3

Abstract

A New Long-Term Record of Volcanic SO₂ Generated From HIRS/2 Satellite Data

Volcanic eruptions that inject ash, sulfur dioxide and other gases into the atmosphere occur on average more than 50 times per year somewhere on Earth. Volcanic eruptions produce a variety of hazards, ranging from short term aircraft hazards, to longer term effects from climate perturbation. However, many important details regarding the magnitude and frequency of sulfur emissions, and chemical and physical processes within volcanic clouds remain unclear. Some key questions may be largely solved if more data were available: issues of excess sulfur from volcanic outgassing; the separation of ash-rich and gas-rich portions of the clouds; the conversion of H₂S to SO₂ in the atmosphere; and the removal rates of SO₂ in the atmosphere.

The most practical means of monitoring large volcanic emissions of sulfur is through satellite remote sensing, and since 1979 the majority of SO₂ measurements have been through NASA's Total Ozone Mapping Spectrometer (TOMS). These measurements rely on reflected ultraviolet radiation and consequently have poor diurnal sampling and low spatial resolution. Two new methods have been proposed which employ infrared channels of the Moderate Resolution Infrared Spectroradiometer (MODIS) sensor, each of which have the ability to quantitatively retrieve SO₂ masses. Our previous analyses of MODIS 7.3 mm data for eruptions at Hekla (2000) and Cleveland (2001) volcanoes have demonstrated that it is possible to estimate upper tropospheric SO₂ layer abundance using the anti-symmetric stretch of the SO₂ molecule.

This proposal focuses on the High Resolution Infrared Radiation Sounder/2 (HIRS/2) sensor, which also includes a 7.3 mm channel, on board the NOAA polar orbiting satellites since 1978. We propose to analyze HIRS radiances within this channel to map and quantify volcanic SO₂ emissions, and construct a long term dataset complementary to TOMS but with potentially improved spatial and temporal resolutions. This additional coverage will greatly enhance our ability to study the dozens of observable eruptions in the past 20 years, as well as provide an important link to data from new sensors. Specifically, we propose to:

(1) acquire and process datasets for volcanic eruptions from HIRS/2, to match with the TOMS and MODIS data we have already archived;

(2) develop and refine our techniques for sulfur dioxide retrievals with the 7.3 mm channel using several key eruptions, under varying environmental and plume conditions; and

(3) create a long-term volcanic SO₂ database from HIRS/2, with MODIS overlap and TOMS data to compare and cross-validate the results.

This proposal responds to the NASA NRA in the following ways: to the Earth Science Enterprise goal of studying the changing Earth, it supports the creation of an independent long term database of global volcanic emissions; by developing this new volcanic record it complements and enhances the usefulness of the existing (TOMS) database funded by the SENH program; by generating more detailed erupted sulfur budgets it will foster many science applications for understanding magmatic processes and cloud fates; and the HIRS/2 methodology provides an additional means of monitoring and understanding global volcanic activity, and consequently, volcanic hazards.

Technical Plan

A New Long-Term Record of Volcanic SO₂ Generated From HIRS/2 Satellite Data

Background and Relevance

The impacts and hazards from volcanic eruptions are potentially large, because their characteristically infrequent bursts of activity have the potential to inject ash, sulfur dioxide and other gases into the upper troposphere and stratosphere. Following large events, the SO₂ gas emplaced in the stratosphere the converts to a sulfate aerosol which in turn affects the radiation budget of the planet by scattering shortwave radiation and absorbing longwave radiation. This generally leads to surface cooling (e.g., Rampino and Self, 1984; Mass and Portman, 1989). Upper tropospheric sulfur dioxide and sulfate are hazards to jet aircraft due to their corrosive nature, and because they are often accompanied by volcanic ash which damages jet turbine engines (Casadevall, 1994). Estimation of SO₂ emissions from volcanoes is also important for diagnosing processes within active volcanic systems, as the amounts and rates of emissions provide clues as to magma movement, degassing properties, and thus potential eruption characteristics (e.g., Symonds et al., 1994; Wallace and Gerlach, 1994).

