<p>This&nbsp / thesis&nbsp / examines&nbsp / the&nbsp / changes&nbsp / that&nbsp / have&nbsp / occurred&nbsp / within&nbsp / West-Coast Renosterveld within&nbsp / the&nbsp / last 350 years, and assesses&nbsp / the viability of&nbsp / the&nbsp / remaining fragments.</p>

Water quality and wetlands represent two vital elements of a healthy coastal ecosystem. Both experienced substantial declines in the U.S. during the 20th century. Overall coastal wetland cover decreased over 50% in the 20th century due to coastal development and water pollution. Management and legislative efforts have successfully addressed some of the problems and threats, but recent research indicates that the diffuse impacts of climate change and non-point source pollution may be the primary drivers of current and future water-quality and wetland stress. In order to respond to these pervasive threats, traditional management approaches need to adopt modern technological tools for more synoptic, frequent and fine-scale monitoring and assessment. In this dissertation, I explored some of the applications possible with new, commercial satellite imagery to better assess the status of coastal ecosystems. Large-scale land-cover change influences the quality of adjacent coastal water. Satellite imagery has been used to derive land-cover maps since the 1960’s. It provides multiple data points with which to evaluate the effects of land-cover change on water quality. The objective of the first chapter of this research was to determine how 40 years of land-cover change in the Tampa Bay watershed (6,500 km2) may have affected turbidity and chlorophyll concentration – two proxies for coastal water quality. Land cover classes were evaluated along with precipitation and wind stress as explanatory variables. Results varied between analyses for the entire estuary and those of segments within the bay. Changes in developed land percent cover best explained the turbidity and chlorophyll-concentration time series for the entire bay (R2 > 0.75, p < 0.02). The paucity of official land-cover maps (i.e. five maps) restricted the temporal resolution of the assessments. Furthermore, most estuaries along the Gulf of Mexico do not have forty years of water-quality time series with which to perform evaluations against land-cover change. Ocean-color satellite imagery was used to derive proxies for coastal water with near-daily satellite observations since 2000. The goal of chapter two was to identify drivers of turbidity variability for 11 National Estuary Program water bodies along the Gulf of Mexico. Land cover assessments could not be used as an explanatory variable because of the low temporal resolution (i.e. approximately one map per five-year period). Ocean color metrics were evaluated against atmospheric, meteorological, and oceanographic variables including precipitation, wind speed, U and V wind vectors, river discharge, and water level over weekly, monthly, seasonal and annual time steps. Climate indices like the North Atlantic Oscillation and El Niño Southern Oscillation index were also examined as possible drivers of long-term changes. Extreme turbidity events were defined by the 90th and 95th percentile observations over each time step. Wind speed, river discharge and El Niño best explained variability in turbidity time-series and extreme events (R2 > 0.2, p < 0.05), but this varied substantially between time steps and estuaries. The background land cover analyses conducted for coastal water quality studies showed that there are substantial discrepancies between the wetland extent estimates mapped by local, state and federal agencies. The third chapter of my research sought to examine these differences and evaluate the accuracy and precision of wetland maps using high spatial-resolution (i.e. two-meter) WorldView-2 satellite imagery. Ground validation data showed that wetlands mapped at two study sites in Tampa Bay were more accurately identified by WorldView-2 than by Landsat imagery (30-meter resolution). When compared to maps produced separately by the National Oceanic and Atmospheric Administration, Southwest Florida Water Management District, and National Wetland Inventory, we found that these historical land cover products overestimated by 2-10 times the actual extent of wetlands as identified in the WorldView-2 maps. We could find no study that had utilized more than six of these commercial images for a given project. Part of the problem is cost of the images, but there is also the cost of processing the images, which is typically done one at a time and with substantial human interaction. Chapter four explains an approach to automate the preprocessing and classification of imagery to detect wetlands within the Tampa Bay watershed (6,500 km2). Software scripts in Python, Matlab and Linux were used to ingest 130 WorldView-2 images and to generate maps that included wetlands, uplands, water, and bare and developed land. These maps proved to be more accurate at identifying forested wetland (78%) than those by NOAA, SWFWMD, and NWI (45-65%) based on ground validation data. Typical processing methods would have required 4-5 months to complete this work, but this protocol completed the 130 images in under 24 hours. Chapter five of the dissertation reviews coastal management case studies that have used satellite technologies. The objective was to illustrate the utility of this technology. The management sectors reviewed included coral reefs, wetlands, water quality, public health, and fisheries and aquaculture.

