The Optimization of Spatial, Spectral, and Temporal Resolution for Constraining Eruption Style on Earth and Io with Thermal Remote Sensing
Volcanic eruptions on Io and Earth are monitored by a variety of thermal remote sensing instruments. While higher resolution data are always desirable, we have developed methodologies to constrain the style of volcanic eruption using low spatial, spectral, and temporal resolution data. For the volcanic moon Io, this is necessitated by the limits of spacecraft and Earth-based telescopic observations. Eruption style can be classified using the concept of "thermal signature" which focuses on the temporal evolution of thermal emission spectra . We find that the ratio of the emission at 2 µm and 5 µm, and how this ratio changes temporally, is often diagnostic of effusive eruption style, even in low spatial resolution data . Tests using ground-based thermal data for terrestrial “ground truth” cases show that this classification system is equally valid for Earth. A square meter of an active lava lake on Io looks very similar to a square meter of an active lava lake on Earth. The same goes for pahoehoe flows. This validation of “thermal signature” means that appropriate physical models can be selected to interpret the data. On Io, the scale of eruptions can utterly dwarf their terrestrial counterparts. “Outburst” eruptions, known to be caused by extensive lava fountaining, can radiate >1013 W. The smallest thermal anomalies detected on Io in thermal infrared data are still larger than any contemporaneous mafic volcanic activity on Earth. The large volumes of lava erupted on Io (e.g., >56 km3 at Pillan in 1997) are an expression of internal tidal heating. It may be that high compressive stresses in the lower lithosphere inhibit magma ascent, and so only relatively large volumes of magma can overcome this “stress barrier” and reach the surface. The results of the “thermal signature” analysis  can be used as an aid in the planning of future space-borne instruments that can be used for volcano monitoring on Io, as well as on Earth. This work was performed at the Jet Propulsion Laboratory-California Institute of Technology, under NASA contract, with support from the NASA Outer Planets Research Program. © 2009. All rights reserved. References:  Davies, A. G., 2007, Volcanism on Io - A Comparison with Earth, Cambridge University Press, 372 pages.  Davies, A. G., Keszthelyi L. P., and Harris, A. J. L., 2009, The Thermal Signature of Volcanic Eruptions on Io and Earth, JVGR, submitted.
MODIS' Performance in Volcano Monitoring in the Northeastern Pacific Ocean
Since the launch of Terra, NASA's first Earth Observing System spacecraft, the accuracy and capability of monitoring volcanoes from the Kuriles to the Cascades has improved due to the increase in spectral and spatial resolution of these sensors. The Alaskan Volcano Observatory uses the Moderate Resolution Imaging Spectroradiometer (MODIS) on the Terra satellite to detect and track eruptions, ash clouds and thermal anomalies of volcanoes in the region using data recorded in visible and infrared wavelengths, and with a pixel size of 250m and 1km, respectively. Volcanoes in the region pose a threat to both global and local air traffic, human health and the infrastructure in general. Therefore, monitoring of these volcanoes is critical to help prevent loss of life and to minimize structural damage and resulting cost. Currently, 23 different geographical regions are being monitored and up to 200 image products per day are produced. This has led to detailed observations and analyses of over 20 eruptions from large Alaskan volcanoes like Shishaldin, Augustine, Okmok or Redoubt and Russian volcanoes including Kluichevskoi, Shiveluch and Bezymianny in the first ten years of MODIS. This project gives an overview of the possibilities MODIS offers volcano monitoring and the results obtained from the analysis. Examples of the most prominent eruptions across the North Pacific have been selected to illustrate MODIS' strengths as well as the areas where future research and analysis of MODIS data can improve volcanic hazard assessment and early warning of volcanic activity. Techniques developed based on MODIS data will prepare us to take advantage of data from the new forthcoming National Polar-orbiting Operational Environmental Satellite System (NPOESS) series of satellites.
