Arctic air temperatures have increased at two to three times the global rate over the past century. As a result, abiotic and biotic responses to climate change are more rapid and pronounced in the Arctic compared to other biomes. One important change detected over the past several decades by satellite studies is a lengthening of the arctic growing season, which is due to earlier onsets and/or delayed ends to growing seasons. A handful of studies also suggest the peak green season (i.e. when the tundra is at maximum leaf-out and maximum carbon uptake potential) is starting earlier in the arctic tundra. The vast majority of studies detecting shifts in the growing season suggest this is due to increasing spring and fall air temperatures, which lead to earlier spring snowmelt and later fall snowfall. Less well understood is how indirect consequences of arctic warming, such as ongoing changes in plant community composition, may also be contributing to these satellite signals. For instance, there is mounting evidence that deciduous shrubs are expanding into previously non-shrub dominated tundra in several parts of the Arctic. Deciduous shrubs may alter tundra canopy phenology and contribute to the regional shifts in timing of phenological events being detected by satellites. Concurrently, in many areas where deciduous shrubs are expanding they are also becoming taller. As taller shrubs become increasingly dominant, arctic landscapes may retain more snow, which could lengthen spring snow cover duration, and offset advances in the start of the growing season that are expected as a result of earlier spring snowmelt. As a consequence, deeper snow and later snowmelt in taller shrub tundra could delay plant emergence, and shorten the period of annual carbon uptake. Thus greater dominance of taller stature deciduous shrubs in the Arctic may actually delay the onset of the growing season, which would suggest that increasing deciduous shrub dominance may not be contributing to satellite signals of an earlier start to the growing season. To contribute to satellite-detected shifts in the onset of the growing and peak seasons, tall deciduous shrubs would need to have accelerated leaf development to compensate for deeper snow packs and later spring snowmelt relative to surrounding tundra. Understanding the drivers of shifts in tundra phenology is important since longer (or shorter) growing and peak green seasons would increase (or decrease) productivity and the period of carbon uptake, which will have implications for landscape-level carbon exchange, and ultimately global carbon balances. Given the rate and magnitude of changes occurring in the face of acute arctic warming, there is a need to monitor, understand, and predict ecological responses over large spatial and temporal scales. However, compared to more southern environments, the arctic tundra is characterized by considerable heterogeneity in vegetation distribution, as well as a short and rapid growing season. In addition, the arctic tundra is relatively vast and inaccessible. These characteristics can make it difficult to monitor and study changes in the Arctic, and make it difficult to develop landscape-level models able to predict changes in ecosystem dynamics and tundra vegetation. The use of airborne and satellite sensors has at least partially fulfilled these needs to monitor, understand, and predict change in the Arctic. The normalized difference vegetation index (NDVI) acquired from these sensors, for instance, has become a widely adopted tool for detecting and quantifying spatial-temporal dynamics in tundra vegetation cover, productivity, and phenology. This suggests that remote sensing technology and vegetation indices may be similarly applied to characterizing patterns of primary and secondary consumers (e.g. arthropods), which would be enormously useful in a region as vast and remote as the Arctic. The research presented in this dissertation provides useful insight into the influence vegetation community composition, particularly increasing deciduous shrub dominance, has on phenology, carbon flux, and canopy arthropod biomass in the arctic foothills region of the Brooks Range, Alaska. Findings in Chapter one suggest that delayed snowmelt in areas dominated by taller shrubs may have a short-lived impact on the timing of leaf development, likely resulting in no difference in duration of peak photosynthetic period between tall and short- stature shrubs. Findings in Chapter two suggest that greater deciduous shrub dominance not only increases carbon uptake due to higher leaf area relative to surrounding tundra, but may also be causing an earlier onset of, and ultimately a net extension of, the period of maximum tundra greenness and further increasing peak season carbon sequestration. Findings in Chapter three suggest that measurements of the NDVI made from air and spaceborne sensors may be able to quantify spatial and temporal variation in canopy arthropod biomass at landscape to regional scales in the arctic tundra.

abstract: The Western Continental United States has a rapidly changing and complex ecosystem that provides valuable resources to a large portion of the nation. Changes in social and environmental factors have been observed to be significantly correlated to usable ground and surface water levels. The assessment of water level changes and their influences on a semi-national level is needed to support planning and decision making for water resource management at local levels. Although many studies have been done in Ground and Surface Water (GSW) trend analysis, very few have attempted determine correlations with other factors. The number of studies done on correlation factors at a semi-national scale and near decadal temporal scale is even fewer. In this study, freshwater resources in GSW changes from 2004 to 2017 were quantified and used to determine if and how environmental and social variables are related to GSW changes using publicly available remotely sensed and census data. Results indicate that mean annual changes of GSW of the study period are significantly correlated with LULC changes related to deforestation, urbanization, environmental trends, as well as social variables. Further analysis indicates a strong correlation in the rate of change of GSW to LULC changes related to deforestation, environmental trends, as well as social variables. GSW slope trend analysis also reveals a negative trend in California, New Mexico, Arizona, and Nevada. Whereas a positive GSW trend is evident in the northeast part of the study area. GSW trends were found to be somewhat consistent in the states of Utah, Idaho, and Colorado, implying that there was no GSW changes over time in these states. / Dissertation/Thesis / Masters Thesis Geography 2018

