• Assessment of LiDAR and spectral techniques for high-resolution mapping of permafrost on the Yukon-Kuskokwim Delta, Alaska

      Whitley, Matthew Allen; Maio, Christopher V.; Frost, Gerald V.; Jorgenson, M. Torre (2017-05)
      The Yukon-Kuskokwim Delta (YKD) is one of the largest and most ecologically productive coastal wetland regions in the pan-Arctic. Formed by the Yukon and Kuskokwim Rivers flowing into the Bering Sea, nearly 130,000 square kilometers of delta support 23,000 Alaskan Natives living subsistence lifestyles. Permafrost on the outer delta commonly occurs on the abandoned floodplain deposits. Ground ice in the soil raises surface elevations on the order of 1-2 meters, creating plateaus on the landscape. Better drainage on the plateaus supports distinct Sphagnum-rich vegetation, which in turn protects the permafrost from rising air temperatures with low thermal conductivity during the summer. This ecosystem-protected permafrost is thus vulnerable to disturbances from rising air temperatures, vegetation mortality, and inland storm surges, which have been known to flood up to 37 km inland. This thesis assesses several novel techniques to map permafrost distribution at high-resolution on the YKD. Accurate baseline maps of permafrost extent are critical for a variety of applications, including long-term monitoring. As air and ground temperatures rise across the Arctic, monitoring landscape change is important for understanding permafrost degradation processes (e.g. thermokarst) and greenhouse gas dynamics from the local to global scales. This thesis separately explored the value of Light Detection And Ranging (LiDAR) and spectral datasets as tools to map permafrost at a high spatial resolution. Furthermore, this thesis sought to automate these processes, with the vision of high-resolution mapping over large spatial extents. Fieldwork was conducted in July 2016 to both parameterize and then validate the mapping efforts. The LiDAR mapping extent assessed a 135 km² area (~15% permafrost cover), and the spectral mapping extent assessed an 8 km² area (~20% permafrost cover). For the LiDAR dataset, the use of a simple elevation threshold informed by field ground truth values provided a permafrost map with 94.9% accuracy. This simple approach was possible because of the extremely flat terrain. For the spectral datasets, an ad-hoc masking technique was developed using a combination of texture analysis, principal component analysis, and morphological filtering. Two contrasting workflows were evaluated with fully-automated and semi-automated methods with mixed results. The highest mapping accuracy was 89.4% and the lowest was 79.1%, though the error of omission in mapping the permafrost remained high (7.02 - 59.7%) for most analyses. The spectral mapping algorithms did not replicate well across different high-resolution images, raising questions about the viability of using spectral methods alone to track thermokarst and landscape change over time. However, incorporating the spectral methods explored in this analysis with other datasets (e.g. LiDAR) has the potential to increase mapping accuracies. Both the methods and the results of this thesis enhance permafrost mapping efforts on the YKD, and provide a good first step to monitoring landscape change in the region.
