Theoretical Study on Volcanic Plume SO and Ash Retrievals ...carboni/doc/TGRS... · Stefano...

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 48, NO. 3, MARCH 2010 1619

Theoretical Study on Volcanic Plume SO2 and AshRetrievals Using Ground TIR Camera: SensitivityAnalysis and Retrieval Procedure Developments.

Stefano Corradini, Cecilia Tirelli, Gabriele Gangale, Sergio Pugnaghi, and Elisa Carboni

Abstract—In this paper, a sensitivity analysis and proceduredevelopment for volcanic-plume sulfur dioxide and ash retrievalsusing ground thermal infrared camera have been carried out.The semiconductor device camera, considered as a reference, hasa spectral range of 8–14 μm with noise equivalent temperaturedifference that is better than 100 mK at 300 K. The camera willbe used to monitor and assess the hazards of Mt. Etna volcanoto mitigate the risk and impact of volcanic eruptions on the civilsociety and transports. A minimum number of filters have beenselected for sulfur dioxide (SO2) and volcanic ash retrievals.The sensitivity study has been carried out to determine the SO2

and volcanic ash minimum concentration detectable by the sys-tem varying the camera geometry and the atmospheric profiles.Results show a meaningful sensitivity increase considering highinstrument altitudes and low camera-elevation angles. For allgeometry configurations and monthly profiles, the sensitivity limitvaries between 0.5 and 2 g · m−2 for SO2 columnar abundanceand between 0.02 and 1 for ash optical depth. Two procedures todetect SO2 and ash, based on the least square fit method and on thebrightness temperature difference (BTD) algorithm, respectively,have also been proposed. Results show that high concentration ofatmospheric water vapor columnar content significantly reducesthe ash-plume effect on the BTD. A water vapor-correction proce-dure introduced improves the ash retrievals and the cloud discrim-ination in every season, considering all the camera geometries.

Index Terms—Ground measurements, Mt. Etna volcano, re-mote sensing, sulfur dioxide (SO2), thermal infrared (TIR) cam-era, volcanic ash.

I. BACKGROUND AND MOTIVATION

VOLCANOES represent one of the most important naturalsources of environmental pollution in the atmosphere both

during eruptions and in the period between them [1]. Goodknowledge of the volatile volcanic emissions in time and spaceis of fundamental importance to understand, on a global scale,

Manuscript received April 1, 2009; revised June 25, 2009. First publishedOctober 30, 2009; current version published February 24, 2010. The work ofE. Carboni was supported by the National Center for Earth Observation(NCEO-NERC) project.

S. Corradini is with the Istituto Nazionale di Geofisica e Vulcanologia(INGV), 00143 Rome, Italy (e-mail: [email protected]).

C. Tirelli is with the University of Rome “La Sapienza,” 00185 Rome, Italy(e-mail: [email protected]).

G. Gangale and S. Pugnaghi are with the Engineering Department of Materi-als and Environment, University of Modena and Reggio Emilia, 41100 Modena,Italy (e-mail: [email protected]; [email protected]).

E. Carboni is with the Atmospheric, Oceanic, and Planetary Physics,Clarendon Laboratory, University of Oxford, OX1 3PU Oxford, U.K. (e-mail:[email protected]).

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/TGRS.2009.2032242

their impact on the world climate and, on a local scale, toprevent problems to human activities and air transport.

The most abundant gases typically released into the at-mosphere from volcanic systems are water vapor (H2O), car-bon dioxide (CO2), and sulfur dioxide (SO2). Volcanoes alsorelease smaller amounts of other gases such as H2S, H2, He,CO, HCl, HF, N and Ar. In addition to the release of a number ofgases, volcanoes emit aerosols and, during explosive eruptions,eject solid-rock fragments and ash into the atmosphere.

The volcanic ash particles are composed of fragments ofpyroclast rocks smaller than 2 mm which are released dur-ing explosive events. Volcanogenic aerosols have a significanteffect on the Earth’s radiative budget even if their emissionstrength is lower than that of anthropogenic sources. Manyvolcanoes degas in the free troposphere, in contrast to anthro-pogenic emissions that are generally restricted to the boundarylayer where species lifetimes are reduced [2], [3]. The radiativeeffects of volcanic aerosol depend on particle-emission rates,particle-size distributions, chemical compositions, and particlemorphologies. Particle radiative effects depend strongly onsize; small particles tend to backscatter incoming shortwaveradiation and large particles tend to absorb outgoing terrestrialradiation [4]. Particle-size distribution and composition deter-mine the ability of particles to act as cloud condensation nucleiand affect the atmospheric lifetime and indirect radiative effectsof the particles [1]. Eruptions that do not penetrate into thestratosphere are usually regarded as being unimportant for long-term or local-climate impacts. However, if such an effusiveeruption persists with for an extended period, it can producea large radiative forcing over local or regional scales. Such tro-pospheric eruptions can have a number of deleterious effects onthe local population including respiratory problems [5], damageto agricultural and industrial productivity [6], and hazards toaviation [7]. The particles with dimension of some millimeterscan damage the aircraft structure (windows, wings, ailerons),while particles less than 10 μm may be extremely dangerous forthe jet engines and undetectable by the pilots during night or inlow-visibility conditions [8]. Theoretical calculations indicatethat volcanic ash clouds with a high concentration of silicateparticles show an opposite spectral behavior with respect towater droplets in the infrared 10–13-μm spectral range [9], [10].Such characteristics can be used to discriminate between vol-canic and meteorological clouds.

Remote sensing is the most suitable technique to detect andretrieve volcanic emissions because of the sporadic nature of

0196-2892/$26.00 © 2009 IEEE

Authorized licensed use limited to: Marco Chini. Downloaded on March 25,2010 at 08:04:09 EDT from IEEE Xplore. Restrictions apply.

1620 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 48, NO. 3, MARCH 2010

volcanic eruptions, the large geographic extent of volcanoes,and the difficulties of direct sampling. The volcanic ash retrievaland cloud-discrimination procedures have been applied usingdifferent satellite platforms, such as the Advanced Very HighResolution Radiometer [9], [11]–[14], Geostationary Opera-tional Environmental Satellite [14]–[16], Total Ozone MappingSpectrometer (TOMS) [13], the MODerate resolution ImagingSpectroradiometer (MODIS) [16]–[21], and the Spin-EnhancedVisible and Infrared Imager (SEVIRI) [21], [22].

