-
ion si es om lida ne wo ks such as ACTRIS/EARLINET
o inco po a ing addi ional sa elli e da a. This would allow
o imp o ed ash emission es ima es o a ious olcanic
e up ions wo ldwide. Fu he mo e, he use o high- esolu ion
wind ield da a (such as Aeolus o u u e wind lida missions)
can enhance he accu acy o dispe sion o ecas s in di e en
geog aphic egions (possibly lacking su icien in o ma ion
om adiosondes), making he me hodology widely applica-
ble o olcanic ash moni o ing and o ecas ing.
3.4.1 A p io i sou ce emissions xa
To cons ain he a iabili y o he e ie ed pa ame e s and o
enhance he s abili y o he in e sion ou come, a p io i emis-
sions a e also used in he in e sion scheme. We de e mine he
a p io i MER o ash pa icles ollowing he app oach ou -
lined by Scollo e al. (2019) by in e ing he obse ed plume
heigh s o e he E na summi c a e om VONA epo s and
ield obse a ions as obse ed by he INGV obse a o y, us-
ing he 1D plume model (Deg uy e and Bonadonna, 2012)
as desc ibed in Sec . 3.3. Addi ionally, he London VAAC
employs he same empi ical ela ionship be ween obse ed
plume heigh s and e up i e mass, as p oposed by Mas in
e al. (2009), assuming a uni o m e ical ash dis ibu ion.
The column heigh s o he ash plume om 12 Ma ch
2021 we e ob ained om he ECV calib a ed came a op-
e a ed by INGV-OE (Cal a i e al., 2021; Co adini e al.,
2018; Scollo e al., 2019) du ing he ime pe iod o 06:30 o
10:30UTC (see Table A1). The ash plume heigh eached
up o 9.0kma.s.l.In o de o calcula e he a p io i emis-
sions, he da a we e esampled a ∼2h in e als, speci i-
cally a 06:00 ( om 06:30 o 07:45UTC), 08:00 ( om 08:00
o 09:45UTC), and 10:00 UTC ( om 10:00 o 10:30 UTC).
Du ing he ini ial hou s o he e up ion (06:30–07:45UTC),
he ash plume was weak (Fig. 1a and Table A1) wi h an
a e age column injec ion heigh o 5.8km, esul ing in an
es ima ed MER o app oxima ely 12000 kgs−1acco ding
o he equa ion by Mas in e al. (2009) (Table 3). A e
07:45UTC, a s onge plume o med, ex ending e ically
abo e he en (Fig. 1b and Table A1). The ash plume ex-
ceeded he ECV came a ield o iew (e.g., mo e han 9.0–
9.5kma.s.l.) and was pa icula ly s ong be ween 08:00 and
09:45UTC (Fig. 1c and Table A1). The MER du ing his pe-
iod a e aged 58800kgs−1, wi h a mean plume heigh o
10kma.s.l.The s anda d de ia ion o he mean MER in-
dica es conside able inconsis ency in he emissions, as he
MER can change apidly du ing an e up ion due o luc-
ua ions in he e up i e dynamics, such as he collapse o
he e up ion column (Table 3). The ash plume heigh be-
gan o dec ease se e al minu es a e he la a oun ain ceased
(Fig. 1d and Table A1), wi h i s disappea ance becoming e -
iden only a e 10:15UTC (Fig. 1d). The MER du ing his
phase (10:00–10:30UTC) was app oxima ely 6300 kgs−1
(Table 3). The maximum plume ele a ion was no eco ded
by he ECV came a due o i s limi ed ield o iew (app ox.
9.0–9.5kma.s.l.as no ed by Scollo e al., 2014). Howe e ,
acco ding o SEVIRI aboa d he geos a iona y Me eosa Sec-
ond Gene a ion sa elli e, he olcanic ash cloud op heigh
(ACTH) be ween 08:15 and 08:45UTC was es ima ed a
11.5kma.s.l.(Cal a i e al., 2021). This highe SEVIRI-
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7358 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
5 Conclusions and discussion
The p esen s udy p esen ed an in e sion me hod o es ima e
he olcanic emission a e p o ile wi h a Lag angian pa icle
dispe sion model and a g ound-based lida sys em. The ech-
nique was applied o he case s udy o he explosi e e up ion
o M . E na, I aly, on 12 Ma ch 2021. To assess he impac
o Aeolus wind assimila ion in olcanic ash dispe sion o e-
cas s, he simula ion was epea ed wice: once wi h Aeolus
da a assimila ed (“w” expe imen ) and once wi hou (“w/o”).
The olcanic ae osol laye s obse ed abo e he PANGEA-
NOA s a ion in An iky he a, along wi h he clea -sky condi-
ions in he days a e he e up ion, made his an ideal es
case. Impo an conclusions om ou wo k a e as ollows:
The PollyXT lida sys em o PANGEA-NOA de ec ed a
dense ae osol laye be ween 8 and 12km, wi h he olcanic
ash plume p ima ily concen a ed be ween 9 and 11km.
FLEXPART simula ions, bo h a p io i (wi h an empi ical
emission p o ile) and a pos e io i (wi h he emission p o ile
p oduced by he in e sion algo i hm) we e conduc ed o de-
i e he modeled plume’s e ical dis ibu ion and concen a-
ion. The a p io i “w” Aeolus simula ion showed a b oade
dispe sion o he ash plume, po en ially due o he o e es-
ima ion o he a p io i ash emissions ob ained by in e ing
he obse ed plume heigh s om he VONA epo s, whe eas
he a pos e io i simula ion, based on he in e sion esul s,
p oduced a mo e e ined and consis en ash plume p o ile,
con ined o a smalle a ea, mos ly a ound An iky he a and
sou he n G eece, which was closely simila o he ash cloud
obse ed by he SEVIRI sa elli e.
In e ms o ash mass concen a ion, he a p io i p o ile
wi h Aeolus wind da a assimila ed shows a good spa io em-
po al ag eemen wi h he lida e ie als bu exhibi ed a sligh
e ical shi o 1km wi h espec o he obse ed ash mass
peaks (Ami idis e al., 2023) along wi h a mis i in mass
concen a ions o abou 50µgm−3, a c i ical ac o o a i-
a ion sa e y. In con as , he a pos e io i ash mass concen-
a ions demons a e a be e ag eemen wi h he obse a-
ions abo e PANGEA-NOA when Aeolus winds a e assimi-
la ed. The maximum ash mass concen a ion is ound close o
255µgm−3a 9.8 km, closely ma ching he peak obse ed by
he lida , depic ing a minimal di e ence o he o de o 2 %
be ween he obse ed and he a pos e io i simula ed ash mass
concen a ions. In con as , he di e ence be ween he lida
obse a ions and he a p io i ash simula ions anged om
28% o 40 %. This consis ency highligh s he obus ness o
he new in e sion algo i hm and he signi ican imp o emen
in he e ical dis ibu ion and he ash mass concen a ion.
Howe e , addi ional independen da ase s, such as g ound-
based, sa elli e emo e sensing da a, o ai bo ne in si u mea-
su emen s along he plume’s ajec o y, would u he en-
hance he alida ion o his me hodology and should be con-
side ed in u u e s udies.
To u he assess he eliabili y o he e ie ed emissions,
a Mon e Ca lo e o p opaga ion analysis was conduc ed, in-
oducing no mally dis ibu ed pe u ba ions o he lida mea-
su emen s. Wi h his me hod, he s anda d de ia ion o he
e ie ed emissions a each heigh le el was es ima ed. The
esul s indica e ha he in e sion ou pu emained highly s a-
ble, wi h minimal a ia ion ac oss Mon e Ca lo ealiza ions,
sugges ing ha he single-s a ion obse a ional se up does
no in oduce signi ican unce ain y. To enhance he sensi-
i i y o he in e sion amewo k and p o ide a mo e com-
p ehensi e unce ain y assessmen , mul iple lida s a ions o
complemen a y emo e sensing echniques a e essen ial.
