Improved prediction and tracking of volcanic ash clouds

Journal of Volcanology and Geothermal Research 186 (2009) 1–9

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research

j ourna l homepage: www.e lsev ie r.com/ locate / jvo lgeores

Improved prediction and tracking of volcanic ash clouds

Peter Webley a,⁎, Larry Mastin b

a Arctic Region Supercomputing Centre (ARSC)/Alaska Volcano Observatory (AVO), University of Alaska Fairbanks, 909 Koyukuk Drive, Fairbanks, Alaska. 99775-6020 USAb United States Geological Survey (USGS), Cascades Volcano Observatory (CVO), 1300 SE Cardinal Court, Building 10, Suite 100. Vancouver Washington. 98683, USA

⁎ Corresponding author. Tel.: +1 907 474 1542; fax: +E-mail addresses: [email protected] (P. Webley)

0377-0273/$ – see front matter © 2008 Elsevier B.V. Adoi:10.1016/j.jvolgeores.2008.10.022

a b s t r a c t

Article history:

Keywords:volcanic ash cloudseruption source parametershazard assessment

ore than 100 airplanes have inadvertently flown through clouds of volcanic ashfrom erupting volcanoes. Such encounters have caused millions of dollars in damage to the aircraft and haveendangered the lives of tens of thousands of passengers. In a few severe cases, total engine failure resultedwhen ash was ingested into turbines and coating turbine blades. These incidents have prompted theestablishment of cooperative efforts by the International Civil Aviation Organization and the volcanologicalcommunity to provide rapid notification of eruptive activity, and to monitor and forecast the trajectories ofash clouds so that they can be avoided by air traffic. Ash-cloud properties such as plume height, ashconcentration, and three-dimensional ash distribution have been monitored through non-conventionalremote sensing techniques that are under active development. Forecasting the trajectories of ash clouds hasrequired the development of volcanic ash transport and dispersion models that can calculate the path of anash cloud over the scale of a continent or a hemisphere. Volcanological inputs to these models, such as plumeheight, mass eruption rate, eruption duration, ash distribution with altitude, and grain-size distribution, mustbe assigned in real time during an event, often with limited observations. Databases and protocols arecurrently being developed that allow for rapid assignment of such source parameters. In this paper, wesummarize how an interdisciplinary working group on eruption source parameters has been instigatingresearch to improve upon the current understanding of volcanic ash cloud characterization and predictions.Improved predictions of ash cloud movement and air fall will aid in making better hazard assessments foraviation and for public health and air quality.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

Volcanic ash is one of the major hazards from volcanic eruptions.Volcanic ash can disrupt local communities, severely damage jetengines, abrade surfaces, clog air filters, obstruct surface transporta-tion, block municipal water and drainage pipes, and shut down majorairports (Blong, 1984; Casadevall, 1993). Volcanic ash drifts into airroutes at altitudes greater than 30,000 ft for roughly 20 days per yearworldwide (Miller and Casadevall, 2000). The effect of ash clouds onaviation is not just local. Casadevall and Krohn (1995) illustrate howthe 1992 Crater Peak eruption cloud of Mount Spurr reached as far asthe Cleveland Air Traffic Control Center, approximately 5000 km fromthe volcano. Schneider et al. (1995) show that Spurr eruption cloudswere seen in satellite images as far away as Greenland. Simpson et al.(2002) describe a possible encounter in Northern California from the2001 Mount Cleveland eruptions, approximately 3000 km from thevolcano. Accurately predicting the movement of ash clouds up to 24 hin advance is critical to aviation safety. Several cases of engine

1 907 450 8603., [email protected] (L. Mastin).

ll rights reserved.

flameout and resulting near-crashes have dramatically demonstratedthe effect of volcanic ash on aircraft (Miller and Casadevall, 2000).

In the past 15 years, the International Civil Aviation Organization(ICAO), through the International Airways VolcanoWatch (IAVW), hasaddressed this threat and established protocols to avoid subsequentash encounters by aircraft (ICAO, 2004a). Explosive eruptions canform an ash cloud that threatens aviation in a matter of minutes. Therapid onset of such hazards has prompted the aviation community topropose an ideal notification time of five minutes following eruptiononset to provide information on the ash clouds location and futuremovement (Foreman, 1994; Hufford et al., 2000; Tupper et al., 2007),which is among the most challenging goals of any natural-hazardswarning system.

In order to mitigate hazards and address scientific questionsregarding interpretation of volcanic behavior, volcanologists andatmospheric scientists have developed an array of increasinglysophisticated instruments, analysis tools, and numerical models. Thestudy of volcanic ash clouds has also become increasingly multi-disciplinary, as volcanologists work with atmospheric scientists, fluiddynamics modelers, and instrument developers. This paper describessome of the most pressing issues regarding ash clouds and theirhazards, and how the contributions in this volume have advanced ourunderstanding of these issues.

2 P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

2. Advances in aviation and community response

Advances in detection, monitoring, and characterization ofvolcanic ash clouds have been followed by improved communicationbetween the volcanological and aviation communities, and betterprotocols that have dramatically reduced, but not eliminated, thenumber of hazardous ash-aircraft encounters.

A cause of engine damage results from the fact that volcanic ashhas a melting point below jet engine operating temperatures withthrust settings above idle. The ash melts in the hot section of theengine and fuses on the high-pressure nozzle guide vanes and turbineblades. In a few severe cases, near-fatal crashes resulted when ash wasingested into turbines, coating turbine blades and causing total enginefailure. The first such case of engine failure to catch the attention of theaviation community occurred on June 24, 1982. A B747 aircraft lostpower on all four engines while flying at 11.3 km (37,000 ft) fromKuala Lumpur, Malaysia to Perth, Australia (ICAO, 2004b). The aircraftdescended without power from 11.3 km to 3.65 km (37,000 ft to12,000 ft) over a 16-min period, at which point the pilot was able torestart three of the engines and make a successful emergency landingat Jakarta, Indonesia. (ICAO, 2004b). Suspicion arose that a volcanicash cloud might be the culprit. Some three weeks later anotheraircraft, a B747 bound for Melbourne, Australia, reported a similarincident. This time power was lost on two engines and the aircraft alsodiverted successfully to Jakarta. To meet this new threat, ICAOdeveloped a set of interim guidelines to assist in the disseminationof information on volcanic ash to pilots and the development ofcontingency arrangements for the diversion of aircraft around affectedareas.

In 1987, ICAO established the International Airways VolcanoWatch(IAVW) (Romero, 2004). Over the next 5 years, several explosiveeruptions all affected aviation. For example, an eruption on December15 1989 at Redoubt volcano, Alaska resulted in a nearly catastrophicencounter between airborne ash and a B747 jet aircraft (Casadevall,1994). ICAO then started the process to designate nine regionalvolcanic ash advisory centers (VAACs): Anchorage, Washington,Buenos Aires, Montreal, London, Toulouse, Tokyo, Darwin and Well-ington. The VAAC's mission is to ensure that volcanic cloud hazardsare identified when they occur (producing advisories to the aviationcommunity) and are an interface between volcano observatories,meteorological watch offices (which issue SIGMET warnings) and airtraffic control centers. VAACs use satellite information, ground andpilot reports, meteorological knowledge and numerical models totrack and forecast ash movements so that aircraft can be routedaround the airborne ash safely.