Since volcanic eruptions are sporadic events and there might be few eruptions in some periods of time, a long observation period is required to provide enough data to assess many volcano-atmosphere issues. There are two substantial eruptions, Pinatubo and El Chichón, and many other smaller eruptions in the past 20 years. Some important questions are still unresolved but may potentially be solved if more data were available. One central focus is the issue of excess sulfur (e.g., Andres et al., 1991; Scaillet et al., 1998). Another problem is the separation of ash-rich and SO₂-rich portions in volcanic clouds (Bluth et al., 1994; Schneider et al., 2000). A further question is the possibility of H₂S-SO₂ conversion in the atmosphere (Bluth et al., 1995; Rose et al., 2000). The removal rate of sulfur dioxide in the atmosphere is also key to understanding effects of volcanogenic emissions (Oppenheimer et al., 1998; Barker et al., 1998).

Measuring and retrieving the burden of SO₂ in volcanic clouds is an important part of studying volcano-atmosphere interactions. Retrieval of volcanic SO₂ using remote sensing technology has been investigated during last decade, most commonly by the Total Ozone Mapping Spectrometer (TOMS) (Bluth et al., 1993; Krueger et al., 2001). However, TOMS uses reflected ultraviolet radiation and consequently is limited by poor diurnal sampling and low spatial resolution. Limited studies at thermal infrared wavelengths include the Thermal Infrared Multispectral Scanner (TIMS) and the NCAR Fourier Transform Spectrometer (Realmuto et al., 1994; Crisp, 1995; Realmuto et al., 1997; Realmuto and Worden, 2000). Microwave measurement of SO₂ using the Microwave Limb Sounder (MLS) instrument aboard Upper Atmosphere Research Satellite (UARS) has also been used for the 1991 Pinatubo eruption (Read et al., 1992; Crisp, 1995).

Analyses of the Moderate Resolution Imaging Spectroradiometer (MODIS) data have resulted in two new methods for sulfur dioxide retrievals. An absorption feature at 8.2 - 9.2 µm can be exploited by the method of Realmuto et al. (1997), which derives SO₂ through comparison with background radiances. Prata et al. (2002) have demonstrated a second, completely independent method using 7.3 µm data for Hekla and Cleveland volcanic eruptions. This work indicates that it is possible to estimate upper-tropospheric SO₂ layer abundance using the strong absorption band centered about 7.2 µm. Despite the existence of strong water vapor features in this channel, sulfur dioxide abundance above ~5km can be estimated owing to the relatively low water contents in the atmosphere above 5 km.

The TIROS Operational Vertical Sounder (TOVS) suite of instruments have a 7.3 mm channel within the High Resolution Infrared Radiation Sounder/2 (HIRS/2) which senses radiation from SO₂ and has produced such data since 1978. The HIRS/2 provides the basic 20 channel IR temperature and humidity soundings of the TOVS system. All HIRS/2 channels are used to provide information on temperature and humidity profiles, surface temperature, cloud parameters and total ozone (http://www.eumetsat.de/en/index.html). The average orbital height of HIRS/2 is 830 km and the spatial resolution is about 17.5 km at nadir and 58.5 x 30 km² for the first and last pixels of a scan line, with each scan taking 6.4 seconds. Full daily coverage is achieved through 14 orbits per day, each with an approximately 2240 km swath width.

Table 1. Key Characteristics of HIRS/2, MODIS and TOMS sensors.

HIRS/2 ^a			MODIS ^a			TOMS ^b
Spatial Resolution: 17.5 km (nadir)			Spatial Resolution = 1 km			Spatial Resolution: 39 km (nadir)
*Channel*	mm	*Uses*	*Channel*	mm	*Uses*	*Channel*	nm	*Uses*
12	6.72	H₂O	27	6.72	H₂O	1	308.6	SO₂, O₃
11	7.33	H₂O, SO₂	28	7.33	H₂O, SO₂	2	313.5	SO₂, O₃
10	8.16	H₂O	29	8.55	SO₂, sulfate, ash	3	317.5	SO₂, O₃
9	9.71	O₃	30	9.73	O₃, sulfate	4	322.3	SO₂, O₃
8	11.11	Ash, ice, sulfate	31	11.03	Ash, ice, sulfate	5	331.2	SO₂, O₃ ash, aerosol
7	13.35	ash	32	12.02	ice, sulfate, ash	6	360.48	ash, aerosol

^aYu and Rose, 2000

^bKrueger et al., 2000; ozone and SO₂ retrievals are derived from varying band combinations.

