A remote sensing system for measuring ocean surface currents is presented in this thesis. Included are a review of the Stepped Frequency Delta K (SFDK) radar system hardware, a description of the system performance, experimental results of ocean current measurements, and comparisons of the measurements with a conventional current probe and a wind driven current model. Important features of this radar system are (1) its capability to perform real time processing of the collected data and (2) its frequency agility, which provides significant improvement of the system's signal-to-clutter ratio. The real time processing capabilities were essential to the collection and monitoring of the ocean currents for long periods of time that is necessary to understand and interpret the results. The SFDK radar participated in two month long experiments. During the first experiment at N. Truro, MA, the SFDK radar demonstrated its ability to make precise phase velocity measurements over long periods of time. During the second experiment on the Chesapeake Light Tower, the radar was able to sense both tidal and wind driven current components. Comparison of radar data and in situ measurements during the CLT experiment shows that the instrument possesses a unique capability to measure near surface currents--not possible with in situ probes.

Digital satellite remote sensing and Geographic Information Systems (GIS) have been used in conjunction with the Universal Soil Loss Equation (USLE) to model soil erosion potential within watersheds. This study compared erosion estimates calculated by the remote sensing method to results obtained in the field by soil conservationists using conventional methods.

Ground truthing actual crop types in an area can be expensive and time-consuming. The California Department of Water Resources attempts to ground truth land use in each county in California every five years. However, this is limited by budgetary constraints and often results in infrequent (more than every ten years) surveying of many counties. An accurate accounting of crops growing in a region is important for a variety of purposes including farm production estimates, groundwater and surface water modeling, evapotranspiration estimation, water planning, research applications, etc. Agricultural land use is continually changing due to development and environmental factors. Currently, USDA NASS provides georeferenced land use maps of regions throughout the U.S. While these are beneficial, the accuracy is not very high for California due to the wide variety of crops grown throughout the state. California has an increasingly complex agricultural system which includes multi-crops changing on an annual and even semiannual basis, long growing seasons, and complex and flexible irrigation schedules. Remotely sensed data from available satellites are used to more accurately classify crop types within the Madera and Merced Counties of California’s Central Valley. An initial classification approach utilizing a simplified decision tree for a data subset of the area considered is presented. In order to accommodate the larger dataset at hand, a computer based approach is applied using the Nearest Neighbor classification algorithm in the computer program eCognition. Iterative analyses were performed to consider a range of scenarios with varying spectral inputs. The results show the methods presented can be beneficial in discriminating 24 of the major crop types from multi-temporal spectral data.

Tidewater glaciers drain a significant proportion of the Greenland Ice Sheet and ice masses of the Canadian Arctic and provide the primary source of icebergs in Canadian waters. However, there remains uncertainty surrounding the processes controlling ice discharge from Canadian Arctic glaciers, the drift paths of icebergs in Canadian waters, and the proximity of icebergs to shipping in the region. This thesis quantifies the processes controlling glacier dynamics from four primary glacier basins on the Prince of Wales (POW) Icefield and using a multi-year dataset of iceberg drift tracks, identifies drift patterns and proximity to ships throughout the eastern Canadian Arctic. On the POW Icefield between 2009 and 2019, Cadogan and Ekblaw glaciers underwent multiyear acceleration and deceleration limited to their lower parts, consistent with characteristics of "pulse-type" glaciers. Trinity and Wykeham glaciers underwent repeating multiyear periods of velocity acceleration between 2009 and 2019 which coincided with significant thinning at their termini. As of 2017, Trinity and Wykeham were each within ~10 m of flotation over their lowermost 4 km. These findings suggest that Trinity and Wykeham glaciers have transitioned to a flow type dominated by dynamic thinning, which is strongly influenced by subglacial topography and may be susceptible to instability of the glacier front and large-scale collapse. Given that both glaciers are grounded below sea level for ~40 km up-glacier from their termini, this process could lead to significant increases in acceleration, retreat, and solid ice discharge. Using a multi-year dataset (2011-2019) of in-situ iceberg drift locations, it was found that icebergs consistently drifted southeast along the east coast of Baffin Island, controlled by a combination of local conditions including short-term wind events, ocean surface currents and semi-diurnal tidal oscillations. A test of the assumption that icebergs drift at 2% of the wind speed indicates that this rule does not apply for the majority of icebergs in this study, which typically exceeded 2% of the wind speed, particularly at low values. The highest median iceberg drift speeds occurred during the winter and spring, reaching up to 2.3 m s⁻¹ in Nares Strait. Icebergs in this study commonly became grounded near eastern Coburg Island and along the SE coast of Baffin Island, where mean residence time exceeded 180 days in all seasons. Through an analysis of a comprehensive database of ship tracks derived from AIS (automatic identification system) data in combination with a subset of iceberg drift locations derived from in-situ satellite trackers and the Canadian Ice Island Drift, Deterioration, and Detection Database (CI2D3), areas of iceberg-ship coexistence throughout Baffin Bay were identified between 2012 and 2019. The regions that saw the largest increases in iceberg-ship coexistence were along the east coast of Baffin Island and east of Bylot Island for dry bulk vessels, and northward into Smith Sound for passenger vessels. As passenger vessels commonly have little ice strengthening, this could pose an elevated hazard to vessels operating in these regions. The results of this study provide a comprehensive examination of the factors controlling glacier terminus dynamics and stability on SE Ellesmere Island, and the drift paths of icebergs once calved. This provides insights into the life cycle of icebergs in Canadian waters, how they may change in a warming climate, and the hazards that they may pose for shipping, particularly given the rapid recent increase in ship transits across the Canadian Arctic.