The accu acy o he FLEXPART a pos e io i simula ion is
highly dependen on he p ecision o he d i ing me eo olog-
ical ields (“w” Aeolus wind ields), as well as on olcano
sou ce pa ame e s such as he plume heigh and mass e up-
ion a es, which a e e ined h ough he in e sion p ocess (a
pos e io i MER).
The ad an ages o Aeolus wind assimila ion o global
NWP models ha e been well documen ed, pa icula ly by
Rennie e al. (2021), who demons a ed signi ican imp o e-
men s in wind ield ep esen a ion, especially in he op-
ics and Sou he n Hemisphe e. Fu he enhancemen s in wind
o ecas s we e obse ed in he s udy o Ami idis e al. (2023),
whe e egional NWP models bene i ed om Aeolus wind as-
simila ion. Ou case s udy alida es hese indings, showing
ha he assimila ion o Aeolus wind p o iles leads o a sig-
ni ican imp o emen in he es ima ion o olcanic emission
a es in he e ical dis ibu ion, op imizing he ag eemen
be ween lida obse a ions and a pos e io i model simula-
ion.
Real- ime applica ions, such as hose o VAACs, demand
a apid esponse o olcanic ash haza ds. Once he plume is
de ec ed and ini ial da a om lida sys ems become a ail-
able, he p esen ed me hod can quickly p o ide he neces-
sa y in o ma ion o calcula e he cu en and u u e posi ion
and ex en o he plume wi hin a ew hou s. This unde sco es
he impe a i e o high-quali y, apidly accessible da a, such
as ha p o ided by o ganized g ound-based lida ne wo ks
employing s anda dized algo i hms and p ocedu es, such as
hose used by EARLINET, a key componen o he ACTRIS
in as uc u e.
Howe e , hei applicabili y o he p oposed me hodol-
ogy depends on he ope a ion o a backsca e -depola iza ion
lida , which cons i u es he p ima y equi emen . In cases
whe e di ec measu emen s o essen ial pa ame e s, such as
lida a ios, a e una ailable, alues om he scien i ic li e -
a u e can be used. A mo e ad anced con igu a ion, inco po-
a ing Raman lida capabili ies, would enhance he accu acy
o e ie ed backsca e and lida a io coe icien s. Addi ion-
ally, o day ime measu emen s, a co-loca ed sun pho ome e
would acili a e he di ec es ima ion o he con e sion ac-
o s equi ed in he in e sion p ocess. Beyond g ound-based
applica ions, he me hodology is also applicable o space-
bo ne ae osol lida s, which p o ide e ical p o iles o he
backsca e coe icien and pa icle linea depola iza ion a-
io, bo h undamen al pa ame e s o he in e sion p ocess.
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7359
Fu he mo e, he me hodology p esen ed he ein can be ap-
plied o cu en o u u e sa elli e missions ha employ lida
measu emen s (e.g., he Ea hCARE mission). While passi e
sa elli es o e nea -global co e age o ash cloud measu e-
men s wi hin minu es o hou s, g ound-based o sa elli e lida
sys ems p o ide mo e accu a e di ec e ie als o he e ical
dis ibu ion wi hin he ash plume column.
Ou me hodology is b oadly applicable and e icien
enough o eal- ime implemen a ion. I can supply ash
o ecas ing models wi h an objec i ely de i ed quan i a i e
sou ce e m, leading o imp o ed o ecas s ha a e c i ical o
he a ia ion sec o . These enhanced o ecas s p o ide mo e
e ec i e eme gency esponses, ensu ing sa e and mo e e i-
cien ligh ope a ions du ing olcanic e up ions, while a he
same ime minimizing he isk o acciden s and he inancial
impac o ligh cancela ions.
Appendix A
The SRR o a size dis ibu ion o olcanic ash pa icles wi h
ou size bins (3, 5, 9, and 21µm diame e ), de i ed om
he FLEXPART model using he “w” Aeolus assimila ed
wind ields, indica e ha he olcanic emissions obse ed
abo e he PANGEA-NOA obse a o y ( ecep o – yaxis)
on 12 Ma ch 2021 ( om 18:30 o 21:30UTC) a he heigh
ange o 6–12km mos ly o igina e om elease heigh s be-
ween 5 and 11.5km abo e he E na olcano (sou ce – xaxis)
(Fig. A2a, c, and e). These sou ce elease heigh s a e con-
sis en wi h he obse ed emissions abo e he PANGEA-
NOA s a ion, pa icula ly when he pa icle elease ime was
06:00–08:00 and 08:00–10:00UTC. The sou ce heigh s o
he ine pa icles align well wi h he e up i e column heigh s,
as epo ed om he INGV-OE calib a ed came as (Fig. 1).
Addi ionally, he in e sion algo i hm was u ilized wi h he
FLEXPART SRR only o hese wo elease imes. In con-
as , he SRR using “w/o” Aeolus assimila ed wind ields
shows ha he olcanic pa icles a i ing abo e he PANGEA
s a ion a heigh s o 8–10km ( ecep o – yaxis) a e ew and
o igina e om elease heigh s o a ound 8–11km abo e E na
(sou ce – xaxis) and only when he pa icle elease ime was
04:00–06:00UTC (Fig. A2b, d, and ). This elease ime is
no accu a e, as he e up ion ac ually began a 06:00 UTC ac-
co ding o he VONA messages om INGV-EO.
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7360 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Figu e A1. Wo k low o he me hodology.
Figu e A2. Sou ce– ecep o sensi i i ies o he ine pa icles (3, 5, 9, and 21µm diame e ) “w” Aeolus assimila ed winds (a,c, and e) and
“w/o” Aeolus simula ion (b,d, and ). The ho izon al axis “x” depic s he pa icle elease heigh (km) abo e E na, and he e ical axis “y”
is he al i ude abo e PANGEA ha he emissions obse ed on 12 Ma ch 2021 (18:30 o 21:30UTC).
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7361
Table A1. Volcanic ash plume heigh s (m) du ing he e up ion ac i i y om 06:30 o 10:30UTC, as eco ded by he ECV came a ope a ed
by INGV-EO (second column) and adjus ed heigh s inco po a ing SEVIRI sa elli e obse a ions whe e applicable ( hi d column).
Time (UTC) Heigh (m) om ECV Heigh (m) inco po a ing
came a (INGV-EO) SEVIRI obse a ions
06:30 4000 4000
06:45 5500 5500
07:00 5500 5500
07:15 6000 6000
07:30 6500 6500
07:45 7000 7000
08:00 7500 7500
08:15 <9000 11500
08:30 <9000 11500
08:45 <9000 11500
09:00 <9000 10000
09:15 <9000 9500
09:30 <9000 9500
09:45 9000 9000
10:00 6500 6500
10:15 5000 5000
10:30 4500 4500
Figu e A3. Wind speed (ms−1) in WRF 18h o ecas s. Ho izon al winds (a) “w” Aeolus assimila ion, (b) “w/o” Aeolus assimila ion, and
(c) wind speed di e ences (“w” – “w/o” Aeolus assimila ion) a 100hPa (∼16km).
