V21A-0377
Lattice Boltzmann model of lava lake convection and solidification
Lava lakes can be viewed as large-scale chemical and thermo-hydrodynamical laboratories, providing unique opportunities to study the coupling between fluid dynamics and liquid-solid phase changes in magmas. Previous studies of the thermal history of lava lakes were mostly based either on heat conduction models, apply a 1D treatment of convective cooling, or use experimental analogs. Here we describe a new numerical tool to solve for the evolution of fluids that can convect and undergo a phase change (solidification in this case). The numerical model is based on the lattice Boltzmann method and includes features such as floor and roof cooling, temperature- and crystallinity-dependent viscosity. Crystallization occurs at the two cooling interfaces (roof and floor), as well as within the subcooled convecting fluid. Convection is driven by the negative buoyancy of high solid fraction magma, leading to driping instabilities originating from the thermal boundary layer under the roof of the lake. Our model permits us to study the effects of convection on the solidification time of lava lakes. We also investigate the evolution of the geometry of the solidification fronts. Finally, we test the importance of the choice of upper thermal boundary condition, namely isothermal versus radiative. The lattice Boltzmann approach presented here can be further modified to study chemical differentiation processes and crystal settling in a magmatic environment subjected to cooling.
V21A-0378
Structural Development and Oxidation of the Takanoobane Rhyolite Lava in Aso Caldera, Japan
The Takanoobane rhyolite lava (hereafter described as the TR lava) is distributed in the western part of Aso caldera, middle Kyushu Island, SW Japan. The TR lava is one of the central cones. The volume, SiO2 contents and K-Ar age are 0.14km3 (Miyabuchi et al., 2004), 71-72% (Furukawa, 2006) and 51+-5ka (Matsumoto et al., 1991), respectively. The TR lava was effused in a subaerial environment. In this study, we show vertical structural variation and the development of the TR lava from the four drilling cores obtained by Aso Volcanological Laboratory in 2001-2002. The TR lava is about 90m thick in the proximal part, and the internal structures are divided into three parts: Alternation of the pumiceous layers and the obsidian layers (the upper part), the crystalline rhyolite layer (the central part), and the obsidian layer (the lower part). This structural variation apparently resembles to that of the Obsidian Dome near long valley caldera in eastern California (Manley and Fink, 1987). The central crystalline rhyolite layer of the TR lava is characterized by the development of the flow structure, which is composed of interconnected minute cavities. The shapes and sizes of the structure are varied from stubby or lens to flattened and from a few mm to above 5 cm in length, respectively. The morphology of the flow structure tends to be flattened with distance from the source region. It is probably due to shear stress caused by the lava movement We described the vertical variation of the mineral assemblage of Fe-Ti oxides. It shows that the highly oxidized Fe-Ti oxides tend to be distributed around the flow structure. Thus, the part is selectively oxidized. It is supported also by the rock magnetic experiments. Above studies and cooling history calculated by a numerical modeling show that the oxidation was caused by the increasing of fO2 at the part. We interpret that the increasing of fO2 was caused by the release of hydrogen from the degassing lava. Hydrogen should be passing through the flow structure, which is composed of interconnected minute cavities, and the part was selectively oxidized.
V21A-0379
The Importance of Surface Features in Long-lived Lava Flows: Preparing the way for Compound Lava Flow Models.
There have been significant advances in the development of lava flow models during the past decade, and their ability to replicate the emplacement of single flow units is very impressive. However, long-lived eruptions often produce compound `a`a flow fields containing many complex features of enigmatic origin, whose significance needs to be understood before such flows can be successfully modelled. The 2001 flank eruption of Mount Etna (17th July – 9th August) provided an excellent opportunity to observe the emplacement of a compound lava flow field. Twenty three days of activity at a single vent on the southern flank produced the lower flow field, which reached its maximum length in 7 days and thereafter grew by the superimposition and juxtaposition of younger flow units, and break-outs from earlier channels. Several sets of data were collected during the eruption, and we have combined the examination of these with detailed analysis of post-emplacement aerial photographs and our own recent field observations, in order to unravel the temporal evolution of the flow field. We are able to distinguish between features reflecting emplacement processes in individual units; those resulting from channel drainage, inflation, squeezing of lava from an older unit when overridden by a younger unit; and post emplacement changes including the emergence of lava in a range of rheological states through flow surfaces, flow margins and levee-channel boundaries. Although no lava tubes were encountered during our field survey, there is clear evidence of preferential thermal pathways within stationary flow units. Had the eruption continued, there is a high probability that a tube system would have developed, as was the case during the 1983, 1991-3 and 2004 eruptions. This would have allowed the flow to extend significantly further, resulting in greater damage to property on this heavily populated side of the volcano. The widespread occurrence of the structures described indicates that the underlying processes responsible for their development are common and repeatable, and therefore amenable to inclusion in future flow models.