Long range path planning for an autonomous ground vehicle with minimal a-priori data is still very much an open problem. Previous research has demonstrated that least cost paths generated from aerial LIDAR and GIS data could play a role in automatically determining suitable routes over otherwise unknown terrain. However, most of this research has been theoretical. Consequently, there is very little literature the effectiveness of these techniques in plotting paths of an actual autonomous vehicle. This research aims to develop an algorithm for using aerial LIDAR and imagery to plan paths for a full size autonomous car. Methods of identifying obstacles and potential roadways from the aerial LIDAR and imagery are reviewed. A scheme for integrating the path planning algorithms into the autonomous vehicle existing systems was developed and eight paths were generated and driven by an autonomous vehicle. The paths were then analyzed for their drivability and the model itself was validated against the vehicle measurements. The methods described were found to be suitable for generating paths both on and off road. / Master of Science

The relationship between microwave radar and optical video imaging of the nearshore region is studied. The remotely sensed data were used to estimate the longshore currents and the surf zone width. Doppler radar relies on small scale surface roughness that scatters the incident electromagnetic radiation so that velocities are obtained from the Doppler shift of the backscattered radiation. Video relies on texture and contrast of scattered sunlight from the sea surface, and velocity estimates are determined using Particle Imaging Velocimetry (PIV). This study compares video PIV-derived and Doppler radar surface velocities over a 1 km alongshore by 0.5 km cross-shore area in the surf zone of a natural beach. The two surface velocity estimates are strongly correlated (R2 ≥ 0:79) over much of the surf zone. Estimates differ at the outer edge of the surf where strong breaking is prevalent, with radar estimated velocities as much as 50% below the video estimates. Both systems observe a strong eddy-like mean flow pattern over 200 m section of coastline with the mean alongshore current changing direction at about the mid surf zone. The radar and PIV velocities at particular locations in the surf zone track each other well over a 6 hour period, showing strong modulations in the mean alongshore flow occurring on 10-20 minute time intervals. The offshore region in the absence of sufficiently strong wind is sometimes barely visible, while the surf zone always appears very bright in radar backscatter images due to persistent surface roughness produced by breaking waves. The backscatter and coherence radar images were used in conjunction with edge detection filters to estimate the surf zone width from radar data. The surf zone width from video data is calculated using the time-stacking techniques. The comparison of surf zone width over 6 hours showed the rms difference of 8.8 m close to the radar location while the radar had the tendency to overestimate the distance for most of the run. The correlation of two measurements was high at 0.89. At locations farther than 600 m away from the radar the surf zone width rms differences were higher, up to 24 m, while correlation remained high. The differences are attributed to the estimate of the shoreline in radar images due to different scattering properties of wet and dry sand. The good spatial and temporal agreement between the two remote measurement techniques which rely on very different mechanisms, suggests that both are reasonably approximating the nearshore processes.