Sulfur dioxide absorption was noted in HIRS/2 spectra of the El Chichon eruption by Susskind (1982). Consequently, an SO₂ detection alert algorithm using MODIS and HIRS/2 data was developed by Crisp (1995), but methods for retrieving and retrospective the SO₂ amount using global HIRS/2 data sets were not investigated. Our work strongly suggests it is possible to analyze HIRS/2 data, using the 7.3 µm channel, to derive the sulfur dioxide concentrations associated with explosive volcanic eruptions (Prata et al., 2002). In this study, we plan to analyze the HIRS/2 radiance data and develop methodologies for SO₂ measurement using the past 20 years of HIRS data. The water vapor channels (10 to 12), split window channel (8) and ozone channel (9) will all be required in data processing and analysis. Exploratory work on selected eruptions will allow us to properly calibrate the retrieval tools required to derive SO₂ from HIRS/2 channel 11 and show how sensitive this method might be (Prata et al., 2002).

Preliminary case studies

Initial, 7.34 mm-based, retrievals of upper tropospheric SO₂ have been done using MODIS data, acquired between 27 and 28 February 2000 over the region 50°N to 80°N and 40°W to 30°E, during the recent Hekla eruption of 26 February, 2000. The results suggest that the infrared absorption in the 7.34 mm MODIS channel (Table 1) can be used to estimate the concentration of sulfur dioxide in the upper troposphere. However, some of this absorption is also due to water vapor.

A scheme for estimating, on a pixel-by-pixel basis, the contribution to the absorption by the background atmosphere was derived (Prata et al., 2002). This is achieved by using MODIS infrared channels with band centers that lie below (channel 27-6.78 m_m) and above (channel 32- 12.02 m_m) the 7.34 m_m channel. For an SO₂-free atmosphere, the radiance in these three channels varies with wavelength in an almost linear manner. Thus for each MODIS image the radiance at 7.34 mm can be estimated by linearly interpolating the radiances at 6.78 m_m and 12.1 m_m. When there is excess absorption in the 7.34 m_m channel, the difference between the estimated and the actual 7.34 mm radiance is due to the presence of SO₂. The reason why the 8.55 mm channel (MODIS channel 29) was not used in the retrieval is that this channel is affected by both SO₂ absorption and some absorption effects due to silicates in volcanic ash. The 6.78 m_m (channel 27) and 12.02 m_m channels (channel 32) are entirely free of SO₂ absorption and to a lesser degree volcanic ash effects, and may require little or no correction. The scheme is applied to MODIS data on a pixel-by-pixel basis (1354 pixels by 2030 lines) and maps of SO₂ are derived at 1 x 1 km² spatial resolution, by weighting retrievals for off-nadir pixels by their respective field-of-view areas.

The SO₂ retrievals for Hekla eruption using three different sensors (MODIS, HIRS/2, and TOMS) are shown in Figure 1 (Prata et al., 2002). The SO₂ plume apparently went to the northeast initially before being transported in a spiral southwards and westwards. In the early hours of 29 February a NASA DC-8 encountered the plume while on an atmospheric chemistry mission (the SOLVE experiment). Measurements of SO₂ were made as the DC-8 traversed the plume. This accidental encounter has allowed us to validate the satellite retrievals for the MODIS image of 29 February at 1115 UT.

The DC-8 encountered the plume at approximately 0511 UT, some 6 hours before the first good MODIS image. By using wind vectors at 200 hPa obtained from the CANERM (Canadian Emergency Response Model) plume dispersion model operated by the Meteorological Service of Canada we have advected the MODIS plume backwards in time and compared the MODIS retrieval with the DC-8 measurements (Figure 1, panels e-h). The in situ measurements indicated great variability (a function of the spatial resolution) and peak amounts of 1 ppmV, whereas the MODIS retrieval has much less variability and smaller peak values. The latitudinal spread of the plume is slightly greater in the MODIS retrieval. The DC-8 was at an altitude of 10.4 km when it encountered the plume, and essentially sampled the plume at that height through a single traverse of the plume. Despite these sampling differences, the integrated SO₂ amounts from the DC-8 (248 ppmV) and from the MODIS retrieval (241 ppmV) are in good agreement. The shapes and locations of the sulfur dioxide cloud at the same approximate times in MODIS, TOMS, and HIRS/2 images are also very similar.

Retrievals were subsequently performed for two other eruptions, using HIRS/2 data: the June, 1991 eruption of Mt. Pinatubo, and the August, 1991 eruption of Cerro Hudson, Chile. Eruption clouds from both eruptions were readily observed (Figure 2).