    • Characterization And Interpretation Of Volcanic Activity At Redoubt, Bezymianny And Karymsky Volcanoes Through Direct And Remote Measurements Of Volcanic Emissions

      Lopez, Taryn M.; Cahill, Catherine; Dehn, Jonathan; Newberry, Rainer; Simpson, William; Werner, Cynthia (2013)
      Surface measurements of volcanic emissions can provide critical insight into subsurface processes at active volcanoes such as the influx or ascent of magma, changes in conduit permeability, and relative eruption size. In this dissertation I employ direct and remote measurements of volcanic emissions to characterize activity and elucidate subsurface processes at three active volcanoes around the North Pacific. The 2009 eruption of Redoubt Volcano, Alaska, produced elevated SO2 emissions that were detected by the Ozone Monitoring Instrument (OMI) satellite sensor for over three months. This provided a rare opportunity to characterize Redoubt's daily SO2 emissions and to validate the OMI measurements. Order of magnitude variations in daily SO2 mass were observed, with over half of the cumulative SO2 emissions released during the explosive phase of the eruption. Correlations among OMI daily SO2 mass, tephra mass and acoustic energies during the explosive phase suggest that OMI data may be used to infer eruption size and explosivity. From 2007 through 2010 direct and remote measurements of volcanic gas composition and flux were measured at Bezymianny Volcano, Kamchatka, Russia. During this period Bezymianny underwent five explosive eruptions. Estimates of passive and eruptive SO2 emissions suggest that the majority of SO2 is released passively. Order of magnitude variations in total volatile flux observed throughout the study period were attributed to changes in the depth of gas exsolution and separation from the melt at the time of sample collection. These findings suggest that exsolved gas composition may be used to detect magma ascent prior to eruption at Bezymianny Volcano. Karymsky Volcano, Kamchatka, Russia, is a dynamic volcano which exhibited four end-member activity types during field campaigns in 2011 and 2012, including: discrete ash explosions, pulsatory degassing, gas jetting, and explosive eruption. These activity types were characterized quantitatively and uniquely distinguished using a multiparameter dataset based on infrasound, thermal imagery, and volcanic emissions. These observations suggest a decoupling between exsolved volatiles and magma at depth. I propose that variations in magma degassing depth influence the flux and proportions of decompression- and crystallization-induced degassing, as well as conduit permeability, can explain the variations in volcanic activity.
    • Investigation On Cirrus Clouds By The Cloud-Aerosol Lidar And Infrared Pathfinder Satellite Observation Data

      Zhu, Jiang; Sassen, Kenneth (2011)
      Understanding and describing the role of clouds in the climate system need intensive and extensive research on cloud properties. The albedo and greenhouse effects of clouds and their relations with the physical properties of clouds are analyzed. Cloud-top height and ice water content are key factors in impacting the longwave and shortwave radiation, respectively. Lidar and infrared radiometer measurement technologies are introduced. Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) level 1 Lidar profile, level 2 cloud layer, and level 2 Lidar/IIR track products are briefly reviewed. The algorithms for identification of cirrus clouds, Linear Depolarization Ratio (LDR), and effective diameter are presented. An average LDR profile is calculated by using the sum of total attenuated backscattering profiles and the sum of perpendicular attenuated backscattering profiles. A weight-average method is applied to calculate the average LDR. A split-window method is applied to estimate the effective diameters of clouds. A set of bulk ice crystal models and a radiative transfer model are applied to produce a look-up table that includes the radiative transfer simulation results. The macro-physical properties of cirrus clouds are analyzed. The frequency of occurrence of cirrus clouds varies with latitude, and strongly relates to the atmospheric circulation. Cirrus clouds are few in high-pressure zones and abundant where seasonal monsoonal circulation occurs. Cloud-top height decreases with increasing latitude. Cloud-top temperature is lower in the tropical regions than in the midlatutude and the polar regions. The measured cloud thickness shows a great diurnal variation.