As aerosol, volcanic SO2 plays an important role in theworld climate [23]–[26]. The study of volcanic SO2 emissionis very important to compare the natural and the anthropogeniccontributions; recent estimates has placed the global volcanicSO2 emissions at 10% to 15% of the total anthropogenicsulfur output [27]. The study of volcanic gaseous emission alsoprovides important insights into subsurface magmatic processes[28], and flux variations can help to predict eruptive activity[29]–[31]. SO2 is also an important volcanic gas because of itseffects on the environment (e.g., acid rain, effects on plants,and public health) and also because once it is high in theatmosphere (> 6 km), it can be transported over long distances,has a greater residence time, and can be oxidized to form sul-fates. Sulfate aerosols in the atmosphere counteract the globalwarming produced by a high concentration of greenhouse gases[32]–[35] and their principal effects are backscattering of so-lar radiation (direct effect) and increased of cloud reflectivity(indirect effect) [36]. Volcanic SO2 columnar abundance andflux can be measured in many ways. On the ground, remotesensing of SO2 in the ultraviolet (UV) and thermal-infrared(TIR) spectral region is used for volcanic-plume monitoring.Instruments most commonly used in the short wavelengths arethe CORrelation SPECtrometer [37]–[39], the Differential Op-tical Absorption Spectroscopy [40], [41], the Light Detectionand Ranging (LIDAR) [42]–[44], the Differential AbsorptionLIDAR [45] and the Fourier Transform Infrared Spectrometer[46]–[48]. Airborne, the TIR Multispectral Scanner [49], [50]and the Multispectral Infrared and Visible Imaging Spectrome-ter (MIVIS) [51] have been used. The first satellite data used forSO2 stratospheric volcanic-eruption retrieval were the TOMSUV measurements [52], [53]. In 1993, Read et al. [54] used themicrowave measurements of the Microwave Limb Sounder, onboard NASA’s Upper Atmosphere Research Satellite, for theMt. Pinatubo volcano-eruption SO2 retrieval. More recently,MODIS has been used to estimate the SO2 of some eruptions[18], [21], and the TIR bands of the Advanced SpaceborneThermal Emission and Reflection Radiometer has been usedto compute an SO2 map of a volcanic plume in the low tro-posphere [55]–[57]. SO2 volcanic-columnar content has beenretrieved using the High Resolution Infrared Spectrometer [58],the Atmospheric Infrared Sounder [59], [60], the Ozone Moni-toring Instrument [61], [62], and SEVIRI [21], [22].

Mt. Etna is the largest active volcano in Europe, and itsposition is hazardous because of its proximity to the airportof Catania and Reggio Calabria. The ash particles emitted byMt. Etna during eruptions, indiscernible from meteorologicalclouds by airplane radar, were able to cause serious damagesto airplane engines [8]. Moreover, the annual passenger trafficvolume at the international Catania–Fontanarossa airport is

more than five million (5.4 during 2006), and the numerousdays of closure during the 2001 and 2006 eruptions causedlarge losses to the Sicilian economy. An effective volcano-monitoring program is essential for public safety and economicreasons.

The aim of this paper is to analyze the possibility of using anon-ground TIR camera system to retrieve SO2 columnar abun-dance and ash in a volcanic plume. This paper presents the the-oretical basis of a future ground-based TIR camera which willaugment the existing monitoring systems on Mt. Etna volcano.We have conducted a theoretical study of the sensitivity of aTIR system to volcanic ash and SO2 and developed methods forretrieving ash and SO2 amounts. The sensitivity study is aimedat computing the minimum SO2 columnar abundance and theminimum aerosol optical depth (AOD) and effective radius (re)detectable by the system varying the instrument altitude andviewing angle. The retrieval procedures are based on the TIRradiative-transfer equation inversion (least square fit) and onthe brightness-temperature difference (BTD) algorithm for SO2

and ash, respectively. As reference, a semiconductor devicecamera, composed of a matrix of uncooled vanadium oxidemicrobolometric sensors has been considered. This instrumenthas a spectral range varying from 8 to 14 μm and a noiseequivalent temperature difference (NETD) that is better than100 mK at 300 K. Three channels centered around 8.7, 11and 12 μm have been chosen within this spectral range; thefirst channel is used for the SO2 retrieval, while the secondand third channels are used for ash retrievals and for clouddiscrimination. In this paper, calculations have been performedfor a number of viewing geometries (altitude and angle) and fordifferent atmospheric mixing ratio, temperature, and pressureprofiles which span the range of mean observed atmosphericconditions throughout the year. To reduce the water vapor’smasking of the BTD effects of an ash plume in the summermonths, we have developed a method for atmospheric watervapor correction. All the simulations used in the sensitivitystudy and the water vapor correction have been performed withthe MODTRAN 4 radiative-transfer model (RTM) [63], [64].

Organization of this paper is as follows: In Section II,the main TIR camera-system characteristics and the criteriaadopted to select the filters for the volcanic species retrievals arepresented. Section III describes the rationale for the sensitivitystudy, the retrievals, and the cloud-discrimination procedure.The SO2 and volcanic ash sensitivity study and retrieval pro-cedures are described in Sections IV and V, respectively. Wepresent our conclusions in Section VI.

II. REFERENCE INSTRUMENT DESCRIPTION

Nowadays the uncooled focal-plane array (FPA) microbo-lometer takes a large part in infrared-imaging applications. Infact, new thermal-imaging applications and new developmentsare focused on improving the sensitivity to enable the possibil-ity of obtaining high-performance small pixel-pitch detectors.1

1Pixel pitch is a specification for pixel-based devices that describes thedistance between dots on the inside of a display screen. It may be measured inlinear units, usually millimeters, with a smaller number meaning closer spacing.


CORRADINI et al.: STUDY ON SO2 AND VOLCANIC ASH PLUME RETRIEVALS 1621

Fig. 1. Black lines represent the transmittance spectra for the indicated species while the gray lines represent the selected channels. (a) Sulfur dioxidetransmittance (5 g · m−2). (b) Volcanic ash transmittance (re = 3.36 μm, AOD = 1). (c) Water-droplet transmittance (re = 3.36 μm, AOD = 1). (d) Totalatmospheric transmittance with all the channels selected.

Uncooled FPA microbolometers are currently available [65] ina range of pixel arrangements (160 × 120, 320 × 240, 640 ×480 pixels), pixel pitch (25, 35, or 45 μm), response in differentbands within the region of 7 to 14 μm, and NETD rangingfrom 100 to 30 mK at 300 K. To perform the sensitivity testsdescribed in the following, a total spectral range of 8–14 μmand an NETD of 100 mK at 300 K have been assumed. Noreduction of the instrument detectivity has been considered fordetector dimension or noise bandwidth due to the used filters.

The minimum radiance (noise equivalent spectral radiance[NESR]) detectable by the camera system, in the 8–14 μm(1250–714 cm−1) spectral range, is the difference between theradiance emitted by a blackbody at TR + NETD = 300.1 K andthat emitted at TR = 300 K

NESR =

∫ ∞0 Φλ [Bλ(TR + NETD) − Bλ(TR)] dλ∫ ∞

0 Φλdλ(1)

where B is the Planck function and Φλ is a generic responsefunction. Using a square instrument-response function whosevalue is one in the 8–14 μm spectral range and zero elsewhere,NESR results to 14 mW · m−2 · sr−1 · μm−1.

The numerator of (1) represents the noise equivalent powerper unit area and unit solid angle, which is able to stimulate thesensor. This value (84 mW · m−2 · sr−1, with no filter, TR =300 K, and NETD = 100 mK) has been assumed to be theminimum value required to get a change in the sensor signalin any condition of temperature background or filter responsefunction. Clearly, to have that minimum radiance, the NETD in(1) will increase if the background (or reference) temperature(TR) decreases, or if a filter, reducing the energy at the sensor,is used.