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7362 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Figu e A4. Wind speed (ms−1) in WRF 18h o ecas s. Ho izon al winds (a) “w” Aeolus assimila ion, (b) “w/o” Aeolus assimila ion, and
(c) wind speed di e ences (“w” – “w/o” Aeolus assimila ion) a 200hPa (∼12km).
Figu e A5. Wind speed (ms−1) in WRF 18h o ecas s. Ho izon al winds (a) “w” Aeolus assimila ion, (b) “w/o” Aeolus assimila ion, and
(c) wind speed di e ences (“w” – “w/o” Aeolus assimila ion) a 500hPa (∼5.5km).
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7363
Code a ailabili y. The in e sion algo i hm was w i en wi h
Py hon p og amming language e sion 3.12 (h ps://www.py hon.
o g/, las access: 27 June 2025) and can be ob ained om he
co-au ho Anna Kampou i ([email p o ec ed]) upon eques . The
WRF model code is publicly a ailable, has a digi al objec iden i ie
(h ps://doi.o g/10.5065/D6MK6B4K, UCAR, 2025; Skama ock e
al., 2019), and can be ob ained ia Gi Hub (h ps://gi hub.com/
w -model/WRF, las access: 27 June 2025). The FLEXPART-
WRF model code is publicly a ailable and can be ob ained om
h ps://gi .nilu.no/ lexpa / lexpa -w (Dingwell, 2025; B ioude e
al., 2013) inpu . The code used o da a p ocessing was w i en wi h
Py hon p og amming language e sion 3.12 (h ps://www.py hon.
o g/, las access: 2 July 2025) and can be ob ained ia Gi Hub
a h ps://gi hub.com/NOA-ReACT/Aeolus_Volcano_2023 (las ac-
cess: 2 July 2025; h ps://doi.o g/10.5281/zenodo.15805552, Kam-
pou i, 2025). The SCC algo i hm used he e is de eloped wi hin
he EARLINET/ACTRIS communi y, and i s sou ce code is no
publicly a ailable. Ins ead, i is used in e nally o p ocess aw li-
da signals in o quali y-assu ed Le el 2 op ical p ope y p o iles,
which a e made publicly a ailable ia he EARLINET da a po al
(h ps://da a.ea line .o g/, las access: 10 July 2025). The e ie als
and he ae osol lida op ical p ope ies a e a ailable om he co-
au ho Anna Giali aki ([email p o ec ed]) upon eques .
Da a a ailabili y. The Aeolus L2A wind da a can be down-
loaded om h ps://apps.ecmw .in /ma s-ca alogue/?class= d&
exp e =hk (ECMWF, 2021b, a). The lida da a om he
PollyXT sys em a he PANGEA-NOA s a ion a e a ailable
in ascii o ma h ough Zenodo unde he ollowing DOI:
h ps://doi.o g/10.5281/zenodo.15805552 (Kampou i, 2025).
These da a we e de i ed using he Single Calculus Chain
(SCC; h ps://scc.imaa.cn .i ) algo i hm (a signal-analysis ool
o aw lida da a p ocessing de eloped wi hin EARLINET
(h ps://www.ea line .o g/, las access: 3 July 2025) and AC-
TRIS (h ps://www.ac is.eu/, las access: 3 July 2025)). SCC
p oduc s (L2 op ical p ope y p o iles) a e publicly a ailable
h ough he EARLINET da a po al (h ps://gi hub.com/ac is-a es/
ac is-ea line -2024_annual_collec ion/ ee/main, ACTRIS
EARLINET, 2024). The WRF and FLEXPART-WRF models’
simula ion esul s a e also a ailable om he co-au ho Anna
Kampou i ([email p o ec ed]) upon eques .
Au ho con ibu ions. AK concep ualized he pape along wi h
VA, PZ, and S S. All au ho s w o e pa s o he pape co espond-
ing o hei wo k and espec i e esul s. AK pe o med he FLEX-
PART and WRF uns and he in e sion algo i hm wi h he suppo
o S S, PZ, and TG. AG and MT p o ided he PollyXT lida e-
ie als. SiS p o ided he INGV-OE came a ma e ial and syne gis-
ic da ase s o he analysis. MR p o ided he ECMWF-IFS da ase s
(“w” and “w/o” Aeolus assimila ion). All au ho s p o ided co ec-
ions and sugges ions o e en ually shape he esea ch, analysis, and
inal pape . AK supe ised and di ec ed he whole p ojec .
Compe ing in e es s. The con ac au ho has decla ed ha none
o he au ho s has any compe ing in e es s.
Disclaime . Publishe ’s no e: Cope nicus Publica ions emains
neu al wi h ega d o ju isdic ional claims made in he ex , pub-
lished maps, ins i u ional a ilia ions, o any o he geog aphical ep-
esen a ion in his pape . While Cope nicus Publica ions makes e -
e y e o o include app op ia e place names, he inal esponsibili y
lies wi h he au ho s.
Acknowledgemen s. Anna Kampou i and he au ho s a ilia ed
o he Na ional Obse a o y o A hens acknowledge he suppo
om he ollowing esea ch p ojec s: he PANORAMA p ojec ,
unded by he Eu opean Union (g an ag eemen 101182795); he
PANGEA4CalVal p ojec , unded by he Eu opean Union (g an
ag eemen 101079201); he Hellenic Founda ion o Resea ch
and Inno a ion (p ojec abb e ia ion: S a oFIRE, p ojec numbe :
3995); he e-shape p ojec , unde he Eu opean Union’s Ho izon
2020 esea ch and inno a ion p og am (g an ag eemen 820852);
he ACTRIS esea ch in as uc u e; and da a and se ices ob ained
om he PANhellenic GEophysical Obse a o y o An iky he a
(PANGEA) o NOA. Addi ionally, Anna Kampou i and Thanasis
Geo giou acknowledge suppo om ESA in he amewo k o he
“Enhancing Aeolus L2A o depola izing a ge s and impac on
ae osol esea ch and NWP” p ojec (4000139424/22/I-NS). Anna
Kampou i acknowledges he suppo om Ioanni Binie oglou, An-
oni Gkika, and Emmanouil P oes aki o hei in aluable assis ance
and insigh ul discussions h oughou he de elopmen o his wo k.
Addi ionally, he au ho s hank he wo anonymous e iewe s o
hei help ul commen s and sugges ions o imp o ing he quali y
o he pape .
Financial suppo . This esea ch has been suppo ed by he Hel-
lenic Founda ion o Resea ch and Inno a ion (P ojec Ac onym:
S a oFIRE, g an no. 3995).
Re iew s a emen . This pape was edi ed by S elios Kazadzis
and e iewed by wo anonymous e e ees.
Re e ences
ACTRIS: PollyXT lida da a a PANGEA s a ion, Na ional Ob-
se a o y o A hens [da a se ], h ps://polly. opos.de/calenda /
loca ion/38, las access: 3 July 2025.
ACTRIS EARLINET: ACTRIS EARLINET 2024 Annual Col-
lec ion – An iky he a (AKY): Ae osol Op ical and Mic o-
physical P ope ies, CNR-IMAA ARES Da a Cen e [da a
se ], h ps://gi hub.com/ac is-a es/ac is-ea line -2024_annual_
collec ion/ ee/main, (las access: 11 July 2025), 2024.
Ami idis, V., Kampou i, A., Gkikas, A., Misios, S., Giali aki, A.,
Ma inou, E., Rennie, M., Benede i, A., Solomos, S., Zanis,
P., Vasa dani, O., Ele he a os, K., Paschou, P., Geo giou, T.,
Scollo, S., Mona, L., Papagiannopoulos, N., Re sche , C., Pa -
inello, T., and S aume, A. G.: Aeolus winds impac on olcanic
ash ea ly wa ning sys ems o a ia ion, Sci. Rep.-UK, 13, 7531,
h ps://doi.o g/10.1038/s41598-023-34715-6, 2023.