Figure 1. (a)-(c). SO₂ retrievals from the 7.34 _m MODIS channel on 27 and 28 February, 2000. (d) TOMS SO₂ retrieval for February 27 at 12 UT. (e)-(f) Plume dispersion results for the Hekla eruption at 06 UT and 12 UT on 28 February at the 200 hPa level. (g)-(h) SO₂ concentration comparison between NASA DC-8 aircraft (measured) and MODIS retrievals.

Figure 1. (i) HIRS/2 SO₂ retrieval for February 27. The missing data scans are the result of calibration sequences.

(a)

(b)

Figure 2. Preliminary study of retrieving SO₂ concentrations using HIRS/2 data for (a) August, 1991 Cerro Hudson Eruption and (b) June, 1991 Pinatubo Eruption.

The pertinent conclusions from our preliminary study are: (a) the HIRS/2 retrieval methodology and database are, by early indications, excellent for retrospective studies; and (b) the HIRS data and imagery are consistent and highly complementary to those from contemporary satellite sensors like TOMS, and from modern instruments like MODIS.

Synthesized Volcanic Cloud Studies at Michigan Tech

In the past few years, our research group has made significant progress in studies of volcano-atmosphere interactions, through a combination of remote sensing, laboratory, and theoretical modeling studies. Much of the research and collaborative work have been the result of NASA SENH funding. Complete conversion of SO₂ to sulfate in the stratosphere occurs at an e-folding rate of about 120 days. While the SO₂ loss from stratospheric volcanic clouds has an apparent e-folding rate of about 35 days, the SO₂ loss rate for volcanic clouds remain in the troposphere or lower stratosphere appears to be much more rapid, on the order of a few days (Bluth et al., 1992; 1997; Barker et al., 1998; Oppenheimer et al., 1998). The latter limits the stratospheric aerosol buildup from smaller eruptions (Bluth et al., 1997) and removal processes are most effective in wet eruptions where ice nucleates and grows on ash particles (Rose et al., 1995).

We can now map and measure both ash and SO₂ using both infrared (Wen and Rose, 1994; Realmuto et al., 1997) and ultraviolet data (Krueger et al., 2001; Krotkov et al., 1997). The infrared retrieval scheme now includes atmospheric corrections (Yu, 2000) and this enables us to look at smaller volcanic clouds (Rose and Mayberry, 2000). We have developed a multispectral IR technique (for the MODIS sensor) which retrieves information on sulfate aerosols as well as ash (Yu and Rose, 2000). Our initial work with modeling volcanic clouds (Guo et al., 2000) has helped us evaluate theories of volcanic cloud evolution. Laboratory measurements relating ash particle geometry to fallout characteristics (C. Riley, Ph.D. thesis in progress) have helped us evaluate the effects on radiative transfer models (Krotkov et al., 1999). Laboratory work on adsorption on fine volcanic ash (Gu, 2001) has helped quantify rates of uptake and release of sulfur dioxide gas under a range of atmospheric conditions. Currently, we are attempting to take advantage of our previous techniques of retrieving both ash and gas species, to study the dozen or so documented cases of species separation in drifting clouds (Schneider et al., 2000).

An international workshop, Remote Sensing of Volcanic Clouds, was held July 29-August 3, 2001, at Michigan Technological University (Rose, 2001). The workshop's goal was to improve and expand the use of satellite-based remote sensing data for hazard mitigation and other research purposes, such as volcano-atmosphere interactions and chemical and meteorological effects on the troposphere and stratosphere. Forty-six researchers attended, representing 11 countries, 9 universities, and several government meteorological and volcanological organizations, as well as the Volcanic Ash Advisory Centers in Washington, D.C., Anchorage, Montreal, Darwin, London, and Toulouse.

The workshop consisted of presentations about volcanic clouds and extended computer laboratory sessions in which attendees worked with actual satellite remote sensing data. The participants covered the current and future status of remote sensing instruments and their detection capabilities by reviewing examples of eruptions; and discussing and demonstrating new techniques for improving detections, such as atmospheric corrections, accurate assessment of initial eruption conditions, environmental conditions during eruptions, and variable volcanic phenomena. Satellite coverage and limitations and integrating trajectory models into mitigation efforts were also discussed, as was how volcanic cloud hazards mitigation efforts could be improved through validation, modeling, multi-sensor comparisons, and different band selections. Common issues discussed were the need for timely, accurate, and diagnostic predictions by agencies involved in monitoring and ash cloud detection and the need for timely dissemination of educational information to responsible agencies and the public. Sponsorship by NASA's Solid Earth Natural Hazards Program helped fund the meeting and the attendance of many graduate students.