A. Channels Definition

The SO2 and ash-retrieval procedures, described in this pa-per, need three different measurements taken in the 8–14 μmrange. These measurements can be obtained by means of threeinterferential filters sequentially located in the optical pathof the camera. In this paper, the spectral response functionof each filter considered is a Gaussian function with unitarytransmittance at the central wavelength and a full-width at half-maximum (FWHM) of 0.5 μm. Nowadays, interferential filterswith similar characteristics are readily available. The choice

TABLE INETD AT 200, 250, AND 300 K FOR EACH CAMERA CHANNEL

of the FWHM value (0.5 μm) is based on the compromisebetween the need to collect a revealable energy amount andthe requirement that the adjacent species influence must beas small as possible. The channel central wavelengths havebeen chosen analyzing the transmittance spectra of the differentspecies considered in this study. All the simulations have beenrealized using MODTRAN 4 RTM for a sensor placed at sealevel pointed in the zenithal direction. Fig. 1(a) shows the SO2

transmittance considering a columnar abundance of 5 g · m−2.The channel selected, widely used in many satellite instruments[18], [49], [51], [55], [57], [66], is centered around the maxi-mum SO2 absorption wavelength.

The retrieval of volcanic ash optical properties (re and AODat 0.55 μm) and the discrimination between volcanic andmeteorological clouds is generally carried out using channelscentered around 11 and 12 μm. In particular, the discriminationbetween clouds is based on the different spectral absorptionbetween volcanic aerosols and water droplets in the 10–13-μmspectral range [9], [10] [see Fig. 1(b) and (c)]. In such aspectral range, the absorption, then the transmittance, result isdecreasing for volcanic ash and increasing for water droplets.Fig. 1(d) shows the total atmospheric-transmittance spectrumwith the superposition of the three selected channels.

The NESR of each channel (ΔRmin) can be computedusing (1), where the numerator has the fix value of 84 mW ·m−2 · sr−1, and knowing the filter-response function at thedenominator. In this paper, all the filter response functions havea Gaussian shape, with unitary transmittance at the centralwavelength and with the same FWHM (0.5 μm); thereforeΔRmin (157.8 mW · m−2 · sr−1μm−1) is the same for all thethree filters. Note also that this value could be computed simplyby multiplying the instrumental total NESR with the ratiobetween the area of the instrument square response function andthe filter Gaussian response function.

In Table I are reported, for each filter, the central wavelengthand three NETD values, obtained by numeric inversion of



TABLE IIPARAMETER SETTING FOR SO2 AND ASH-PLUME MODTRAN

SIMULATIONS. THE PARTICLES’ EFFECTIVE RADII AND THE

OPTICAL-DEPTH VALUES HAVE BEEN SELECTED TO BE

EQUISPACED IN A LOGARITHMIC SCALE

Fig. 2. February and August monthly mean precipitable water and temper-ature atmospheric profiles computed considering 23 years of Trapani WMOMeteo station measurements.

(1) with the aforementioned NESR and three different (TR)background temperature values: 200, 250, and 300 K. The trendof the NETD in the range 200–300 K is not linear, but thevalues, computed at the edges of this quite realistic temperatureinterval, can give a good idea of its variation.

III. RATIONALE OF SENSITIVITY ANALYSIS

AND RETRIEVAL SCHEMES

The sensitivity study for SO2 and ash-plume retrievals hasbeen carried out by means of RTM computations simulating theradiance measured by the camera looking through the volcanicplume. The simulations have been obtained varying the cameraaltitude and viewing angle and the plume characteristics as SO2

columnar abundance and ash optical depth considering differenteffective radii (see Table II). For each camera configuration(altitude viewing angle), the minimum SO2 columnar abun-dance (see Section IV-A) or ash optical depth and effectiveradius (see Section IV-B) able to produce a signal detectableby the camera channels (see Section II-A) has been obtained.All the simulations have been realized using MODTRAN 4RTM. To take into account the strong atmospheric-profile sea-sonal variation and, in particular, the influence of water vaporin the spectral range of measurements, the RTM simulationshave been carried out considering all the monthly atmosphericprofiles. The mean monthly atmospheric profiles of pressure,temperature and humidity (PTH) were derived from the analysisof 23 years of radiosounding measurements made at TrapaniWMO Meteo Station. Trapani is located at the western tip ofSicily and is the closest meteorological station to the Mt. Etnavolcano. Fig. 2 shows an example of the mean precipitable

water and atmospheric-temperature profiles for February andAugust, the months with the lowest and the highest water vaporcontent, respectively.

The ash optical properties (single-scattering albedo, extinc-tion coefficients, and asymmetry parameter) necessary for theash-plume characterization have been computed using the Ox-ford University Atmospheric, Oceanic, and Planetary PhysicsMie Code [67]. The aerosol-size distribution has been consid-ered log-normal (with fixed spread = 1.77 and different valuesof mode radius between 0.34 and 4.42 μm in order to obtainthe effective radius in Table II) and the refractive index derivesfrom Volz (1973) [68].

The procedure proposed for the SO2 columnar abundance re-trieval is based on the least square fit method (see Section V-A)considering the channel centered around 8.7 μm, while theretrieval of the ash optical depth and effective radius, as wellas the ash and water plumes discrimination, are based on theBTD procedure (see Section V-B) considering the channelscentered around 11 and 12 μm. Results obtained in Section V-Bemphasize that the atmospheric water vapor radiative effectcounteracts the radiative effect of volcanic ash on BTD, and,in case of high water vapor (as in summer season), reducemeaningfully the ash-plume detection and retrieval. A watervapor correction has been carried out in order to make effectivethe ash retrievals and the cloud discrimination, also, in case ofhigh-precipitable water content in the atmosphere.

IV. SENSITIVITY ANALYSIS

In this section, the SO2 columnar abundance and ash-retrieval sensitivity study will be presented.

As described previously, the sensitivity study has been per-formed by means of MODTRAN 4 simulations varying thecamera geometry, the mean monthly atmospheric profiles, theSO2 columnar abundance, the ash-particle effective radius,and the ash optical depth. For the SO2 columnar abundancesensitivity analysis, the top plume altitude has been set at3400 m and the depth at 100 m. This geometry describes wellthe Mt. Etna volcano during its quiescent state. For the ash sen-sitivity analysis, the top plume altitude has been set at 5000 mand the depth at 1000 m. This scenario simulates properlythe Mt. Etna volcano during a tropospheric eruption. All thesimulations have been convoluted with the response function ofthe instrument selected channel (see Section II-A).

A. SO2 Sensitivity Study

Fig. 3 shows a zoom (only up to 5 g/m2) of the sensitivitycurves varying the instrument view angles for the differentaltitudes for February and August atmospheric profiles; allthe other monthly sensitivity curve results were included. Theradiance on the ordinate axis represents the difference betweenthe radiance computed with the SO2 columnar content indicatedin the abscissa (Rx) and the one obtained without SO2 plume(R0). The horizontal line is the minimum radiance variationdetectable by the camera channels (ΔRmin, see Section II-A).

The presence of a SO2 plume clearly increases the radianceat the camera with respect to the case without it. The extra



Fig. 3. SO2 sensitivity curves for instrument channel 1 considering February and August mean atmospheric profiles. Each plot represents the sensitivity for afixed instrument altitude and varying instrument viewing angle. The horizontal line in every plot indicates the instrument channel sensitivity.