Ansmann, A., Tesche, M., Sei e , P., G oß, S., F euden hale ,
V., Api uley, A., Wilson, K. M., Se iko , I., Linné, H.,
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7364 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Heinold, B., Hiebsch, A., Schnell, F., Schmid , J., Ma is, I.,
Wandinge , U., and Wiegne , M.: Ash and ine-mode pa i-
cle mass p o iles om EARLINET-AERONET obse a ions
o e cen al Eu ope a e he e up ions o he Eyja jallajökull
olcano in 2010, J. Geophys. Res.-A mos., 116, D00U02,
h ps://doi.o g/10.1029/2010JD015567, 2011a.
Ansmann, A., Pe zold, A., Kandle , K., Tegen, I., Wendisch,
M., Mülle , D., Weinzie l, B., Mülle , T., and Hein zen-
be g, J.: Saha an Mine al Dus Expe imen s SAMUM–1 and
SAMUM–2: wha ha e we lea ned?, Tellus B, 63, 403,
h ps://doi.o g/10.1111/j.1600-0889.2011.00555.x, 2011b.
Ansmann, A., Sei e , P., Tesche, M., and Wandinge , U.: P o-
iling o ine and coa se pa icle mass: case s udies o Saha-
an dus and Eyja jallajökull/G ims ö n olcanic plumes, A -
mos. Chem. Phys., 12, 9399–9415, h ps://doi.o g/10.5194/acp-
12-9399-2012, 2012.
Baa s, H., Sei e , P., Engelmann, R., and Wandinge , U.:
Ta ge ca ego iza ion o ae osol and clouds by con inu-
ous mul iwa eleng h-pola iza ion lida measu emen s, A mos.
Meas. Tech., 10, 3175–3201, h ps://doi.o g/10.5194/am -10-
3175-2017, 2017.
Becke , F., Rossi, E., De enish, B., Wi ham, C., and Bonadonna,
C.: Modelling he size dis ibu ion o agg ega ed ol-
canic ash and implica ions o ope a ional a mosphe ic dis-
pe sion modelling, A mos. Chem. Phys., 22, 3409–3431,
h ps://doi.o g/10.5194/acp-22-3409-2022, 2022.
B ioude, J., A nold, D., S ohl, A., Cassiani, M., Mo on, D., Seib-
e , P., Ange ine, W., E an, S., Dingwell, A., Fas , J. D., Eas e ,
R. C., Pisso, I., Bu kha , J., and Wo awa, G.: The Lag angian
pa icle dispe sion model FLEXPART-WRF e sion 3.1, Geosci.
Model De ., 6, 1889–1904, h ps://doi.o g/10.5194/gmd-6-1889-
2013, 2013.
Cal a i, S., Bonacco so, A., and Ganci, G.: Ana omy o a
Pa oxysmal La a Foun ain a E na Volcano: The Case o
he 12 Ma ch 2021, Episode, Remo e Sens.-Basel, 13, 3052,
h ps://doi.o g/10.3390/ s13153052, 2021.
Chen, F. and Dudhia, J.: Coupling and ad anced land su ace-
hyd ology model wi h he Penn S a e-NCAR MM5 modeling
sys em. Pa I: Model implemen a ion and sensi i i y, Mon.
Wea he Re ., 129, 569–585, 2001.
Cla isse, L., P a a, F., Lacou , J., Hu mans, D., Cle baux, C., and
Coheu , P.: A co ela ion me hod o olcanic ash de ec ion using
hype spec al in a ed measu emen s, Geophys. Res. Le ., 37,
2010GL044828, h ps://doi.o g/10.1029/2010GL044828, 2010.
Cla kson, R. and Simpson, H.: Maximising ai space use du -
ing olcanic e up ions: ma ching engine du abili y agains
ash cloud occu ence, in: NATO STO Specialis s’ Mee ing
on Impac o Volcanic Ash Clouds on Mili a y Ope a ions
(STO-MP-AVT-272), Melbou ne, Aus alia, 16–18 May
2017, NATO Science and Technology O ganiza ion, 15–
17, h ps://www. esea chga e.ne /publica ion/317617735_
Maximising_Ai space_Use_Du ing_Volcanic_E up ions_
Ma ching_Engine_Du abili y_agains _Ash_Cloud_Occu ence
(las access: 10 July 2025), 2017.
Co adini, S., Gue ie i, L., Lomba do, V., Me ucci, L., Musac-
chio, M., P es i ilippo, M., Scollo, S., Sil es i, M., Spa a, G.,
and S eli ano, D.: P oximal Moni o ing o he 2011–2015 E na
La a Foun ains Using MSG-SEVIRI Da a, Geosciences, 8, 140,
h ps://doi.o g/10.3390/geosciences8040140, 2018.
Dac e, H. F., G an , A. L. M., Hogan, R. J., Belche , S. E., Thom-
son, D. J., De enish, B. J., Ma enco, F., Ho , M. C., Haywood,
J. M., Ansmann, A., Ma is, I., and Cla isse, L.: E alua ing
he s uc u e and magni ude o he ash plume du ing he ini ial
phase o he 2010 Eyja jallajökull e up ion using lida obse a-
ions and NAME simula ions, J. Geophys. Res., 116, D00U03,
h ps://doi.o g/10.1029/2011JD015608, 2011.
D’Amico, G., Amodeo, A., Baa s, H., Binie oglou, I., F euden-
hale , V., Ma is, I., Wandinge , U., and Pappala do, G.:
EARLINET Single Calculus Chain – o e iew on me hod-
ology and s a egy, A mos. Meas. Tech., 8, 4891–4916,
h ps://doi.o g/10.5194/am -8-4891-2015, 2015.
Deg uy e , W. and Bonadonna, C.: Imp o ing on mass low
a e es ima es o olcanic e up ions, Geophys. Res. Le ., 39,
2012GL052566, h ps://doi.o g/10.1029/2012GL052566, 2012.
Dingwell, A.: FLEXPART-WRF code, NILU Gi Lab eposi-
o y [code], h ps://gi .nilu.no/ lexpa / lexpa -w , las access:
10 July 2025.
Dubo ik, O., Sinyuk, A., Lapyonok, T., Holben, B. N., Mishchenko,
M., Yang, P., Eck, T. F., Vol en, H., Muñoz, O., Veihelmann, B.,
Van De Zande, W. J., Leon, J., So okin, M., and Slu ske , I.: Ap-
plica ion o sphe oid models o accoun o ae osol pa icle non-
sphe ici y in emo e sensing o dese dus , J. Geophys. Res., 111,
2005JD006619, h ps://doi.o g/10.1029/2005JD006619, 2006.
Du an , A. J., Bonadonna, C., and Ho well, C. J.: A mosphe ic
and En i onmen al Impac s o Volcanic Pa icula es, Elemen s,
6, 235–240, h ps://doi.o g/10.2113/gselemen s.6.4.235, 2010.
Eckha d , S., P a a, A. J., Seibe , P., S ebel, K., and S ohl, A.: Es i-
ma ion o he e ical p o ile o sul u dioxide injec ion in o he
a mosphe e by a olcanic e up ion using sa elli e column mea-
su emen s and in e se anspo modeling, A mos. Chem. Phys.,
8, 3881–3897, h ps://doi.o g/10.5194/acp-8-3881-2008, 2008.