Historical series and Near Real Time data analysis produced within ASI-SRV project infrastructures
ASI-Sistema Rischio Vulcanico (SRV) project is devoted to the development of a pre-operative integrated system managing different Earth Observation (EO) and Non EO data to respond to specific needs of the Italian Civil Protection Department (DPC) and improve the monitoring of Italian active volcanoes. The project provides the capability to maintain a repository where the acquired data are stored and generates products offering a support to risk managers during the different volcanic activity phases. All the products are obtained considering technical choices and developments of ASI-SRV based on flexible and scalable modules which take into account also the new coming space sensors and new processing algorithms. An important step of the project development regards the technical and scientific feasibility of the provided products that depends on the data availability, accuracy algorithms and models used in the processing and of course the possibility to validate the results by means of comparison with non-EO independent measurements. The ASI-SRV infrastrucutre is based on a distributed client/server architecture which implies that different processors need to ingest data set characterized by a constant and common structure. ASI-SRV will develop, in its final version, a centralized HW-SW system located at INGV which will control two complete processing chains, one located at INGV for Optical data, and the other located at IREA for SAR data. The produced results will be disseminated through a WEB-GIS interface which will allow the DPC to overview and assimilate the products in a compatible format respect to their local monitoring system in order to have an immediate use of the provided information. In this paper the first results producing ground deformation measurement via Differential Interferometric SAR (DInSAR) techniques by using SAR data and via the application of the Small BAseline Subset (SBAS) technique developed at IREA, are reported. Moreover different products obtained using optical data (ASTER, MODIS, HYPERION, and AVHRR) are also reported. The processing modules for EO Optical sensors data are based on procedures those allow to estimate a number of parameters which include: surface thermal proprieties, concentration and flux of sulphur dioxide (SO2), water vapor and volcanic aerosol optical thickness, ash emissions and to characterize the volcanic products in terms of composition and geometry. For the analysis of the surface thermal characteristics, the available algorithms allow to extract information useful to detect small changes in the retrieved parameters. ASI-SRV foresees the generation of significant information also for the definition of the new lava and ash cover distribution after the end of an eruption.
DISTRIBUTION OF THE TEMPERATURE IN THE ASH-GAS FLOW DURING KORYAKSKY VOLCANO ERUPTION IN 2009
The observations of the ash-gas plumes during the Koryaksky eruption in March 2009 by the high resolution thermovision camera allowed obtaining thermal distributions inside the ash-gas flows. The plume structure is formed by single emissions. They rise at the rate of 5.5-7 m/s. The plume structure in general is represented as 3 zones: 1. a zone of high heat exchange; 2. a zone of floating up; 3. a zone of lateral movement. The plume temperature within the zone of lateral movement exceeds the atmospheric temperature by 3-5 oC, within the zone of floating up it exceeds by 20 oC. Its rate within the zone of floating up comprises 5-7 m/s. At the boundary between the zones of high heat exchange and floating up where we know the plume section, from heat balance equation we can estimate steam rate and heat power of the fluid thermal flow. Power of the overheated steam was estimated as Q=35 kg/s. It forms the ash-gas plume from the eruption and has temperature equal to 450 oC. The total volume of water steam produced during 100 days of eruption was estimated 3*105 t, its energy - 109 MJ.
Volcanoes activity Early warning system based on MSG SEVIRI data
Low-spectral, high-temporal resolution sensors on board geosynchronous satellites played an important role for monitoring known source of thermal anomalies. With the launch of the Meteosat Second Generation Spin Enhanced Visible and Infrared Imager (MSG-SEVIRI) it is now possible to follow with high-temporal resolution, changes of volcanoes summit thermal behaviour from space. This work is addressed to the analysis of volcanic area by means of thermal infrared data of Spinning Enhanced Visible and Infra-Red Imager (SEVIRI) onboard the geostationary satellite Meteosat Second Generation (MSG). In this paper the results obtained investigating datasets regarding the Mt Etna and Stromboli volcanoes both in the Sicily (South Italy) are presented. The Spinning Enhanced Visible and Infrared Imager sensor (SEVIRI) mounted on the Meteosat Second Generation (MSG) satellite produces 11 channels images every 15 minutes and the image processor, presented in this paper, provides radiance data with the same time rate. The system used is dedicated to the preprocessing, calibration, coregistration and storing of MSG-SEVIRI data. The MSG data processor collects different stripes of data coming from MSG SEVIRI and creates calibrated data files for the desired region. The overall architecture of the system has been home-developed in order to be used both for the continuous reception of data or to extract information from a single SEVIRI frame. This work emphasizes the capability of MSG-SEVIRI data for the estimation of parameter suitable for the volcanic monitoring. Measurements are made every 15 min, following the event evolution in near real time. The spectral radiance emitted by hot spots, reaches its maximum in the region of mid infrared (MIR, around 3 micron). Consequently, MSG 4th channel (centered at 3.9 micron), can be widely used for surface thermal monitoring, despite its relatively ground resolution cell (3 km nadir view). Investigating different data sets, regarding Mt Etna eruption (September 2004, July 2006, November 2006, November 2007) and Stromboli volcano eruption (February 2007), it has been pointed out that the beginning of new eruption is marked by a sudden increase of radiance value (up ten times) associated to the pixels centered over the crater (July 2006, November 2007, February 2007). The 3.9 micron MIR channel is saturated at 2.4 Watt/sr*m2*micron. The system proposed allows us to set up an early warning system dedicated to the impending eruption.