Fig. 4. Volcanic ash sensitivity curves for instrument channel 2 and 3 considering February and August mean atmospheric profiles. Each plot represents theminimum retrievable AOD considering different effective radii, for a fixed instrument altitude and varying instrument viewing angle. The lower curves indicategreater sensitivity.

radiance contribution due to the SO2 plume presence dependson many parameters and, in particular, on the viewing angle andon the atmospheric transmittance underneath the plume. Fig. 3shows that, in all our scenarios, the lower the camera elevation(minimum 15◦) the greater the added radiance, i.e., the greaterthe sensitivity. A small SO2 sensitivity increase occurs also forhigh instrument altitudes. No meaningful seasonal dependencehas been revealed. Note that both the lower camera elevationsand higher camera altitudes produce a similar sensitivity im-provement even if the path through the atmosphere (the watervapor effect) is increasing and decreasing, respectively. Thecontradiction is only apparent because, in the first case, togetherwith an increase of the path through the atmosphere, thereis also an increase of the path through the plume. Until 15◦

of elevation, the plume contribution (plume radiative effect)overlaps the water vapor effect. Further simulations indicatethat for a defined camera altitude, there is a camera elevationangle below which the atmospheric water vapor contributionoverlaps the plume contribution; such angle results always

to less than 15◦. Both for February and August and for allthe geometries, the SO2 sensitivity varies between 0.5 and2 g · m−2.

B. Ash Sensitivity Study

Fig. 4 shows the volcanic ash sensitivity curves considering0 and 2400 m camera altitudes for channels 2 and 3, varyingthe viewing angle for February and August, respectively, inlogarithmic scale. Each curve shows, for the effective radiusin the abscissa, the value of the minimum AOD at 0.55 μmdetectable by the considered channel. It means that a plumeof ash particles with the indicated effective radius and AODproduces a radiance variation, with respect to the same at-mosphere with clear sky, which is equal to NESR. This sen-sitivity analysis has been performed for channels 2 and 3 thatare the ones involved in the BTD algorithm. Considering thatthe sensitivity depends on the central wavelength of the usedchannel, for every configuration, the worst sensitivity between



the two has been considered. The graphs show in the ordinatethe minimum aerosol optical depth detectable by the instrumentfor each value of the particle effective radius in the abscissa.This sensitivity increases for lower viewing angles, increasingcamera altitudes, and strongly, for the particle effective radiithat are greater than 2 μm. The figures also show that thechannel 3 minimum retrievable AOD results are greater thanchannel 2 for each configuration. Taking into account that theash effective radius and AOD retrieval is performed using thecombination between the two channels (see Section V-B),the sensitivity of channel 3 has been considered as reference.For all configurations and seasonal atmospheric profiles, theAOD sensitivity results are better than 0.1 for particle effectiveradius that are greater than 2 μm. Conversely, for smallerparticles, the minimum retrievable AOD varies between 0.2 and1 considering 15◦ and 90◦, respectively.

V. RETRIEVAL PROCEDURE DEVELOPMENT

A. SO2 Retrieval

The SO2 retrieval scheme proposed is based on a leastsquares fit procedure [18], [49], [57] using the channel 1camera measurements and the simulated on-ground radiancesfor different SO2 columnar abundances. The function to beminimized may be written as

χ2cs

=⌊

Lm,λ − Ls,λ(cs)Lm,λ

⌋2

(2)

where

λ the central wavelength of channel 1;Lm,λ the measured radiance;Ls,λ (cs) the simulated radiance varying the SO2 columnar

abundance cs.

The simulated radiances will be computed from MODTRANRTM using as input the atmospheric profiles (PTH) of the dayof measurements.

B. Ash Retrievals and Discrimination Between Ash andWater-Droplet Volcanic Plumes

The aim of this section is to verify the possibility to retrievethe ash characteristics (AOD at 0.55 μm and re) from thecamera measurements as well as to discriminate between ashand water-droplet volcanic plumes. Note that we are consid-ering the extreme cases from which only ash or only waterdroplets are present in the volcanic plume. The effect of thecontemporaneous presence of ash and water droplets is out ofthe aim of this paper.

The ash retrievals are computed from the inverted arch curvesof the BTD versus brightness temperature at 11-μm curvesby varying AOD and re [12]–[16], [69]. All the simulationsneeded for the brightness temperature computation have beenmade by using MODTRAN RTM. The ash and water-dropletplumes discrimination is made by using the BTD technique,based on the difference between the brightness temperaturescomputed from the 11- and 12-μm channels [9], [10], which

Fig. 5. Inverted arch curves for volcanic ash and water-droplet plumesconsidering an altitude of 2400 m, an elevation of 90◦, and February meanatmospheric profiles. The upper inverted arch curve is relative to volcanicash while the lower to water droplets. The solid lines represent the differenteffective radii, and the dashed lines represent AOD at 0.55 μm.

will have opposite signs by considering ash and water droplets;such difference will be positive for ash clouds and negative forwater-droplet plume clouds. Note that the sign of the differenceresults are inverse with respect to the BTD sign consideringsatellite remote sensing. The reason is that, looking at thesky from ground in the TIR spectral range and neglecting thecases of thermal inversion above the plume, the plume acts asemitter at the plume temperature. The greater the absorption,the greater the emission, then, greater radiation is reaching thesensor, i.e., positive BTD. Conversely, looking at the groundfrom a satellite, the plume acts as absorber for the radiationcoming from the surface; therefore, the greater the absorption,the lower is the radiation reaching the sensor, i.e., negativeBTD. Fig. 5 represents the way the results will be presented; theplot concerns a camera placed at 2400 m, looking at the zenithfor February atmospheric profile. The inverted arch curves forvolcanic ash and water-droplet plumes lay in the upper andlower part of the figure (over and under zero BTD), respectively.The solid lines represent different effective radii decreasingtoward the zero BTD line, while the dashed lines represent theAOD at 0.55 μm increasing from left to right. As the figureshows, the ash and water-droplet plumes results are very welldiscriminable. Solid- and dashed-line grid allows estimatingboth the effective radius and the AOD.

Fig. 6 shows the inverted arch curves considering the Febru-ary and August mean atmospheric profiles for camera altitudesof 0 and 2400 m and camera viewing angle of 90◦ and 15◦. Asthe plates show, the lower the altitude and the lower the viewangle (i.e., the greater the atmospheric water vapor abundancein the camera line of sight), the worse is the plume discrim-ination and the effective radius and AOD retrievals. The twoinverted arch curves results are well defined and separated forthe 2400-m 90◦ February mean-profile configuration (best case)but are extremely compressed for the 0-m 15◦ August mean-profile configuration (worst case). The results indicate that theatmospheric water vapor deteriorates both the plume discrim-ination and the particle retrievals, i.e., the highly precipitable



Fig. 6. Inverted arch curves for volcanic ash and water-droplet plumes considering camera altitudes of 0 and 2400 m and camera elevations of 15◦ and 90◦ forFebruary and August mean atmospheric profiles. The upper inverted arch curve concerns to ash while the lower to water-droplet plumes. The solid lines representdifferent effective radii, and the dashed lines represent the different AOD at 0.55 μm.

water in the summer season dramatically reduces the radiativeeffects on BTD of both volcanic ash and water droplets.