Project Objectives

This project focuses on three main objectives:

(1) Data acquisition: We will construct a set of HIRS data to coincide with known volcanic eruption observations by both TOMS and MODIS sensors.

Complementary HIRS/2, TOMS and MODIS data sets need to be acquired first. The MODIS and TOMS data sets (Hekla and Cleveland eruptions) are largely already in our database, but will updated as new eruptions occur. We need to get complete HIRS/2 datasets for several important eruptions in the past 20 years (e.g., Pinatubo, El Chichon, Cerro Hudson, Mt. St. Helens). A computer facility with large disk space has been built up in our research group, and will be used for storage and processing the past 20 years of HIRS data for this study.

(2) Methodology. A robust methodology will be developed to map and retrieve SO₂ masses from a variety of different environmental conditions.

The techniques will be developed and checked based on the analysis of those important eruptions. The preliminary study of Hekla eruption using MODIS and HIRS/2 data indicates that it is possible to use IR sensing to retrieval the volcanic SO₂ concentration. The more detailed methods and techniques will be developed based on the study of different eruptions (key events of the past 20 years) as well as different environmental conditions.

(3) Applications. We will generate a long term, volcanic SO₂ database from the HIRS/2 sensor, to complement existing and future data from TOMS and MODIS; this information will form the basis of many applications of plume chemistry and fates.

Sensitivity studies will be performed by comparing the results from different sensors and different real conditions. MODIS overlap and TOMS database will be used to compare and correct the results. The techniques will be modified, a more validated, advanced, and completed method will be developed. A database of HIRS retrievals will be produced, for comparison to the TOMS and MODIS datasets. The development of a complementary SO₂ database will provide important additional data for volcanic cloud studies: particularly, reaction-based studies which require more frequent mass retrievals (such as separation, conversion and removal processes) will benefit from more frequent data availability.

Technical Approach

There are a total of 36 bands measured by the MODIS instrument. The bands used to detect the volcanic cloud (volcanic ash and gases) are band 27 to band 32 (Table 1). The 7.34 mm channel (band 28) of the MODIS instrument is currently used to measure the upper tropospheric water vapor. This channel also adsorbs radiation from other gases, especially for SO₂. Figure 3 shows the line strengths of SO₂ taken from the HITRAN96 database (Rothman et al., 1987) and indicates that the SO₂ absorptions are in the n1-band, between 8.2 and 9.2 mm, and the n3-band, between 7.2 and 7.5 mm (MODIS bands 27, 28 and 29). Figure 3b gives atmospheric transmittance in a vertical path to space due to all atmospheric gases and especially for water vapor - the main absorber/emitter in this region of the infrared spectrum. Figure 3c shows the brightness temperatures at the top-of-the-atmosphere, deduced using a monochromatic radiative transfer program, for an atmosphere with background SO₂ concentrations (black line) and the SO₂ concentration of 35 m atm-cm (red line) where 1 m atm-cm = 2.678 x 10 molecules cm^-2. The difference between these two calculations is shown in Figure 3d. Figure 3d illustrates that the SO₂ absorption for n3-band is significantly stronger than the n1-band, but there is also greater water vapor absorption across the n3-band, as shown in Figure 3b. Previous studies (Realmuto et al., 1997; Realmuto and Worden, 2000) used the n1-band (centered at _ 8.6 mm) to retrieve the SO₂ concentration (because of the lack of interference from water vapor for this band), or used very high resolution spectroscopic measurements (Mankin et al., 1992; Goldman et al., 1992) to exploit the 'micro' windows between 1300 and 1400 cm^-1 (7.14 - 7.69 mm). It is generally believed that water vapor absorption makes it difficult to determine SO₂ from broad band measurements in the n3-band. This is true for SO₂ in the lower troposphere where most of the atmospheric water vapor resides. But for higher altitude (above 6 km or so), based on global climatological considerations, there is less than 5% of the total precipitable water above that layer. Therefore, broadband measurements (n3-band) of the excess absorption of 7.34 mm infrared radiation in the upper troposphere and lower stratosphere can be used to infer SO₂ concentrations (essentially, at 7.34 mm, the sensor cannot detect the ground; it uses emissions from water vapor below the plume as a source of electromagnetic radiation).