Satellite based observation and interpretation of the 2009 Sarychev Peak eruption on Matua Island of Kuril Islands, Russia
Sarychev Peak, which occupies the northwest of Matua Island with a summit elevation of 1496 m, is one of the most active volcanoes of the Kuril Islands. Violent volcanic eruptions at Sarychev Peak were detected by satellite imagery from June 12 to June 18, 2009. Kuril Islands are on the air traffic path between North America and East Asia, the volcanic ash clouds from Sarychev Peak forced some aircraft to reroute. ASTER composite image of before and after the eruption shows vegetated areas covered by ash, mud flow and new land platforms outside the old coast line. Total area of the new land platforms is 1.4 km2. Elevation changes are calculated from the digital elevation models generated from PRISM stereo images taken before and after the eruption. We found two remarkable areas that have large elevation changes. One is bow-shaped area with 2 km long and 200 m width from the summit to the northern flank. The thickness of the area is 10 to 30 m except for the most northern part up to 40 m. The other one is also bow-shaped area with 2.3 km long and 150 m width form the summit to eastern flank. Thermal anomalies on the bow-shaped areas are also detected nighttime ASTER thermal image observed on June 20. These bow-shaped areas are seemed to be trace of new lava flows. Plume heights are estimated two different methods using ASTER image observed on June 16. First one is temperature method (Kienele and Shaw, 1979). The lowest temperature of -53 C was detected, which corresponding to 10 km altitude, at about 10 km north-northeast of the vent. Then the plume moved to farther down to north-northeast with slightly lower altitude of 7 to 9 km. Second one is stereo image method (Urai, 2004). Plume heights of 7 points are estimated using the stereo image method. These heights are 1.8 to 3.5 km higher than plume heights estimated by the temperature method. SO2 is one of major component of volcanic gases and can be monitored using Ozone Monitoring Instrument (OMI) onboard Aura satellite. Total area that has more than 1, 2 and 10 DU of column SO2 amounts in the range from 40N to 52.5N and from 130E to 150W are 1,100,000 km2, 720,000 km2 and 186,000 km2, respectively.
The use of satellite-based remote sensing to quantify the contribution of volcanoes to the global SO2 budget
Volcanic degassing is a major contributor to the global sulphur dioxide (SO2) budget, characterised by permanent quiescent emissions in the lower troposphere punctuated with sporadic and spatially variable explosive eruptions into the upper troposphere and lower stratosphere (UTLS). The Moderate Resolution Imaging Spectroradiometer (MODIS), the Atmospheric Infrared Sounder (AIRS) and the Ozone Monitoring Instrument (OMI) on the A-Train suite of satellites, have the combined ability to measure both passive and explosive volcanic emissions of SO2. Near-coincident thermal infrared (TIR; MODIS and AIRS) and ultraviolet (UV; OMI) observations with different vertical sensitivity allow SO2 emissions in the UTLS to be delineated from lower tropospheric degassing, with ramifications for climate impacts. This study compares MODIS and AIRS TIR and OMI UV retrievals of SO2 for a number of recent, large volcanic eruptions on a case-by-case basis in order to compare the performance of each of the sensors for different eruptive scenarios (latitude, plume altitude, season etc.). Using this approach, we attempt to quantify rates of SO2 loading, residence times, and the overall performance of each of the sensors in a range of settings in order to produce a global estimate of volcanic SO2 emissions to the atmosphere.
