Michigan Tech professors and students presented papers at the American Geophysical Union annual meeting. Geology and geological engineering students presented papers on: Gas and ash emissions, Remote Sensing, Seismic Monitoring, Lahar hazards, Volcanic Aerosols, Dome morphology. Travel for the presenters to the AGU Meeting in San Francisco was thanks to: MTU Graduate School, Remote Sensing Institute, and the GMES Department.

		Shown at left is part of the group that attended, these being the volcanology contingent
	Shown in this series of photos above and left, is the Michigan Tech booth at the AGU convention.

Observing Popocatepetl's Volcanic Clouds Using MODIS Infrared Data
Matiella, M , Michigan Tech

Delgado-Granados, H, Instituto de Geofisica, Universidad Nacional Autonoma de Mexico, UNAM, Ciudad Universitaria, Coyoacan, Mexico DF, Mexico

Rose, W I, Michigan Tech

Watson, I, Michigan Tech

Popocatepetl Volcano, Mexico, is a tropical volcano with significant and persistent
emissions of SO₂ and ash. These emissions pose significant hazards to the
large population in close proximity to the volcano and are an important indicator of
eruptive activity (Love et al., 1998, Nature 396: 563-566). Moderate Resolution
Imaging Spectroradiometer (MODIS) satellite imagery provides us with a synoptic
perspective of volcanic emissions and atmospheric interactions, information
unavailable from ground-based or aircraft studies, which can be useful for hazard
mitigation. We report on MODIS data during December 2000 and January 2001
coincident with abundant emissions based on COSPEC data. SO₂ masses are
retrieved using the 7.3 mm and 8.6 mm absorption features of SO₂
(7.3 $\mu$m method after Prata et al, in press, AGU Volcanism Atmospheric
Monograph; 8.6 mm method after Realmuto et al, 1997, J. Geophys Res 102:
15057-15072). Ash masses are retrieved using silicate absorption features at 11mm and 12 mm (Wen \& Rose, 1994, J Geophys Res 99: 5421-5431).
Furthermore, this data set tests the accuracy of the algorithms and the conditions
under which the algorithms work best, and can be compared to COSPEC
measurements taken by Popocatepetl's monitoring team. We found that MODIS data
often showed more than one volcanic cloud. For example, one MODIS image collected
January 23rd, 2001, at 0450 UT, shows four large eruptions that have dispersed
volcanic clouds over an extensive area of Mexico. Using upper air data and monitoring
records, the movements of the 4 ash clouds are fit with eruption times and winds,
and using retrieval data for SO₂ and ash we can derive a time based SO₂
and fine ash emission record. The results of these retrievals compliment
ground-based measurements which cannot measure large scale eruptions.

Using Internet-Based Automated Software to Process GPS Data at Michigan Tech University
Crook, A, Michigan Tech

Diehl, J F, Michigan Tech

The Michigan Tech University GPS monument was made operational in October of
2002. The monument, which consists of a concrete pillar extending approximately 10
feet below the surface and protrudes 5 feet above ground, is located at the
Houghton County Memorial Airport (47.171803 degreesN, 88.498361 degrees W). The
primary purpose of the monument is to measure the velocity of the North American
Plate at this location. A Trimble 4000ssi geodetic receiver with a Trimble Zephyr
antenna is used to collect GPS data. The data are sent to a PC where they are
processed using Auto-GIPSY, an internet-based GPS processing utility, which makes
it possible to process GPS data, via email, without having knowledge of how the
software works. Two Perl scripts were written to facilitate automation and to simplify
processing of the GPS data even further. Twelve months of GPS data were
processed, using Auto-GIPSY, which produced a velocity of -24 - 5 mm/yr and
-4 - 6 mm/yr for the X and Y components respectively with an azimuth of
261 degrees with respect to the ITRF2000. This calculated result compares well with
the NNR-NUVEL1A velocity of -17 mm/yr and -1 mm/yr for the X and Y components
respectively with an azimuth of 267 degrees. The results from an alternative online
processing service, the Scripps Coordinate Update Tool (SCOUT) that uses GAMIT,
will also be presented as a comparative method.