ECMWF: Synop ic-scale a iabili y in he Medi e anean, ECMWF
Technical Repo No. 60, Eu opean Cen e o Medium-Range
Wea he Fo ecas s, Reading, UK, h ps://www.ecmw .in /en/
elib a y/78648-synop ic-scale- a iabili y-medi e anean (las
access: 9 July 2025), 2010.
ECMWF: ECMWF s a s assimila ing Aeolus wind da a, Eu o-
pean Cen e o Medium-Range Wea he Fo ecas s, Reading,
UK, h ps://www.ecmw .in /en/abou /media-cen e/news/2020/
ecmw -s a s-assimila ing-aeolus-wind-da a (las access: 9 July
2025), 2021a.
ECMWF: Aeolus L2A wind da a, Eu opean Cen e o Medium-
Range Wea he Fo ecas s [da a se ], h ps://apps.ecmw .in /
ma s-ca alogue/?class= d&exp e =hk (las access: 3 July
2025), 2021b.
Engelmann, R., Kani z, T., Baa s, H., Heese, B., Al hausen, D.,
Skupin, A., Wandinge , U., Komppula, M., S achlewska, I. S.,
Ami idis, V., Ma inou, E., Ma is, I., Linné, H., and Ansmann,
A.: The au oma ed mul iwa eleng h Raman pola iza ion and
wa e - apo lida PollyXT: he neXT gene a ion, A mos. Meas.
Tech., 9, 1767–1784, h ps://doi.o g/10.5194/am -9-1767-2016,
2016.
EUROCONTROL: Fo ecas Upda e 2022–2028, Eu opean
Fligh Mo emen s and Se ice Uni s, EUROCONTROL,
B ussels, Belgium, h ps://www.eu ocon ol.in /publica ion/
eu ocon ol- o ecas -upda e-2022-2028 (las access: 9 July
2025), 2022.
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7365
Fang, S., Dong, X., Zhuang, S., Tian, Z., Chai, T., Xu, Y.,
Zhao, Y., Sheng, L., Ye, X., and Xiong, W.: Oscilla ion-
ee sou ce e m in e sion o a mosphe ic adionuclide
eleases wi h join model bias co ec ions and non-
smoo h compe ing p io s, J. Haza d. Ma e ., 440, 129806,
h ps://doi.o g/10.1016/j.jhazma .2022.129806, 2022.
Fea nley, C. J., Bi d, D. K., Haynes, K., McGui e, W. J., and
Jolly, G. (Eds.): Obse ing he Volcano Wo ld: Volcano C i-
sis Communica ion, Sp inge In e na ional Publishing, Cham,
h ps://doi.o g/10.1007/978-3-319-44097-2, 2018.
Flou si, A. A., Baa s, H., Engelmann, R., Al hausen, D., Ansmann,
A., Bohlmann, S., Heese, B., Ho e , J., Kani z, T., Haa ig, M.,
Ohneise , K., Radenz, M., Sei e , P., Skupin, A., Yin, Z., Abdul-
lae , S. F., Komppula, M., Filioglou, M., Giannakaki, E., S ach-
lewska, I. S., Janicka, L., Bo oli, D., Ma inou, E., Ami idis,
V., Giali aki, A., Mamou i, R.-E., Ba ja, B., and Wandinge , U.:
DeLiAn – a g owing collec ion o depola iza ion a io, lida a io
and Ångs öm exponen o di e en ae osol ypes and mix u es
om g ound-based lida obse a ions, A mos. Meas. Tech., 16,
2353–2379, h ps://doi.o g/10.5194/am -16-2353-2023, 2023.
F ancis, P. N., Cooke, M. C., and Saunde s, R. W.: Re ie al o
physical p ope ies o olcanic ash using Me eosa : A case s udy
om he 2010 Eyja jallajökull e up ion, J. Geophys. Res., 117,
2011JD016788, h ps://doi.o g/10.1029/2011JD016788, 2012.
Gas eige , J., G oß, S., F euden hale , V., and Wiegne , M.: Vol-
canic ash om Iceland o e Munich: mass concen a ion e-
ie ed om g ound-based emo e sensing measu emen s, A -
mos. Chem. Phys., 11, 2209–2223, h ps://doi.o g/10.5194/acp-
11-2209-2011, 2011.
Giannakaki, E., P ülle , A., Ko honen, K., Mielonen, T., Laakso,
L., Vakka i, V., Baa s, H., Engelmann, R., Beukes, J. P., Van
Zyl, P. G., Josipo ic, M., Tii a, P., Chiloane, K., Pike h, S., Li-
ha ainen, H., Leh inen, K. E. J., and Komppula, M.: One yea
o Raman lida obse a ions o ee- oposphe ic ae osol lay-
e s o e Sou h A ica, A mos. Chem. Phys., 15, 5429–5442,
h ps://doi.o g/10.5194/acp-15-5429-2015, 2015.
Giles, D. M., Sinyuk, A., So okin, M. G., Scha e , J. S., Smi no ,
A., Slu ske , I., Eck, T. F., Holben, B. N., Lewis, J. R., Campbell,
J. R., Wel on, E. J., Ko kin, S. V., and Lyapus in, A. I.: Ad ance-
men s in he Ae osol Robo ic Ne wo k (AERONET) Ve sion 3
da abase – au oma ed nea - eal- ime quali y con ol algo i hm
wi h imp o ed cloud sc eening o Sun pho ome e ae osol op-
ical dep h (AOD) measu emen s, A mos. Meas. Tech., 12, 169–
209, h ps://doi.o g/10.5194/am -12-169-2019, 2019.
Goloub, P., Li, Z., Dubo ik, O., Bla el, L., Pod in, T., Jankowiak,
I., Lecoq, R., De oo, C., Cha ene , B., Mo el, J. P., Cue as, E.,
and Ramos, R.: PHOTONS/AERONET sunpho ome e ne wo k
o e iew: desc ip ion, ac i i ies, esul s, SPIE P oc., 69360V-
69360V–15, h ps://doi.o g/10.1117/12.783171, 2007.
G oß, S., F euden hale , V., Wiegne , M., Gas eige , J., Geiß, A.,
and Schnell, F.: Dual-wa eleng h linea depola iza ion a io o
olcanic ae osols: Lida measu emen s o he Eyja jallajökull
plume o e Maisach, Ge many, A mos. En i on., 48, 85–96,
h ps://doi.o g/10.1016/j.a mosen .2011.06.017, 2012.
G oß, S., Esselbo n, M., Weinzie l, B., Wi h, M., Fix, A., and Pe -
zold, A.: Ae osol classi ica ion by ai bo ne high spec al eso-
lu ion lida obse a ions, A mos. Chem. Phys., 13, 2487–2505,
h ps://doi.o g/10.5194/acp-13-2487-2013, 2013.
Gue ie i, L., Co adini, S., Theys, N., S eli ano, D., and Me ucci,
L.: Volcanic Clouds Cha ac e iza ion o he 2020–2022 Se-
quence o M . E na La a Foun ains Using MSG-SEVIRI and
P oduc s’ C oss-Compa ison, Remo e Sens.-Basel, 15, 2055,
h ps://doi.o g/10.3390/ s15082055, 2023.
Gu an i, M., Ewe , J. W., Gallina, G. M., Blu h, G. J. S., and Swan-
son, G. L.: Volcanic-ash haza d o a ia ion du ing he 2003–2004
e up i e ac i i y o Ana ahan olcano, Commonweal h o he
No he n Ma iana Islands, J. Volcanol. Geo h. Res., 146, 241–
255, h ps://doi.o g/10.1016/j.j olgeo es.2004.12.011, 2005.