    • Mass Balances And Dynamic Changes Of The Bering, Malaspina, And Icy Bay Glacier Systems Of Alaska, United States, And Yukon, Canada

      Muskett, Reginald R.; Lingle, Craig (2007)
      The Bering and the Malaspina Glacier systems of south-central Alaska, U.S.A., and southwest Yukon Territory, Canada, in the Saint Elias Mountains constitute the two largest temperate surge-type piedmont glaciers on Earth. This is largest region of glaciers and icefields in continental North America. Determining and understanding the causes of wastage of these two glaciers is important to understanding the linkages of glacier mass balance to climate change, glacier dynamics, and the contributions of the glaciers of northwestern North America to rising sea level. Presented are the first detailed estimate of the net mass balances of the Bering and Malaspina Glacier systems, the effects of glacier dynamics on their accumulation areas, and the wastage of the tidewater glaciers of Icy Bay. The combined wastage of the Bering and Malaspina Glacier systems from 1972 to 2003, 254 +/- 16 km 3 water equivalent over a glacier area of 7734 km2, is equivalent to an area-average mass balance of -1.06 +/- 0.07 m/y over that time period. This represents a contribution to global sea-level rise of 0.70 +/- 0.05 mm, 0.023 +/- 0.002 mm/yr from 1972 to 2003. This is roughly 0.8% of the modern sea-level rise as estimated from tide-gauges and satellites, and roughly 9% of the contribution from non-polar glaciers and ice caps. Glacier wastage has been caused by climate warming (negative mass balance) superimposed on the effects of glacier dynamics. Near-concurrent surge of the three largest glaciers of the Malaspina Glacier piedmont were observed during 1999 to 2002. In addition, the tidewater Tyndall Glacier, whose retreat since 1910 was interrupted in 1964 by a major surge, also surged during 1999 to 2002. These four surges have occurred roughly 23 years after the 1976/77 shift of the Pacific Decadal Oscillation to its current warm-wet phase. Despite the increase of high-elevation snow accumulation observed on Mt. Logan, the accumulation areas of the Bering and Malaspina Glacier systems are being drawn down by the effects of glacier dynamics. Wastage has accelerated since 2000.
    • Merging remotely sensed data with geophysical models

      Searcy, Stanley Craig; Layer, Paul; Stringer, William; Dean, Ken; Niebauer, Joe; Goering, Doug; Weller, Gunter (1996)
      Geophysical models are usually derived from the idealistic viewpoint that all required external parameters are, in principle, measurable. The models are then driven with the best available data for those parameters. In some cases, there are few measurements available, because of factors such as the location of the phenomena modeled. Satellite imagery provides a synoptic overview of a particular environment, supplying spatial and temporal variability as well as spectral data, making this an ideal source of data for some models. In other cases, although frequent satellite image observations are available, they are of little use to the modeler, because they do not provide values for the parameters demanded by the model. This thesis contains two examples of geophysical models that were derived expressly to utilize measurements and qualitative observations taken from satellite images as the major driving elements of the model. The methodology consists of designing a model such that it can be 'run' by numerical data extracted from image data sets, and using the image data for verification of the model or adjustment of parameters. The first example is a thermodynamic model of springtime removal of nearshore ice from an Arctic river delta area, using the Mackenzie River as a study site. In this example, a multi-date sequence of AVHRR images is used to provide the spatial and temporal patterns of melt, allowing the required physical observations in the model to be parameterized and tested. The second example is a dynamic model simulating the evolution of a volcanic ash cloud under the influence of atmospheric winds. In this case, AVHRR images are used to determine the position and size of the ash cloud as a function of time, allowing tuning of parameters and verification of the model.
    • Oceanic emissions of sulfur: Application of new techniques

      Jodwalis, Clara Mary; Benner, Richard L. (1998)
      Sulfur gases and aerosols are important in the atmosphere because they play major roles in acid rain, arctic haze, air pollution, and climate. Globally, man-made and natural sulfur emissions are comparable in magnitude. The major natural source is dimethyl sulfide (DMS) from the oceans, where it originates from the degradation of dimethysulfonioproprionate (DMSP), a compound produced by marine phytoplankton. Global budgets of natural sulfur emissions are uncertain because of (1) the uncertainty in the traditional method used to estimate DMS sea-to-air flux, and (2) the spatial and temporal variability of DMS sea-to-air flux. We have worked to lessen the uncertainty on both fronts. The commonly used method for estimating DMS sea-to-air flux is certain to a factor of two, at best. We used a novel instrumental technique to measure, for the first time, sulfur gas concentration fluctuations in the marine boundary layer. The measured concentration fluctuations were then used with two established micrometeorological techniques to estimate sea-to-air flux of sulfur. Both methods appear to be more accurate than the commonly used one. The analytical instrument we used in our studies shows potential as a direct flux measurement device. High primary productivity in high-latitude oceans suggests a potentially large DMS source from northern oceans. To begin to investigate this hypothesis, we have measured DMS in the air over northern oceans around Alaska. For integrating and extrapolating field measurements over larger areas and longer time periods, we have developed a model of DMS ocean mixing, biological production, and sea-to-air flux of DMS. The model's main utility is in gaining intuition on which parameters are most important to DMS sea-to-air flux. This information, along with a direct flux measurement capability, are crucial steps toward the long-term goal of remotely sensing DMS flux. A remote sensing approach will mitigate the problems of spatial and temporal variability. The new developments in methodology, field sampling, and modeling put forth in this thesis are tools we have used to better understand and quantify sulfur gas emissions from northern oceans, which appear to be a significant source of sulfur to the global atmosphere.