Elevated Uranium in Aquifers of the Jacobsville Sandstone
Sherman, H, Michigan Tech

Gierke, J, Michigan Tech

The EPA has announced a new standard for uranium in drinking water of 30 parts per
billion (ppb). This maximum contaminant level (MCL) takes effect for community water
supplies December 2003. The EPA's ruling has heightened awareness among
residential well owners that uranium in drinking water may increase the risk of kidney
disease and cancer and has created a need for a quantified, scientific understanding
of the occurrence and distribution of uranium isotopes in aquifers. The authors are
investigating the occurrence of elevated uranium in northern Michigan aquifers of the
Middle Proterozoic Jacobsville sandstone, a red to mottled sequence of sandstones,
conglomerates, siltstones and shales deposited as basin fill in the 1.1 Ga Midcontinent
rift. Approximately 25% of 300 well water samples tested for isotopic uranium have
concentrations above the MCL. Elevated uranium occurrences are distributed
throughout the Jacobsville sandstone aquifers stretching across Michigan's Upper
Peninsula. However, there is significant variation in well water uranium concentrations
(from 0.01 to 190 ppb) and neighboring wells do not necessarily have similar
concentrations. The authors are investigating hydrogeologic controls on ground water
uranium concentrations in the Jacobsville sandstone, e.g. variations in lithology,
mineralogy, groundwater residence time and geochemistry. Approximately 2000' of
Jacobsville core from the Amoco St. Amour well was examined in conjunction with the
spectral gamma ray log run in the borehole. Spikes in equivalent uranium (eU)
concentration from the log are frequently associated with clay and heavy mineral
layers in the sandstone core. The lithology and mineralogy of these layers will be
determined by analysis of thin sections and x-ray diffraction. A portable
spectrometer, model GRS-2000/BL, will be used on the sandstone cliffs along Lake
Superior to characterize depositional and lithologic facies of the Jacobsville sandstone
in terms of concentrations and ratios of eU, eTh and K. Equipped with borehole
accessories, the spectrometer will be used to log residential drinking wells to
determine a relationship between the uranium concentration of well water and the eU
concentration in the sandstone. Tritium/helium-3 dating will be used to determine
whether ground water uranium concentrations increase with residence time.
PHREEQCI will be used to model dominate aqueous species of uranium and saturation
indices of uranium minerals.

Analysis of the impact of soil heterogeneity on optimal policies for groundwater remediation
Bau, D, Michigan Tech,

Mayer, A S, Michigan Tech

Typical groundwater remediation problems involve the design of the number, location
and flow rate schedule of pumping and injection wells. Simulation models combined
with optimization models are used to rank alternatives while considering management
objectives, e.g. minimizing remediation cost and/or maximizing cleanup efficiency, and
constraints, e.g. the maximum permissible concentrations at selected compliance
sites. Mostly due to both high computational effort required and lack of data, the
simulation models often are based on simplified 2D homogeneous hydrogeologic
settings. The purpose of this work is to investigate how simplifying hypotheses may
affect the final optimal remediation policy. In particular, the analysis addresses the
case of a heterogeneous layered aquifer versus an homogeneous one with an
equivalent (lumped) hydraulic conductivity. To simulate groundwater flow and
contaminant transport, use is made of a fully 3D finite element unsaturated flow
model along with a particle tracking transport code. The flow and transport code is
then coupled to a genetic algorithm model to optimize the specified objective
function. The problem considered is the remediation of a hypothetical
aquifer-contaminant system using pump and treat. The objective is to minimize the
cost of the remediation system. The cost function is a nonlinear function of decision
variables (pumping rates) and state variables (hydraulic heads and contaminant
concentrations). Constraints include limits on hydraulic head and the contaminant
mass remaining in the aquifer at the end of the remediation. The results of
homogeneous and heterogeneous simulations are compared in terms of cost and
values of the decision variables.

Santiaguito Volcano, Guatemala: A Study of Vent Dynamics Over the Past 50 Years
Head, E M, Michigan Tech

Dalton, M P, Michigan Tech

Bluth, G J, Michigan Tech

Rose, W I, Michigan Tech

October 1902 marked one of the Earth's largest historic eruptions at Santa Maria
volcano, Guatemala. Since 1922, volcanic dome extrusion has been occurring on the
southwest flank of Santa Maria from the continuously active Santiaguito vent
system. Dangerous activity including vertical ash eruptions, avalanches, lava flows,
and pyroclastic flows has been persistent from Santiaguito, requiring observations of
the volcano to be carried out from a distance. Consequently, our understanding of
subsurface processes, such as the shape, size, and position of the magma body and
conduit systems must be inferred by remote measurements. There may be a
correlation at Santiaguito between the effects of subsurface conduit dynamics on
summit vent morphology and subsequent volcanic activity. Our recent observations
have identified the presence of a ring-shaped vent, approximately 150m wide, within
the Santiaguito summit dome, which is thought to represent the surface expression of
a series of fractures generated from the conduit at depth. Thus, a time series of
surface observations of vent geometry and observations of volcanic activity could
help constrain the near surface conduit evolution. An archive of aerial photos from
1947 to 2000, photographs of vent activity from 1969 to present, and 11 hours of
digital video from 2002 and 2003 volcanic activity at Santiaguito will be analyzed to
discover how these observations are related. The objective of this project will be to
better understand the conduit evolution at Santiaguito volcano by documenting
changes in vent morphology over the last 50 years and to correlate these findings
with eruptive history. These results can be further correlated with gas, thermal,
seismic, and other geophysical studies, calculations of dome volume and dome
dimensions, and predictive modeling of Santiaguito's volcanic activity.