Radar sounders (RS) are active instruments that have proved to be able to profile the subsurface of planetary bodies. The design of RS instruments, as well as the interpretation of the acquired data, is a non-trivial task due to the complexity of the scenario of acquisition and the limited amount of information on the targets (especially in planetary exploration). In this context, data simulations are necessary to support the design of the radar, the development of the related processing chain, and the definition of algorithms for the automatic analysis of data. However, state-of-the-art RS simulation methods are characterized by different trade-offs between simulation accuracy and computational costs. On the one hand, numerical methods, such as the Finite-Difference Time-Domain (FDTD) technique, allow to accurately model the wave-target interaction by exactly solving Maxwell's equations at the cost of very high computational requirements. On the other hand, optical methods, such as the ray-tracing based Multi-layer Coherent Simulator (MCS), rely on approximated solution of Maxwell's equations that allow for a better usage of computational resources at the cost of a less accurate modeling. Moreover, simulators produce raw or range-compressed only data, making it difficult to interpret and analyze them via direct comparison with the real data, which are typically processed also for azimuth compression. In this thesis, we present four main contributions related to the simulation of RS data to address the above-mentioned limitations. The first and second contributions thus present 3D simulations of selected targets of two new RS instruments, i.e., the Radar for Icy Moon Exploration (RIME) and the EnVision Subsurface Radar Sounder (SRS). The simulations are performed with the FDTD and MCS simulators. Despite producing good results in terms of detection probability of the selected targets, these two contributions highlight the above-mentioned gaps in the literature of simulation of RS data. The first main limitation is the lack of methods that can accurately model both large and small-scale scattering phenomena at relatively low computational costs. This problem is addressed by the third contribution of this thesis, which presents a novel integrated simulation technique that models both large and small-scale surface scattering phenomena by combining the advantages of the FDTD and MCS techniques, in an accurate and computationally efficient way. The second problem identified is the lack of SAR processing techniques to be applied to the simulated radargrams. This is addressed in the fourth contribution which presents a range-Doppler method for focusing raw radar sounder data simulated with 3D coherent electromagnetic simulators. The method is general and can be applied to any electromagnetic simulator, and is demonstrated for both the FDTD and MCS methods. The results presented throughout the thesis indicate that the proposed methods advance the state-of-the-art of techniques for both generating and processing simulated RS data.

Globally, species are experiencing geographical range shifts as a result of increased frequency and severity of extreme weather events exceeding their realized thermal niche boundaries. Using thermal limit approximations, relative heat indices can predict species extinction-colonization patterns over broad spatial scales. Locally, microclimate refugia can act as buffers against short term thermal extremes and improve species persistence probabilities. Opportunities to explore the role of microclimates in local species extinctions have recently emerged with advances in unmanned aerial vehicle (UAV) and thermal imaging technologies. My first chapter proposed a UAV-based methodology facilitating direct and accurate air temperature measurements at biologically relevant scales for butterfly species. These high-resolution microclimate measurements enabled broad-scale thermal limit approximation model applications to patch-level measurements using a verified thermal positioning index. In my second chapter, I evaluated the applicability of broad-scale models for predictions of local species distributions and abundances. The methodology proposed in Chapter 1 was used to generate patch-specific thermal position indices for butterfly species observed and surveyed in our study patches. Patch-level measurements of thermally tolerable area (overheating index) helped predict aspects of butterfly abundance, presence, and overall species richness, along with other environmental metrics that are relevant for butterfly biology. This thesis explores a frontier of direct UAV-based microclimate measurements and underscores the importance of considering thermal extremes to understand butterfly distribution and abundance, even in protected habitats.

No description available.

<p> This thesis provides a means of measuring the type and number of livestock on a given farm in southern Ontario using as a data base aerial photographs of medium scale. The feasibility of making such measurements is shown to be an extension of past work in the field of agricultural air photo interpretation. The methodology is presented in the form of a model. The inputs from aerial photographs are crop acreages, building type and dimensions, and silo sizes. Average yields in each study area, in combination with crop acreage gives feed available. A comparison is made between feed weight and housing space available and required feed weights and stabling facilities (taken as constants in southern Ontario) to support cattle of a given type. This comparison technique, within the model, yields an output of farm type and actual numbers of cattle. The accuracy of prediction so obtained is high and is independent of the location or attributes of the widely spaced sample areas.</p> / Thesis / Master of Arts (MA)