V21A-0380
Generation of Hummocky Flow Morphology Revealed through Ground-based LiDAR Measurements of Actively Inflating Pahoehoe Lavas
Although the extrusion of pahoehoe lava flows is one of the most dominant planetary surface-forming processes in the solar system, emplacement models remain controversial, and affect our ability to understand the implications of continental effusive eruptions. To study the detailed growth patterns of an actively inflating hummocky pahoehoe field in Hawaii, we used a Riegl LMSZ420i ground-based light detection and ranging (LiDAR) system that captures topographic data at unprecedented resolutions and speed, and co-registers the x, y and z coordinates with the RGB values of true color high-resolution (12 megapixel) photographs from an externally-mounted camera. Over a 3-day period (February 21-23, 2007) we acquired 4 surveys of surface topography over a ~200 x 200 m area within the Pu'u O'o flow field that contained actively inflating pahoehoe flows emplaced over older, hummocky pahoehoe lavas. Total scan times ranged from 6 to 19 minutes, with topographic points collected at a 0.05-0.08 degree spacing. Each scan obtained between 1.6 and 5.1 million x, y, and z data points. We acquired topographic data at a rate of 12,000 points/second, permitting repeatable digital elevation model (DEM) generation with 5mm accuracy. We differenced successive DEMs generated from our topographic data to determine the magnitude and patterns of growth. We documented uneven rates of inflation over the area, ranging from less than 0.5 m to 3.9 m, with several tumuli forming over the 3-day time period. These results are the first detailed measurements that help us constrain the movement of lava between upper and lower flow crusts.
V21A-0381
Pahoehoe, 'A'a and Transitional Lava at Makapuu Point, SE Oahu
At Makapuu Point (SE Oahu, Hawaii) three lava flows over-lie each other in a 2.3-m-thick section. Although being emplaced on identical slopes, at identical distances from the likely source (the SE rift zone of Koolau volcano), and stack up one upon the other, the three flows are entirely different. The base of the sequence comprises pahoehoe, the middle section comprises a single 'a'a flow unit, and the top comprises a flow that is transitional between pahoehoe and 'a'a. We have carried out macroscopic field studies of this sequence, supported by laboratory-based microscopic studies of the vesicle and crystal populations to (1) define the differences between these three flow types, and (2) determine what makes a lava of the same composition emplaced at the same location so different in morphology. The pahoehoe forms a 0.5-m-thick pile of seven lobes. These have densities of 1.25-1.75 g m-3 which, using a dense rock value of 2.65 g m-3, yield vesicularities of 34-53 %. Macroscopically, the vesicles show a classic S-Type pahoehoe distribution, with a population of smaller spherical bubbles at each lobe periphery, and larger spherical bubbles towards each lobe center. The 'a'a flow comprises a surface and basal clinker, 0.42 and 0.23 m thick, respectively, separated by a 0.58-m-thick massive core. Densities are 1.57-2.09 g m-3 giving vesicularities of 21-41 %. Locally, the direction of shear (as indicated by the long axis orientation of vesicles) opens around shear-parting structures internal to the core. Finally, the 0.5-m-thick transitional flow has 0.08-0.42 m thick upper and lower zones with paheohoe characteristics, and a 0.17-m-thick core with 'a'a characteristics. The base has a population of small, spherical, bubbles and has a density of 1.63 g m-3 to give a vesicularity of 38 %. This basal layer also toes into the underlying 'a'a clinker. This lower zone is separated from the core by a sharp contact, with the core containing sheared vesicles and having a density of 2.1 g m-3 and vesicularity of 21 %. Microscopic analysis of bubble shape, as well as the content of phenocrysts and microphenocrysts, will be used to further define the differences between these three flow types, and to examine the likely differences in cooling history and rheology, as well as effusion rate, which likely led to each flow inheriting their observed morphology.
V21A-0382
Morphology and Emplacement of the Muliwai a Pele Lava Channel, Mauna Ulu (Kilauea, Hawai'i)
The 5.6 km-long Muliwai a Pele lava channel formed during one of the last stages of Kilauea's 1969 - 1974 Mauna Ulu eruption, specifically during a fountaining event of May 29 - June 2, 1974. We have mapped the entire channel to define spatial and temporal variations in channel construction, and their influence on the final channel form. The levee sequence comprises a basal 'a'a unit covered by pahoehoe overflow units. Morphologically the channel can be split into five down-channel sections: (1) 0 - 1.1 km: Tubed - The channel initiates as a notch in Mauna Ulu's south wall. Along this section, a braided tube system has developed; its line marked by surface collapse depressions and skylights. (2) 1.1 - 3.1 km Proximal - Along this section the channel is 3 to 11 m wide (mean = 9 m), and 2 to 6 m deep (mean = 3 m). The mean depth compares with a mean flow thickness of 4 m. The channel floor comprises undrained lava within which a tube developed and locally collapsed. Channel overflows are dominated by thin (1 - 20 cm), smooth surfaced, vesicular (53 - 58 %), pahoehoe sheets. These are of limited extent, extending no more than a few 10's of meters from the channel, but were able to bury the basal 'a'a entirely. (3) 3.1 - 4.1 km Medial I - Along this section blockages become common, with 8 occurring within this 1 km distance. Blockages are composed of levee chunks plucked from the up-flow channel walls. They are porous so lava flowed through them as well as around them. Behind each blockage, relatively viscous and dense (40-48 % vesicles) lava became ponded. In places this overflowed to feed flows with spiny-pahoehoe-to-'a'a surfaces. (4) 4.1 - 5.1 km Medial II - Along this section surge-fed overflows evolve into denser (40 - 50 % vesicularity), more viscous-appearing surface forms. Overflows are no longer smooth, and instead have spiny textures that transition into localized 'a'a swirls. The drained channel depth (mean = 10 m) is deeper than the sequence thickness (mean = 9 m), and the channel width varies between narrow (7-14 m wide) and broad (15-25 m wide) sections as the underlying slope begins to steepen. The over-deepening may be the result of thermal and mechanical erosion. (5) 5.1 - 5.6 km Distal - Over this section the channel descends the steep (20-30 °) slopes of Poliokeawe Pali, at the base of which the channel dies out to feed ~2.8 km of dispersed 'a'a flow. The channel features indicate variable supply over the course of emplacement, resulting in a three-stage construction process. First, a high effusion rate phase emplaced a basal 'a'a unit, that crops out along most of the medial section. Within this basal unit the channel developed. For the majority of the channel's life, lower-flux (<10 m3 s-1) flow dominated, and localized tubes developed under and between ponds of viscous lava collecting between blockages. Occasionally, short-lived (~100 3 s-1) surges emplaced fluid, vesicular lava to build pahoehoe overflow units. These surges also transported blockage material.