Photometric Observations of Aerosol Plumes From Volcanoes in Guatemala, El Salvador, and Nicaragua
Lopez, T M, Michigan Tech

Watson, I, Michigan Tech

Rodriguez, L A, Michigan Tech

Branan, Y K, Michigan Tech

Rose, W I, Michigan Tech

Bluth, G J, Michigan Tech

Visible to near infrared sun-photometers were used to measure spectral optical
depths in order to infer particle size distributions of volcanic aerosols in plumes from
volcanoes in the Central American arc. Data were taken from the following volcanoes
on the listed dates in 2002: Pacaya, Guatemala on January 14, 16, 20 and 21; Santa
Ana, El Salvador on January 24 and 26; San Miguel, El Salvador on January 28; and
San Cristobal, Nicaragua on February 3. These volcanoes were chosen for study
because of: good calibration of the sun-photometer (Pacaya), the presence (and
effects) of a crater lake (Santa Ana) and a paucity of previous measurements (San
Miguel and San Cristobal). The optical properties of these tropospheric volcanic
aerosols will be retrieved after the removal of the background optical depth. Through
the application of the Angstrom equation and a King-type inversion, the Angstrom
coefficients, the particle size distribution, and the effective radius (Reff) will be
determined (Watson and Oppenheimer, 2000; 2001). Through these methods we hope
to increase the understanding of emission and conversion processes of tropospheric
volcanic aerosols. We hope the interpretation of these data will help elucidate
environmental and climatic effects of these aerosols on local to global scales, and
provide insight into modulation of aerosol emissions through the presence of a crater
lake.

Detection and Fate of the August 18 and 28, 2000 Eruption Clouds of Miyakejima, Japan: An Analysis Using TOMS, MODIS, AVHRR, GMS, and ASTER

McCarthy, E, Michigan Tech

Bluth, G, Michigan Tech

Watson, I, Michigan Tech

Rose, W, Michigan Tech

Tupper, A, Darwin Volcanic Ash Advisory Centre, Commonwealth Bureau of Meteorology, Northern
Territory Regional Office, Casuarina, NT 0811 Australia

Kamada, Y, Tokyo Volcanic Ash Advisory Center, Japan Meteorological Agency, Tokyo, Japan

Volcanic eruptions eject ash, sulfur dioxide, and other gases into the atmosphere.
These volcanic products create a variety of hazards, including health and aviation
hazards and changes in climate. Some of the major questions regarding the mitigation
of these hazards include: the mass, extent, and height of the cloud, how long it
takes for SO₂ to convert to sulfuric acid, and how long aerosols remain in the
atmosphere. Remote sensing techniques allow: long-term tracking of volcanic clouds
because data are acquired on a regular basis, analysis of eruptions in isolated areas,
and measurements of an entire eruption cloud. New techniques for retrievals have
been developed and new sensors have been launched, however, there has been no
systematic comparison to understand the capabilities of the sensors under varying
environmental and volcanological conditions. In this case study of Miyakejima, Japan,
data from five different satellite sensors are compared and used to produce
constraints on the masses and distributions of ash, SO₂, and aerosols released
by the August 18 and 28, 2000 eruptions. Satellite data include: 6 TOMS, 6 MODIS,
11 AVHRR, and 2 ASTER images, as well as 8 days of hourly GMS images. Preliminary
results, comparing TOMS and MODIS, show the August 18 ash and gas cloud drifting
south. Retrievals using the two sensors suggest an ash mass of at least 26 kilotonnes
(kt). The August 28 eruption cloud drifted to the northeast and contained
approximately 11 kt of SO₂.