The following equations are the transmission function model (Prata et al., 2002), t of the n3 SO₂ band as a double-exponential function of SO₂ absorber amount, U_abs_.

The scaled absorber amount is W_abs, T is atmospheric temperature, p is atmospheric pressure, T₀ and P₀ are values at Standard Temperature and Pressure (STP) conditions. The coefficients Ci, a, n and m are wavelength dependent and have been determined through comparisons with detailed line-by-line radiative transfer calculations (McKeen et al., 1984). Figure 4a gives the relationship between absorber amount and transmission, the transmission as a function of absorber amount for a modeled SO₂ plume located at 10 km with a vertical spread of 2 km. Figure 4 also indicates that the relationship between transmission and absorber amount is approximately linear over the range of absorber amounts 0-100 m atm cm. The contribution to the radiation reaching a radiometer viewing the atmosphere from above over a narrow band of

Figure 3. Panel a. Line strengths for SO₂. Also shown are the nominal filter response functions for MODIS bands 27, 28 and 29. Panel b. Atmospheric transmission as a function of wave number (cm__) through a vertical path of an atmosphere containing gases and no cloud. Panel c. Top of the atmosphere brightness temperatures as a function of wave number for a standard atmosphere with background concentrations of SO₂ and, for the same atmosphere but with significantly elevated levels of SO₂ (red line). Panel d. Brightness temperature differences between the background and elevated SO₂ atmospheres.
Figure 4

Figure 4. (a) Transmission as a function of absorber amount for the SO₂-band. (b) Weighting functions for SO₂ and water vapor.

Figure 5. Medium resolution (2 cm__) radiative transfer calculations of the brightness temperature deficit-defined as difference between the background atmosphere and an atmosphere containing 0.08 ppmV of SO₂ -as a function of wave number for (a) the SO₂ __-band, and (b) the SO₂ __-band. Each line shows the deficit obtained for a plume 1 km thick centered at the indicated height above the surface.
wave numbers (the radiometer channel), from different portions of the atmosphere is described mathematically by a weighting function. Denoting the Planck function by B[T(p), n], where T(p)is the vertical temperature profile in the atmosphere, and p_s is the surface pressure, the weighting function, W, can be calculated from

The weighting functions for the SO₂ n3-band (MODIS channel 28) and for the lower troposphere water vapor channel (MODIS channel 27) are shown in Figure 4b. The peak of the weighting function for channel 28 occurs around 350 hPa, and indicates that this channel is sensitive to upper tropospheric sulfur dioxide.

Radiative transfer calculations using a 2 cm^-1 spectral resolution model (MODTRAN-3b) for a sulfur dioxide concentration of 0.8 ppmV at various heights in the atmosphere are shown in Figure 5 for the SO₂ n1-band (left panel) and for the SO₂ n3- band (right panel). These calculations show the temperature differences (background SO₂ - anomaly) expected across each of the bands. The biggest differences occur when the plume is situated around 9.9 km; smaller differences occur when the plume is either higher or lower. These results are only slightly sensitive to the plume vertical extent.

Expected Significance

There are a number of important results and applications of this work that are specific to the NASA NRA, and that will benefit the scientific community in general.

(1) Generation of a robust database of volcanic sulfur dioxide emissions. The HIRS dataset extends back to 1978, and thus provides complementary coverage to the TOMS sensors. Current eruptions can also be studied by the TOMS (which unfortunately is suffering from some degradation), MODIS, and HIRS/2. There are now two separate techniques for deriving SO₂ masses from MODIS, which extends it's capabilities. The TOMS (1 per day) and MODIS (1 per two days) temporal resolutions will be enhanced by the HIRS/2 daily coverage provided by multiple NOAA platforms. We expect that there will be cases where short-lived eruptions which are not observed by TOMS or MODIS will be detected by the HIRS/2. Thus, the long-term global coverage of explosive volcanism by satellite sensors will be vastly improved.

(2) An additional, spectrally unique methodology for analysis of volcanic clouds. The TOMS sensor has been the mainstay of remote sensing for volcanic SO₂ emissions since 1979. While observing over a hundred individual eruption clouds, TOMS detection and retrievals are hindered by its low temporal (once per day) and spatial (39 km at nadir) resolutions, and daytime-only viewing. The addition of MODIS to observe and quantify clouds using infrared wavelengths has provided an important means for cross-validation and improved spatial resolution, but with lower sampling frequency (every two days). The HIRS/2 approach is therefore extremely valuable because of the ability for retrospective studies with TOMS, and that it provides another independent means of detecting and studying volcanic SO₂ clouds. The development of the HIRS/2 SO₂ retrieval methodology provides important improvements in spatial and temporal resolution, and thus far has shown to be consistent with both TOMS and MODIS retrievals for near-simultaneous datasets.