V21A-0383
Comparison of Pre- and Post-Eruption Surveys of the 2005-2006 East Pacific Rise Volcanic Event: Implications for Fast-Spreading Mid-Ocean Ridge Eruption Processes.
We present two high-resolution DSL-120A sidescan data sets collected before (2001) and after (2007) a recent eruption along the East Pacific Rise (EPR) that image the ridge crest between 9° 45`N and 9° 57`N at 2-m resolution. These data provide a unique perspective on the volcanic processes and seafloor features that developed during the 2005-06 eruptions. Co-registration of the two data sets is facilitated by bottom-moored transponder navigation and an exceptional correspondence in the morphology of the bounding walls of the axial summit trough (AST), which has uniquely shaped, small-scale offsets that, in many places, remain unchanged since 2001. By cataloging changes in the acoustic characteristics of the seafloor between the datasets, we are able to refine our estimates of the 2005-2006 eruption extent. The 2007 sidescan data, along with correlative bottom imaging using the WHOI TowCam system and ROV Jason2, allow us to interpret the heterogeneity and distribution of surface morphologies that develop within a single eruptive unit, and illustrate the mechanisms that distribute lava across the ridge crest. The most easily identified change resulting from the 2005-2006 eruption is the development of an extensive network of lava channels, primarily between 9° 49`-52`N, which distributed new lava more than 2 km from the AST. The sidescan data are complemented by phase bathymetry and SM2000 multibeam bathymetry collected with the DSL-120A at 2-m resolution, as well as near-bottom, sub-meter resolution SM2000 bathymetry data collected with ROV Jason 2. Comparison of these data with pre-eruption bathymetric data reveal localized changes over portions of the eruption area, which help constrain the overall emplacement of lava. Comparison of pre- and post-eruption magnetic field intensity profiles over the ridge crest provide additional insight into patterns of lava deposition during the eruption.
V21A-0384
Thermal Imaging Applied For Studying The Spreading Of A'a' Lava Flows During The 2002- 2003 Stromboli Effusive Eruption
On 28 December 2002 Stromboli Volcano (Aeolian Islands, Italy) was site of a new effusive activity. Lava flows emplacing onto the steep slopes of the Sciara del Fuoco (Calvari et al. 2005), formed two lava flow fields, separated in space and time and fed by different vents. The first flow field formed in the middle of the Sciara del Fuoco and was fed by a vent located at 550 m a.s.l. until 15 February 2003. This resulted in a compound flow field of numerous overlapping 'a'a flow units. The second flow field emplaced mainly in the upper eastern part of the Sciara del Fuoco and was fed by a main vent located at 670 m a.s.l. Activity here lasted until the end of the eruption on 21 July 2003, and formed a thick compound lava shield containing a complex lava tube network (Lodato et al., 2007). Daily handheld Forward Looking InfraRed (FLIR) thermal camera helicopter surveys allowed us to record: (1) the development of lava channels and tubes, (2) the formation of tumuli, (3) the opening and closure of ephemeral vents, and (4) the discrimination of active lava flows. Using these data, it was possible to track the evolution of the lava flow field and associated channel-tube system, including the establishment of the main lava tube, expansion of the network resulting from branching of the tube system, and the shut down of the system during July 2003. Thermal measurements also permitted lava tube mapping and the discrimination of three orders of tumuli (Lodato et al., 2007). The development and stability of these lava flow field structures are the result of the interaction between different parameters including effusion rate and topographic gradient.
V21A-0385
Lava Effusion Rates from Satellite Remote Sensing for Kliuchevskoi Volcano, Kamchatka- Russia in 2007
Kliuchevskoi volcano is the largest and one of the most active volcanoes on the Kamchatka- Peninsula. The most recent eruption from January to August 2007 produced several strombolian eruptions, lava flows, mudflows, and ash plumes. Field-based observations can be dangerous during such activity. Satellite imagery provides safe and relatively inexpensive views of the eruption. To monitor the progress of an eruption effusions rates are calculated from thermal infrared imagery. A volume of lava at a given temperature emits a fixed amount of thermal radiation measured at the satellite. By constraining temperature, volume with time (effusion rate) can be calculated. Effusion rates can give an indication of changes during an eruption, rather than after the activity has decreased or ceased. Effusion rates were consistently greater than 3.0m3/s and reaching to 15.1m3/s. Calculations of effusion rates of Kliuchevskoi were conducted using a two component analysis from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and compared to those made with coarser spatial resolution sensors: Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS). By constraining effusion rates in near real-time, the potential hazards from the volcano can be effectively mitigated.