In addition to the creation of the IAVW and the VAACS, twointernational conferences on volcanic ash and aviation safety haveserved to increase awareness of ash hazards to aviation among thescientific and aviation communities. A recently completed handbookof the IAVW (ICAO, 2004a) provides documentation on what isrequired by each party and the actions to be taken during a volcaniccrisis. The last twenty years have certainly seen the emergence ofvolcanic ash as a serious hazard and financial cost to aircraft andaerodrome operations and through ICAO, the IAVW has beendeveloped to respond to this threat (ICAO, 2004b). The United Stateshas developed a, interagency national volcanic ash operations plan foraviation in support of the ICAO IAVW (US Department of Commerce/NOAA, 2007).

3. Improved characterization of volcanic cloud development atthe vent

Volcanic clouds most commonly originate as plumes from volcanicvents. The height of these plumes reach is of immediate concern to airtraffic. Also important is the rate ofmass loading of the atmospherewithash, and the fraction of ash that is fine enough to remain in the cloud for

hours or days. A long history of analytical work (Morton et al., 1956;Settle, 1978), one-dimensional numerical calculations (Woods, 1988,1993), and empirical compilations (Wilson et al., 1978; Sparks et al.,1997, Chapter 5) show that plumeheight (H) correlates roughlywith thefourth root of mass eruption rate (M), however there is considerablescatter in empirical plots of H versus M (Mastin et al., 2009-thisvolume). This scatter likely reflects bothmeasurement error and a rangeof physical processes and conditions such as elutriation of pyroclastic-flow debris, variably humid air, and wind. Reducingmeasurement errorand developing realistic models of these processes and conditions arekeys to reducing risk and improving understanding.

Measurements of volcanic cloud height, by visual observation,radar reflectivity, and several satellite techniques, have sometimesdisagreed by kilometers from one another depending on theobservational method and the observer (e.g. Tupper et al., 2007).The discrepancies are not simply an artifact of the technique, asdifferent observers can record different heights using the sametechnique. In the first systematic compilation of its type, Tupper andWunderman (2009-this volume) compare hundreds of plume heightestimates made by various methods at Indonesian and New Guineanvolcanoes in the past five years. Surprisingly, they find that ground-based observations are strongly skewed toward lower heights thansatellite or radar observations. They propose a systematic method ofcompiling and averaging multiple observations during large eruptionsto improve real-time cloud-height accuracy.

Estimates of mass eruption rate, given the eruption duration, cansimilarly vary bymore than a factor of two for a given eruption, as theyare based on measurements of the thickness distribution of a tephradeposit and some method of integrating thickness within and beyondthe mapped area (Pyle, 1989; Fierstein and Nathenson, 1992; Rose,1993; Bonadonna et al., 1998). These are from empirical methods.Forward modeling methods have been developed, Carey and Sparks,1986; Wilson and Walker, 1987, and also rise heights and masseruption rates have been calculated from studies on the change inmass in give size fractions (Bursik et al., 1992) Such estimates can onlybe made well after an eruption, at a well-studied volcano, and are atbest an indirect indicator of ash-cloud mass loading rate, as only thefinest ash fraction, less than perhaps several tens of microns (µm),remains in the cloud after the first half hour (Rose, 1993). In somecases mass loading rates have been estimated from the growth rate oflarge umbrella clouds in satellite images (Sparks et al., 1986, 1997;Koyaguchi and Tokuno, 1993, Chapter 11), although the uncertainty inthese estimates is large and such estimates have similarly been madeonly in post-analysis, after an eruption is long over. To date, estimatesof the rate of mass loading during an eruptionmust rely on theoreticaland empirical correlations between plume height and eruption rate,with full knowledge of the range of uncertainty inherent in suchrelationships.

The fine ash fraction of the total eruptedmass has been the focus ofrelatively recent work. Several studies have attempted to measure thegrain-size distribution of an entire erupted mass, generally bymeticulously sampling an entire erupted deposit, measuring grain-size distributions of all samples, and performing a weighted sum ofthe results (Murrow et al., 1980; Carey and Sigurdsson, 1982;Bonadonna and Houghton, 2005; Rose et al., 2007; Durant et al., inreview). These results show that the mass fraction of ash less than 63μm can range can range from a few percent from small basalticeruptions to more than 50% for large silicic ones, especially those thatcontain elutriated pyroclastic-flow debris (Mastin et al., 2009-thisvolume; Rose and Durant, this volume).

4. Improved understanding of physical processes that affectash-clouds and their hazards

Many factors affect the plume height, particle size distribution, rateof particle fallout, and ultimate lifetime of volcanic ash clouds. Among

3P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

the most important are water condensation, ice nucleation, wind, andturbulence effects. Several studies in this volume have improved ourunderstanding of these processes.

4.1. Water condensation

In humid and especially tropical environments, condensation ofentrained water vapor in the eruptive plume is thought to boostconvection, increasing the height of eruption columns by kilometersabove their height in dry air. One-dimensional plumemodels (Woods,1993; Sparks et al., 1997, Chapter 4; Mastin, 2007) suggest a boost inplume height of up to a few kilometers, primarily for small tomedium-sized events (b~106 kg/s). Tupper et al. (in press) presentresults from the multidimensional, multi-component ATHAM code(Oberhuber et al., 1998) which suggest a still stronger effect on plumeheight, causing even small tropical plumes to rise to the tropopause,potentially affecting air traffic. Their results also suggest that thetropopause controls plume height more strongly than predicted byone-dimension models, holding plumes to this elevation over morethan an order of magnitude range in eruption rate. Tupper et al. (inpress) also discuss how the variation in tropopause height withlatitude can cause eruptions of similar magnitude to have verydifferent maximum plume heights.

4.2. Ice nucleation

Like water condensation, the growth of ice crystals in a plumereleases latent heat that can help buoy the plume to greater height. Icecrystallization adds only about 10% as much energy to a plume aswater condensation and hence does not strongly affect plume height(Woods, 1993). It does affect rates of particle aggregation and fallout;may mask the spectral characteristics of ash in satellite images; and,for large eruptions, may mitigate the atmospheric effects of sulfurdioxide (SO2) by encasing sulfur aerosols (Rose et al., 1995a). Thenucleation and growth of ice is more complicated than watercondensation. Liquid water easily nucleates on microscopic surfaces,ensuring that the amount of liquid in a cloud at above-freezingtemperature is roughly the equilibrium value. Once the temperaturedrops below freezing however, ice may not begin to form untiltemperatures drop below−10 °C. Between about−10 °C and−40 °C,liquid water and ice coexist. The fraction of disequilibrium liquid orvapor that may exist in a below-freezing cloud depends on itsresidence time at that temperature, the abundance of nucleation sites,the type of nuclei and the ice growth rate, which could be suppressedby components such as nitric acid (Gao et al., 2004). One-dimensionalplume models either do not consider ice nucleation (Woods, 1993), oruse a simple linear relationship with temperature for mass fractions ofcoexisting liquid and ice (Khairoutdinov and Randall, 2003; Mastin,2007). The multidimensional ATHAM model uses an exponentialfunction of temperature to calculate crystal number density (Herzoget al., 1998, Eq. (7)) and explicitly considers snow scavenging butignores ash-ice interaction.