(3) Science applications. We have noted four key question regarding explosive volcanic emissions that have the potential of being solved given sufficient data: causes and characteristics of excess sulfur emissions; determination of potential H₂S emission and conversion to SO₂ in erupted clouds; the separation of ash and gas-rich phases in erupting as well as drifting clouds; and removal rates of sulfur dioxide in the troposphere and stratosphere. The potential for significant progress towards these issues is great. Thus far we have typically lacked sufficient data to make much headway towards understanding reaction-based processes in volcanic clouds. With the HIRS/2 data we have the potential to more than double our current archives of cloud images and data, based on our initial results.

(4) Outreach. NASA Earth Science Enterprise has stated goals of broadening its applications and influence throughout the global scientific community. This project is therefore very timely, following the recent international workshop on Remote Sensing of Volcanic Clouds hosted by Michigan Tech. Although much of the workshop focused on ash retrievals and tracking, some of the key recommendations from that conference (Rose, 2001) are pertinent to this study:

-making recommendations for instrumental improvements for volcanic cloud remote sensing applications particularly for improvements of the Geostationary Operational Environmental Satellite (GOES) instruments after GOES-Q (~2009).

-providing constructive input to NOAA's forum for information about Natural Hazards Information Strategy.

-encouraging Moderate Resolution Imaging Spectroradiometer (MODIS) use for volcanic clouds.

-developing a strategy to provide better source term information for volcanic cloud trajectory models.

-building constructive partnership/linkages and input from the informal group of workshop attendees to other professional organizations worldwide (IAVCEI, AGU, CEOS, WMO, ICAO, AMS, etc.).

A second international workshop on the remote sensing of volcanic clouds is being planned for summer 2003, at Michigan Technological University. In this next workshop, we will try to improve on our initial efforts by encouraging representation from all of the Volcanic Ash Advisory Centers. We hope to have greater representation from volcano observatories and at least some representation from the aircraft industry and dispatchers. We also will encourage and recruit graduate students to attend and plan to increase the size of the workshop to 50-75 participants.

References

Andres, R.J., W.I. Rose, P.R. Kyle, S. de Silva, P.W. Francis, M.C. Gardeweg, and H. Moreno, 1991, Excessive sulfur dioxide emissions from Chilean volcanoes. J. Volc. Geotherm. Res., 46, 323-329.

Barker, S.L., G.J.S. Bluth, and A.J. Krueger, 1998, Examining the removal rate of volcanogenic SO₂ in the atmosphere. Eos Trans. AGU, 79, F935.

Bluth, G.J.S., S.D. Doiron, A.J. Krueger, L.S. Walter, and C.C. Schnetzler (1992) Global tracking of the SO₂ clouds from the June, 1991 Mount Pinatubo eruptions. Geophys. Res. Lett., 19, 151-154.

Bluth, G.J.S., C.C. Schnetzler, A.J. Krueger, and L.S. Walter, 1993, The contribution of explosive volcanism to global atmospheric sulphur dioxide concentrations. Nature, 366, 327-329.

Bluth, G.J.S., T.J. Casadevall, C.C. Schnetzler, S.D. Doiron, L.S. Walter, A.J. Krueger, and M. Badruddin (1994) Evaluation of sulfur dioxide emissions from explosive volcanism: the 1982-1983 eruptions of Galunggung, Java, Indonesia. J. Volc. Geotherm. Res., 63, 243-256.

Bluth, G.J.S., C.J. Scott, I.E. Sprod, C.C. Schnetzler, A.J. Krueger, and L.S. Walter (1995) Explosive SO2 emissions from the 1992 eruptions of Mount Spurr, Alaska. U.S. Geological Survey Bulletin 2139, 37-45.

Bluth, G. J. S., W.I. Rose, I.E. Sprod, and A.J. Krueger, 1997, Stratospheric loading from explosive volcanic eruptions, J. Geology, 105, 671-683.

Casadevall, T.J., 1994, The 1989-1990 eruptions of Redoubt volcano, Alaska: Impacts on aircraft operations. J. Volc. Geotherm. Res., 62, 301-316.