V21A-0386
Thermal infrared investigation of the pyroclastic flow deposits and dome region of Bezymianny volcano, Kamchatka, Russia
Bezymianny (Kamchatka, Russia) is an active stratovolcano that contains a summit lava dome and pyroclastic flow (PF) sheet to the southeast. Two recent eruptions (24 December 2006 and 11 May 2007) generated fresh pyroclastic flows on the southeastern flank. During the winter of 2006, the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) collected several day- and night-time images to monitor the December eruption deposits and subsequent events. A field campaign in August 2007 was conducted to investigate recent changes at the lava dome and to map the new PF deposits. Within the cloud-free night time ASTER image from 30 June 2007, seven ASTER thermal infrared (TIR) 90 m pixels were identified as being thermally-anomalous and were investigated in the field. Both handheld Forward Looking Infrared Radiometer (FLIR) and thermocouple probe data were obtained and compared to the satellite TIR data. Helicopter- and ground-based FLIR surveys revealed thermally-elevated PF deposits that contained warm blocks and fumaroles. The maximum fumarole temperature within the December 2006 PF deposit was 377C at a distance of five kilometers from the lava dome. At the terminus of the 24 December 2006 PF, seven kilometers from the dome, the maximum temperature recorded was 228C. This suggests that eight months after the December 2006 eruption, the deposit was of a sufficient thickness in this area to retain heat. In addition, the thickness of the deposit probably increased as the slope angle decreased approximately 4 kilometers from dome, which may explain the high temperatures observed. We present spaceborne, airborne and ground-based thermal data in order to compare direct and remote thermal observations of ongoing activity and provide the first ground-based TIR data of actively cooling PF deposits at Bezymianny.
V21A-0387
Thermal observations of the degassing of the December 2006 and May 2007 pyroclastic flow deposits at Bezymianny Volcano, Kamchatka
Bezymianny Volcano (Kamchatka, Russia) has been periodically active since its reawakening in 1956, most recently erupting on December 24, 2006 and May 11, 2007. Currently, one or two eruptive cycles occur each year, consisting of spine emplacement followed by an explosive eruption accompanied by pyroclastic flows, concluding with extrusion of a lava flow. During August 2007, systematic depth profiles of temperatures of the December 2006 and May 2007 pyroclastic flow deposits (PFDs) were obtained in seven satellite-derived target locations using an Omega HH509 thermocouple. In addition, 38 fumarole temperatures were obtained and Forward Looking Infrared Radiometer (FLIR) data were collected throughout the target areas. Degassing of the PFDs was observed through large clusters and linear arrays of fumaroles located in depressions with distinct rocky centres. Hot depressions were also observed without active fumaroles, suggesting that hydrothermal activity due to interaction with the underlying snow layer after PFD emplacement was possibly not forceful enough to breach the surface in these locations or that the activity had diminished since the eruption. The hottest fumarole observed reached a temperature of 377C, 5 km from the lava dome. Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data will be used to characterise the two deposits and assess cooling rates and emplacement temperatures. Preliminary data show that the two areas with hottest fumarole temperatures do not correspond to the hottest areas of the deposits. This is the first time extensive direct temperature measurements of warm PFDs have been obtained at Bezymianny.
V21A-0388
Monitoring of the 2007 Eruption of Kluichevskoi Volcano Using the ASTER Urgent Request Protocol
The automated rapid response system and urgent request protocol of the Earth-orbiting ASTER instrument have provided an unprecedented time series of high-resolution, multispectral data throughout the 2007 eruption of Kluichevskoi Volcano, Russia. The Kamchatka Volcanic and Eruption Response Team (KVERT) first reported a weak thermal anomaly within the summit crater on 14 December 2006. By mid-January 2007, Strombolian explosions initiated along with associated ash, gas/steam plumes, and an increase in volcanic tremor. Immediately thereafter, the ASTER urgent request protocol was initiated, capturing day and night paired data approximately once every ten days, resulting in 40 datasets from January through June. During this time, small Strombolian explosive events punctuated longer-lived effusive stages, which produced three lava flows (~ 4 km in length) on the northern (26 April), northwestern (28 May), and southeastern (6 June) flanks. Lahars, associated with flow emplacement over snow/ice, extended up to 3 km from the flow terminus. Initially, maximum temperatures of the summit thermal anomaly were approximately 10 C above background (-30 C). However, steadily increasing temperatures coincident with the onset of lava effusion from the summit on 26 April resulted in the first occurrence of saturated thermal infrared (TIR; 97 C) and subsequently the short-wave infrared (SWIR; 467 C) pixels. Incandescence of the active lava flows was also observed in the visible-near infrared (VNIR) data and solar-corrected daytime datasets yielded maximum lava flow temperatures between 856 and 905 C. Consistent saturation of both the TIR and SWIR data between April and June suggests that vigorous effusive activity from the summit crater was continuous and well established in open channels at that time. The lava flow time series in conjunction with ASTER's digital elevation model (DEM) capabilities provide a framework for extracting additional information such as lava effusion rates. Using the pre-flow slope calculated from the 18 March ASTER-derived DEM and the down-slope advance during the 13 hours between image pairs on 5 and 6 June, the average flow velocity was calculated to be approximately 26 m/hr. However, image acquisitions conducted following the 5 and 6 June datasets suggest effusive activity ceased shortly thereafter. Validation in the field was conducted during August 2007 using a high-spatial resolution airborne thermal imaging camera. These data revealed continual cooling of the lava flows with the maximum temperatures not in excess of 20 C. Future work includes the determination of compositional, textural, heat flux, and cooling rate of the three lava flows. This study also provides a basis from which to study other active volcanic systems using spaceborne infrared remote sensing.