Durant et al. (2008) report on freezing experiments that showlittle sensitivity of ice nucleating ability with ash composition, andsuggest that volcanic clouds are sufficiently ash-rich to nucleate alarge number density of small-diameter ice crystals in a fairly narrowtemperature range of ~250–260 K. Durant and co-workers (Durantand Rose, 2009-this issue; Durant et al., in review) also provideevidence for hydrometeor-enhanced sedimentation at Mount St.Helens in 1980 and Mount Spurr, Alaska, in 1992, pointing outamong other things that this process may produce the downward-bulging mammatus features visible in some ground-based photo-graphs of the Mount St. Helens 1980 ash cloud. The evaporation orsublimation of hydrometeors during descent in the atmosphere mayenhance fallout without producing rain or snow at ground level. Theconditions under which tephra fallout is not enhanced by hydro-

meteors is of particular interest, as such conditions may result inespecially long-lived ash clouds. Althoughmost ash clouds dissipate ina matter of days, others stay in the atmosphere for weeks, damagingaircraft as long as 20 days after eruptive events (Tupper et al., 2003).

4.3. Wind and turbulence

Although the effects of wind on ash-cloud transport may seemobvious, many aspects of this process are just now being explored.Volcanic plumes that rise at a rate much greater than the averageatmospheric wind speed tend to produce strong plumes consisting ofvertical columns topped by umbrella clouds. Plumes that rise at a rateslower than the averagewind speedwill form bent-over weak plumes.Relationships between depositional patterns, the plume height, andaverage wind speed were established in the 1980s (Carey and Sparks,1986), but for the most part assume constant horizontal wind. Realwind patterns are known to vary with altitude. Bursik et al. (2009-thisvolume) interpret long, narrow tephra deposit distributions atAvachinsky, Kamchatka, Russia and South Deadman Dome (InyoVolcanic Chain) to have resulted from entrainment of the volcanicplume into the jet stream. Because the jet stream is used by commercialaircraft to reduce flight times and conserve fuel, ash-cloud entrainmentin the jet stream can increase the probability of aircraft encounters.Moreover, plumes that intersect high-velocity winds near the tropo-pause elevation but extend higher into the stratosphere may have ashclouds whose leading edge lies at the at tropopause elevation ratherthan near the plume top, as observed at Mount St. Helens on May 18,1980 (Sarna-Wojcicki et al., 1981). Bursik et al. (2009-this volume)point out that, like water condensation processesmentioned above, jet-stream winds can constrain the plume height to the tropopauseelevation overmore than an order of magnitude range of eruption rates.

The effects of turbulence have been relatively poorly studied inash-cloud models, although turbulence influences the rate of atmo-spheric dispersal as well as the rate of particle fallout. Tephra falloutmodels generally assume that particles fall at their own terminalvelocity in still air, (for example ASHFALL; Hurst, 1994) although theysimultaneously assume that particles are well-mixed by turbulencewithin the umbrella cloud. Based on experiments of particle fallout inturbulentwater, Koyaguchi et al. (2009-this volume) argue thatparticlesshould be homogeneously distributed in the turbulent cloud if its rootmean square (RMS) velocity exceeds the terminal fall velocity of theparticles. Three-dimensional plume models suggest that RMS velocityfluctuations in a plinian umbrella cloud near the margins of the verticalcolumnmay be only great enough to homogenize particles smaller thana fewmillimeters in diameter. Larger particles may settle to the bottomof the cloud and eventually fall out. Turbulent velocity fluctuations mayalso contribute to the growth of raindrops or hail by accretion, whichmay affect fallout rates.

5. Improvements in detection and monitoring of volcanicash clouds

In general, immediate notification of eruptive activity requires thepresence of geophysical (mostly seismic) instruments; or nearbyground observers, daylight, good weather, and good communication.Most of the world's volcanoes however still lack local seismicnetworks, and evenwhere they are present, the rapid onset of seismicactivity can't be confirmed to record an eruption without other obser-vations. Radar, lightning detection systems (Hoblitt, 1994), dropoutsin radio telemetry signals (Qamar et al., in press), ground-based vi-sual observations and satellite imagery have all been used to verifyeruptive activity that was suggested by seismic signals. At remote,non-instrumented volcanoes, the appearance of an ash cloud isusually detected first through satellite imagery. The detection timeof eruptions from satellite images varies widely depending on thesatellite's orbital path, the remote sensing platform, methods of image

4 P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

processing, and the vigilance of humans reviewing the images, amongother factors.

5.1. Satellite detection and monitoring

Volcanic ash clouds can be detected with satellite data over verylarge distances. Schneider et al. (1995) show for the August 1992Crater Peak, Mount Spurr Alaska eruptive event that the ash cloudscould be tracked, using the split window method (Prata, 1989a,b), upto three days after the onset of the eruption. Remote sensing satellitedata are used by VAACs and volcano observatories to detect volcanicash clouds, both those that are spectrally opaque and translucent.Operationally, data can be used from the National Oceanic andAtmospheric Administration (NOAA) Polar Operational Environmen-tal Satellite (POES), National Aeronautical and Space Administration(NASA) Aqua and Terra satellites, the Multi-Functional TransportSatellite (MTSAT), and the Geostationary Operational EnvironmentalSatellite (GOES). For spectrally opaque clouds, thermal infraredsatellite data, centralized at 10.5 µm, are used for ash cloud detection(Dean et al., 2004). Satellite data in this range are sensitive to the coldtemperatures of eruption clouds, and hence are the preferred band todetermine cloud-top temperatures. Once an ash cloud is spectrallytranslucent, then the split window, Prata (1989a,b), detectiontechnique is used. Tupper et al. (2004) examine the differenttechniques for ash-cloud detection from satellite data and state thatthere is no single method of observation or sensor able to detect andtrack every case of ash clouds, but that the split window method iscurrently the best method. However, Simpson et al. (2000) and Prataet al. (2001) discuss some occasions where the split window isunsuccessful. Dean et al. (2002) provide a description of the AVOsystem, which provides near-real-time automated detection of ashclouds. However, Webley et al. (2009-this volume) discuss some ofthe non-volcanic features that can produce a split-window signal inthe satellite data and provide some possible methods to eliminatethese from automated detection.