abstract: Big data that contain geo-referenced attributes have significantly reformed the way that I process and analyze geospatial data. Compared with the expected benefits received in the data-rich environment, more data have not always contributed to more accurate analysis. “Big but valueless” has becoming a critical concern to the community of GIScience and data-driven geography. As a highly-utilized function of GeoAI technique, deep learning models designed for processing geospatial data integrate powerful computing hardware and deep neural networks into various dimensions of geography to effectively discover the representation of data. However, limitations of these deep learning models have also been reported when People may have to spend much time on preparing training data for implementing a deep learning model. The objective of this dissertation research is to promote state-of-the-art deep learning models in discovering the representation, value and hidden knowledge of GIS and remote sensing data, through three research approaches. The first methodological framework aims to unify varied shadow into limited number of patterns, with the convolutional neural network (CNNs)-powered shape classification, multifarious shadow shapes with a limited number of representative shadow patterns for efficient shadow-based building height estimation. The second research focus integrates semantic analysis into a framework of various state-of-the-art CNNs to support human-level understanding of map content. The final research approach of this dissertation focuses on normalizing geospatial domain knowledge to promote the transferability of a CNN’s model to land-use/land-cover classification. This research reports a method designed to discover detailed land-use/land-cover types that might be challenging for a state-of-the-art CNN’s model that previously performed well on land-cover classification only. / Dissertation/Thesis / Doctoral Dissertation Geography 2019

Thesis (M.Sc. (Soil Science)) --University of Limpopo, 2001 / Refer to document

Ground-level and overhead images provide complementary viewpoints of the world. This thesis proposes methods which leverage dense overhead imagery, in addition to sparsely distributed ground-level imagery, to advance traditional computer vision problems, such as ground-level image localization and fine-grained urban mapping. Our work focuses on three primary research areas: learning a joint feature representation between ground-level and overhead imagery to enable direct comparison for the task of image geolocalization, incorporating unlabeled overhead images by inferring labels from nearby ground-level images to improve image-driven mapping, and fusing ground-level imagery with overhead imagery to enhance understanding. The ultimate contribution of this thesis is a general framework for estimating geospatial functions, such as land cover or land use, which integrates visual evidence from both ground-level and overhead image viewpoints.

"May 2004" / Bibliography: leaves 243-275. / xvi, 275, [24] leaves : ill., maps (some col.), plates (col.) ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Looks at land use/land cover change detection, identification, analysis and prediction using remote sensing and GIS techniques in the downstream are of the Ci Tarum watershed and its surroundings in West Java, Indonesia. Supervised Maximum Likelihood classification of PCS and NDVI transformed images are used to classify and identify land use/land cover categories. A post-classification comparison approach was used to detect land use/land cover changes, and a Markov Cellular automata model is used to predict possible future land use/land cover patterns in the study area. / Thesis (Ph.D.)--University of Adelaide, School of Social Sciences, Discipline of Geographical and Environmental Studies, 2004

Satellite technology represents the only technique for measuring sea surface temperatures (SSTs) on a global scale. SSTs are important as boundary conditions for climate and atmospheric boundary layer models which attempt to describe phenomena of all scales, ranging from local forecasts to predictions of global warming. Historical use of infrared satellite measurements for SST determination has been based on a theory which assumes that the atmosphere is 'thin', i.e., that atmospheric absorption of infrared radiation emitted from the sea surface has very little effect on the radiant intensity that is measured by satellites. However, a variety of independent radiative transfer models point to the possibility that the so-called 'thin approximation' is violated for humid atmospheres such as those found in the tropics, leading to errors in the retrieved SST that would be unacceptable to those who make use of such products. Furthermore, such tropical regions represent a significant portion of the globe, where coupled ocean-atmosphere disturbances can have global effects (e.g., the tropical Pacific El Nino-Southern Oscillation events). This study evaluates the thin approximation empirically, by combining radiative transfer theory and satellite data from the Eastern Atlantic ocean region studied during the Atlantic Statocumulus Transition Experiment (ASTEX). Six months of satellite data from May, June, and July of 1983 and 1984 are analyzed. To the degree that the data may be considered representative of globally valid relationships between measured variables, it is shown that the thin approximation is not appropriate for the tropics. This suggests that new methods are necessary for retrieving SSTs from the more humid regions of the globe. / Graduation date: 1993