Forecasting large explosions using thermal satellite data at Bezymianny volcano, Kamchatka
Volcanic ash injected into aircraft routes poses a severe risk to both life and cargo, with an estimated economic risk in the US alone of $70 million annually. Here we present a new method of forecasting explosive eruptions based on the spectral radiance received by orbiting Advanced Very High Resolution Radiometers (AVHRR). Volcanic dome building episodes commonly show increases in extrusion rate prior to erupting explosively, which will result in an increase in the radiated thermal flux that can be detected by satellites and therefore used in forecasting. From mid-1993 to mid-2008 Bezymianny exhibited 20 large (ash plumes>6 km a.s.l.) explosions, and three phases of dome growth without reported accompanying explosions. AVHRR data are available surrounding 19 of the explosions. Three types of precursory activity are observed before explosions: (I) in two cases precursory thermal anomalies clustered around the mode of the dataset (8.5°C), (II) three explosions were preceded by major thermal activity causing sensor saturation and (III) fourteen explosions were preceded by minor precursory thermal activity in which an upward trend in thermal anomaly values was detected 1-5 days before an explosion. A pattern recognition algorithm based on the trends observed prior to known explosions uses contextual, temporal and fixed threshold approaches to analyze slope and intercept values of straight lines fitted through 30-day moving windows of AVHRR thermal data. Using type II and III precursory patterns, the algorithm triggered at least one alert in the 30 days preceding all of the 17 explosions that show precursory increases in pixel-integrated radiant temperature. The alerts issued by the algorithm are color coded: yellow, orange and red alerts indicate probabilities of an explosion within the next 30 days of 43%, 64% and 83% respectively. This study highlights that it is possible to develop a computationally simple but successful algorithm to forecast explosive behavior in near real-time based on thermal changes. This algorithm will provide alerts of changes in the time series that would not be obvious to analysts looking at a single image and it can serve as a trigger to evaluate other available geophysical datasets (e.g. seismic data) alongside the thermal data. Precursory thermal data from future explosions can be used to update and adjust the algorithm as required, potentially resulting in even greater forecasting accuracy. It will be beneficial to try this technique at other dome-forming volcanoes around the world.
RETRIEVAL OF MULTISPECTRAL INFRARED EMISSIVITY FROM THERMALLY-MIXED VOLCANIC SURFACES FOR MORE ACCURATE COMPOSITIONAL MAPPING
The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor was launched in December 1999 as one of five instruments on the Terra satellite, part of NASA’s Earth Observing System (EOS), and has proven effective for the detection and monitoring of volcanic eruptions and their associated products. However, continuous advancement in analytical techniques remains essential to better understand the data acquired from active volcanoes. These images commonly contain features that are below the spatial resolution of the data; are more indicative of the state of volcanic unrest; and tend to saturate thermal infrared (TIR) sensors due to their high emitted radiance. In addition, compositional, textural, and thermal heterogeneities can vary greatly over the area of just one 90m ASTER TIR pixel, and without more advanced techniques the accurate retrieval of the temperature and composition of that surface becomes impossible. Previous studies of isothermal, compositionally heterogeneous pixels have shown that the emitted radiance mixes linearly, such that the spectrum is a combination of the energy from each component in proportion to its areal percentage. However, where thermal mixing occurs, the linear approach is no longer valid. We present a new approach for these thermally mixed pixels in an ASTER TIR scene using predetermined thermal components in order to model the associated errors in the emissivity spectra. These results have been used to determine temperature thresholds and corrections for basalt spectra from the active flow fields of Kilauea volcano, Hawaii. ASTER data of Kilauea acquired during an active effusive phase in October of 2006 have been deconvolved with the new approach, which identifies the thermally mixed pixels and then separates them into their hot and cool thermal components using the higher spatial resolution short wave infrared (SWIR) bands. Results provide more accurate non saturated temperature estimates as well as corrections to the emissivity for better compositional and textural mapping of the surface. This approach also serves as rapid means for identifying surface breakouts and minimizes the processing time, therefore allowing critical information to be disseminated expeditiously. It could prove to be an invaluable tool for understanding other high temperature processes and hazards, which are commonly obscured in low to medium spatial resolution orbital data sets.