V21A-0389
Ground-Based Thermal Infrared Monitoring of Dome Growth at Soufrière Hills Volcano, Montserrat
Activity at Soufrière Hills Volcano, Montserrat, has included the growth and partial collapse of a lava dome or dome lobe complex. During the latest period of magma eruption (Aug 2005-Apr 2007), thermal infrared monitoring was conducted from various ground and aerial positions using a hand-held Mikron Model TH7515 forward-looking infrared (FLIR) thermal imaging camera. The camera monitors at 8.0-14.0 μm, yielding apparent temperatures of <500C. Each pixel typically represented <1 m to 3 m resolution depending on observation position. Thermal observations were directed at the dome lobe complex, associated pyroclastic flow and rockfall deposits, and ash and gas vents. Early-stage dome growth showed pancake morphology with rubbly top surfaces at near-ambient to 100C apparent temperatures (T, uncorrected for emissivity and atmospheric conditions) and flanks at 150-300C. Dome morphology evolved after 1-3 months into shear lobes, including short-lived spines, fins, and rare whalebacks, and unstable oversteepened dome faces. Most surfaces of active dome lobes were relatively cool (T=35-100C). Isolated hot areas of 200-450C marked spall zones and expansion cracks in early stages, or pfs, rockfalls, and cracks in later stages. Active spines and fins typically had surfaces of 70-250C that reached 300C where cracked or spalling. Hotter (+25C to +75C) subhorizontal zones 2-10 m thick developed in the lower portion of the active shear lobe face in early 2006, perhaps as expansion cracks during endogeneous growth partly unconfined by crater walls. Overall, most thermal flux was not from the dome itself, but rather from rockfall and pf deposits at T>200C, and max T=300-400C. The major dome collapse on 20 May 2006 was preceded by a significant qualitative decrease in thermal flux in the preceding 2 weeks, both from the dome and surrounding rockfall and pf apron. The data suggest collapse triggered by increased internal dome pressure following a switch to more endogenous growth.
V21A-0390
Time Series Radar Observations of a Growing Lava Dome
Exogenous growth of Peléean lava domes occurs by addition of lava from a central summit vent and mass wasting on the flanks as rockfalls and pyroclastic flows, forming an apron of talus. We observed this process at the Soufrière Hills Volcano, Montserrat between 30 March and 10 April 2006 using a ground-based imaging mm-wave radar, AVTIS, to measure the shape of the dome surface.From a time series of range and intensity measurements at a distance of six kilometres we measured the topographic evolution of the lava dome. The locus of talus deposition moved to the southeast with time and the talus surface grew upwards on average at about 2 metres per day. The AVTIS measurements show an acceleration in lava extrusion rate on 5 April, with a 2-day lag in the equivalent rockfall seismicity record. We account for the budget of lava addition and dispersal during the eleven days of measurements using: AVTIS range measurements to measure the talus growth (7.2 Mm3, 67%), AVTIS range and intensity measurements to measure the summit lava growth (1.7 Mm3, 16%), and rockfall seismicity and visual observations to measure the pyroclastic flow deposits (1.8 Mm3, 17%). This gives an overall dense rock equivalent extrusion rate of about 9.7 m3s-1. These figures demonstrate how efficient non-explosive lava dome growth can be in generating large volumes of primary clastic deposits, and how this process could also reduce the propensity for large hazardous pyroclastic flows. http://www.st- andrews.ac.uk/~mmwave/mmwave/avtis.shtml
V21A-0391
Effect of Input Parameters on a Lava Flow Emplacement Model. Application to the 1999-2002 Lava Flow of Santiaguito, Guatemala.
The emplacement of a lava flow depends on external conditions such as effusion rate, topography, and ambient
temperature, as well as on the characteristics of the lava such as chemical composition, crystal content,
vesicularity, core temperature, crust thickness, etc. Hence these input values control the quality of any modeling.
Where most of these variables are known, it may be possible to constrain an unknown factor if the final
emplacement history is known. Here we explore the application of this technique to determining extrusion rates at
the active dacitic Santiaguito lava dome complex, Guatemala. Caliente is the current active dome with continuous
growth through an extrusive lava front and frequent ash explosions Like the other domes, it presents some more
effusive periods of activity such as the 4 km long lava flow (1999-2004).