Once a volcanic ash cloud is detected, techniques such as describedby Wen and Rose (1994) and Yu and Rose (1999) can be applied forash retrievals, providing information on particle size, mass loadingsand ash cloud volumes. The mass loadings and cloud volumes reflectthe mass and volume of the cloud containing particles that aredetectable by the satellite data, 1–10 µm. However, as Prata and Grant(2001) state atmospheric correction is required as high levels of watervapor may hide the split-window signal and so under-detect the ashcloud.

Satellite remote sensing data can also be used for plume heightestimations by correlating infrared temperatures from opaque ashclouds with the temperature at a given elevation obtained byradiosonde profiles (Holasek et al., 1996). This information isinvaluable in the early development of ash clouds and can be used ascloud heights in volcanic ash dispersion and transport modelinitialization. Issues can occurwhen an ash cloud reaches stratosphericheights or if the vertical profile shows a consistent temperature withheight. Accurate and consistent height estimates are still one of themajor issues for volcanic ash clouds. Once the ash clouds are spectrallytranslucent and the split window technique applicable, plume heightestimates from infrared data are less reliable. Fig. 1 shows an exampleof an ash cloud as detected from the international space station (ISS)and the MODIS satellite signal seen once data was available. For thisevent, the remote sensing data analyzed within hours of the ISSobservation. Here the ash cloud was detectable with the split windowmethod as the cloud had become spectrally translucent.

5.2. Ground-based detection and monitoring

In addition to the satellite data, ground observations are used todetect andmonitor volcanic ash clouds. Both specialized and standard

weather radar systems have been used to detect new and developingvolcanic ash clouds. Schneider et al. (2008) showed how NEXRADweather radar was used to detect the volcanic ash clouds fromMountAugustine in 2006, providing information on plume height, plume-rise velocity and particle size. They showed that fine ash, of approx10 µm size, was detectable, if in clear air mode, and was tracked forseveral hours after the start of the eruption. Rose et al. (1995b)showed how C-band radar (5 cm wavelength) can be used todetermine ash-plume heights and provide information on ashparticles, sized from 1 mm to centimeters. However, they note thatthe C-band radar underestimates the height, as it was unable todetect the finer particles which rose higher, and that it does notdetect the distal (also fine) ash, which has a residence time of greaterthan 30 min. Schneider et al. (2008) state that the NEXRAD radar hasa height resolution of ±1.5 km. However, another important issue isthat the data is not a true cross section of the plume, but is dependenton the scan angle used.

Recently, Sassen et al. (2007) and Webley et al. (2008b) showedthat LIDAR observations can be used to detect distal ash clouds inBarrow Alaska, from the 2006 Mount Augustine eruption, more than1000 km from the volcano. Also, thermal imaging cameras are nowbeing developed for the detection and discrimination of volcanic ashand sulfur dioxide from volcanoes. Prata and Bernando (2009-thisvolume) show that these cameras, through the same process as usedin the split window technique for satellites, can be used to detect newvolcanic ash clouds, provide information on particle size. Well-positioned cameras of this type can rapidly detect ash clouds atnight, providing a tool for quick detection of eruptive activity.

Other ground-based data that can be used to monitor eruptionclouds and their subsequent dispersion include in-situ ashfallsampling, such as Cahill et al. (in prep), as well respective ashfalldistribution sampling, Bonadonna and Houghton (2005). Thesetypes of datasets allow us to understand the ashfall particle sizes,their temporal and spatial distributions and relate these back to themodel operational forecasts. Improvements in understanding theseprocesses will help models accurately predict ash cloud movementin future, as well as providing information that can be related backto initial ash cloud volumes and particle-size distributions.

6. Modeling and forecasting ash-cloud paths

Predicting ash-cloud behavior is necessary in order to mitigate ashhazards to aviation (Peterson et al., 2008). Near real-time satelliteremote sensing cannot be solely relied upon to provide all necessaryinformation on ash-cloud properties and no remote sensing techniqueprovide a forecast for the ash-cloud trajectory. To help forecast ash-cloud movement, volcanic ash transport and dispersion (VATD)models have been developed as an additional tool to comprehensivelymonitor volcanic ash clouds. These models developed initially fortracking explosion clouds from nuclear events, Pudykiewicz 1988,with the HYSPLIT model (Draxler and Hess, 1998) being developed fortracking a suite of atmospheric applications (for example ash,pollution, nuclear and chemical). The primary users include opera-tional forecasters of potential volcanic hazards, located in VAACs andmeteorological watch offices.

6.1. Operational VATD models

These models are necessary in forecasting the movement ofvolcanic ash clouds during an eruption. The model forecasts arecomplemented and validated by remote sensing data and visualobservations from the ground and aircraft. The models are importantfor analyzing data after a volcanic eruption in an attempt to betterunderstand the event. Globally, several models are in use forpredicting the ash clouds from erupting volcanoes by both VAAC'sand volcano observatories. The Alaska Volcano Observatory, U.S. Air

Fig. 1. (a) Astronaut photograph (ISS013E24186) of the May 23, 2006 eruption of Cleveland volcano. Image from Science and Analysis Laboratory, NASA-Johnson Space Center. (b) MODIS visible satellite image from 2358 UTC on May 23, 2006(image from Rick Wessels of AVO) and (c) MODIS split window signal image from 2358 UTC on May 23, 2006 (a1.06143.2358).

5P.W

ebley,L.Mastin

/Journal

ofVolcanology

andGeotherm

alResearch

186(2009)

1–9

Fig. 2. (a) Puff, (b) CaNERM and (c) HYSPLIT model simulation outputs for the January 2006 Mount Augustine eruption. (Adapted from Peterson et al., 2008).

6P.W

ebley,L.Mastin

/Journal

ofVolcanology

andGeotherm

alResearch

186(2009)

1–9

7P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

Force Weather Agency (AFWA), and Anchorage, Wellington andWashington VAACs use the Puff (Searcy et al., 1998) model. The TokyoVAAC uses an adapted version of the Puff model. The Buenos Aires,Darwin, Anchorage and Washington VAAC's both use the HybridSingle-Particle Lagrangian Integrated Trajectories (HYSPLIT, Draxlerand Hess, 2004) model. The Montreal VAAC uses the CanadianEmergency Response Model (CanERM, Servranckx et al., 1996), theLondon VAAC use the Numerical Atmospheric dispersion ModelingEnvironment (NAME, Ryall andMaryon,1998) and the Toulouse VAACuse the MEDIA (Sandu et al., 2003) modeling environment. Fig. 2shows an example from the Puff, HYSPLIT and CaNERM modelsimulations from the January 2006 eruption at Mount Augustine,Alaska, adapted from Peterson et al. (2008). These models requiremeteorological numerical weather prediction data to provide thewind field source for the dispersion of the ash clouds. As a result, themodel predictions are only was reliable as the forecast wind fields andso errors in these wind fields can propagate into the VATD modelforecasts.