    • Optical Remote Sensing Of Snow On Sea Ice: Ground Measurements, Satellite Data Analysis, And Radiative Transfer Modeling

      Zhou, Xiaobing; Li, Shusun; Stamnes, Knut; Sharpton, Buck; Jeffries, Martin O.; Echelmeyer, Keith (2002)
      The successful launch of the Terra satellite on December 18, 1999 opened a new era of earth observation from space. This thesis is motivated by the need for validation and promotion of the use of snow and sea ice products derived from MODIS, one of the main sensors aboard the Terra and Aqua satellites. Three cruises were made in the Southern Ocean, in the Ross, Amundsen and Bellingshausen seas. Measurements of all-wave albedo, spectral albedo, BRDF, snow surface temperature, snow grain size, and snow stratification etc. were carried out on pack ice floes and landfast ice. In situ measurements were also carried out concurrently with MODIS. The effect of snow physical parameters on the radiative quantities such as all-wave albedo, spectral albedo and bidirectional reflectance are studied using statistical techniques and radiative transfer modeling, including single scattering and multiple scattering. The whole thesis consists of six major parts. The first part (chapter 1) is a review of the present research work on the optical remote sensing of snow. The second part (chapter 2) describes the instrumentation and data-collection of ground measurements of all-wave albedo, spectral albedo and bidirectional reflectance distribution function (BRDF) of snow and sea ice in the visible-near-infrared (VNIR) domain in Western Antarctica. The third part (chapter 3) contains a detailed multivariate correlation and regression analysis of the measured radiative quantities with snow physical parameters such as snow density, surface temperature, single and composite grain size and number density. The fourth part (chapter 4) describes the validation of MODIS satellite data acquired concurrently with the ground measurements. The radiances collected by the MODIS sensor are converted to ground snow surface reflectances by removing the atmospheric effect using a radiative transfer algorithm (6S). Ground measured reflectance is corrected for ice concentration at the subpixel level so that the in situ and space-borne measured reflectance data are comparable. The fifth part (chapter 5) investigates the single scattering properties (extinction optical depth, single albedo, and the phase function or asymmetry factor) of snow grains (single or composite), which were calculated using the geometrical optical method. A computer code, GOMsnow, is developed and is tested against benchmark results obtained from an exact Mie scattering code (MIE0) and a Monte Carlo code. The sixth part (chapter 6) describes radiative transfer modeling of spectral albedo using a multi-layer snow model with a multiple scattering algorithm (DISORT). The effect of snow stratification on the spectral albedo is explored. The vertical heterogeneity of the snow grain-size and snow mass density is investigated. It is found that optical remote sensing of snow physical parameters from satellite measurements should take the vertical variation of snow physical parameters into account. The albedo of near-infrared bands is more sensitive to the grain-size at the very top snow layer (<5cm), while the albedo of the visible bands is sensitive to the grain-size of a much thicker snow layer. Snow parameters (grain-size, for instance) retrieved with near-infrared channels only represent the very top snow layer (most probably 1--3 cm). Multi-band measurements from visible to near-infrared have the potential to retrieve the vertical profile of snow parameters up to a snow depth limited by the maximum penetration depth of blue light.