Correlation of SO₂ Gas Emissions, Seismicity and Thermal Signals at Santiaguito, Guatemala
Branan, Y K, Michigan Tech

Watson, I , Michigan Tech

Harris, A J, University of Hawai'i, Hawai'i Institute of Geophysics and Planetology

Rose, W, Michigan Tech

Bluth, G J, Michigan Tech

Chigna, G, Insituto Nacional de Sismologia, Vulcanologia, Meteorologia y Hidrologia, Guatemala City, Guatemala

Mota, M, Insituto Nacional de Sismologia, Vulcanologia, Meteorologia y Hidrologia, Guatemala City, Guatemala

With vertical explosions occurring approximately every 40-50 minutes, the Santiaguito
dome at Santa Maria Volcano is an ideal system for examining short-term data
patterns. A 3-week long field experiment was performed in January 2003 at the
Santiaguito Volcano Observatory in order to record high temporal resolution
measurements of volcanic activity. We collected digital seismic data from a single
vertical component seismometer located approximately 4 km southeast of the active
Caliente vent. A portable infrared thermal monitoring unit was deployed daily to
record the temperature of the plume as it left the vent at an acquisition rate of 300
measurements per minute. A miniature ultraviolet spectrometer (MUSE) was also
deployed daily to measure the SO$_{2}$ gas emissions just above the vent. This
instrument is based on the differential optical absorption spectroscopy (DOAS)
technique and allowed for continuous readings at a rate of 36 measurements per
minute from approximately 6.5 km south of the Caliente vent. At abstract time, the
seismic data is not analyzed, but there is a strong correlation between the SO$_{2}$
emission and thermal data showing that the expulsed gas heats the dome extensively
as it is emitted, with a possibility of different signatures indicating certain types of
activity such as pyroclastic flows. It is expected that, with the addition of seismic
data and the application of analysis of periodicity using Fourier Transforms, the data
will elucidate conduit processes, providing additional vital constraints to sub-surface
models.

SO₂ loss rates at Lascar volcano, Chile: preliminary results and interpretations from 2002 measurements
Rodriguez, L A, Michigan Tech

Watson, I, Michigan Tech

Tassi, F, Dipartimento di Scienze della Terra, Universita degli Studi di Firenze, Firenze,
Italy

Viramonte, J, Instituto GEONORTE, Facultad de Ciencias Naturales, Universidad Nacional de Salta, Buenos Aires, Salta, Argentina

Poodts, M, Instituto GEONORTE, Facultad de Ciencias Naturales, Universidad Nacional de Salta, Buenos Aires, Salta, Argentina

Rose, W I, Michigan Tech

Bluth, G J, Michigan Tech

The measurement of volcanic SO₂ emission rates is commonly performed on
cross sections of the plume some distance downwind from the active vent, some
time after its interaction with other volcanogenic gases (primarily water), particles
and droplets of volcanogenic and/or meteoric origin, and atmospheric gases. The
emission rates being measured therefore do not necessarily represent the real fluxes
emitted by the volcano; instead they are underestimated due to the conversion of
SO₂ to SO₄ (Oppenheimer et al., 1998). Near source plume
chemistry is not well understood, but can have significant effects on climatologically
active species, which is why it is important to quantify volcanic SO₂$
conversion rates as a function of meteorological environment and plume age. A
mini-UV spectrometer and a Microtops II sun photometer were used to measure
SO₂ emission rates and aerosol particle size distributions at Lascar volcano,
Central Andes, during the months of October and November 2002. Direct gas
measurements from fumaroles in the crater were made on November 1. These
represent the first gas samples ever collected directly in the crater. Lascar volcano
represents one of the end-members of the environmental spectrum, being a high
volcano (summit altitude at 5600 meters above sea level) in a dry atmosphere
(average 15). Its location allows for simultaneous near-vent and downwind
measurements at similar altitudes. Here we present our preliminary results from
November 2, 2002, taken at a variety of azimuthal angles (effective distances from
the plume) from the vent downwind to about 20 km, during a two-hour period
(0900-1100 local time). This represents a time downwind of up to 42 minutes,
based on a plume speed of 8 m/sec. Initial interpretations and correlations with
the direct measurements are also presented. The data obtained from a Kestrel 4000
weather station will help clarify the effects of Lascar's high, dry, and extremely
transmissive atmosphere upon SO₂ conversion rates.

Landsat TM and ETM+ Time Sequence of Lahar Hazards on Fuego Volcano, Guatemala
Reif, S L, Michigan Tech