Crisp, J., 1995, Volcanic SO₂ Alert, NASA EOS IDS Volcanology Team Data Product document, Product # 3288 (http://eos.pgd.hawaii.edu/docs/dpd.html).

Goldman, A., F.J. Murcray, C.P. Rinsland, R.D. Blatherwick, S.J. David, F.H. Murcray, and D.G. Murcray, 1992. Mt. Pinatubo SO₂ column measurements from Mauna Loa, Geophys. Res. Lett., 19, 183-186.

Gu, Y., 2001, Laboratory study of SO₂ adsorption onto fine volcanic ash. M.S. Thesis, Michigan Technological University.

Guo, S., W.I. Rose, G.J.S. Bluth, C. Textor, and H.-F. Graf (2000) ATHAM model simulation of the 15 June 1991 Pinatubo volcanic plume. Eos Trans. AGU, 81, F1287.

Krueger, A.J., S. Schaefer, N. Krotkov, G. Bluth, and S. Barker (2000) Ultraviolet remote sensing of volcanic emissions and applications to aviation hazard mitigation. Geophysical Monograph Series 116, 25-43.

Krotkov, N.A., A.J. Krueger and P.K. Bhartia, 1997, Ultraviolet model of volcanic clouds for remote sensing of ash and sulfur dioxide. J. Geophys. Res., 102, 21,891-21,904.

Krotkov, N.A., D.E. Flittner, A.J. Krueger, A. Kostinski, C. Riley, W. Rose, and O. Torres, 1999, Effect of particle non-sphericity on satellite monitoring of drifting volcanic ash clouds. J. Quant. Spect. Rad. Trans., 63, 613-630.

Mankin, W. G., M.T. Coffey, and A. Goldman, 1992, Airborne observations of SO₂, HCl, and O₃ in the stratospheric plume of the Pinatubo volcano in July 1991, Geophys. Res. Lett., 19, 179-182.

Mass, C.F., and D.A. Portman, 1989, Major volcanic eruptions and climate: A critical evaluation. J. Climate, 2, 566-593.

McKeen, S.A., S.C. Liu, and C.S. Kiang, 1984, On the chemistry of stratospheric SO₂ from volcanic eruptions: J. Geophys. Res., 89 (D3), 4873-4881.

Oppenheimer, C., P. Francis, and J. Stix, 1998, Depletion rates of sulfur dioxide in tropospheric volcanic plumes. Geophys. Res. Lett., 25, 26671-2674.

Prata, A.J., M. Watson, W.I. Rose, V. Realmuto, G.J.S. Bluth, R. Servranckx, 2002, Volcanic sulfur dioxide concentrations derived from infrared satellite measurements, in prep.

Rampino, M.R., and S. Self, 1984, Sulfur-rich volcanic eruptions and stratospheric aerosols. Nature, 310, 677-679.

Read, W.G., Froidevaux, L., and J.W. Waters, 1993, Microwave Limb Sounder measurement of stratospheric SO₂ from the Mt. Pinatubo volcano. Geophys. Res. Lett., 20, 1299-1302.

Realmuto, V.J., M.J. Abrams, M.F. Buongiorno, and D.C. Pieri, 1994, The use 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., 99, 481-488.

Realmuto, V.J., A.J. Sutton and T. Elias, 1997, Multispectral thermal infrared mapping of sulfur dioxide plumes: A case study from the East Rift Zone of Kilauea Volcano, Hawaii. J. Geophys. Res., 102, 15,057-15,072.

Realmuto, V.J., and H.M. Worden, 2000, Impact of atmospheric water vapor on the thermal infrared remote sensing of volcanic sulfur dioxide emissions: A case study from the Pu'u 'O'o vent of Kilauea Volcano, Hawaii, J. Geophys. Res., 105(B9), 21,497-21,508.

Rothman, L.S., R.R. Gamache, A. Goldman, L.R. Brown, R.A. Toth, H.M. Pickett, R.L. Poynter, J-M. Flaud, C. Camy-Peyret, A. Barbe, N. Husson, C.P. Rinsland, and M.A.H. Smith, 1987, The HITRAN database: 1986 Edition, Appl. Opt., 26, 4058-4097.

Rose, W.I., D.J. Delene, D.J. Schneider, G.J.S. Bluth, A. J. Krueger, I. Sprod, C. McKee, H.L. Davies, and G.G.J. Ernst, 1995, Ice in the 1994 Rabaul eruption cloud: implicat