Multispectral Thermal Infrared Remote Sensing of Volcanic SO2 Plumes with NASA’s Earth Observing System
The instruments aboard NASA’s series of Earth Observing System satellites provide a rich suite of measurements for the mapping of volcanic plumes and clouds. This presentation will focus on applications of thermal multispectral infrared (TIR) data acquired with the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Moderate-Resolution Imaging Spectrometer (MODIS), and Atmospheric Infrared Sounder (AIRS) to the recent eruptions of Augustine and Sarychev volcanoes in Alaska and the Russian Kuril Islands, respectively. ASTER, MODIS, and AIRS provide complimentary information on the quantity and distribution of sulfur dioxide (SO2), silicate ash, and sulfate (SO4) aerosols within plumes. In addition, data from the Multi-angle Imaging SpectroRadiometer (MISR) are used to derive estimates of aerosol loading, cloud-top altitude, wind direction, and wind speed. MODIS is our workhorse for plume mapping projects. There are MODIS instruments on the Terra and Aqua platforms, ensuring at least two MODIS passes per day over most volcanoes and four passes per day over many volcanoes. The spatial resolution of MODIS TIR radiance measurements is 1 km (at nadir) over a ground swath of 2330 km. MODIS can detect both the 7.3 and 8.5 μm bands of SO2, although the 7.3 μm band is often obscured by water vapor absorption when plumes are altitudes below ~ 4 km. ASTER has five channels in the TIR, and can detect the 8.5 μm SO2 band. The high spatial resolution (90 m) of ASTER TIR radiance measurements results in high sensitivity to SO2 within a narrow ground swath (60 km). AIRS has over 2700 spectral channels between 3.7 and 15.4 μm, allowing us to make unambiguous identifications of SO2, SO4 aerosols, and ash over a ground swath of ~2330 km. AIRS can detect the 7.3 μm SO2 band, and the strength of this band partially offsets the coarse spatial resolution of this instrument (~17 km at nadir). The key to multi-sensor mapping is the availability of a standard set of tools for the processing of data from different instruments. MAP_SO2 is a graphic user interface to the MODTRAN radiative transfer model that provides tools for the estimation of emissivity spectra, water vapor and ozone correction factors, surface temperature, and concentrations of SO2. To date we have used the MAP_SO2 toolkit to analyze data from MODIS, ASTER, AIRS, and a variety of airborne instruments. Our plans for future refinements of MAP_SO2 include the incorporation of AIRS-based profiles of atmospheric temperature, water vapor and ozone, and MISR-based maps of plume-top altitude into the plume mapping procedures.
Spaceborne observations from AVHRR and GOES weather satellites during the November 2002 eruption of Reventador Volcano, Ecuador
Satellite data were used to document the sequence of events during the November 2002 eruption of Reventador Volcano, Ecuador. Single-, multi-band and brightness temperature difference ash detection and monitoring techniques using data from the Advanced Very High Resolution Radiometer (AVHRR) and Geostationary Operational Environmental Satellite (GOES) revealed two primary stages of the eruption: a paroxysmal explosive phase between 3 and 4 November 2002 (Phase I), and an intermittent and mildly explosive phase between 5 and 25 November 2002 (Phase II). Initial phreatic activity at Reventador was observed by GOES at 1045z on 3 November, and by 1412z the volcano had entered a Vulcanian explosion that was not observed from space until the GOES scan at 1415z. This eruption generated a 16-km high, near-circular umbrella cloud 40 km in diameter characterized by long period gravity waves that were discernable from space. Similar processes were observed following the eruptions of Mt. St. Helens (1980) and Mt. Pinatubo (1991). The initial spreading rate of Reventador’s umbrella region was measured at 8.6 ms-1, and had a minimum cloud top-temperature of approximately 195°K (-78° C). After the initial 90 minutes, the radial umbrella cloud evolved into an ellipse due to entrainment by regional winds. Our analyses of plume-top temperatures, plume front velocities, and shadow clinometry indicate that the umbrella cloud transitioned into two elongated ash clouds due to wind shear effects at the Tropopause. The ash reached approximately 14 km (Troposphere) and 16 km (Stratosphere) above sea level and advanced westward and eastward, respectively. Over a period of 60 hours, the ash clouds advanced at 5 - 11 ms-1 and traveled 1900 km W and 400 km E of Reventador before falling below satellite detection thresholds. Volcanic ash dispersion model simulations from HYSPLIT and PUFF independently corroborated the observed transport velocities and trajectories. Ground-based data showed good correlations with satellite-derived observations, but also evidenced the inherent limitations that space-based monitoring platforms can have (i.e.: spatial resolution, scanning intervals, occlusion due to 2-dimensional perspective, scan angle). Overall, the GOES and AVHRR satellite data provided valuable synoptic multispectral measurements that allowed for the detection and characterization of the 2002 eruption of Reventador Volcano.