C. Atmospheric Water Vapor Correction

To make easier both the ash and water plume discriminationand the AOD and re retrievals, an atmospheric water vaporcorrection has been introduced. In literature, the water vaporcorrections proposed for satellite measurements [14], [20], [69]are based on the experimental relationship between the BTDand the 11-μm channel brightness temperature varying with theatmospheric water vapor. The correction proposed in this paper,however, is based on the separation between the atmosphericwater vapor and plume terms considering a simplified radiative-transfer equation in the TIR spectral range. Considering aground instrument looking at the sky through a volcanic plume,the radiative-transfer equation can be approximated by the sumof the contribution of the atmosphere under the plume and thecontribution by the plume itself

Lλ = La,λ + Lp,λτa,λ (3)

whereλ the central wavelength of channel 2 or 3;Lλ the radiance reaching the camera;La,λ the downwelling atmospheric radiance underneath the

plume;Lp,λ the downwelling plume radiance;τa,λ the atmospheric transmittance underneath the plume.

In (3), the contribution due to the atmosphere above theplume has not been inserted because it is negligible comparedwith the considered terms. To minimize the effect due to theapproximation described, La,λ and τa,λ have been consideredas the total downwelling path radiance and the total atmospherictransmittance. Separating the water vapor terms from the plumeterms, (3) can be written as

Lp,λ =Lλ − La,λ

τa,λ(4)

Lp,λ is not affected by atmospheric water vapor absorption.Note that (4) is valid within the 10–13-μm spectral rangein which the atmospheric transmittance varies approximatelybetween 0.4 to 0.8 considering the extreme cases of very highand low atmospheric water vapor.

Fig. 7 shows the inverted arch curves derived applying (4)for the same configurations shown in Fig. 6. The atmosphericcorrection terms La,λ and τa,λ have been computed usingMODTRAN RTM and the February and August mean at-mospheric profiles. As the plates show, the inverted arch curveresults are well defined and separated by the zero BTD, i.e., thewater vapor correction procedure meaningfully improves eitherthe plume discrimination or the particle effective radius andAOD retrieval. As can be noted, for every inverted arch curves,three different regions have been emphasized. The region de-limited by the green lines is the region where the retrievals arepossible. It means that for the same effective radius, each AODvariation produces a radiance variation greater than NESR, andit can be detected by the sensor in both channels. The regionsdelimited by the red and blue lines are the regions where theretrievals are not permitted. The red regions represent the lowestlimit of detection (NESR) according to the sensitivity study pre-sented in Section IV-B, i.e., the ash signal computed consider-ing the effective radii and AOD included in the red area is lowerthan the instrument limit of detection and not retrievable. Theblue regions represent the effect of the instrument sensitivity forhigh AOD values. In this region, there is a reduction of AODresolution, i.e., a greater AOD variation is required to producea signal variation greater than the instrument NESR. Fig. 7also shows that in the case of August atmospheric profiles,with a camera altitude of 0 m and an elevation of 15◦, thediscrimination is effective because the two curve results arewell separated, but the retrieval is not permitted because thered region borders the blue region. The AOD and re retrievalsare preserved by simply placing the camera at 800 m.

The results obtained applying the water vapor correctionshow that both the discrimination and retrievals are possible inevery season by placing the instrument at least at 800 m.



Fig. 7. Inverted arch curves computed considering the water vapor correction procedure for volcanic ash and water-droplet particles considering camera altitudesof 0 and 2400 m and camera elevations of 15◦ and 90◦ for February and August mean atmospheric profiles. The region delimited by the green lines is the regionwhere the retrievals are possible, while the region delimited by the red and blue lines are the regions where the retrievals are not permitted.

VI. CONCLUSION

In this theoretical work, a sensitivity study and a proceduredevelopment for SO2 and volcanic ash retrievals using a newon-ground TIR camera system have been carried out. Thecamera will be used to improve Mt. Etna volcano’s monitoring,i.e., to mitigate the volcanic eruption impact on civil society andtransport (the highway and international airport of Catania areabout 20 km from the craters).

Three channels have been chosen within the camera spectralrange (8–14 μm) to allow the detection of SO2 and volcanicash. A typical NESR has been computed assuming for the totalrange a common NETD of 100 mK at 300 K.

The sensitivity study has been realized for different camerageometries and monthly atmospheric profiles (PTH) obtainedby the analysis of 23 years of measurements made at TrapaniWMO Meteo Station. The simulations have been obtainedby MODTRAN 4 RTM varying the SO2 columnar abun-dance, volcanic ash particles and water-droplet effective radiusand AOD.

In all the scenarios considered here, the SO2 columnarabundance sensitivity on channel 1 improve, decreasing theelevation and the instrument altitude, while no meaningfulseasonal dependence has been revealed. Considering a 15◦ and90◦ camera viewing angle, the SO2 sensitivity varies between0.5 and 2 g · m−2 for all the geometries. Similarly, the volcanicash AOD sensitivity, computed for channels 2 and 3, increases,lowering the viewing angle, increasing the camera altitude, andconsidering particles greater than 2-μm effective radius. For allconfigurations, the AOD sensitivity results are better than 0.1for the particle effective radius greater than 2 μm. In the typicalrange of Mt. Etna SO2 emission, the sensitivity gets slightlyworse, increasing the columnar abundance. The same happensfor AOD sensitivity for high values of AOD.

An SO2 columnar abundance retrieval procedure, based onthe least square fit method, has been proposed.

The ash and water-droplet plume discrimination, the ef-fective radius, and the AOD retrievals has been realized by

means of BTD procedure, based on the difference betweenthe brightness temperatures of channels 2 and 3. The resultsshow the strong effect of atmospheric water vapor on plumediscrimination and particle retrievals. In particular, during thesummer season, the high water vapor content significantlyreduces the volcanic ash and water-droplet radiative effects. Toimprove both the cloud discrimination and the particle retrieval,a water vapor correction, based on the separation between theatmospheric water vapor and plume terms on the radiative-transfer equation, has been developed. The sensitivity studyprovides the minimum AOD variation required to produce asignal variation greater than the instrument noise. In case ofhigh integrated water vapor (usually during summer season)for low altitudes and low camera elevations, the AOD and re

retrievals are nearly impossible. Notwithstanding, placing thecamera at least at 800-m altitude makes effective a contempo-rary SO2 volcanic ash retrievals and cloud discrimination.

ACKNOWLEDGMENT

We thank Mauro Coltelli for suggesting the use of the TIRcamera to monitor the plume of Mt. Etna volcano. E. Carboni’swork was supported by National Center for Earth Observation(NCEO-NERC) project.

REFERENCES

[1] C. Oppenheimer, D. M. Pyle, and J. Barclay, Volcanic Degassing, vol. 213.Bath, U.K.: Geol. Soc. Publ. House, 2003, ser. Geol. Soc. Lond.Spec. Pub.

[2] H. F. Graf, B. Langmann, and J. Feichter, “The contribution of Earthdegassing to the atmosphere sulfur budget,” Chem. Geol., vol. 147, no. 1,pp. 131–145, May 1998.

[3] D. S. Stevenson, C. E. Johnson, W. J. Collins, and R. G. Derwent, “Thetropospheric sulphur cycle and the role of volcanic SO2,” in VolcanicDegassing, vol. 213, C. Oppenheimer, D. M. Pyle, and J. Barclay, Eds.Bath, U.K.: Geol. Soc. Publ. House, 2003, ser. Geol. Soc. Lond. Spec.Pub., pp. 295–305.

[4] R. G. Grainger, A. Lambert, F. W. Taylor, J. J. Remedios, C. D. Rodgers,C. D. Corney, and B. J. Kerridge, “Infrared absorption by volcanicstratospheric aerosols observed by ISAMS,” Geophys. Res. Lett., vol. 20,no. 12, pp. 1283–1286, 1993.