Bluth, G J, Michigan Tech

Rose, W I, Michigan Tech

Matias, O, INSIVUMEH, Guatemala, Guatemala

Volcanic hazards pose a threat to a large number of the world's population, especially
secondary hazards due to remobilization of volcanic material such as landslides and
lahars. Many hazard-prone areas would benefit by remote sensing tools for hazard
mitigation. In this study, we propose to use remote sensing and GIS techniques to
map these hazard prone areas around Fuego volcano, Guatemala and provide
information to local organizations to assist in mitigation. Fuego is a steep sided
volcano with a history of large eruptive events, including the well-studied 1974
eruption, that have extruded a large amount of material onto the upper reaches of its
watersheds. The volcano is well studied, but historically more emphasis has been
placed on eruption processes. A study of the way material moves down Fuego and to
the extent that it moves is needed to help mitigate the range of potential hazards.
We propose an in-depth remote sensing survey to map the hazard-prone areas. The
study will consist of processing 20 years (15 cloud-free images) of Landsat TM and
ETM+ data to look at changes in landforms and vegetation. Vegetation indices will be
calculated to locate areas devoid of vegetation and a masking process will be used
These area changes will be related to field measurements to create GIS to measure
the area of these zones. layers denoting geometry changes in the channels around
Fuego. These changes will be loaded into a GIS, along with regional climate data,
DEMs, hydrologic data, infrastructure, and information about the known volcanic
activity recorded in the area by the local volcanologists. Modeling of lahars using
LAHARZ and climate data will also be done to determine an estimate of the amount of
material moved and to what distances it can be transported. A field survey
undertaken in January 2003 acquired GPS ground truth data of landslide boundaries
and channel volumes for the GIS. The deposits that were seen in the channels 10 km
from the source of the sediment ranged in thickness from 1.5-8 m.

Forward modeling of volcanic aerosols transmissions at different latitudes; quantifying the effects of varying tropospheric water vapor on ash detection.

Watson, I M, Michigan Tech

Rose, W I, Michigan Tech

Realmuto, V J, Visualization and Scientific Animation Group, Jet Propulsion Laboratory, Pasadena, CA

Bluth, G J, Michigan Tech

Current algorithms for volcanic ash cloud detection by satellite, based upon a
difference in transmissivity of silicate ash at 11 and 12 microns, are subject to
significant and yet poorly quantified modulation by atmospheric water vapor. In order
to investigate these effects we have written an aerosol forward model, based upon
Mie-scattering code, and embedded it in a MODTRAN-based atmospheric radiative
transfer model. This facilitates investigation of errors associated with inverse
solutions derived from satellite data, and more importantly for this study, can be used
to vary the atmosphere within which a hypothetical volcanic ash cloud is contained.
We have taken images of the most well parameterized ash clouds of the satellite age,
the sequence of eruptions of Mt. Spurr in mid to late 1992, and applied the forward
model to predict transmission spectra of the ash clouds. Calculations were performed
for the ash clouds both within the cloud's original atmosphere and within that of a
sequence of eruptions of Soufriere Hills Volcano, Montserrat, in the British West
Indies in the late 1990s. Preliminary results indicate a +1-3.5 K brightness
temperature difference (BTD) effect associated with increasing the temperature and
water vapor content of the atmosphere. This translates to a significant loss of
detectability in terms of cloud area; up to half the area of the clouds no longer have
a negative brightness temperature difference, and an effect on reducing the optical
depth and mass of the cloud. In most cases there was very good agreement between
the 'clear ocean' BTD and the modeled effect, suggesting a potential source of
in-image calibration for detection algorithms. In terms of mass and optical depth
retrievals however, it is obvious that both 11 and 12 micron channel radiances are
affected by the presence of water vapor, clearly indicating the need for atmospheric
correction before quantifying the size or concentration of ash in clouds from satellite
data.

Volcanic SO2 Emissions vs. Seismicity - July 2002 LP Swarm, Soufriere Hills Volcano, Montserrat
Shannon, J, Michigan Tech

Bluth, G, Michigan Tech

Edmonds, M, Montserrat Volcano Observatory, Fleming, Montserrat, W. Indies

Thompson, G, British Geological Survey, Keyworth, Nottingham, NG125GG United Kingdom

Volcanic sulfur dioxide (SO2) measurements of passive plumes have recently improved
with the application of Differential Optical Absorption Spectroscopy (DOAS). In
January 2002, the Montserrat Volcano Observatory installed two fixed DOAS
instruments which collect rapid, continuous measurements of SO2 emissions. For the
first time, SO2 fluxes are being collected on a time scale of minutes, allowing
short-term changes to be evaluated with respect to atmospheric transport, surface
activity, and magmatic source mechanisms. In this study, we investigate relationships
between SO2 emissions and seismicity for July 2002 at Montserrat. The seismic data
consist of rockfall, long-period rockfall, hybrid, and long-period (LP) waveforms.
Seismic energy increased slowly throughout July, culminating in a swarm of
long-period earthquakes. The SO2 data used for this study consist of measurements
collected every 4-5 minutes from approximately 8 a.m. to 5 p.m. daily by a DOAS
instrument positioned at Lovers Lane (~4.5 km west of vent). From July 1-18,
seismicity is dominated by rockfalls with intermittent hybrids and LPs. During this
time, SO2 fluxes generally increase and coincide with an increasing trend in the
number of rockfalls per day. Average daily SO2 fluxes range from 140 to 1256
tonnes/day with an average of 616 tonnes/day. On July 19, however, the average
daily SO2 flux decreases from a monthly maximum of 1256 tonnes/day to 227
tonnes/day and continues to average only 326 tonnes/day from July 20-30. July 19
marks a seismic transition into a LP earthquake swarm that continues into August.
After the LP swarm begins and SO2 fluxes decrease, however, the number of daily
rockfalls remains relatively high. Therefore, it appears that LPs are more strongly
linked to SO2 emissions than rockfall activity. Future work will involve investigating
short-term (minutes/hours) relationships between seismicity and SO2 emissions.