    • Permafrost Distribution Mapping And Temperature Modeling Along The Alaska Highway Corridor, Interior Alaska

      Panda, Santosh K.; Prakash, Anupma (2011)
      An up-to-date permafrost distribution map is critical for making engineering decisions during the planning and design of any engineering project in Interior Alaska. I used a combination of empirical-statistical and remote sensing techniques to generate a high-resolution spatially continuous near-surface (< 1.6 m) permafrost map by exploiting the correlative relationships between permafrost and biophysical terrain parameters. A Binary Logistic Regression (BLR) model was used to establish the relationship between vegetation type, aspect-slope and permafrost presence. The logistic coefficients for each variable class obtained from the BLR model were supplied to respective variable classes mapped from remotely sensed data to estimate permafrost probability for every pixel. The BLR model predicts permafrost presence/absence at an accuracy of 88%. Near-surface permafrost occupies 37% of the total study area. A permafrost map based on the interpretation of airborne electromagnetic (EM) resistivity data shows 22.5 -- 43.5% of the total study area as underlain by permafrost. Permafrost distribution statistics from both the maps suggest near-surface permafrost distribution in the study area is sporadic (10 -- 50 % of the area underlain by permafrost). Changes in air temperature and/or winter snow depth are important factors responsible for permafrost aggradation or degradation. I evaluated the effects of past and recent (1941-2008) climate changes on permafrost and active-layer dynamics at selected locations using the Geophysical Institute Permafrost Laboratory model. Results revealed that active-layer thickness reached 0.58 and 1.0 m, and mean annual permafrost temperature increased by 1.6 and 1.7 �C during 1966-1994 at two sites in response to increased mean annual air temperature, mean summer air temperature and winter snow depth. The study found that active-layer thickness is not only a function of summer air temperature but also of mean annual air temperature and winter snow depth. Model simulation with a projected (2008-2098) climate scenario predicts 0.22 m loss of near-surface permafrost at one site and complete permafrost disappearance at another site by 2098. The contrasting permafrost behaviors at different sites under similar climate scenarios highlight the role of soil type and ground ice volume on permafrost dynamics; these factors determine permafrost resilience under a warming climate.
    • Permafrost Dynamics In 20Th And 21St Centuries Along The East-Siberian And Alaskan Transects

      Sazonova, Tatiana Sergeevna; Romanovsky, Vladimir (2003)
      High latitude ecosystems where the mean annual ground surface temperature is around or below 0�C are highly sensitive to global warming. This is largely because these regions contain vast areas of permafrost, which will begin to degrade when the mean annual ground temperatures will rise above 0�C. The Alaskan and East Siberian transects, centered on the 155� WL and 135� EL, were chosen for evaluation of permafrost---atmosphere interactions. The analysis of measured data shows a significant increase in air and ground temperatures that started from the late 1960s within both these transects and the magnitude of this increase is from 0.2 to 0.5�C per decade. This trend is comparable to trends predicted by majority of global warming scenarios. A simple and accurate model for evaluating the permafrost dynamics was developed in Geophysical Institute Permafrost Lab (GIPL). The GIPL model is a fusion of the modified Kudryavtsev's approach, which is a set of analytical formulas for active layer thickness (ALT) and mean annual ground temperature (MAGT) calculations, with the Geographic Information System (GIS). The evaluation of the GIPL performance showed that when applied to long-term (decadal or longer time scale) averages, this model achieves an accuracy of +/-0.2--0.4�C for the mean annual ground temperatures and +0.1--0.3 m for the active layer thickness calculations. The GIPL model was used for the hindcast of the permafrost dynamics in the 20th century, using a combination of observationally-based and simulated monthly grids of surface air temperature. The results showed that during the 20th century there were a number of relatively cold and warm periods. These climatic variations produced 1 to 3�C changes in MAGT and up to 1 m in the ALT. The forecast for the period of 2000--2100 was performed using climatic parameters from six Global Climate Models provided by Arctic Climate Impact Assessment program. The results showed that by the end of 21st century mean annual ground temperatures will be 2 to 6�C warmer and the ALT from 0.2 to 1 m deeper. During this period, in many areas within both transects the degradation of permafrost from the surface will start. By 2100, the area with actively degrading permafrost will cover about 10--15% of the Siberian transect and up to 30% and more within the Alaskan transect.