Experimental measurements suggest a viscosity about 108 to 109 Pa.s in the range of temperature 850 to
800°C for fully remelted dacites. Actual viscosities will be around 10 times higher due to the crystal
content, estimated to be 40% (including 30% phenocrysts), almost exclusively plagioclase. Calculations and
petrographic observations indicate that crystallinity remains constant within the flow. The vesicularity is about 9
volume %. Using the FLOWGO model (Harris et al., 2001 Bull Volc), we predict that eruption temperatures must
exceed 835°C in order for long flows to form (i.e. to at least reach the base of the dome edifice). The
calculated extrusion rate is then in the range of 0.8 to 1.6 m
V21A-0392
Forecasting Lava Flow Hazards by Using a Combined Stochastic/Thermo-Rheological Approach
There are many methods for predicting the area most likely to be inundated with lava during a volcanic eruption. At its simplest, such forecasting may involve the application of volcano-specific empirical length/effusion rate relationships to estimate flow length. At their most complex, such predictions may involve iteratively solving a system of equations that characterize the effects that cooling-induced changes in rheology have on the ability of lava to flow downhill, given spatial variations in slope determined from a digital elevation model. Here we describe results obtained by combining two recently published flow simulation models that provide an efficient means to predict the final dimensions (i.e. potential lengths and widths) of a lava flow. The model of Favalli et al., (2005) uses a stochastic approach to predict possible flow paths by calculating drainage paths using a vent location and a DEM. A family of possible flow paths is produced by iteratively adding random noise to the DEM and, at each step, re-running the drainage model. For large numbers of iterations the resulting family of paths results in impressive simulation of likely flow widths as a function of distance from the vent, but does not constrain flow length. The FLOWGO model presented by Harris and Rowland (2001) predicts the maximum potential length of a cooling-limited flow by estimating the effect that cooling-induced changes in rheology have on flow velocity. While FLOWGO allows for prediction of flow length, it provides no information regarding the potential area over which the lava may expand. Both models are computationally simple. Here, we present results obtained by combining these two approaches. By using the stochastic approach to predict all the areas that a lava flow could possibly invade (in effect forecasting the maximum potential flow width at all distances from the vent) and the thermo-rheological model to terminate these paths once the predicted cooling-limited length for the flow has been attained, this hybrid approach allows for computationally efficient simulations of lava flow hazards. A variety of case studies pertaining to Mount Etna will be used to demonstrate the potential of this approach. We also show how this model can be driven by satellite-derived lava effusion rates. As these data can be obtained in near-real-time, such an approach will allow flow simulations to be updated in response to changing eruption conditions.
V21A-0393
Conveying Lava Flow Hazards Through Interactive Computer Models
As part of an Information Sciences senior class project, a software package of an interactive version of the FLOWGO model was developed for the Island of Hawaii. The software is intended for use in an ongoing public outreach and hazards awareness program that educates the public about lava flow hazards on the island. The design parameters for the model allow an unsophisticated user to initiate a lava flow anywhere on the island and allow it to flow down-slope to the shoreline while displaying a timer to show the rate of advance of the flow. The user is also able to modify a range of input parameters including eruption rate, the temperature of the lava at the vent, and crystal fraction present in the lava at the source. The flow trajectories are computed using a 30 m digital elevation model for the island and the rate of advance of the flow is estimated using the average slope angle and the computed viscosity of the lava as it cools in either a channel (high heat loss) or lava tube (low heat loss). Even though the FLOWGO model is not intended to, and cannot, accurately predict the rate of advance of a tube- fed or channel-fed flow, the relative rates of flow advance for steep or flat-lying terrain convey critically important hazard information to the public: communities located on the steeply sloping western flanks of Mauna Loa may have no more than a few hours to evacuate in the face of a threatened flow from Mauna Loa's southwest rift whereas communities on the more gently sloping eastern flanks of Mauna Loa and Kilauea may have weeks to months to prepare for evacuation. Further, the model also can show the effects of loss of critical infrastructure with consequent impacts on access into and out of communities, loss of electrical supply, and communications as a result of lava flow implacement. The interactive model has been well received in an outreach setting and typically generates greater involvement by the participants than has been the case with static maps and photos and other lava flow demonstrations.