Ha ey, N. J., Dac e, H. F., Webs e , H. N., Taylo , I. A., Khanal, S.,
G ainge , R. G., and Cooke, M. C.: The Impac o Ensemble Me-
eo ology on In e se Modeling Es ima es o Volcano Emissions
and Ash Dispe sion Fo ecas s: G íms ö n 2011, A mosphe e, 11,
1022, h ps://doi.o g/10.3390/a mos11101022, 2020.
Hess, M., Koepke, P., and Schul , I.: Op ical P ope ies o
Ae osols and Clouds: The So wa e Package OPAC, B. Am.
Me eo ol. Soc., 79, 831–844, h ps://doi.o g/10.1175/1520-
0477(1998)079<0831:OPOAAC>2.0.CO;2, 1998.
Houchi, K., S o elen, A., Ma seille, G. J., and De Kloe,
J.: Compa ison o wind and wind shea clima olo-
gies de i ed om high- esolu ion adiosondes and he
ECMWF model, J. Geophys. Res., 115, 2009JD013196,
h ps://doi.o g/10.1029/2009JD013196, 2010.
Iacono, M. J., Delame e, J. S, Mlawe , E. J, Shepha d, M.
W., Clough, S. A, and Collins, W. D.: Radia i e o cing by
long-li ed g eenhouse gases: Calcula ions wi h he AER adia-
i e ans e models, J. Geophys. Res.-A mos., 113, D13103,
h ps://doi.o g/10.1029/2008JD009944, 2008.
ICAO: No h A lan ic Sys ems Planning G oup – Fi y-second
mee ing epo (NAT SPG/52), 20–24 June 2016, Pa is,
F ance, Eu opean and No h A lan ic (EUR/NAT) O ice, In-
e na ional Ci il A ia ion O ganiza ion, Mon eal, Canada,
h ps://www.icao.in /EURNAT/EUR%20and%20NAT%
20Documen s/NAT%20Documen s/NAT%20SPG%20Repo s/
NAT%20SPG_52%20(2016)%20Repo .pd (las access: 9 July
2025), 2016.
INGV: Osse a o io E neo, Is i u o Nazionale di Geo isica e Vul-
canologia (INGV), h ps://www.c .ing .i /, las access: 2 July
2025.
Janjic, Z. I.: Nonsingula implemen a ion o he Mello -Yamada
le el 2.5 scheme in he NCEP Meso model, NCEP O -
ice No e 437, Na ional Cen e s o En i onmen al P edic ion
(NCEP), Camp Sp ings,61 pp., h ps:// eposi o y.lib a y.noaa.
go / iew/noaa/11409 (las access: 3 July 2025), 2002.
Janjic, Z. I.: A nonhyd os a ic model based on a new
app oach, Me eo ol. A mos. Phys., 82, 271–285,
h ps://doi.o g/10.1007/s00703-001-0587-6, 2003.
Jones, A., Thomson, D., Ho , M., and De enish, B.: The U. K.
Me O ice’s Nex -Gene a ion A mosphe ic Dispe sion Model,
NAME III, in: Ai Pollu ion Modeling and I s Applica ion
XVII, Bos on, MA, 580–589, h ps://doi.o g/10.1007/978-0-387-
68854-1_62, 2007.
Kampou i, A.: akampou i/Volcanic-emission-
es ima es_Kampou i_e _al.2025: Volcanic emission es ima es
om he in e sion o ACTRIS lida obse a ions and hei use
o quan i a i e dispe sion modeling ( 1.0.0), Zenodo [code and
da a se ], h ps://doi.o g/10.5281/zenodo.15805552, 2025.
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
7366 A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions
Kampou i, A., Ami idis, V., Solomos, S., Giali aki, A., Ma inou, E.,
Spy ou, C., Geo goulias, A. K., Ak i idis, D., Papagiannopoulos,
N., Mona, L., Scollo, S., Tsichla, M., Tsikoudi, I., Py ha oulis, I.,
Ka acos as, T., and Zanis, P.: In es iga ion o Volcanic Emissions
in he Medi e anean: “The E na–An iky he a Connec ion”, A -
mosphe e, 12, 40, h ps://doi.o g/10.3390/a mos12010040, 2020.
Kampou i, A., Ami idis, V., Geo giou, T., Solomos, S., Binie oglou,
I., Giali aki, A., Ma inou, E., Gkikas, A., P oes akis, E., Ren-
nie, M., Benede i, A., Scollo, S., Mona, L., Papagiannopoulos,
N., and Zanis, P.: In e sion Techniques on E na’s Volcanic Emis-
sions and he Impac o Aeolus on Quan i a i e Dispe sion Mod-
eling, in: 16 h In e na ional Con e ence on Me eo ology, Cli-
ma ology and A mosphe ic Physics – COMECAP 2023, In e -
na ional Con e ence on Me eo ology, Clima ology and A mo-
sphe ic Physics, 12–16 June 2023, Thessaloniki, G eece, 187,
h ps://doi.o g/10.3390/en i onscip oc2023026187, 2023.
Koepke, P., Gas eige , J., and Hess, M.: Technical No e: Op ical
p ope ies o dese ae osol wi h non-sphe ical mine al pa i-
cles: da a inco po a ed o OPAC, A mos. Chem. Phys., 15, 5947–
5956, h ps://doi.o g/10.5194/acp-15-5947-2015, 2015.
Kons a, D., Tseke i, A., Solomos, S., Siomos, N., Giali aki, A.,
Te oni, E., Lopa in, A., Goloub, P., Dubo ik, O., Ami idis,
V., and Nas os, P.: The Po en ial o GRASP/GARRLiC Re-
ie als o Dus Ae osol Model E alua ion: Case S udy du -
ing he P eTECT Campaign, Remo e Sens.-Basel, 13, 873,
h ps://doi.o g/10.3390/ s13050873, 2021.
K is iansen, N. I., S ohl, A., P a a, A. J., Rich e , A., Eckha d , S.,
Seibe , P., Ho mann, A., Ri e , C., Bi a , L., Duck, T. J., and
S ebel, K.: Remo e sensing and in e se anspo modeling o he
Kasa ochi e up ion sul u dioxide cloud, J. Geophys. Res., 115,
2009JD013286, h ps://doi.o g/10.1029/2009JD013286, 2010.
K is iansen, N. I., S ohl, A., P a a, A. J., Bukowiecki, N., Dac e, H.,
Eckha d , S., Henne, S., Ho , M. C., Johnson, B. T., Ma enco, F.,
Neininge , B., Rei ebuch, O., Seibe , P., Thomson, D. J., Web-
s e , H. N., and Weinzie l, B.: Pe o mance assessmen o a ol-
canic ash anspo model mini-ensemble used o in e se model-
ing o he 2010 Eyja jallajökull e up ion, J. Geophys. Res., 117,
2011JD016844, h ps://doi.o g/10.1029/2011JD016844, 2012.
K is iansen, N. I., A nold, D., and S ohl, A.: P ocedu es o ol-
cano in e sions using FLEXPART, No wegian Ins i u e o Ai
Resea ch (NILU) VAST Repo , Kjelle , No way, h ps://nilu.
com/wp-con en /uploads/dnn/KST-VAST_ inal_ epo .pd (las
access: 3 July 2025), 2014.
Lensky, I. M., Dayan, U., and Helman, D.: Synop ic Ci -
cula ion Impac on he Nea -Su ace Tempe a u e Di e -
ence Ou weighs Tha o he Seasonal Signal in he Eas e n
Medi e anean, J. Geophys. Res.-A mos., 123, 11,333-11,347,
h ps://doi.o g/10.1029/2017JD027973, 2018.