[5] C. J. Horwell and P. J. Baxter, “The respiratory health hazards of volcanicash: A review for volcanic risk mitigation,” Bull. Volcanol., vol. 69, no. 1,pp. 1–24, 2006.

[6] C. Stewart, D. M. Johnston, G. S. Leonard, C. J. Horwell, T. Thordarson,and S. J. Cronin, “Contamination of water supplies by volcanic ashfall:A literature review and simple impact modelling,” J. Volcanol. Geotherm.Res., vol. 158, no. 3/4, pp. 296–306, Nov. 2006.

[7] T. J. Casadewall, “Volcanic ash and aviation safety,” U.S. Geol. Surv. Bull.,vol. 2047, pp. 1–6, 1994.

[8] G. L. Hufford, L. J. Salinas, J. J. Simpson, E. G. Barske, and D. C. Pieri,“Operational implications or airborne volcanic ash,” Bull. Amer. Meteorol.Soc., vol. 81, no. 4, pp. 745–755, Apr. 2000.

[9] A. J. Prata, “Observation of volcanic ash clouds using AVHRR-2 radi-ances,” Int. J. Remote Sens., vol. 10, no. 4/5, pp. 751–761, Apr. 1989.

[10] A. J. Prata, “Radiative transfer calculations for volcanic ash clouds,”Geophys. Res. Lett., vol. 16, no. D11, pp. 1293–1296, 1989.

[11] L. E. Holasek and W. I. Rose, “Anatomy of 1986 Augustine volcano erup-tions as recorded by multispectral image processing of digital AVHRRweather satellite data,” Bull. Volcanol., vol. 53, no. 6, pp. 420–435,Aug. 1991.

[12] S. Wen and W. I. Rose, “Retrieval of sizes and total masses of particles involcanic clouds using AVHRR bands 4 and 5,” J. Geophys. Res., vol. 99,no. D3, pp. 5421–5431, 1994.

[13] D. J. Shneider, W. I. Rose, L. R. Coke, G. J. S. Bluth, I. Sprod, andA. J. Krueger, “Early evolution of stratospheric volcanic eruption cloud asobserved with TOMS and AVHRR,” J. Geophys. Res., vol. 104, no. D4,pp. 4037–4050, 1999.

[14] T. Yu, W. I. Rose, and A. J. Prata, “Atmospheric correction for satellite-based volcanic ash mapping and retrievals using split window IR datafrom GOES and AVHRR,” J. Geophys. Res., vol. 107, no. D16, p. 4311,2002.

[15] W. Rose and G. C. Mayberry, “Use of GOES thermal infrared imagery foreruption scale measurements, Soufriere Hills, Montserrat,” Geophys. Res.Lett., vol. 27, no. 19, pp. 3097–3100, 2000.

[16] G. P. Ellrod, B. H. Connell, and D. W. Hillger, “Improved detectionof airborne volcanic ash using multispectral infrared satellite data,”J. Geophys. Res., vol. 108, no. D12, p. 4356, Jun. 2003.

[17] D. W. Hillger and J. D. Clark, “Principal component image analysis ofMODIS for volcanic ash. Part I: Most important bands and implica-tions for future GOES imagers,” J. Appl. Meteorol., vol. 41, no. 10,pp. 985–1001, Oct. 2002.

[18] I. M. Watson, V. J. Realmuto, W. I. Rose, A. J. Prata, G. J. S. Bluth,Y. Gu, C. E. Bader, and T. Yu, “Thermal infrared remote sensing ofvolcanic emissions using the moderate resolution imaging spectrora-diometer,” J. Volcanol. Geotherm. Res., vol. 135, no. 1/2, pp. 75–89,Jul. 2004.

[19] A. Tupper, S. Carn, J. Davey, Y. Kamada, R. Potts, and F. Prata, “Anevaluation of volcanic cloud detection techniques during recent significanteruption in the western Ring of Fire,” Remote Sens. Environ., vol. 91,no. 1, pp. 27–46, May 2004.

[20] S. Corradini, C. Spinetti, E. Carboni, C. Tirelli, M. F. Buongiorno,S. Pugnaghi, and G. Gangale, “Mt. Etna tropospheric ash retrieval andsensitivity analysis using Moderate Resolution Imaging Spectroradiome-ter measurements,” J. Appl. Remote Sens., vol. 2, no. 1, p. 023 550, 2008.

[21] S. Corradini, L. Merucci, and A. J. Prata, “Retrieval of SO2 from thermalinfrared satellite measurements: Correction procedures for the effects ofvolcanic ash,” Atmos. Meas. Tech., vol. 2, no. 1, pp. 177–191, 2009.

[22] A. J. Prata and J. Kermann, “Simultaneous retrieval of volcanic ash andSO2 using MSG-SEVIRI measurements,” J. Geophys. Res., vol. 112,p. D20 204, 2007.

[23] R. E. Stoiber, S. N. Williams, and B. Huebert, “Annual contribution ofsulfur dioxide to the atmosphere by volcanoes,” J. Volcanol. Geotherm.Res., vol. 33, no. 1–3, pp. 1–8, Aug. 1987.

[24] P. A. Spiro and D. J. Jacob, “Global inventory of sulfur emission with1◦ × 1◦ resolution,” J. Geophys. Res., vol. 97, no. D5, pp. 6023–6036,1992.

[25] J. T. Kiehl and B. P. Briegleb, “The relative roles of sulfate aerosolsand greenhouse gases in climate forcing,” Science, vol. 260, no. 5106,pp. 311–314, Apr. 1993.

[26] T. P. Karl, R. W. Knight, G. Kukla, and J. Gavin, “Evidence for radiativeeffects of anthropogenic sulfate aerosols in the observed climate record,”in Aerosol Forcing of Climate, R. J. Charlson and J. Heintzenberg, Eds.Hoboken, NJ: Wiley, 1995, pp. 363–382.

[27] M. M. Halmer, H. U. Schmincke, and H. F. Graf, “The annual volcanicgas input into the atmosphere, in particular into the stratosphere: A globaldata set for the past 100 years,” J. Volcanol. Geotherm. Res., vol. 115,no. 3/4, pp. 511–528, Jun. 2002.

[28] T. M. Gerlach and E. J. Graeber, “Volatile budget of Kilauea volcano,”Nature, vol. 313, no. 6000, pp. 273–277, Jan. 1985.

[29] L. L. Malinconico, Jr., “Fluctuation in SO2 emission during recent erup-tion of Etna,” Nature, vol. 278, no. 5699, pp. 43–45, Mar. 1979.

[30] L. L. Malinconico, Jr., “On the variation of SO2 emission from vol-canoes,” J. Volcanol. Geotherm. Res., vol. 33, no. 1–3, pp. 231–237,Aug. 1987.

[31] T. Caltabaiano, R. Romano, and G. Budetta, “ SO2 flux measurementsat Mount Etna (Sicily),” J. Geophys. Res., vol. 99, no. D6, pp. 12 809–12 819, 1994.

[32] E. G. Dutton and J. R. Christy, “Solar radiative forcing at selectedlocations and evidence for global lower tropospheric cooling followingthe eruption of El Chichon and Pinatubo,” Geophys. Res. Lett., vol. 19,no. 23, pp. 2313–2316, 1992.