Quantifying Volcanic Emissions of Trace Elements to the Atmosphere: Ideas Based on Past Studies
Rose, W I, Michigan Tech

Extensive data exist from volcanological and geochemical studies about exotic
elemental enrichments in volcanic emissions to the atmosphere but quantitative data
are quite rare. Advanced, highly sensitive techniques of analysis are needed to
detect low concentrations of some minor elements, especially during major eruptions.
I will present data from studies done during low levels of activity (incrustations and
silica tube sublimates at high temperature fumaroles, from SEM studies of particle
samples collected in volcanic plumes and volcanic clouds, from geochemical analysis
of volcanic gas condensates, from analysis of treated particle and gas filter packs)
and a much smaller number that could reflect explosive activity (from fresh ashfall
leachate geochemistry, and from thermodynamic codes modeling volatile emissions
from magma). This data describes a highly variable pattern of elemental enrichments
which are difficult to quantify, generalize and understand. Sampling in a routine way
is difficult, and work in active craters has heightened our awareness of danger, which
appropriately inhibits some sampling. There are numerous localized enrichments of
minor elements that can be documented and others can be expected or inferred.
There is a lack of systematic tools to measure minor element abundances in volcanic
emissions. The careful combination of several methodologies listed above for the
same volcanic vents can provide redundant data on multiple elements which could
lead to overall quantification of minor element fluxes but there are challenging issues
about detection. For quiescent plumes we can design combinations of measurements
to quantify minor element emission rates. Doing a comparable methodology to
succeed in measuring minor element fluxes for significant eruptions will require new
strategies and/or ideas.

Integration of Seismic and Video Records of Eruptive Activity on Santiaguito Volcano, Guatemala
Dalton, M P, Michigan Tech

Head, E M, Michigan Tech

Bluth, G J, Rose, Michigan Tech

W I, Molina, E, INSIVUMEH, Guatemala City, Guatemala

Toorongian, J E, Michigan Tech

Santiaguito is a long-lived, dacitic volcanic dome complex, which has now been
continuously active since 1922. The volcano's longevity is thought to be related to a
large magma body and the pattern of extrusion is notably unsteady. Direct
observations of activity are inhibited by logistics, such as cloudy weather at proximal
vantage points after mid morning. Activity observable at Santiaguito each morning
includes multiple vertical explosions, dome and flow collapses leading to block and
ash flows or rock avalanches, steam exhalations and fumarolic discharges. These
phenomena were recorded on digital video from an excellent vantage point (the
summit of Santa Maria) located 1200 m above and 2.5 km NE of the active Caliente
Vent of Santiaguito for about three hours on the morning of January 11, 2003, and
from El Brujo Vent, located $\sim$200 m below and 1.2 km W for about four hours on
the morning of January 9, 2003. The volcano is also ismically monitored, and we have
synchronized the video with digital seismic data processed through the Seisan
computer program. Our expected result is a correlated record of seismic data and
visually-identified eruptive activity, allowing us to evaluate eruption characteristics.
This will be in the form of a digital video of surface activity on Santiaguito with
corresponding seismic signals, which will be applicable for education and scientific
analysis. By correlating seismic signals to eruptive activity, monitoring agencies that
are distant from the volcano can better interpret the incoming seismic data, and
communicate a more complete picture of volcanic activity to the surrounding
communities. It also allows us to go back through the seismic record from Santiaguito
and catalog past eruptive activity. A synchronized seismic and eruptive activity
video is also a valuable educational/ outreach tool. With the technique for correlating
video and seismic records established, additional types of data (such as temperature
and gas readings) can be compared as well.