Li, X., Li, H., Liu, Y., Xiong, W., and Fang, S.: Join elease a e
es ima ion and measu emen -by-measu emen model co ec ion
o a mosphe ic adionuclide emission in nuclea acciden s: An
applica ion o wind unnel expe imen s, J. Haza d. Ma e ., 345,
48–62, h ps://doi.o g/10.1016/j.jhazma .2017.09.051, 2018.
Lopa in, A., Dubo ik, O., Chaiko sky, A., Goloub, P., Lapyonok,
T., Tan é, D., and Li ino , P.: Enhancemen o ae osol cha ac-
e iza ion using syne gy o lida and sun-pho ome e coinciden
obse a ions: he GARRLiC algo i hm, A mos. Meas. Tech., 6,
2065–2088, h ps://doi.o g/10.5194/am -6-2065-2013, 2013.
Lopa in, A., Dubo ik, O., Fue es, D., S enchiko , G., Lapyonok,
T., Veselo skii, I., Wienhold, F. G., She chenko, I., Hu, Q.,
and Pa ajuli, S.: Syne gy p ocessing o di e se g ound-based e-
mo e sensing and in si u da a using he GRASP algo i hm: ap-
plica ions o adiome e , lida and adiosonde obse a ions, A -
mos. Meas. Tech., 14, 2575–2614, h ps://doi.o g/10.5194/am -
14-2575-2021, 2021.
Mamou i, R.-E. and Ansmann, A.: Po en ial o pola iza ion/Raman
lida o sepa a e ine dus , coa se dus , ma i ime, and an h o-
pogenic ae osol p o iles, A mos. Meas. Tech., 10, 3403–3427,
h ps://doi.o g/10.5194/am -10-3403-2017, 2017.
Mas in, L. G., Gu an i, M., Se anckx, R., Webley, P., Ba -
so i, S., Dean, K., Du an , A., Ewe , J. W., Ne i, A.,
Rose, W. I., Schneide , D., Siebe , L., S unde , B., Swan-
son, G., Tuppe , A., Volen ik, A., and Way homas, C. F.:
A mul idisciplina y e o o assign ealis ic sou ce pa am-
e e s o models o olcanic ash-cloud anspo and dispe -
sion du ing e up ions, J. Volcanol. Geo h. Res., 186, 10–21,
h ps://doi.o g/10.1016/j.j olgeo es.2009.01.008, 2009.
Mazzocchi, M., Hanss ein, F., and Ragona, M.: The 2010 olcanic
ash cloud and i s inancial impac on he Eu opean ai line in-
dus y, CESi o Fo um, 11, 92–100, h ps://hdl.handle.ne /10419/
166397 (las access: 9 July 2025), 2010.
Mülle , D., Ansmann, A., Ma is, I., Tesche, M., Wandinge , U.,
Al hausen, D., and Pisani, G.: Ae osol- ype-dependen lida a-
ios obse ed wi h Raman lida , J. Geophys. Res.-A mos., 112,
D16202, h ps://doi.o g/10.1029/2006JD008292, 2007.
Näslund, E. and Thaning, L.: On he Se ling Veloci y in a
Nons a iona y A mosphe e, Ae osol Sci. Tech., 14, 247–256,
h ps://doi.o g/10.1080/02786829108959487, 1991.
Ox o d Economics: The economic impac s o ai a el es ic ions
due o olcanic ash, p epa ed o Ai bus, Ox o d Economics
L d, Ox o d, UK, h ps://con o e ses.minespa is.psl.eu/public/
p omo10/p omo10_G11/da a/documen s/Volcanic-Upda e.pd
(las access: 3 July 2025), 2012.
Pappala do, G., Amodeo, A., Mona, L., Pandol i, M., Pe gola,
N., and Cuomo, V.: Raman lida obse a ions o ae osol emi -
ed du ing he 2002 E na e up ion, Geophys. Res. Le ., 31,
2003GL019073, h ps://doi.o g/10.1029/2003GL019073, 2004.
Pa olonis, M. J., Heidinge , A. K., and Siegla , J.: Au oma ed e-
ie als o olcanic ash and dus cloud p ope ies om upwelling
in a ed measu emen s, J. Geophys. Res.-A mos., 118, 1436–
1458, h ps://doi.o g/10.1002/jg d.50173, 2013.
Pe e sen, G. N., Bjo nsson, H., A ason, P., and on Löwis, S.: Two
wea he ada ime se ies o he al i ude o he olcanic plume
du ing he May 2011 e up ion o G íms ö n, Iceland, Ea h Sys .
Sci. Da a, 4, 121–127, h ps://doi.o g/10.5194/essd-4-121-2012,
2012.
Pisso, I., Sollum, E., G y he, H., K is iansen, N. I., Cas-
siani, M., Eckha d , S., A nold, D., Mo on, D., Thomp-
son, R. L., G oo Zwaa ink, C. D., E angeliou, N., Sode-
mann, H., Haimbe ge , L., Henne, S., B unne , D., Bu kha ,
J. F., Fouilloux, A., B ioude, J., Philipp, A., Seibe , P., and
S ohl, A.: The Lag angian pa icle dispe sion model FLEX-
PART e sion 10.4, Geosci. Model De ., 12, 4955–4997,
h ps://doi.o g/10.5194/gmd-12-4955-2019, 2019.
Pouge , S., Bu sik, M., Webley, P., Dehn, J., and Pa olonis, M.:
Es ima ion o e up ion sou ce pa ame e s om umb ella cloud
o downwind plume g ow h a e, J. Volcanol. Geo h. Res.,
A mos. Chem. Phys., 25, 7343–7368, 2025 h ps://doi.o g/10.5194/acp-25-7343-2025
A. Kampou i e al.: Volcanic emission es ima es om he in e sion o ACTRIS lida obse a ions 7367
258, 100–112, h ps://doi.o g/10.1016/j.j olgeo es.2013.04.002,
2013.
P a a, A. J.: Sa elli e de ec ion o haza dous olcanic clouds
and he isk o global ai a ic, Na . Haza ds, 51, 303–324,
h ps://doi.o g/10.1007/s11069-008-9273-z, 2009.
P a a, A. J. and P a a, A. T.: Eyja jallajökull olcanic ash concen-
a ions de e mined using Spin Enhanced Visible and In a ed
Image measu emen s, J. Geophys. Res., 117, 2011JD016800,
h ps://doi.o g/10.1029/2011JD016800, 2012.
P a a, A. T., G ainge , R. G., Taylo , I. A., Po ey, A. C., P oud,
S. R., and Poulsen, C. A.: Unce ain y-bounded es ima es o ash
cloud p ope ies using he ORAC algo i hm: applica ion o he
2019 Raikoke e up ion, A mos. Meas. Tech., 15, 5985–6010,
h ps://doi.o g/10.5194/am -15-5985-2022, 2022.
P a a, F. and Lynch, M.: Passi e Ea h Obse a ions o Vol-
canic Clouds in he A mosphe e, A mosphe e, 10, 199,
h ps://doi.o g/10.3390/a mos10040199, 2019.
Rennie, M. P., Isaksen, L., Weile , F., De Kloe, J., Kani z, T., and
Rei ebuch, O.: The impac o Aeolus wind e ie als on ECMWF
global wea he o ecas s, Q. J. Roy. Me eo . Soc., 147, 3555–
3586, h ps://doi.o g/10.1002/qj.4142, 2021.
Robock, A.: Volcanic e up ions and clima e, Re . Geophys., 38,
191–219, h ps://doi.o g/10.1029/1998RG000054, 2000.