[33] G. L. Stenchikov, L. Georgiy, I. Kirchner, A. Robock, H. F. Graf,J. C. Antuna, R. G. Grainger, A. Lambert, and L. Thomason, “Radiativeforcing from the 1991 Mount Pinatubo volcanic eruption,” J. Geophys.Res., vol. 103, no. D12, pp. 13 837–13 857, Jun. 1998.

[34] A. Robock, “Volcanic eruptions and climate,” Rev. Geophys., vol. 38,no. 2, pp. 191–219, 2000.

[35] R. G. Grainger and E. J. Highwood, “Changes in stratospheric compo-sition chemistry, radiation and climate caused by volcanic eruptions,”in Volcanic Degassing, vol. 213, C. Oppenheimer, D. M. Pyle, andJ. Barclay, Eds. Bath, U.K.: Geol. Soc. Publ. House, 2003, ser. Geo-logical Society, London, Special Publication, pp. 329–347.

[36] S. Twomey, “The influence of pollution on the shortwave albedo ofclouds,” J. Atmos. Sci., vol. 34, no. 7, pp. 1149–1152, Jul. 1977.

[37] A. J. Moffat and M. M. Millan, “The applications of optical correlationtechniques to the remote sensing of SO2 plumes using sky light,” Atmos.Environ., vol. 5, no. 8, pp. 677–690, Aug. 1971.

[38] R. E. Stoiber, L. L. Malinconico, Jr., and S. N. Williams, “Use of thecorrelation spectrometer at volcanoes,” in Forecasting Volcanic Events,H. Tazieff and J. C. Sabroux, Eds. Amsterdam, The Netherlands:Elsevier, 1983, pp. 425–444.

[39] T. Caltabiano and R. Romano, “Messa a punto di metodologie di misuracon apparecchiatura COSPEC del flusso di SO2 da vulcani attivi italiani,”Boll. GNV , pp. 133–145, 1988.

[40] T. J. Casadevall, W. I. Rose, W. H. Fuller, W. H. Hunt, M. A. Hart,J. L. Moyers, D. C. Woods, R. L. Chuan, and J. P. Friend, “Sulfur dioxideand particles in quiescent volcanic plumes from Poas, Arenal, and Colimavolcanoes, Costa Rica and Mexico,” J. Geophys. Res., vol. 89, no. D6,pp. 9633–9641, 1984.

[41] P. Weibring, H. Edner, S. Svanberg, G. Cecchi, L. Pantani, R. Ferrara,and T. Caltabiano, “Monitoring of volcanic sulfur dioxide emissions us-ing differential absorption Lidar (DIAL), differential optical absorptionspectroscopy (DOAS), and correlation spectroscopy (COSPEC),” Appl.Phys. B, Lasers Opt., vol. 67, no. 4, pp. 419–426, Oct. 1998.

[42] M. G. Villani, L. Mona, A. Maurizi, G. Pappalardo, A. Tiesi, M. Pandolfi,M. D’Isidoro, V. Cuomo, and F. Tampieri, “Transport of volcanic aerosolin the troposphere: The case study of the 2002 Etna plume,” J. Geophys.Res., vol. 111, no. D21, p. D21 102, Nov. 2006.

[43] G. S. Kent and G. M. Hansen, “Multiwavelength Lidar observations of thedecay phase of the stratospheric aerosol layer produced by the eruption ofMount Pinatubo in June 1991,” Appl. Opt., vol. 37, no. 18, pp. 3861–3872,Jun. 1998.

[44] S. Kenneth, J. Zhu, P. Webley, K. Dean, and P. Cobb, “Volcanic ash plumeidentification using polarization Lidar: Augustine eruption, Alaska,” Geo-phys. Res. Lett., vol. 34, no. 8, p. L08 803, Apr. 2007.

[45] H. Edner, P. Ragnarson, S. Svanberg, E. Wallinder, R. Ferrara, R. Cioni,B. Raco, and G. Taddeucci, “Total fluxes of sulfur dioxide from theItalian volcanoes Etna, Stromboli and Vulcano measured by differentialabsorption Lidar and passive differential optical absorption spectroscopy,”J. Geophys. Res., vol. 99, no. D9, pp. 18 827–18 838, 1994.

[46] P. Francis, M. R. Burton, and C. Oppenheimer, “Remote measurementsof volcanic gas compositions by solar occultation spectroscopy,” Nature,vol. 396, no. 6711, pp. 567–570, Dec. 1998.

[47] L. Horrocks, M. Burton, P. Francis, and C. Oppenheimer, “Stable gasplume composition measured by OP FTIR spectroscopy at MasayaVolcano, Nicaragua 1998,” Geophys. Res. Lett., vol. 26, no. 23, pp. 3497–3500, 1999.

[48] M. Burton, P. Allard, F. Mur, and C. Oppenheimer, “FTIR remotesensing of fractional magma degassing at Mt. Etna, Sicily,” in VolcanicDegassing, vol. 213, C. Oppenheimer, D. M. Pyle, and J. Barclay, Eds.Bath, U.K.: Geol. Soc. Publ. House, 2003, ser. Geol. Soc. Lond. Spec.Pub., pp. 281–293.

[49] V. J. Realmuto, M. J. Abrams, M. F. Buongiorno, and D. C. Pieri, “Theuse of multispectral thermal infrared image data to estimate the sulfur



dioxide flux from volcanoes: A case study from Mount Etna, Sicily,July 29, 1986,” J. Geophys. Res., vol. 99, no. B1, pp. 481–488, 1994.

[50] V. J. Realmuto, A. J. Sutton, and T. Elias, “Multispectral thermal infraredmapping of sulfur dioxide plumes: A case study from the East Rift Zone ofKilauea volcano, Hawaii,” J. Geophys. Res., vol. 102, no. B7, pp. 15 057–15 072, 1997.

[51] S. Teggi, M. P. Bogliolo, M. F. Buongiorno, S. Pugnaghi, and A. Sterni,“Evaluation of SO2 emission from Mount Etna diurnal and nocturnalmultispectral IR and visible imaging spectrometer thermal IR remotesensing images and radiative transfer models,” J. Geophys. Res., vol. 104,no. B9, pp. 20 069–20 079, 1999.

[52] A. J. Krueger, “Sighting of El Chichon sulfur dioxide clouds with theNimbus 7 total ozone mapping spectrometer,” Science, vol. 220, no. 4604,pp. 1377–1379, Jun. 1983.

[53] A. J. Krueger, L. S. Walter, C. C. Schnetzler, and S. D. Doiron, “TOMSmeasurement of the sulfur dioxide emitted during the 1985 Nevado delRuiz eruptions,” J. Volcanol. Geotherm. Res., vol. 41, no. 1–4, pp. 7–15,Jul. 1990.

[54] W. G. Read, L. Froidevaux, and J. W. Waters, “Microwave Limb Sounder(MLS) measurements of SO2 from Mt. Pinatubo volcano,” Geophys. Res.Lett., vol. 20, no. 12, pp. 1299–1302, 1993.

[55] S. Corradini, S. Pugnaghi, S. Teggi, M. F. Buongiorno, and M. P. Bogliolo,“Will ASTER see the Etna SO2 plume?,” Int. J. Remote Sens., vol. 24,no. 6, pp. 1207–1218, Mar. 2003.