V21A-0394
Expected Behavior of Basaltic Magma with the Proposed High Level Nuclear Repository at Yucca Mountain, Nevada
The expected series of eruptive events for a future igneous event in Yucca Mountain within the next 1 MY is comparable to that at Lathrop Wells basalt center and other Crater Flats Quaternary volcanoes. Lathrop Wells and Crater Flats Quaternary volcanoes are, in general, comprised of a single scoria cone with one or two lava flow fields extending from the base. The lava flow fields associated with scoria cones all appear to extend from the base of a scoria cone and to be comprised of lava terraces. A three-dimensional model of the plumbing system for a possible future igneous event is presented in this paper, based on the characteristic features of eruptive deposits at Lathrop Wells and other Crater Flats Quaternary volcanoes. Also described in the model are consequences related to the interaction between magma and the repository. The repository is expected to be 200-300 m below the surface and comprised of parallel drifts 5 m in diameter, 0.5-1.0 km in length and spaced 85 m. Each drift is to be filled with a series of 1.8 m diameter waste packages made of Alloy 22 stainless steel. The conceptual model of the plumbing system and related consequences are described in six stages. Stage 1 Intersection of dike with drift: One dike will intersect the repository. The width of a future dike in YMR is expected to vary along the length with a maximum value of < 4.0 m at repository depths. The number of drifts that will be intersected by the dike will be 6-24 depending on the lateral extent of the dike through the repository. Stage 2 Initial stage magma-drift interaction: The lateral variation in magma properties will produce, in general, two different styles of expected activity upon entering a drift: a mixture of gas and fragments of magma characteristic of a lava fountain at wide portions of the dike, and crystallizing magma relatively depleted in volatiles at the narrowest part of the dike. A spray of pyroclastics is expected inside a drift from a lava fountain that will bombard and coat waste packages with magma. Crystallizing magma relatively depleted in volatiles will be a slow moving crystallizing flow and is expected to behave like a plug sealing the drift. Stage 3 Surface activity: initial cone building stage: Magma that is not diverted into the drift will follow the crack tip and make its way to the surface and erupt at the surface along the fissure as a curtain of lava fountains. A conduit or cone building part of the eruption will develop at the widest part of the dike. Stage 4 Second stage of magma drift interaction: Strombolian activity at the repository depth is expected to occur in only one drift; the drift that is intersected by the widest part the dike. This drift will be inundated with pyroclastic material associated with the early cone building Strombolian events. Lava is expected to enter an adjacent drift as discrete pulses. Stage 5 Surface activity: final cone building phase: At the surface, activity will transition to a more violent Strombolian style. Additional discrete pulses of lava may occur. Stage 6 Final stage magma drift interaction: Magma entering drifts at this stage will be either a pyroclastic flow or lava.
V21A-0395
Physical Volcanological and Petrogenetic Implications of Intra-lava Flow Geochemical Heterogeneity in the Columbia River Flood Basalt Province, USA.
Continental flood basalt lava flows are widely assumed to represent compositionally uniform and rapidly erupted products of large well-mixed magma reservoirs. However, this study presents new data to illustrate systematic element and isotope variations within the flow field formed by an individual flood basalt eruption, both vertically within each sheet lobe and laterally between the constituent lobes. Such variation is significant in chemostratigraphic correlation of flood basalt lava units, in identifying source variability during one eruption, and in petrogenetic modeling. We investigate the extent and cause of compositional variation through tracing lava sheet lobes in a 2,660 cubic kilometer pahoehoe flow field formed during a single eruption in the Columbia River Basalt Province, USA. This is based on features related to emplacement by the inflation mechanism. This method of emplacement is supported by small but statistically significant and systematic major and trace element variation e.g. MgO 3.09- 4.55 wt%, Ni 17.5-25.6 ppm, indicative of fractional crystallisation. Re-Os isotopes indicate progressive crustal contamination of the magma over the timescale of a single flood basalt eruption. By establishing this physical volcanological framework, we determine a temporal link with the supply of lava from the vent(s) and apply it to investigate sequential magmatic evolution during the timescale of one eruption.
V21A-0396
Distribution and Stratigraphy of Basaltic Lavas in the Southwest Portion of the Quaternary Big Pine Volcanic Field, California
The Pleistocene Big Pine Volcanic Field (BPVF), located in the Owens Valley of eastern California, is dominated by basaltic cinder cones and lavas, and poses a potential volcanic hazard to local infrastructure, in particular Highway 395 and the Los Angeles Aqueduct. However, despite numerous petrologic studies, the volcanic history and distribution of products from individual BPVF eruptions are poorly known. Using detailed field mapping and petrology, we have determined the distribution and stratigraphy of basaltic lavas in the southwest portion of the BPVF, which contains the largest number exposed vents and lava. In the Aberdeen area, individual cinder cones and fissure vents are aligned along N-S trending lineaments, with local clustering of vents and lavas. Approximately 19 cones, and at least 14 lava flows occur in the area. Two cinder cones located near the valley floor are partly buried by younger lavas and alluvium. In most cases, lavas are aa in character and traveled ca. 7 km down alluvial fans (8% gradient) towards the Owens River. On average, individual lavas cover ca. 10-12 km2, with average individual volumes of ca. 0.065 km3. Two general groups of basaltic lavas characterize the Aberdeen area: 1) xenocryst-rich and 2) xenolith-poor basalts. Xenolith-rich basalts contain variable amounts of ultramafic, mafic, granitic, and metamorphic lithologies, whereas xenolith-poor lavas are dominated by olivine phenocrysts. Overlapping flow margins define relative ages between adjacent basalts. In both north and south portions of the Aberdeen area, flows composing the base of the volcanic stratigraphy are the xenolith-rich variety, and are typically overlain by xenolith-poor flows. In general, these younger xenolith-poor lavas are approximately 25% larger in volume than the older xenolith-rich lavas. Several vents record changes in lava type during individual eruptions, suggesting transitions in magma discharge rate. At one vent cluster, pahoehoe is restricted to within ca. 1.5 km of its fissure vent and overlies more extensive aa from the same vent, suggesting a decrease in discharge rate of basalt over the course of the eruption. In contrast, lavas from a cinder cone on the eastern portion of BPVF comprise thin (<0.5 m) pahoehoe capped by thicker (1-2 m) aa flows, suggesting increasing discharge rate over the course of this eruption.