Rousi, E., Mimis, A., S amou, M., and Anagnos opoulou, C.: Clas-
si ica ion o ci cula ion ypes o e Eas e n medi e anean us-
ing a sel -o ganizing map app oach, J. Maps, 10, 232–237,
h ps://doi.o g/10.1080/17445647.2013.862747, 2014.
Scollo, S., Col elli, M., Bonadonna, C., and Del Ca lo, P.:
Teph a haza d assessmen a M . E na (I aly), Na . Haza ds
Ea h Sys . Sci., 13, 3221–3233, h ps://doi.o g/10.5194/nhess-
13-3221-2013, 2013.
Scollo, S., P es i ilippo, M., Peco a, E., Co adini, S., Me ucci, L.,
Spa a, G., and Col elli, M.: E up ion column heigh es ima ion
o he 2011–2013 E na la a oun ains, Ann. Geophys., 57, 3,
h ps://doi.o g/10.4401/ag-6396, 2014.
Scollo, S., P es i ilippo, M., Bonadonna, C., Cioni, R., Co adini, S.,
Deg uy e , W., Rossi, E., Sil es i, M., Biale, E., Ca pa elli, G.,
Cassisi, C., Me ucci, L., Musacchio, M., and Peco a, E.: Nea -
Real-Time Teph a Fallou Assessmen a M . E na, I aly, Re-
mo e Sens.-Basel, 11, 2987, h ps://doi.o g/10.3390/ s11242987,
2019.
Seibe , P. and F ank, A.: Sou ce- ecep o ma ix calcula ion wi h
a Lag angian pa icle dispe sion model in backwa d mode, A -
mos. Chem. Phys., 4, 51–63, h ps://doi.o g/10.5194/acp-4-51-
2004, 2004.
Seibe , P., K is iansen, N. I., Rich e , A., Eckha d , S., P a a, A. J.,
and S ohl, A.: Unce ain ies in he in e se modelling o sulphu
dioxide e up ion p o iles, Geoma ics, Na u al Haza ds and Risk,
2, 201–216, h ps://doi.o g/10.1080/19475705.2011.590533,
2011.
Sica d, M., Ba on, A., Ranai ombola, M., Gan ois, D., Mille , T.,
Selli o, P., Bègue, N., Benche i , H., Payen, G., Ma ques au ,
N., and Du lo , V.: Radia i e impac o he Hunga s a osphe ic
olcanic plume: ole o ae osols and wa e apo o e Réu-
nion Island (21°S, 55° E), A mos. Chem. Phys., 25, 367–381,
h ps://doi.o g/10.5194/acp-25-367-2025, 2025.
Skama ock, W. C., Klemp, J. B., Dudhia, J., Gill, D. O., Ba ke ,
D. M., Duda, M. G., Huang, X.-Y., Wang, W., and Powe s,
J. G.: A desc ip ion o he Ad anced Resea ch WRF Ve sion
4, NCAR Technical No e NCAR/TN-556+STR, Na ional Cen-
e o A mosphe ic Resea ch (NCAR), Boulde , CO, 145 pp.,
h ps://doi.o g/10.5065/1d h-6p97, 2019.
S ein, A. F., D axle , R. R., Rolph, G. D., S unde , B. J. B., Co-
hen, M. D., and Ngan, F.: NOAA’s HYSPLIT A mosphe ic
T anspo and Dispe sion Modeling Sys em, B. Am. Me eo-
ol. Soc., 96, 2059–2077, h ps://doi.o g/10.1175/BAMS-D-14-
00110.1, 2015.
S o elen, A., Ma seille, G. J., Bou ie , F., Vasilje ic, D., De Haan,
S., and Ca dinali, C.: ADM-Aeolus Dopple wind lida Obse -
ing Sys em Simula ion Expe imen , Q. J. Roy. Me eo . Soc., 132,
1927–1947, h ps://doi.o g/10.1256/qj.05.83, 2006.
S o elen, A., Benede i, A., Bo de, R., Dabas, A., Flaman , P.,
Fo sy he, M., Ha des y, M., Isaksen, L., Källén, E., Kö nich, H.,
Lee, T., Rei ebuch, O., Rennie, M., Riishøjgaa d, L.-P., Schy-
be g, H., S aume, A. G., and Vaughan, M.: Wind P o ile Sa elli e
Obse a ion Requi emen s and Capabili ies, B. Am. Me eo ol.
Soc., 101, E2005–E2021, h ps://doi.o g/10.1175/BAMS-D-18-
0202.1, 2020.
S ohl, A., Fo s e , C., F ank, A., Seibe , P., and Wo awa, G.:
Technical no e: The Lag angian pa icle dispe sion model
FLEXPART e sion 6.2, A mos. Chem. Phys., 5, 2461–2474,
h ps://doi.o g/10.5194/acp-5-2461-2005, 2005.
S ohl, A., P a a, A. J., Eckha d , S., Cla isse, L., Du an , A.,
Henne, S., K is iansen, N. I., Minikin, A., Schumann, U., Seib-
e , P., S ebel, K., Thomas, H. E., Tho s einsson, T., Tø se h, K.,
and Weinzie l, B.: De e mina ion o ime- and heigh - esol ed
olcanic ash emissions and hei use o quan i a i e ash dis-
pe sion modeling: he 2010 Eyja jallajökull e up ion, A mos.
Chem. Phys., 11, 4333–4351, h ps://doi.o g/10.5194/acp-11-
4333-2011, 2011.
S aume-Lindne , A. G., Pa inello, T., Von Bisma ck, J., Bley,
S., We nham, D., Kani z, T., Al a ez, E., Fischey, P., De
Lau en is, M., Feh , T., Ehle s, F., Duc T an, V., K isch,
I., Rei ebuch, O., and Renni, M.: ESA’S Wind Mission Ae-
olus – O e iew, S a us and Ou look, in: IGARSS 2021–
2021 IEEE In e na ional Geoscience and Remo e Sensing
Symposium, B ussels, Belgium, 11–16 July 2021, 755–758,
h ps://doi.o g/10.1109/IGARSS47720.2021.9554007, 2021.
Tacke , J. L., Ka , J., Vaughan, M. A., Ge zewich, B. J., Kim,
M.-H., Ve nie , J.-P., Oma , A. H., Magill, B. E., Pi s, M. C.,
and Winke , D. M.: The CALIPSO e sion 4.5 s a osphe ic
ae osol sub yping algo i hm, A mos. Meas. Tech., 16, 745–768,
h ps://doi.o g/10.5194/am -16-745-2023, 2023.
Thompson, G., Field, P. R., Rasmussen, R. M., and Hall, W. D.:
Explici Fo ecas s o Win e P ecipi a ion Using an Imp o ed
Bulk Mic ophysics Scheme. Pa II: Implemen a ion o a New
Snow Pa ame e iza ion, Mon. Wea he Re ., 136, 5095–5115,
h ps://doi.o g/10.1175/2008MWR2387.1, 2008.
UCAR: Wea he Resea ch & Fo ecas ing Model (WRF), UCAR
[code], h ps://doi.o g/10.5065/D6MK6B4K, 2025.
Wagne , J., Ansmann, A., Wandinge , U., Sei e , P., Schwa z, A.,
Tesche, M., Chaiko sky, A., and Dubo ik, O.: E alua ion o he
Lida /Radiome e In e sion Code (LIRIC) o de e mine mic o-
physical p ope ies o olcanic and dese dus , A mos. Meas.
Tech., 6, 1707–1724, h ps://doi.o g/10.5194/am -6-1707-2013,
2013.
h ps://doi.o g/10.5194/acp-25-7343-2025 A mos. Chem. Phys., 25, 7343–7368, 2025