The cooling rate of an aa lava flow determined using an orbital imaging spectrometer
The surface temperature of an active lava flow is an important property to measure. Through its influence on lava crystallinity, cooling exerts a fundamental control on lava rheology. Remotely sensed thermal radiance data acquired by multispectral sensors such as Landsat Thematic Mapper and the Terra Advanced Spaceborne Thermal Emission and Reflection Radiometer, are of insufficient spectral and radiometric fidelity to allow for realistic determination of lava surface temperatures from Earth orbit. This presentation describes results obtained from the analysis of active lava flows using hyperspectral data acquired by NASA’s Earth Observing-1 Hyperion imaging spectrometer. The contiguous nature of the measured radiance spectrum in the 0.4-2.5 micron region means that, although sensor saturation most certainly occurs, unsaturated radiance data are always available from even the hottest, and most radiant, active lava flow surfaces. The increased number of wavebands available allows for the assumption of more complex flow surface temperature distributions in the radiance-to-temperature inversion processes. The application of such data to the analysis of a time-series of three Hyperion images of an active lava flow, acquired during a four day period at Mount Etna, Sicily, is demonstrated. The results provide insights into the temperature-radiance mixture modeling problem that will aid in the analysis of data acquired by future hyperspectral remote sensing missions, such as NASA’s proposed HyspIRI mission. By also proving radiance data on the opposite limb of the planckian emittance curve (i.e. the MIR and TIR), HyspIRI will allow us to improve upon these antecedent results.
Advantages of SO2 evaluation in the wavelength range 360-390 nm
Differential Optical Absorption Spectroscopy (DOAS) is a well established spectroscopic method to determine trace gases in the atmosphere. During the last decade, passive DOAS, which uses solar radiation scattered in the atmosphere as a light source, has become a standard tool to determine SO2 emission fluxes from volcanoes and other large sources. However, there are several problems limiting the accuracy of this technique. Especially for high atmospheric SO2 concentrations, the observed optical depth observed in the UV is not a linear function of the atmospheric column density any-more. While these ‘saturation’ effects can in principle be corrected, here we propose to use an alternative wavelength region (around 380nm) for the DOAS evaluation of SO2, which we show to have considerable advantages, in particular when high SO2 column densities prevail or when measurements at large distances from the emission source are desired. We will present example evaluations for both, the standard as well as the “new” wavelength range for ground based and satellite measurements. Additionally to spectroscopic issues, model results investigating the influence of radiative transfer in both regions will be discussed.
Monitoring passively degassing volcanoes from space: A comparison between ASTER and OMI retrievals of lower tropospheric SO2
Passively degassing volcanoes contribute a climatologically significant quantity of sulfur dioxide (SO2) to the atmosphere. Both the Advanced Thermal Emission and Reflection Radiometer (ASTER) and the Ozone Monitoring Instrument (OMI) are capable of detecting emissions from volcanoes in a non-eruptive state. There are fundamental differences between the sensors that affect their sensitivity to SO2. OMI operates in the ultraviolet with a 13x24 km nadir footprint and a 2600 km swath width, providing daily global coverage and retrievals of SO2 at all altitudes from the planetary boundary layer to the stratosphere (Carn et al, 2008). In contrast, ASTER operates in the infrared (specifically the 8.6 µm region of the thermal infrared for SO2 detection) with 90 m spatial resolution and a 60 km swath width. Hence the temporal resolution and geographic coverage of ASTER is somewhat less than OMI, with one ASTER scene acquired every 5-16 days for a given location. However, the higher spatial resolution of ASTER provides more information on the structure of tropospheric SO2 plumes. Six volcanoes were selected based on their differing climates, altitudes and SO2 emission rates: Mt Etna, Sicily; Pacaya, Guatemala; Masaya, Nicaragua; Popocatepetl, Mexico; Nyiragongo, DR Congo; and Kilauea, Hawaii. These volcanoes are continually active and typically emit in excess of 1000 metric tons per day of SO2. ASTER and OMI data were acquired for each volcano and processed in order to compare the satellite SO2 retrievals under different conditions. Our goal is to determine the optimum conditions for lower tropospheric SO2 retrievals using each instrument, and constrain the lower limit of volcanic SO2 emission rate that can be detected and monitored from space.