    • Remote sensing of surface albedo and cloud properties in the Arctic from AVHRR measurements

      Han, Wei; Stamnes, Knut; Bowling, Sue Ann; Harrison, William; Li, Shusun; Lubin, Dan; Watkins, Brenton (1996)
      Based on a comprehensive radiative transfer model, algorithms suitable for arctic conditions are developed to retrieve broadband surface albedo and water cloud properties from National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) narrowband measurements. Reflectance anisotropy of snow surfaces is first simulated by an discrete ordinates radiative transfer formulation, and is then included in the comprehensive model for the retrieval. Ground-based irradiance measurements made by NOAA Climate Monitoring and Diagnostics Laboratory (CMDL) in Barrow, Alaska are compared with retrieved albedo and downwelling irradiances computed from retrieved cloud optical depth and effective radius. Good agreement is found between satellite estimates and ground-based measurements, which indicate that the retrieval algorithms proposed in this thesis are suitable for arctic conditions. It is found that the effects of snow bidirectional reflectance on the retrieval of the broadband albedo are significant, and that the Lambertian approximation could lead to a 30% underestimate of the surface albedo. It is also found that cloud effective radius in the Arctic is generally smaller as compared with mid- and low-latitudes.
    • Satellite remote sensing of active wildfires in Alaska's boreal forest

      Waigl, Christine F.; Stuefer, Martin; Prakash, Anupma; Verbyla, David; Ichoku, Charles (2017-12)
      This research addresses improvements to the detection and characterization of active wildfires in Alaska with satellite-based sensors. The VIIRS I-band Fire Detection Algorithm for High Latitudes (VIFDAHL) was developed and evaluated against existing active fire products from the Visible Infrared Imaging Radiometer Suite (VIIRS) and the Moderate Resolution Imaging Spectroradiometer (MODIS). This new algorithm is based on VIIRS 375 m spatial resolution imagery and was tuned using fires in Alaska's boreal forest. It provides improved fire detection of low-intensity fires, especially during daytime and at sensor zenith angles smaller than approximately 50° off nadir. Low-intensity active fires, which represent residual combustion present after the passage of a high-intensity fire front, are not very well detected by existing active fire products. A second topic was fire remote sensing with ~30 m resolution imaging spectrometer (or hyperspectral instrument), the Hyperion sensor on NASA's EO-1 spacecraft, which was in use from 2000 to 2016. Hyperion had a much higher spectral resolution than VIIRS or MODIS, but no repeat imagery of the same active fire was available in Alaska. The investigation relied on absorption and emission features in the radiance spectra acquired at every pixel location. Three fire detection methods were evaluated using archived Hyperion data from three fires in interior Alaska from 2004 and 2009: A version of the Hyperspectral Fire Detection Algorithm (HFDI) produced excellent active fire maps; an approach that relies on a shortwave infrared carbon dioxide absorption feature and associated Continuum Interpolated Band Ratio (CO₂ CIBR) proved to be useful, but was affected by sensor noise and clouds; finally, a potassium emission feature from biomass burning was not detectable in the Hyperion data. Fire temperatures were determined using the Hyperion shortwave infrared spectra between 1400 nm and 2400 nm. The temperatures of active fire, the corresponding partial pixel areas, and the pixel areas occupied by unburned and already-burned vegetation, respectively, were modeled within each fire pixel. A model with two reflected background components and two temperature endmembers, applied to the same three study scenes, yielded an excellent fit to Hyperion spectral radiance data. Fire temperatures ranged from approximately 500-600 K to approximately 800-900 K. The retrieved lower fire temperatures are within the range of smoldering combustion; high-temperature values are limited by Hyperion's saturation behavior. High-temperature fire occupying 0.2% of a pixel (2 m²) was detectable. Sub-pixel fire area and temperature were also retrieved using VIIRS 750 m (M-band) data, with comparable results. Uncertainties were evaluated using a Monte Carlo simulation. This work offers insight into the sensitivity of fire detection products to time of day (largely due to overpass timing), spatial distribution over the study area (largely due to orbital properties) and sensor zenith angle. The results are relevant for sensor and algorithm design regarding the use of new multi- and hyperspectral sensors for fire science in the northern high latitudes. Data products resulting from this research were designed to be suitable for supporting fire management with an emphasis on real-time applications and also address less time-sensitive questions such as retrievals of fire temperature and time series of fire evolution.