6.2. Research VATD models

In addition to the operational VATD models, more sophisticatedfluid-dynamics models are used to study eruption processes that arenot included in the VATD models. An example of such a model isATHAM, Active Tracer High Resolution Atmospheric model (Graf et al.,2002; Textor et al., 2006a,b). Tupper et al. (in press) use ATHAM toexplore the sensitivity of plume height to tropospheric instability andshow how plume height can vary with latitude and time of year forsimilar weak or strong eruptions. Such models are invaluable forunderstanding the dynamical rise of volcanic plumes and effects ofaggregation, hydrometeor interaction and coalescence on plumemovement and particle fallout. In operational VATD models, suchprocesses are either not included or are parameterized to allow quickmodel results. Although the analysis of plume movement using suchmodels is time consuming and impractical operationally, the resultsdo provide an understanding that can be fed back into the operationalmodels.

6.3. Model validation

Recent studies have focused on validating VATD model predic-tions by comparing them with ground-based datasets and satelliteremote sensing data. Such studies include Dean et al. (2004) for the2001 eruption of Mount Cleveland, Alaska and Webley et al.(2008a) for the 2006 eruption of Mount Augustine, Alaska. Webleyet al. (2009-this volume) describe how the split window techniquewas applied to the 1992 eruption of Mount Spurr, Alaska to analyzethe sensitivity of model forecasts to the initial conditions. However,they note that the satellite data may not detect all the ash cloud,especially the finer ash particles. Moreover, Fero et al. (2009-thisvolume) analyze the ability of VATD models to provide thecomplete picture of the 1991 Mount Pinatubo eruption and thetephra deposit, through comparisons with satellite data andpublished reports of the ashfall deposits. These validation experi-ments determine the accuracy of the modeled forecasts, where anydiscrepancies exist, and allow analysis of how the model forecastscan be improved for both volcanic ash dispersion and tephra fall.Analysis of past ashfall distributions to validate with the opera-tional model predictions (Durant and Rose, 2009-this volume) hasbeen approached to understand the ashfall predictions that theVATD model produce. Webley et al. (2009-this issue) illustrate atechnique for assessing model predictions directly from thesatellite data. This validation study evaluates model accuracy,inter-model discrepancies, and helps identify potential modelimprovements for both volcanic ash-dispersion and tephra-depositforecasting.

7. Development of Eruption source parameters for VATD models

In the late 1990s, the IAVW identified a need for improved volcanicsource parameters to be used in model forecasts of ash clouds duringvolcanic eruptions. These parameters include plume height, masseruption rate, duration, and the fraction of the erupted massconsisting of fine ash that can remain in the cloud for hours or days.These parameters are difficult to assess during an eruption, especiallyin the first hours when few observations are available. The U.S.Geological Survey responded to this need in 2007 by organizing the“Eruption Source Parameter Working Group”, consisting of volcanol-ogists and meteorologists familiar with analyzing and modelingtephra eruptions. This group has established source parameters foreleven categories of eruptions, from small silicic to large mafic and hasassigned most likely eruption types and source parameters to theroughly 1500 Holocene volcanoes of the world based on past behavior(Mastin et al., 2009-this volume). The source parameters for eacheruption type are based onwell-documented eruptions in which suchparameters have been well constrained by observations. For example,the June 17, 1996 eruption of Ruapehu volcano, New Zealand (BGVN,1996) is assigned as a type eruption, as is the October 14, 1974eruption of Fuego Volcano, Guatemala (Rose et al., 1978). These typicaleruptions havewell-studied eruption source information and are usedas base parameters for all the volcanoes with this eruption type. Thecategories may also be used to revise source parameters during aneruption based on changes in observed plume height or duration.

8. Summary of the special volume

This volume derives from a session at the American GeophysicalUnion (AGU) fall conference entitled “Observations and Techniques toImprove Prediction andTracking of Volcanic AshClouds”. The papers inthis volume cover many important aspects of the physical properties,processes, and monitoring techniques of ash clouds; and research intoimproved tracking of ash-cloudmovement and behavior. Volcanic ash-cloud detection and prediction is a rapidly evolving research field andnew ideas and tools are being developed to improve forecasts andbetter understand our representation of volcanic processes withinVATD models. Future operational VATD models will likely incorporatesome processes currently in research models, such as particleaggregation, and may combine with tephra-fall models to accuratelypredict both ash-cloud movement and tephra deposit distribution.

Acknowledgements

We thank Marianne Guffanti at the U. S. Geological Survey andRene Servranckx at the Canadian Meteorological Centre and MontrealVAAC for their liaisons between the eruption source parametersworking group and the International Airways Volcano Watch Opera-tions Group. Also, we thank two USGS reviewers for useful andinformative comments. Finally, we thank all the presenters at the V31session of the AGU 2007 fall conference and those authors whosubmitted papers to this special volume: Volcanic Ash Clouds for theJournal of Volcanology and Geothermal Research.

References

Blong, R.J., 1984. Volcanic Hazards: A sourcebook on the effects of eruptions. AcademicPress, Sydney. 424 pp.

Bonadonna, C., Houghton, B.F., 2005. Total grain-size distribution and volume of tephra-fall deposits. Bulletin of Volcanology 67, 441–456.

Bonadonna, C., Ernst, G.G.J., Sparks, R.S.J., 1998. Thickness variations and volumeestimates of tephra fall deposits: the importance of particle Reynolds number.Journal of Volcanology and Geothermal Research 81, 173–187.

Bulletin of Global Volcano Network,1996. Ruapehu VolcanoMonthly report: September1996. BGVN 20:09.

Bursik, M., Carey, S., Sparks, R., 1992. A gravity current model for the May 18, 1980Mount St. Helens Plume. Geophysical Research Letters 19 (16), 1663–1666.

8 P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

Bursik, M.I., Kobs, S.E., Burns, A., Braitseva, O.A., Bazanova, L.I., Melekestsev, I.V.,Kurbatov, A., Pieri, O.C., 2009. Volcanic plumes and wind: Jetstream interactionexamples and implications for air traffic. Journal of Volcanology and GeothermalResearch 186, 60–67 (this volume).

Cahill, C.F., Rinkleff, P.G., Dehn, J., Webley, P.W., Cahill, T.A., Barnes, D.E., 2009. AerosolMeasurements from a Recent Volcanic Eruption: Implications for Volcanic AshTransport Predictions. Journal of Volcanology and Geothermal Research. In prep.

Carey, S., Sigurdsson, H., 1982. Influence of particle aggregation on deposition of distaltephra from the May 18, 1980, eruption of Mount St. Helens volcano. Journal ofGeophysical Research 87 (B8), 7061–7072.

Carey, S., Sparks, R.S.J., 1986. Quantitative models of the fallout and dispersal of tephrafrom volcanic eruption columns. Bulletin of Volcanology 48, 109–125.