Thermal Infrared Remote Sensing of the Yellowstone Geothermal System
The Yellowstone National Park (YNP) geothermal system is one of the largest in the world, with thousands of individual thermal features ranging in size from a few centimeters to tens of meters across, (e.g., fumaroles, geysers, mud pots and hot spring pools). Together, large concentrations of these thermal features make up dozens of distinct thermal areas, characterized by sparse vegetation, hydrothermally altered rocks, and usually either sinter, travertine, or acid sulfate alteration. The temperature of these thermal features generally ranges from ~30 to ~93 oC, which is the boiling temperature of water at the elevation of Yellowstone. In-situ temperature measurements of various thermal features are sparse in both space and time, but they show a dynamic time-temperature relationship. For example, as geysers erupt and send pulses of warm water down slope, the warm water cools rapidly and is then followed by another pulse of warm water, on time scales of minutes. The total heat flux from the Park’s thermal features has been indirectly estimated from chemical analysis of Cl- flux in water flowing from Yellowstone’s rivers. We are working to provide a more direct measurement, as well as estimates of time variability, of the total heat flux using satellite multispectral thermal infrared (TIR) remote sensing data. <P>Over the last 10 years, NASA’s orbiting ASTER and MODIS instruments have acquired hundreds and thousands of multispectral TIR images, respectively, over the YNP area. Compared with some volcanoes, Yellowstone is a relatively low-temperature geothermal system, with low thermal contrast to the non-geothermal surrounding areas; therefore we are refining existing techniques to extract surface temperature and thermal flux information. This task is complicated by issues such as, during the day, solar heated surfaces may be warmer than nearby geothermal features; and there is some topographic (elevation) influence on surface temperatures, even at night. Still we have been able to obtain temperature and heat flux values from small scale geothermal features with ASTER and some larger scale thermal areas with MODIS. The latest results of this study will be presented; including MODIS time-series data and examples of using higher spatial resolution ASTER data for identifying hot spots.
Absorption Coefficients for SO2 between 280 and 330 nm and 200 and 300 K
A consistent set of absorption coefficients have been measured for SO2 between 280 and 330 nm. Measurements were taken at medium resolution (0.015 nm) and low pressure (~1 Torr) at temperatures of 295, 273, 258, 246, 217, and 197 K. Absorption cross-sections of SO2 at lower temperatures showed a consistent increase in the center of bands and a decrease in the wings.
SO2 absorption cross-section at 258 K
Detection of thermal changes possibly associated with volcanic activity and discrimination of faint changes from MODIS
There are many reports in natural disasters such as volcanic activity and earthquakes. To mitigate these disasters, monitoring of the crustal activity is important. But it is difficult to monitor all volcanoes on the ground. On the other hand, satellite remote sensing on temperature anomaly is one of the most effective methods to monitor volcanic activity, because it can monitor in a wide area at a time and with a high frequency on a day. We use the MODerate Resolution Imaging Spectroradiometer (MODIS) sensor on board AQUA satellite for volcanic activity monitoring in this paper. MODIS has 36 different bands. Band 20 of MODIS has a character of less influence from atmosphere but a tendency to give lower temperature for cloud surface. Bands 31 and 32 of MODIS are more sensitive to clouds. Therefore, these bands are considered the best tool for monitoring volcanic activity for MODIS. The previous studies focus on the variation of temperature around the targets such as a crater and there is a possibility to include spurious changes due to meteorological condition, seasonal factors, and so on. Since a brightness temperature observed by MODIS is affected by various factors. Therefore, in order to extract a brightness temperature by a specific cause, it is effective to consider a difference between adequate reference points. In this paper, differential brightness temperature is computed for extracting regional or global changes in an image. Then, the differential brightness temperature is evaluated in singularity over all the analyzed period in time and space. The purpose of this paper is to present the developed algorithm for detecting thermal temperature anomaly associated with a volcanic eruption by MODIS and demonstrate the effectiveness of practical application to eruptions of Mt. Merapi volcano and Mt. Asama volcano. In the case of Mt. Merapi volcano, monitoring daily maps and time variation of the brightness temperature, it is found that the increase of brightness temperature at both the summits and there are no increases at the pixels far from the summits. Using the historical data of band31 and band 32, the criterion of cloud pixel is successfully defined and it causes the stable result for anomalous changes in brightness temperature. Singularity of the differential brightness temperature seems to have a high correlation with volcanic activities. In the case of Mt. Asama volcano, Japan, differential brightness temperature shows the thermal anomalies associated with volcanic activities. But the relationship between singularity and volcanic activity is not high. The reason is that clouds influences still remain in the case of Mt. Asama volcano. It is very important to develop the methodology to remove cloud effects from differential temperature data in the temperate zone such as Japan.