    • The 1993-1994 Surge Of Bering Glacier, Alaska Observed With Satellite Synthetic Aperture Radar

      Roush, James Joseph (1996)
      Sequential synthetic aperture radar (SAR) images acquired by the First European Remote Sensing Satellite (ERS-1) were employed for observation of the 1993-'94 surge of Bering Glacier, Alaska. Evidence of accelerated motion became visible in late April 1993. Subsequently the surge front propagated down-glacier at a mean velocity of 90 m/day between 19 May and 25 August, reaching most of the 34 km perimeter of the terminus by shortly after 25 August. The calving terminus then advanced rapidly into proglacial Vitus Lake at a maximum rate, during 9 August to 18 October, of 19 m/day in its central area. The propagating surge front consisted of a distributed region of undulations and bulges on the glacier surface having heights, estimated from SAR data, of 40 to 110 m and widths varying from 0.7 to 1.5 km. The measurements were made using terrain-corrected, geocoded and coregistered images. <p>
    • Transport And Formation Processes For Fine Airborne Ash From Three Recent Volcanic Eruptions In Alaska: Implications For Detection Methods And Tracking Models

      Rinkleff, Peter G.; Cahill, Catherine F.; Dehn, Jonathan; Dean, Kenneson G.; Beget, James E. (2012)
      Airborne fine volcanic ash was collected during the eruptions of Augustine Volcano in 2006, Pavlof Volcano in 2007, and Redoubt Volcano in 2009 using Davis Rotating Unit for Measurement (DRUM) cascade impactors to observe atmospheric processes acting on ash as an atmospheric particle. During the Redoubt eruption, samples were also collected by Beta Attenuation Mass (BAM-1020) and Environmental Beta Attenuation Mass (EBAM) monitors. BAM-1020s and EBAMs provided real-time mass concentration data; DRUM samplers provided samples for post-eruptive analysis. DRUM samples were retrospectively analyzed for time-resolved mass concentration and chemistry. EBAM and BAM-1020s reported near real-time, time-resolved mass concentrations. Scanning Electron Microscopy with Energy Dispersive Spectroscopy was conducted to determine particle size, shape, and composition. Image processing methods were developed to determine particle size distributions and shape factors. Ash occurred as single grains, ash aggregates, and hybrid aggregates. Ash aggregates occurred in plumes from pyroclastic flows and were found in a discrete aerodynamic size range (2.5-1.15 microm). Hybrid ash was common in all samples and likely formed when downward mixing ash mingled with upward mixing sea salt and non-sea salt sulfate. The mass concentration of sulfate did not vary systematically with ash which indicated that the source of sulfate was not necessarily volcanic. Ash size distributions were log-normal. Size distribution plots of ash collected from the same plume at different transport distances showed that longer atmospheric residence times allowed for more aggregation to occur which led to larger but fewer particles in the plume the longer it was transported. Ash transport and dispersion models forecasted ash fall over a broad area, but ash fall was only observed in areas unaffected by topographic barriers. PM10 (particulates &le; 10 microm in aerodynamic diameter or OA) ash was detected closer to the volcano when no PM2.5 (particulates &le; 2.5 microm O A) ash was observed. Further downwind, PM2.5 ash was collected which indicated that the settling rates of PM10 and PM2.5 influenced their removal rates. Diurnal variations in ash mass concentrations were controlled by air masses rising due to solar heating which transported ash from the sampling site, or descending due to radiative cooling which brought ash to the sampling site. Respirable (PM2.5) ash was collected when there were no satellite ash detections which underscored the importance of ash transport and dispersion models for forecasting the presence of ash when mass concentrations are below satellite detection limits.