V21A-0397
Volcano Flank Structures on Earth and Mars
Shield volcanoes on Earth and Mars share common features, including calderas and pit crater chains. A set of structures present on the sides of several of the large shields on Mars are not regarded as having Earth analogues, however. Flank terraces are topographically subtle structures, characterised by a gentle convex profile and a distinctive "fish scale" imbricate distribution pattern. Magma chamber inflation, lithospheric flexure, flank relaxation, or gravitational slumping have been suggested as terrace formation mechanisms. Terraces on both Mars and Earth are clearly visible only in slope maps, and may thus escape visual detection in the field. We show that both Mauna Loa (Hawaii) and Etna (Sicily) display the same characteristic "fish scale" terrace pattern. This pattern delineates structures that we contend are terrestrial flank terraces. Heterogeneities in volcano geometry, due to buttressing or extension, result in terrace distributions that are not as evenly circumferential as those on Mars. Plan and cross-sectional profiles, however, parallel those of the Martian structures. These structures may also be present on Alayta (Ethiopia), Santa Cruz (Galapagos), and Tendürek Dagi (Turkey). Another type of structure, larger and steeper than flank terraces but sharing a similar plan-view morphology, is also present on Mauna Lau and Etna. These "flank bulges" appear to correlate with structures on Piton de la Fournaise (La Réunion), Cosiguina (Nicaragua), and Karthala (Comoros) on Earth, and Apollinaris Patera and Tharsis Tholus on Mars. Elsewhere (Paul K. Byrne et al., this volume) we argue that lithospheric flexure is a likely formation mechanism for Martian terraces. Flexure is active beneath Mauna Loa, and possibly under Etna, and so may also be responsible for terrestrial flank terraces. Scaled analogue models suggest that the larger flank bulges are due to magma intrusions derived from large chambers within these edifices. There is thus a strong morphological link between deformation, intrusion, and surface topography. Additionally, as these structures occur on both planets, the governing tectonic and volcanological processes are fundamentally the same for Earth and Mars.
V21A-0398
Mechanical Interactions in South Iceland Pleistocene Basalt-Hyaloclastite Sequences
In southern Iceland, a series of Pleistocene eruptive sequences, likely derived from proto-Laki eruptions, displays unusual basalt-hyaloclastite interaction (Bergh and Sigvaldason, 1991). These sequences are exposed over an E-W distance of 75 km in paleo-seacliffs and display alternating layers of basalt lava and basaltic hyaloclastite, with lava thicknesses reaching up to 50m and overlying massive hyaloclastite thicknesses reaching over 100m. The lavas are composed of a "standard" sequence of columnar lava overlain by cube-jointed lava, with columnar lava absent in some areas. Throughout the exposed cliff sections, cube-jointed lava injects into and is disaggregated within the overlying hyaloclastite, yielding pillows, pods, flame structures, and dikes. The geometry and scale of these structures vary dramatically by location. Bergh and Sigvaldason (1991) proposed simultaneous subaqueous eruption of both lava and hyaloclastite with entrainment of lava accompanying mass flow. Emplacement of lava flows preceding hyaloclastite, rather than as sills, is supported by a relative lack of vesiculation in the lava, indicating considerable post-eruption degassing. We propose that the lavas were erupted from an ice-free, subaerial portion of the fissure. As the eruption propagated from subaerial eruption to subglacial, the recently-erupted lava flows were quickly buried by hyaloclastite deposited from a jokulhlaup generated by eruption-induced glacial melting. Heat was introduced from contact with the lava, still molten beneath a thin crust. Fracturing of the crust led to a volume expansion of the water in the hyaloclastite, leading to a pressure field that locally weakened the hyaloclastite, allowing injection to occur. Freezing of lava margins once the lava propagated upward into the hyaloclastite allowed the liquid lava to continue vertical propagation. Rheological variation in hyaloclastite related to water content, temperature, and material sorting likely contribute to the variation in style of injection and disaggregation. Approximately 250 EMP and LA-ICPMS analyses of fresh glass from hyaloclastite samples from 3 representative locations revealed uniform compositions, consistent with a common source. The major element composition of about 0.3 wt% K2O, 2.5 wt% TiO2, 7 wt% MgO, 2.5wt% Na2O, and 13 wt% FeO is similar to modern Laki elemental values (Lacasse et al., 1998; Metrich et al. 1991). In contrast, 35 analyses of glass from lava at the base of the section in its easternmost exposure display elemental signatures that are markedly different from the hyaloclastites, including the one that overlies it. Compared to the hyaloclastite glasses, this glass displays distinctively higher SiO2, FeO, and Na2O, 2x to 4x higher REE, TiO2, K2O, and P2O5, and lower MgO and Al2O3. This puzzling relationship might be a consequence of one of the following: (a) The lava was sourced from fractionated melt at the top of a magma chamber and erupted subaerially; the less-evolved magma that immediately followed erupted subglacially, from a slightly different location, formed the hyaloclastite, and was transported over the recent lava; (b) The lava and hyaloclastite were sourced simultaneously from a laterally variable magma chamber; or (c), The lava and the hyaloclastite were sourced from identical magma (though at separate locations along the fissure), but the hyaloclastite melt was immediately quenched to preserve the initial liquid composition, whereas the lava fractionated during transit from the eruption site to yield