Modeling Of LARSE II Refraction Data In The Western Mojave Desert, California
Lim, K, Michigan Tech

Lim, K., USGS, 345, Menlo Park, CA

Baher, S, USGS, 3 Menlo Park, CA

Fuis, G, USGS, Menlo Park, CA

The major goals of the LARSE II (Los Angeles Region Seismic Experiment, Part II) are
to image the subsurface in Los Angeles region utilizing reflection and refraction
surveys, to understand the detailed structure of the subsurface in relation to
earthquakes. LARSE II data acquisition was accomplished in October 1999 along the
150km long main line that extended from Santa Monica Bay to the western Mojave
Desert. We modeled LARSE II refraction data in the western Mojave Desert using a
forward modeling technique, fitting the first arrivals in most cases less than 0.05
second. In general, layers were modeled to match the breaks in slope of the first
arrival branches. Three layers over basement were modeled. They have velocities or
velocity ranges (in km/s) of 1.0 or less (Layer 1), 2.0-3.4 (Layer 2) and 3.3-3.9
(Layer 3). Preliminary interpretations are as follow: Layer 1, up to 0.2km thick, is
unsaturated, unconsolidated Quaternary sediments. Layer 2 and 3, which thicken to
a total of 1.5km in the central Mojave Desert, are Quaternary and Tertiary
sedimentary rocks. An oil-test well in the southern Mojave Desert bottoms at the
base of Layer 3 (1.04 km depth) in the Tertiary rocks. An oil-test well in the northern
Mojave Desert bottoms in the basement (quartz diorite) at the depth of 1.01km.
Upward steps in basements can be seen on the north and south sides of the western
Mojave Desert, at or near mapped faults. In summary, maximum depth to basement is
1.7 km in the western Mojave Desert.

TOMS and Volcanic SO₂: an Important aid to the Understanding of Volcanism and the Atmosphere
Rose, W I, Michigan Tech

Bluth, G J, Michigan Tech

It is impossible to measure gas fluxes from the ground in a major volcanic event, but
the TOMS instrument provided the first quantitative measurements of individual
stratospheric eruptions, because SO₂ could be measured as well as O₃.
The measurements were quickly noticed by scientists, because the masses of sulfur
erupted often far exceeded what they expected to find, based on petrology and its
supposed constraints, by surprising factors of 10 to 100. This result is still not well
understood, and is an important driving idea for volcanologic research. TOMS was
applied globally and the explosive volcanic flux of SO₂ to the atmosphere was
compiled for the first time using direct measurements --an important input to earth
systems analysis. Comparison of TOMS volcanic cloud SO₂ maps with infrared
volcanic ash cloud maps showed that there is often spatial separation of gas-rich
volcanic clouds emplaced higher in the atmosphere and ash-rich clouds which are
lower and which drift in different directions because of windshears. Sequential
examination of TOMS data showed that SO₂ masses in volcanic clouds
increases for 24 hours or more after eruption. The best explanation of this increase is
that ice which forms early in volcanic clouds captures SO₂ which is then
released again as the stratospheric ice sublimes. The presentation will document all
of the best examples of the discoveries listed above. Volcanologists and those
interested in the mitigation of volcanic cloud hazards have repeatedly suggested that
geostationary SO₂ and ash sensing capability at higher spatial resolution would
provide important new science opportunities. The sensors of the next remote sensing
era (MODIS, ASTER, SEVIRI, OMI, ABI) bring us closer to achieving these goals.

The Role of TOMS in Understanding the Fates of Volcanic Emissions
Bluth, G J, Michigan Tech

Rose, W I, Michigan Tech

Guo, S, Michigan Tech

Carn, S , JCET, University of Maryland Baltimore County, Baltimore, MD

The Total Ozone Mapping Spectrometer (TOMS) has observed over 100 eruption s
during the past 25 years, from both explosive and effusive sources. The four TOMS
instruments (Nimbus-7, Meteor, ADEOS, and Earth Probe) have gen erated an
unprecedented archive of eruption data and allowed us to look at processes, in
addition to describing individual events. The TOMS sensors are less affected by
atmospheric water vapor and optical thickness than i nfrared techniques, and thus
are able to return information on the cloud f rom it's early, most concentrated form,
to dilute cloudmasses several days later. From the many observations and advances
generated from TOMS data, here we document those which pertain to the first few
days following emis sion into the atmosphere. Following discrete eruption events, we
have observed that the mass of retr ieved SO2 often increases for 1-2 days,
independent of any volcanic contri bution. Combining TOMS with other sensor data
suggests that significant S O2 is sequestered by ice in the rising plume. Ablation of
the ice slowly re-releases SO2, which results in the apparent increase. After 2-3
days, this process appears to be largely complete, and SO2 removal then follows an
exponential decay rate. TOMS-derived removal rates of SO2 have ranged from
approximately 25 days (e-folding time) for Pinatubo-sized eruptions, to less than one
day for smaller or tropospheric eruptions. Within the sa me eruption this rate may
vary, as SO2 removal is strongly affected by ads orption onto co-existing ash and ice
particles. The removal processes can also be linked to an eruption height threshold,
separating eruptions whic h are emplaced above the tropopause and produce
potentially long-lasting a tmospheric impacts from those which are rapidly removed
from the atmospher e. We have also observed that many eruptions produce a vertical
separatio n of gas-rich and ash-rich phases. However, other events have produced
no separation, suggesting that the separation may be linked to the eruption dynamics
or an early, gas-enriched pulse rather than a post-eruption, grav itational process.