    • Unsupervised multi-scale change detection from SAR imagery for monitoring natural and anthropogenic disasters

      Ajadi, Olaniyi A.; Meyer, Franz; Webley, Peter; Tape, Carl; Cahill, Catherine (2017-08)
      Radar remote sensing can play a critical role in operational monitoring of natural and anthropogenic disasters. Despite its all-weather capabilities, and its high performance in mapping, and monitoring of change, the application of radar remote sensing in operational monitoring activities has been limited. This has largely been due to: (1) the historically high costs associated with obtaining radar data; (2) slow data processing, and delivery procedures; and (3) the limited temporal sampling that was provided by spaceborne radar-based satellites. Recent advances in the capabilities of spaceborne Synthetic Aperture Radar (SAR) sensors have developed an environment that now allows for SAR to make significant contributions to disaster monitoring. New SAR processing strategies that can take full advantage of these new sensor capabilities are currently being developed. Hence, with this PhD dissertation, I aim to: (i) investigate unsupervised change detection techniques that can reliably extract signatures from time series of SAR images, and provide the necessary flexibility for application to a variety of natural, and anthropogenic hazard situations; (ii) investigate effective methods to reduce the effects of speckle and other noise on change detection performance; (iii) automate change detection algorithms using probabilistic Bayesian inferencing; and (iv) ensure that the developed technology is applicable to current, and future SAR sensors to maximize temporal sampling of a hazardous event. This is achieved by developing new algorithms that rely on image amplitude information only, the sole image parameter that is available for every single SAR acquisition. The motivation and implementation of the change detection concept are described in detail in Chapter 3. In the same chapter, I demonstrated the technique's performance using synthetic data as well as a real-data application to map wildfire progression. I applied Radiometric Terrain Correction (RTC) to the data to increase the sampling frequency, while the developed multiscaledriven approach reliably identified changes embedded in largely stationary background scenes. With this technique, I was able to identify the extent of burn scars with high accuracy. I further applied the application of the change detection technology to oil spill mapping. The analysis highlights that the approach described in Chapter 3 can be applied to this drastically different change detection problem with only little modification. While the core of the change detection technique remained unchanged, I made modifications to the pre-processing step to enable change detection from scenes of continuously varying background. I introduced the Lipschitz regularity (LR) transformation as a technique to normalize the typically dynamic ocean surface, facilitating high performance oil spill detection independent of environmental conditions during image acquisition. For instance, I showed that LR processing reduces the sensitivity of change detection performance to variations in surface winds, which is a known limitation in oil spill detection from SAR. Finally, I applied the change detection technique to aufeis flood mapping along the Sagavanirktok River. Due to the complex nature of aufeis flooded areas, I substituted the resolution-preserving speckle filter used in Chapter 3 with curvelet filters. In addition to validating the performance of the change detection results, I also provide evidence of the wealth of information that can be extracted about aufeis flooding events once a time series of change detection information was extracted from SAR imagery. A summary of the developed change detection techniques is conducted and suggested future work is presented in Chapter 6.