[56] M. Urai, “Sulfur dioxide flux estimation from volcanoes using advancedspaceborne thermal emission and reflection radiometer—A case studyof Miyakejima volcano, Japan,” J. Volcanol. Geotherm. Res., vol. 134,no. 1/2, pp. 1–13, Jun. 2004.

[57] S. Pugnaghi, G. Gangale, S. Corradini, and M. F. Buongiorno, “Mt. Etnasulfur dioxide monitoring using ASTER-TIR data and atmospheric ob-servations,” J. Volcanol. Geotherm. Res., vol. 152, no. 1/2, pp. 74–90,Apr. 2006.

[58] A. J. Prata, W. I. Rose, S. Self, and D. M. O’Brien, “Global, long-termsulphur dioxide measurements from TOVS data: A new tool for studyingexplosive volcanism and climate,” in Volcanism and Earths Atmosphere,Geophysics Monograph, vol. 139. Washington, DC: Amer. Geophys.Union, 2003, pp. 75–92.

[59] A. J. Prata and C. Bernardo, “Retrieval of volcanic SO2 total columnfrom AIRS data,” J. Geophys. Res., vol. 112, no. D20, p. D20 204,2007.

[60] A. J. Prata, G. Gangale, and L. Clarisse, “Particle size and compositionalretrievals of the Chaiten volcanic ash from spaceborne, high spectral reso-lution infrared AIRS and IASI measurements,” Eos Trans. AGU, vol. 89,no. 53, 2008, Fall Meet. Suppl., Abstract V42C-04.

[61] N. A. Krotkov, S. A. Carn, A. J. Krueger, P. K. Bhartia, and K. Yang,“Band residual difference algorithm for retrieval of SO2 from the auraOzone Monitoring Instrument (OMI),” IEEE Trans. Geosci. RemoteSens., vol. 44, no. 5, pp. 1259–1266, May 2006.

[62] N. A. Krotkov, B. McClure, R. R. Dickerson, S. A. Carn, C. Li,P. K. Bhartia, K. Yang, A. J. Krueger, Z. Li, P. F. Levelt, H. Chen, P. Wang,and D. Lu, “Validation of SO2 retrievals from the Ozone MonitoringInstrument (OMI) over NE China,” J. Geophys. Res., vol. 113, no. D16,p. D16 S40, 2008.

[63] A. Berk, L. S. Bernstein, and D. C. Robertson, “MODTRAN: A moderateresolution model for LOWTRAN7,” Air Force Geophys. Lab., Bedford,MA, Rep. GL-TR-89-0122, 1989.

[64] G. P. Anderson, F. X. Kneizys, J. H. Chetwynd, J. Wang, M. L. Hoke,L. S. Rithman, L. M. Kimball, R. A. McClatchey, E. P. Shettle,S. A. Clought, W. O. Gallery, L. W. Abreu, and J. E. A. Selby,“FASCODE/MODTRAN/LOWTRAN: Past/present/future,” in Proc.18th Annu. Rev. Conf. Atmos. Transm. Models, Jun. 6–8, 1995.

[65] B. Fieque, J. L. Tissot, C. Trouilleau, A. Crastes, and O. Legras, “Un-cooled microbolometer detector: Recent developments at Ulis,” InfraredPhys. Technol., vol. 49, no. 3, pp. 187–191, 2007.

[66] Z. Wan and J. Dozier, “A generalized split-window algorithm for retriev-ing land-surface temperature from space,” IEEE Trans. Geosci. RemoteSens., vol. 34, no. 4, pp. 892–905, Jul. 1996.

[67] Oxford University AOPP Mie Code. [Online]. Available: http://www.atm.ox.ac.uk/code/mie/index.html

[68] F. E. Volz, “Infrared optical constants of ammonium sulfate, Sahara dust,volcanic pumice and fly ash,” App. Opt., vol. 12, no. 3, pp. 564–568,Mar. 1973.

[69] A. J. Prata and I. F. Grant, “CSIRO Atmospheric Research TechnicalPapers, 48,” Commonw. Sci. Ind. Res. Organ., Melbourne, Australia,2001.

Stefano Corradini received the B.S. degree inphysics from the University of Modena and ReggioEmilia, Modena, Italy, in 1999, and the Ph.D. de-gree in geophysics from the University of Genova,Genova, Italy, in 2004.

He is currently a Researcher with the ItalianNational Institute of Geophysics and Volcanology(INGV), Rome, Italy. He is involved in projectssupported by the European Space Agency, the ItalianSpace Agency, and the Italian Civil Protection ded-icated to the volcanic-eruptions risk mitigation on

society and transports. His research is focused on atmospheric remote sensingand in particular on volcanic-plume sulfur dioxide and ash retrievals in thethermal infrared spectral range using satellite platforms.

Cecilia Tirelli received the B.S. degree in physicsfrom the University of Modena e Reggio Emilia,Modena, Italy, in 2007. She is currently work-ing toward the Ph.D. degree in remote sensingat the University of Rome “La Sapienza,” Rome,Italy. Her studies are focused on remote sensingof tropospheric aerosol with active and passiveinstruments.

She is involved in Italian projects dedicated toaerosol remote sensing in the urban area of Rome, inthe central Mediterranean, and in the Arctic regions.

Gabriele Gangale received the B.S. degree inphysics from the University of Modena and ReggioEmilia, Modena, Italy, in 2005, where he is currentlyworking toward the Ph.D. degree in Earth SystemSciences Doctorate School.

He collaborates with the Italian National Insti-tute of Geophysics and Volcanology (INGV), Rome,Italy within the ASI-SRV project, dedicated to themitigation of volcanic-eruption risk. He also collab-orates with the Norwegian Institute for Air Research(NILU) in a project dedicated to plume detection

from the AIRS satellite.

Sergio Pugnaghi received the B.S. degree in physicsfrom the University of Modena and Reggio Emilia,Modena, Italy, in 1975.

He has been a Physicist, Senior Researcher withthe Science Faculty, University of Modena andReggio Emilia, Modena, Italy since 1980. He isresponsible for the atmospheric measurements andthermal infrared data analysis of the “Mitigationof Volcanic Risk by Remote Sensing,” EuropeanProject, for many others projects of the Italian SpaceAgency (ASI), and also for projects of the Italian

National Group of Volcanology (INGV), Rome, Italy.

Elisa Carboni received the B.S. degree in physicsand the Ph.D. degree in physical modeling for envi-ronmental protection from the University of BolognaBologna, Italy, in 1998 and 2004, respectively.

Since 2004, she has been a Postdoctoral. Re-search Assistant with the Atmospheric, Oceanic, andPlanetary Physics Department, University of Oxford,Oxford, U.K. She is currently working on volcanicatmospheric emission, SO2, and ash, supported bythe U.K. National Center for Earth Observation.Her experiences include the following: aerosol re-

trieval from satellite spectral radiance measurements (MSG SEVIRI, ATSR/2,AATSR, GOME and SCIAMACHY data) in the UV, VIS, and IR spectralrange; development and implementation of radiative-transfer model for aerosoland cloud; and multispectral sun photometer measurements.


Theoretical Study on Volcanic Plume SO and Ash Retrievals ...carboni/doc/TGRS... · Stefano...

Documents

Transcript of Theoretical Study on Volcanic Plume SO and Ash Retrievals ...carboni/doc/TGRS... · Stefano...