We map the regional physiography and surficial lithology (Surficial Geologic Habitat or SGH) over the continental margin of Oregon. This thesis develops, describes, and implements an iterative interpretive method to map seafloor habitat types from disparate geological and geophysical datasets including: bathymetric images, sidescan sonar images, seismic reflection profiles, sediment samples, geologic maps of structure, and observations from submersibles. An indirect technique for the assessment of map accuracy or habitat type misidentification error is also explored and used to derive supplemental maps of varying interpretative confidence, or "quality". The geological and geophysical datasets used to produce the SGH maps of the Oregon margin are by their nature patchy, and form an irregular mosaic of variable data density and quality. Uniform sampling of continental margins does not yet exist, thus these maps are an attempt to glean as much information as possible from the framework of existing data. In any given area the quantity and quality of data available varied considerably, and required a flexible method of interpretation based on this availability. The integrated interpretative GIS techniques are developed to facilitate mapping geologic habitat types over this region of discontinuous and patchy seafloor data. The SGH map and thematic map accuracy assessment support improved habitat-based inventory and assessment methods. They also serve as habitat reference materials for marine resources management and planning activities at local to national scales. SGH and data quality maps are incorporated as thematic layers within a broader habitat geodatabase for west coast groundfish and are directly applied for modeling Essential Fish Habitat (EFH) for these species. / Graduation date: 2004

Many ecological systems follow a seasonal cycle affecting primary production, carbon flux, and vegetative gas emissions. The seasonal variation of ecological systems are both affected by and have effects upon climatic factors. A quantitative estimate of the seasonal variation of vegetation is required to characterize ecological systems and their interaction with climate. Monitoring the spaliotemporal variation of foliar biomass density (FBD) over one year will provide a quantitative estimate of the annual cycle and regional variation of photosynthetic activity. FBD is a quantitative measure of leafy material per unit of area produ\:ed by photosynthetically active vegetation. This seasonal variation in FBD is an important parameter for global and other large scale investigations of ecological, hydrological, and biogeochemical systems which require data and expertise from a variety of sources and disciplines. Therefore, FBD is potentially of great utility for ecologists, hydrologists, climatologists, and atmospheric scientists. Recent regional scale investigations of ecological systems concluded that the repetitive coverage and synoptic view of remotely sensed measurements provide data to monitor the seasonal variation of biomass. A method to estimate the seasonal variation of FBD at global scales has not been developed. The objective of this research is to develop a methodology that could be used to estimate the seasonal variation of FBD for the entire terrestrial biosphere. By coupling global satellite data, measured field data, and a vegetation classification, a model was developed to estimate the global spatiotemporal variation of FBD. Comparisons between literature estimates of FBD and estimated FBD from this model were made as a means of validation. A more specific comparison was conducted between grasslands based on work conducted in the Senegalese Sahel region in Africa. Finally, a sensitivity analysis was performed to characterize the potential propagation of error associated with the literature FBD estimates used to drive this model. / Graduation date: 1992

This research has involved the conduct of a series of laboratory measurements of the centimeter-wavelength opacity of water vapor along with the development of a hybrid radiative transfer ray-tracing simulator for the atmosphere of Jupiter which employs a model for water vapor opacity derived from the measurements. For this study an existing Georgia Tech high-sensitivity microwave measurement system (Hanley and Steffes , 2007) has been adapted for pressures ranging from 12-100 bars, and a corresponding temperature range of 293-525°K. Water vapor is measured in a mixture of hydrogen and helium. Using these measurements which covered a wavelength range of 6--20 cm, a new model is developed for water vapor absorption under Jovian conditions. In conjunction with our laboratory measurements, and the development of a new model for water vapor absorption, we conduct sensitivity studies of water vapor microwave emission in the Jovian atmosphere using a hybrid radiative transfer ray-tracing simulator. The approach has been used previously for Saturn (Hoffman, 2001), and Venus (Jenkins et al., 2001). This model has been adapted to include the antenna patterns typical of the NASA Juno Mission microwave radiometer (NASA/Juno -MWR) along with Jupiter's geometric parameters (oblateness), and atmospheric conditions. Using this adapted model we perform rigorous sensitivity tests for water vapor in the Jovian atmosphere. This work will directly improve our understanding of microwave absorption by atmospheric water vapor at Jupiter, and improve retrievals from the Juno microwave radiometer. Indirectly, this work will help to refine models for the formation of Jupiter and the entire solar system through an improved understanding of the planet-wide abundance of water vapor which will result from the successful opreation of the Juno Microwave Radiometer (Juno-MWR).