Optimization of Engineering Design of Subsurface Environmental Remediation Systems: Development and Testing of Community Benchmark Problems
Mayer, A S, Michigan Tech

Miller, C T, Center for the Advanced Study of the Environment Department of Environmental Sciences
and Engineering University of North Carolina, Chapel Hill, NC

It is well established that the design of economically efficient subsurface remediation
systems can benefit from the joint use of formal optimization and simulation models.
It is also well known that obtaining the optimal solution for such designs is usually
difficult and computationally expensive, due to the characteristic nonlinear,
nonconvex nature of the solution spaces. We believe that more rapid progress on
optimal design methods might result from both improved methods of evaluation and
comparison of existing methods on realistic problems and from the investigation of
novel methods not yet studied in subsurface remediation field. This work responds to
these needs. We have designed a set of systematic test problems to be attacked by
the engineering and mathematics community, as a means for benchmarking and
comparing optimization approaches. The test problems pose many of the difficulties
anticipated in solving real-world problems such as (a) mixed continuous and integer,
nonlinear objective functions, (b) the combination of boundary conditions and system
parameters gives rise to complex relationships between the objective function, the
decision variables, the constraints, and the state variables, (c) evaluation of the
objective function is based on solving model equations that are difficult to solve
accurately and quickly; and (d) the number and range of decision variables is
potentially enormous. The physical problems include water supply design problems in
freshwater and freshwater-saltwater systems, a contaminant plume capture zone
design problem, and a contaminant plume pump-and-treat design problem. Problem
domains are specified in terms of hydraulic conductivity distributions- from
homogeneous domains to spatially-correlated random fields- and in terms of confined
vs. unconfined conditions. Each problem is specified completely in mathematical and
numerical terms, but sufficient flexibility is allowed to provide for a wide range of
problem solution approaches. For example, the user is free to select a flow and
transport simulator and to specify spatial and temporal discretizations. This paper will
describe the test problems and initial results.

Lost in Loess: Paleomagnetic investigation into loess and tephra deposits in interior Alaska
Semler, L, Michigan Tech.,

Arnold, K E, Dept. of Geology, Hamilton College, Clinton, NY

Williams, D, Dept. of Geography and Geology, Indiana State Univ., Terre Haute, IN

Morton, J, Dept. of Geology, College of William and Mary, Williamsburg, VA

Layer, P W, Geophysical Institute, Univ. of Alaska Fairbanks, Fairbanks, AK

Stone, D B, Geophysical Institute, Univ. of Alaska Fairbanks, Fairbanks, AK

Beget, J, Geophysical Institute, Univ. of Alaska Fairbanks, Fairbanks, AK

Schaefer, J , Alaska Div. of Geological and Geophysical Surveys, Fairbanks, AK

As a part of a NSF-funded Research Experience for Undergraduates (REU) program
at the University of Alaska Fairbanks, loess and tephra samples were collected at a
road cut near Tok, Alaska, to determine if the site was suitable for paleoclimate
reconstruction of the Pleistocene. Oriented cubes and cores were obtained from a
section of loess just below the Sheep Creek tephra dated at 190 $\pm$ 20 ka,
through the Tetlin tephra (1.2 meters below the Sheep Creek tephra) dated at 630
$\pm$ 50 ka to two meters below the Tetlin tephra. The accumulated amount of
loess between dated tephra layers suggests a span of more than one million years
for the whole section assuming a constant sedimentation rate for the loess. The
samples were measured for natural remanent magnetization (NRM) and magnetic
susceptibility before magnetic cleaning techniques (Alternating Field Demagnetization
and Thermal Demagnetization) were employed. Samples were analyzed using a
cryogenic magnetometer. The demagnetizations show stable magnetic vectors and
demonstrate that loess is a reliable paleomagnetic recorder. At Tok, we found that
the entire section is of normal polarity and shows no sign of the 780 ka
Brunhes-Matuyama polarity reversal, which was expected to be 42 cm below the
Tetlin tephra. Absence of the reversal may be due to discontinuities or other
changes in the sedimentation rate in the loess, a problem with the sampling
methods, or incorrect ages of the tephras. Based on our studies, we feel that
changing rates of loess deposition or other unseen discontinuities are the reason the
reversal was not found. Because of the uncertainty of the depositional history of the
Tok loess, this section is not useful for obtaining a continuous record of ancient
climate.