Casadevall, T.J., 1993. Volcanic hazards and aviation safety, lessons of the past decade.FAA Aviation Safety Journal 2, 1–11.

Casadevall, T.J., 1994. The 1989–1990 eruption of Redoubt Volcano Alaska: impacts onaircraft operations. Journal of Volcanology and Geothermal Research 62 (30),301–316.

Casadevall, T.J., Krohn, D.,1995. Effects of the 1992 Crater Peak eruptions on airports andaviation operations in the United States and Canada. U.S. Geological Survey Bulletin2139, 205–220.

Dean, K., Dehn, J., Engle, K., Izbekov, P., Papp, K., Patrick, M., 2002. Operational satellitemonitoring of volcanoes at the Alaska Volcano Observatory. Advances inEnvironmental Monitoring and Modelling 1 (1), 3–35.

Dean, K.G., Dehn, J., Papp, K.P., Smith, S., Izbekov, P., Peterson, R., Kearney, C., Steffke, A.,2004. Integrated satellite observations of the 2001 eruption of Mt. Cleveland,Alaska. Journal of Volcanology and Geothermal Research 135, 51–73.

Draxler, R.R.,Hess,G.D.,1998. Anoverviewof theHysplit 4modeling systemfor trajectories,dispersion and deposition. Australian Meteorological Magazine 47, 295–308.

Draxler, R.R., Hess, G.D., 2004. Description of the HYSPLIT 4 modeling system. NOAATechnical Memorandum ERL ARL-224. 28 pp.

Durant, A.J., Rose, W.I., 2009. Sedimentological constraints on hydrometeor-enhancedparticle deposition: 1992 Eruptions of Crater Peak, Alaska. In: Mastin, Larry,Webley,Peter (Eds.), Journal of Volcanology and Geothermal Research: Special Issue onVolcanic Ash Clouds, 186, pp. 40–60 (this volume).

Durant, A.J., Shaw, R.A., Rose, W.I., Mi, Y., Ernst, G.G.J., 2008. Ice nucleation andoverseeding of ice in volcanic clouds. Journal of Geophysical Research 113.doi:10.1029/2007JD009064.

Durant, A.J., Rose, W.I., Sarna-Wojcicki, A.M., Carey, S., Volentik, A.C., in review.Hydrometeor-enhanced tephra sedimentation: Constraints from the 18 May 1980eruption of Mount St. Helens (USA). Journal of Geophysical Research.

Fero, J., Carey, S.N., Merrill, J.T., 2009. Simulating the dispersal of tephra from the 1991Pinatubo eruption: Implications for the formulation of widespread ash layers. In:Mastin, Larry, Webley, Peter (Eds.), Journal of Volcanology and GeothermalResearch: Special Issue on Volcanic Ash Clouds, 186, pp. 121–132 (this volume).

Fierstein, J., Nathenson, M., 1992. Another look at the calculation of fallout tephravolumes. Bulletin of Volcanology 54, 156–167.

Foreman, P., 1994. Warning systems and pilot actions. In: Casadevall, T.J. (Ed.), VolcanicAsh and Aviation Safety: Proceedings of the First International Symposium onVolcanic Ash and Aviation Safety. USGS Bulletin 2047. U.S. Government PrintingOffice, Washington, D.C., pp. 164–168.

Gao, R.S., Popp, P.J., Fahey, D.W., Marcy, T.P., Herman, R.L.,Weinstock, E.M., Baumgardner,D.G., Garrett, T.J., Rosenlof, K.H., Thompson, T.L., Bui, P.T., Ridley, B.A., Wofsey, S.C.,Toon, O.B., Tolbert, M.A., Karcher, B., Peter, T.h., Hudson, P.K., Weinheimer, A.J.,Heymsfield, A.J., 2004. Evidence that nitric acid increases relative humidity in low-temperature cirrus clouds. Science 303, 516–520.

Graf, H.-F., Herzog, M., Oberhuber, J.M., Textor, C., Trentmann, J., 2002. ATHAM – ActiveTracer High Resolution Atmospheric Model manual. Max-Planck-Institute forMeteorology, Hamburg, Germany. 33 pp.

Herzog, M., Graf, H.-F., Textor, C., Oberhuber, J.M., 1998. The effect of phase changes ofwater on the development of volcanic plumes. Journal of Volcanology andGeothermal Research 87, 55–74.

Hoblitt, R.P., 1994. An experiment to detect and locate lightning associated witheruptions of Redoubt Volcano. Journal of Volcanology and Geothermal Research 62(1–4), 499–517.

Holasek, R.E., Self, S.,Woods, A.W.,1996. Satellite observations and interpretation of the 1991Mount Pinatubo plumes. Journal of Geophysical Research 101 (B12), 27,365–27,655.

Hufford, G.L., Salinas, L.J., Simpson, J.J., Barske, E.G., Pieri, D.C., 2000. Operationalimplications of airborne volcanic ash. Bulletin of the American MeteorologicalSociety 81 (4), 745–755.

Hurst, A.W., 1994. ASHFALL — a computer program for estimating volcanic ash fallout.Report and users guide, Institute of Geological and Nuclear Sciences Report 94/23.Institute of Geological and Nuclear Sciences, Wellington. 22 pp.

International Civil Aviation Organization, ICAO, 2004a. Handbook on the InternationalAirways Volcano Watch (IAVW). Operational procedures and contact list,International Civil Aviation Organization Doc 9766-AN/968, 2nd ed. 49 pp.

International Civil Aviation Organization, ICAO, 2004b. Manual on Volcanic Ash,Radioactive Material and Toxic Chemical Clouds. International Civil AviationOrganization Doc 9691. 122 pp.

Khairoutdinov, M.F., Randall, D.A., 2003. Cloud resolving modeling of the ARM summer1997 IOP: model formulation, results, uncertainties, and sensitivities. Journal of theAtmospheric Sciences 60 (4), 607–625.

Koyaguchi, T., Tokuno, M., 1993. Origin of the giant eruption cloud of Pinatubo, June 15,1991. Journal of Volcanology and Geothermal Research 55, 85–96.

Koyaguchi, T., Ochiai, K., Suzuki, Y.J., 2009. The effect of intensity of turbulence inumbrella cloud on tephra dispersion during explosive volcanic eruptions:Experimental and numerical approaches. In: Mastin, Larry, Webley, Peter (Eds.),

Journal of Volcanology and Geothermal Research: Special Issue on Volcanic AshClouds, 186, pp. 69–79 (this volume).

Mastin, L.G., 2007. A user-friendly one-dimensional model for wet volcanic plumes. G3.Geochemistry, Geophysics, Geosystems 8 (3). doi:10.1029/2006GC001455 (Tech-nical Brief, March 24, 2007).