Theoretical validation of ASTER_SW algorithm used for the monitoring of hot volcanic lakes
Volcanic lakes act as calorimeters trapping most of the heat released by the magmatic-hydrothermal system. Their temperatures are reflecting the balance between heat input from hydrothermal fluids and heat output by radiation and evaporation of the lake surface to the atmosphere. The lake surface temperature is one of the key parameters used to detect any changes occurring in the activity of the volcano. Many volcanic lakes are located in remote areas with difficult accessibility; these lakes are rarely visited or monitored. For these lakes remote sensing by satellite sensors can provide very useful information at relatively low cost. To retrieve surface temperatures of volcanic lakes, ASTER TIR images were analyzed with a recently developed algorithm based on a Split-Window method: ASTER_SW. The difference in brightness temperatures between bands 13 and 14 (BT13-BT14) is used to remove the atmospheric effects. The use of two TIR channels enables a differential absorption measurement in order to remove the effects of atmospheric vapor and other absorbing constituents. Further validation of the ASTER_SW algorithm was completed by applying it to a set of radiance simulations using a line-by-line radiative transfer code (Atmosphit). Parameters used for the simulations included: surface temperatures, atmospheric models and surface altitudes. The obtained spectra were integrated on the spectral response functions of ASTER and converted with the inverse of Planck’s law to get the Simulated Brightness Temperatures (SBT). The ASTER_SW algorithm was applied to the SBT. The coherence between ASTER_SW-derived temperatures and model surface temperatures was examined to test the validity and robustness of the algorithm. ASTER_SW proved to be accurate in most circumstances, revealing no systematic bias in any peculiar atmosphere or altitude. However, for very warm lakes (T>50°C), a small dependency on surface altitude is appearing in tropical atmospheres. The low frequency of ASTER imaging (at best 1-2 image per week, depending on latitude and priority acquisition) can limit its interest for real time monitoring of crater lake temperature, but this shortcoming should disappear with the future HYSPIRI sensor, that will have a wider image swath and a continuous coverage.
Detection of Low Temperature Volcanogenic Thermal Anomalies with ASTER
Predicting volcanic eruptions is a thorny problem, as volcanoes typically exhibit idiosyncratic waxing and/or waning pre-eruption emission, geodetic, and seismic behavior. It is no surprise that increasing our accuracy and precision in eruption prediction depends on assessing the time-progressions of all relevant precursor geophysical, geochemical, and geological phenomena, and on more frequently observing volcanoes when they become restless. The ASTER instrument on the NASA Terra Earth Observing System satellite in low earth orbit provides important capabilities in the area of detection of volcanogenic anomalies such as thermal precursors and increased passive gas emissions. Its unique high spatial resolution multi-spectral thermal IR imaging data (90m/pixel; 5 bands in the 8-12um region), bore-sighted with visible and near-IR imaging data, and combined with off-nadir pointing and stereo-photogrammetric capabilities make ASTER a potentially important volcanic precursor detection tool. We are utilizing the JPL ASTER Volcano Archive (http://ava.jpl.nasa.gov) to systematically examine 80,000+ ASTER volcano images to analyze (a) thermal emission baseline behavior for over 1500 volcanoes worldwide, (b) the form and magnitude of time-dependent thermal emission variability for these volcanoes, and (c) the spatio-temporal limits of detection of pre-eruption temporal changes in thermal emission in the context of eruption precursor behavior. We are creating and analyzing a catalog of the magnitude, frequency, and distribution of volcano thermal signatures worldwide as observed from ASTER since 2000 at 90m/pixel. Of particular interest as eruption precursors are small low contrast thermal anomalies of low apparent absolute temperature (e.g., melt-water lakes, fumaroles, geysers, grossly sub-pixel hotspots), for which the signal-to-noise ratio may be marginal (e.g., scene confusion due to clouds, water and water vapor, fumarolic emissions, variegated ground emissivity, and their combinations). To systematically detect such intrinsically difficult anomalies within our large archive, we are exploring a four step approach: (a) the recursive application of a GPU-accelerated, edge-preserving bilateral filter prepares a thermal image by removing noise and fine detail; (b) the resulting stylized filtered image is segmented by a path-independent region-growing algorithm, (c) the resulting segments are fused based on thermal affinity, and (d) fused segments are subjected to thermal and geographical tests for hotspot detection and classification, to eliminate false alarms or non-volcanogenic anomalies. We will discuss our progress in creating the general thermal anomaly catalog as well as algorithm approach and results. This work was carried out at the Jet Propulsion Laboratory of the California Institute of Technology under contract to NASA.