Mastin, L.G., Guffanti, M., Servranckx, R., Webley, P.W., Barsotti, S., Dean, K., Denlinger,R., Durant, A., Ewert, J.W., Gardner, C.A., Holliday, A.C., Neri, A., Rose, W.I., Schneider,D., Siebert, L., Stunder, B., Swanson, G., Tupper, A., Volentik, A., Waythomas, C.F.,2009. A multidisciplinary effort to assign realistic source parameters to model ofvolcanic ash-cloud transport and dispersion during eruptions. In: Mastin, Larry,Webley, Peter (Eds.), Journal of Volcanology and Geothermal Research: SpecialIssue on Volcanic Ash Clouds, 186, pp. 10–21 (this volume).

Miller, T.P., Casadevall, T.J., 2000. Volcanic ash hazards to aviation. In: Sigurdsson, H.,Houghton, B.F., McNutt, S.R., Rymer, H., Stix, J. (Eds.), Encyclopedia of Volcanoes.Academic Press, San Diego, pp. 915–930.

Morton, B.R., Taylor, G.I., Turner, J.S., 1956. Turbulent gravitational convection frommaintained and instantaneous sources. Proceedings of the Royal Society of London.Series A 234, 1–23.

Murrow, P.J., Rose,W.I., Self, S., 1980. Determination of the total grain size distribution ina vulcanian eruption column, and its implications to stratospheric aerosolperturbation. Geophysical Research Letters 7 (11), 893–896.

Oberhuber, J.M., Herzog, M., Graf, H.F., Schwanke, K., 1998. Volcanic plume simulationon large scales. Journal of Volcanology and Geothermal Research 87 (1), 29–53.

Peterson, R., Webley, P.W., D'Amours, R., Servranckx, R., Stunder, R., Papp, K., 2008.Volcanic Ash Cloud Dispersion Models. Chapter 7 in Volcanoes Satellite Monitoringof Volcanoes: Spaceborne Images of North Pacific. In Review.

Prata, A.J., 1989a. Observations of volcanic ash clouds in the 10–12 µm window usingAVHRR/2 data. International Journal of Remote Sensing 10, 751–761.

Prata, A., 1989b. Infrared radiative transfer calculations for volcanic ash clouds.Geophysical Research Letters 16 (11), 1293–1296.

Prata, A.J., Bernardo, C., 2009. Retrieval of volcanic ash particle size, mass and opticaldepth from a ground-based thermal infrared camera. In: Mastin, Larry, Webley,Peter (Eds.), Journal of Volcanology and Geothermal Research: Special Issue onVolcanic Ash Clouds, 186, pp. 92–108 (this volume).

Prata, A.J., Grant, I.F., 2001. Retrieval of mircophyiscal and morphological properties ofvolcanic ash plumes from satellite data: application to Mt. Ruapehu, New Zealand.Quarterly Journal of the Royal Meteorological Society 122, 1–25.

Prata, F., Bluth, G., Rose, W., Schneider, D., Tupper, A., 2001. Comments on “failures indetecting volcanic ash from a satellite-based technique”. Remote Sensing ofEnvironment 78, 341–346.

Pudykiewicz, J., 1988. Numerical simulation of the transport of radioactive cloud fromthe Chernobyl nuclear accident. Tellus 40B, 241–259.

Pyle, D.M., 1989. The thickness, volume and grain size of tephra fall deposits. Bulletin ofVolcanology 51, 1–15.

Qamar, A.I., Malone, S.D., Moran, S.C., Steele,W.P., Thelen,W., in press. Chapter 3. Near-Near-Real-Time Information Products for Mount St. Helens—Tracking the Ongoing Eruption.In: D. Sherrod andW.E. Scott (Editors), AVolcano Rekindled: The Renewed Eruption ofMount St. Helens, 2004–2006. U.S. Government Printing Office, Washington, D.C.

Romero, R., 2004. The International Airways Volcano Watch (IAVW). Proceedings of 2ndInternational ConferenceonVolcanicAshandAviationSafety. June21–24,Virginia,USA.

Rose, W.I., 1993. Comment on ‘another look at the calculation of fallout tephra volumes’by Judy Fierstein and Manuel Nathenson. Bulletin of Volcanology 55, 372–374.

Rose, W.I., Durant, A., 2009. Fine ash content of explosure erruption. In: Mastin, Larry,Webley, Peter (Eds.), Journal of Volcanology and Geothermal Research: SpecialIssue on Improved Prediction and Tracking of Volcanic Ash Clouds, 186, pp. 32–39(this volume).

Rose, W.I., Anderson, A.T., Woodruff, L.G., Bonis, S.B., 1978. The October 1974 basaltictephra from Fuego volcano: description and history of the magma body. Journal ofVolcanology and Geothermal Research 4 (1–2), 3–53.

Rose, W.I., Delene, D.J., Schneider, D.J., Bluth, G.J.S., Krueger, A.J., Sprod, I., McKee, C.,Davies, H.L., Ernst, G.G.J., 1995a. Ice in the 1994 Rabaul eruption cloud: implicationsfor volcano hazard and atmospheric effects. Nature 375 (6531), 477–479.

Rose, W.I., Kostinski, A.B., Kelley, L., 1995b. Real-time C-Band radar observations of 1992eruption clouds from Mt. Spurr volcano, Alaska. U. S. Geological Survey Bulletin2139, 19–26.

Rose,W.I., Self, S., Murrow, P.J., Bonadonna, C., Durant, A.J., Ernst, G.G.J., 2007. Nature andsignificance of small volume fall deposits at composite volcanoes: insights from theOctober 14, 1974 Fuego eruption, Guatemala. Bulletin of Volcanology. doi:10.1007/s00445-007-0187-5.

Ryall, D.B., Maryon, R.H., 1998. Validation of the UK Met. Office's NAME model againstthe ETEX dataset. Atmospheric Environment 32 (24), 4265–4276.

Sandu, I., Bompay, F., Stefan, S., 2003. Validation of atmospheric dispersionmodels usingETEX data. International Journal of Environment and Pollution 19 (4), 367–389.

Sarna-Wojcicki, A.M., Shipley, S., Waitt, R., Dzurisin, D., Wood, S.H.,1981. Areal distribution,thickness, mass, volume, and grain size of air-fall ash from the six major eruptions of1980. In: Lipman,P.W.,Christiansen, R.L. (Eds.), The 1980EruptionsofMount St.Helens,Washington; USGS Professional Paper 1250. U.S. Geological Survey, pp. 577–601.

Sassen, K., Zhu, J., Webley, P.W., Dean, K., Cobb, P., 2007. Volcanic ash plumeidentification using polarization lidar: Augustine Eruption, Alaska. GeophysicalResearch Letters 34 (8). doi:10.1029/2006GL027237.

Schneider, D.J., Rose, W.I., Kelley, L., 1995. Tracking of 1992 eruption clouds from CraterPeak vent of Mount Spurr Volcano, Alaska, using AVHRR. U. S. Geological SurveyBulletin 2139, 27–36.

Schneider, D., Wallace, K., Marzano, F.S., 2008. NEXRADWeather Radar Observations ofthe 2006 Augustine Volcanic Eruption Clouds. U.S. Geological Survey ProfessionalPaper. In Review.

9P. Webley, L. Mastin / Journal of Volcanology and Geothermal Research 186 (2009) 1–9

Searcy, C., Dean, K.G., Stringer, W., 1998. PUFF: a volcanic ash tracking and predictionmodel. Journal of Volcanology and Geothermal Research 80, 1–16.

Servranckx, R., D'Amours, R., Jean, M., Toviessi, J.-P., Trudel, S., 1996. Volcanic ashforecasting at the Canadian Meteorological Centre. Environmental EmergencyResponse Division, Environment Canada. Presented and Published from Pan PacificHazards '96, July 29 - August 2 1996, Vancouver BC.

Settle, M., 1978. Volcanic eruption clouds and the thermal power output of explosiveeruptions. Journal of Volcanology and Geothermal Research 3, 309–324.

Simpson, J.J., Hufford, G., Pieri, D., Berg, J., 2000. Failures in detecting volcanic ash fromsatellite based technique. Remote Sensing of Environment 72, 191–217.

Simpson, J.J., Hufford, G.L., Pieri, D., Servranckx, R., Berg, J., Bauer, C., 2002. The February2001 eruption of Mount Cleveland, Alaska: case study of an aviation hazard.Weather and Forecasting 17 (4), 691–704.

Sparks, R.S.J., Moore, J.G., Rice, C.J., 1986. The initial giant umbrella cloud of the May18th, 1980, explosive eruption of Mount St. Helens. Journal of Volcanology andGeothermal Research 28 (3–4), 257–274.

Sparks, R.S.J., Bursik, M.I., Carey, S.N., Gilbert, J.E., Glaze, L., Sigurdsson, H., Woods, A.W.,1997. Volcanic Plumes. John Wiley and Sons, Chichester. 574 pp.

Textor, C., Graf, H.-F., Herzog, M., Oberhuber, J.M., Rose, W.I., Ernst, G.G.J., 2006a.Volcanic particle aggregation in explosive eruption columns. Part I: Parameteriza-tion of the microphysics of hydrometeors and ash. Journal of Volcanology andGeothermal Research 150, 259–377.

Textor, C., Graf, H.-F., Herzog, M., Oberhuber, J.M., Rose, W.I., Ernst, G.G.J., 2006b.Volcanic particle aggregation in explosive eruption columns. Part II: Numericalexperiments. Journal of Volcanology and Geothermal Research 150, 259–377.

Tupper, A., Wunderman, R., 2009. Reducing discrepancies in ground and satellite-observed eruption heights. In: Mastin, Larry, Webley, Peter (Eds.), Journal ofVolcanology and Geothermal Research: Special Issue on Volcanic Ash Clouds, 186,pp. 22–31 (this volume).

Tupper, A.C., Davey, J.P., Potts, R.J., 2003. Monitoring volcanic eruptions in Indonesia andthe Southwest Pacific. Proceedings of Symposium on Researching Eruption Cloudson Volcanic Island Chains, Kagoshima University, 9–10 November 2002. OccasionalPapers No. 37. Kagoshima University, pp. 153–163.

Tupper, A., Carn, S., Davey, J., Kamada, Y., Potts, R., Prata, F., Tokuno, M., 2004. Anevaluation of volcanic cloud detection techniques during recent significanteruptions in the western ‘Ring of Fire’. Remote Sensing of the Environment 91,27–46.

Tupper, A., Itikarai, I., Richards, M., Prata, F., Carn, S., Rosenfeld, D., 2007. Facing thechallenges of the international airways volcano watch: The 2004/05 eruptions ofManam, Papua New Guinea. Weather and Forecasting 22 (1), 175–191.

Tupper, A., Textor, C., Herzog, M., Graf, H.F., in press. Tall clouds from small eruptions:modeling the sensitivity of eruption height and fine ash fallout to troposphericinstability. Natural Hazards.

U.S. Department of Commerce/National Oceanic and Atmospheric Administration,2007. National Volcanic Ash Operations Plan for Aviation. FCM-P35-2007.Washington, D. C., August 2007. 68 pp.

Webley, P.W., Dean, K.G., Dehn, J., Bailey, J.E., Peterson, R., 2008a. Volcanic AshDispersion Modeling of the 2006 Eruption of Augustine Volcano. USGS ProfessionalPaper: Augustine Volcano 2006 eruption, In Review.

Webley, P.W., Dean, K., Collins, R., Fochesatto, J., Sassen, K., Atkinson, D., Cahill, C., Prata,A., 2008b. Validation of a volcanic ash dispersion model during late January/EarlyFebruary 2006 Eruption of Mount Augustine Volcano. Bulletin of AmericanMeteorological Society. doi:10.1175/2008BAMS2579.1.

Webley, P.W., Dehn, J., Lovick, J., Dean, K.G., Bailey, J.E., Valcic, L., 2009. Near Real TimeVolcanic Ash Cloud Detection: Experiences from the Alaska Volcano Observatory.In: LarryMastin, Larry, PeterWebley, Peter (Eds.), Journal of Volcanology andGeothermal Research: Special Issue on Volcanic Ash Clouds, 1862009, pp. 80–91(this volume).

Webley, P.W., Stunder, B.J.B.,Dean,K.G., 2009. Preliminarysensitivity studyof eruption sourceparameter for operational volcanic ash cloud transport and dispersion models – A casestudy of the August 1992 eruption of the Crater Peak Vent, Mount Spurr, Alaska. In:LarryMastin, Larry, PeterWebley, Peter (Eds.), Journal of Volcanology and GeothermalResearch: Special Issue on Volcanic Ash Clouds, 186, pp. 109–120 (this volume).

Wen, S., Rose, W.I., 1994. Retrieval of sizes and total masses of particles in volcanicclouds using AVHRR bands 4 and 5. Journal of Geophysical Research 99 (D3),5421–5431.

Wilson, L., Walker, G.P.L., 1987. Explosive volcanic eruptions: VI. Ejecta dispersal inplinian eruptions: the control of eruption conditions and atmospheric properties.Geophysical Journal of The Royal Astronomical Society 89, 657–679.

Wilson, L., Sparks, R.S.J., Huang, T.C., Watkins, N.D., 1978. The control of volcanic columnheights by eruption energetics and dynamics. Journal of Geophysical Research 83(B4), 1829–1836.

Woods, A.W., 1988. The fluid dynamics and thermodynamics of eruption columns.Bulletin of Volcanology 50, 169–193.

Woods, A.W., 1993. Moist convection and the injection of volcanic ash into theatmosphere. Journal of Geophysical Research 98 (B10), 17627–17636.

Yu, T.X., Rose,W.I., 1999. Retrieval of sulfate and silicate ashmasses in young (1 to 4 daysold) eruption clouds using multiband infrared HIRS/2 data. AGU Monograph onRemote Sensing of Active Volcanism, vol. 116. American Geophysical Union,Washington, DC, pp. 87–100.