Despite being an indicator of modernization and macro-economic growth, urbanization in regions such as Sub-Saharan Africa is tightly interwoven with poverty and deprivation. This has manifested physically as slums, which represent the worst residential urban areas, marked by lack of access to good quality housing and basic services. To effectively combat the slum phenomenon, local slum conditions must be captured in quantitative and spatial terms. However, there are significant hurdles to this. Slum detection and mapping requires readily available and reliable data, as well as a proper conceptualization of measurement and scale. Using Bamenda, Cameroon, as a test case, this dissertation research was designed as a three-pronged attack on the slum mapping problematic. The overall goal was to investigate locally optimized slum mapping strategies and methods that utilize high resolution satellite image data, household survey data, simple machine learning and regionalization theory. The first major objective of the study was to tackle a "measurement" problem. The aim was to explore a multi-index approach to measure and map local slum conditions. The rationale behind this was that prior sub-Saharan slum research too often used simplified measurement techniques such as a single unweighted composite index to represent diverse local slum conditions. In this study six household indicators relevant to the United Nations criteria for defining slums were extracted from a 2013 Bamenda household survey data set and aggregated for 63 local statistical areas. The extracted variables were the percent of households having the following attributes: more than two residents per room, non-owner, occupying a single room or studio, having no flush toilet, having no piped water, having no drainage. Hierarchical variable clustering was used as a surrogate for exploratory factor analysis to determine fewer latent slum factors from these six variables. Variable groups were classified such that the most correlated variables fell in the same group while non-correlated variables fell in separate groups. Each group membership was then examined to see if the group suggested a conceptually meaningful slum factor which could quantified as a stand-alone "high" and "low" binary slum index. Results showed that the slum indicators in the study area could be replaced by at least two meaningful and statistically uncorrelated latent factors. One factor reflected the home occupancy conditions (tenancy status, overcrowded and living space conditions) and was quantified using K-means clustering of units as an ‘occupancy disadvantage index’ (Occ_D). The other reflected the state of utilities access (piped water and flush toilet) and was quantified as utilities disadvantage index (UT_D). Location attributes were used to examine/validate both indices. Independent t-tests showed that units with high Occ_D were on average closer to nearest town markets and major roads when compared with units of low Occ_D. This was consistent with theory as it is expected that typical slum residents (in this case overcrowded and non-owner households) will favor accessibility to areas of high economic activity. However, this situation was not the same with UT_D as shown by lack of such as a strong pattern. The second major objective was to tackle a "learning" problem. The purpose was to explore the potential of unsupervised machine learning to detect or "learn" slum conditions from image data. The rationale was that such an approach would be efficient, less reliant on prior knowledge and expertise. A 2012 GeoEye image scene of the study area was subjected to image classification from which the following physical settlement attributes were quantified for each of the 63 statistical areas: per cent roof area, percent open space area, per cent bare soil, per cent paved road surface, per cent dirt road surface, building shadow-roof area ratio. The shadow-roof ratio was an innovative measure used to capture the size and density attributes of buildings. In addition to the 6 image derived variables, the mean slope of each area was calculated from a digital elevation dataset. All 7 attributes were subject to principal component analysis from which the first 2 components were extracted and used for hierarchical clustering of statistical areas to derive physical types. Results show that area units could be optimally classified into 4 physical types labelled generically as Categories 1 – 4, each with at least one defining physical characteristic. Kruskal Wallis tests comparing physical types in terms of household and locations attributes showed that at least two physical types were different in terms of aggregated household slum conditions and location attributes. Category 4 areas, located on steep slopes and having high shadow-to-roof ratio, had the highest distribution of non-owner households. They were also located close to nearest town markets. They were thus the most likely candidates of slums in the city. Category 1 units on other hand located at the outskirts and having abundant open space were least likely to have slum conditions. The third major objective was to tackle the problem of "spatial scale". Neighborhoods, by their very nature of contiguity and homogeneity, represent an ideal scale for urban spatial analysis and mapping. Unfortunately, in most areas, neighborhoods are not objectively defined and slum mapping often relies in the use of arbitrary spatial units which do not capture the true extent of the phenomenon. The objective was thus to explore the use of analytic regionalization to quantitatively derive the neighborhood unit for mapping slums. Analytic neighborhoods were created by spatially constrained clustering of statistical areas using the minimum spanning tree algorithm. Unlike previous studies that relied on socio-economic and/or demographic information, this study innovatively used multiple land cover and terrain attributes as neighborhood homogenizing factors. Five analytic neighborhoods (labeled Regions 1-5) were created this way and compared using Kruskal Wallis tests for differences in household slum attributes. This was to determine largest possible contiguous areas that could be labeled as slum or non-slum neighborhoods. The results revealed that at least two analytic regions were significantly different in terms of aggregated household indicators. Region 1 stood apart as having significantly higher distributions of overcrowded and non-owner households. It could thus be viewed as the largest potential slum neighborhood in the city. In contrast, regions 3 (located at higher elevation and separated from rest of city by a steep escarpment) was generally associated with low distribution of household slum attributes and could be considered the strongest model of a non-slum or formal neighborhood. Both Regions 1 and 3 were also qualitatively correlated with two locally recognized (vernacular) neighborhoods. These neighborhoods, "Sisia" (for Region 1) and "Up Station" (for Region 3), are commonly perceived by local folk as occupying opposite ends of the socio-economic spectrum. The results obtained by successfully carrying the three major objectives have major implication for future research and policy. In the case of multi-index analysis of slum conditions, it affirms the notion the that slum phenomenon is diverse in the local context and that remediation efforts must be compartmentalized to be effective. The results of image based unsupervised mapping of slums from imagery show that it is a tool with high potential for rapid slum assessment even when there is no supporting field data. Finally, the results of analytic regionalization showed that the true extent of contiguous slum neighborhoods can be delineated objectively using land cover and terrain attributes. It thus presents an opportunity for local planning and policy actors to consider redesigning the city neighborhood districts as analytic units. Quantitively derived neighborhoods are likely to be more useful in the long term, be it for spatial sampling, mapping or planning purposes.

Philosophiae Doctor - PhD / South Africa

Deep learning has proven a powerful tool for image analysis during the past two decades. With the rise of high resolution overhead imagery, an opportunity for automatic geospatial 3D-recreation has presented itself. This master thesis researches the possibil- ity of 3D-recreation through deep learning based image analysis of overhead imagery. The goal is a model capable of making predictions for three different tasks: heightmaps, bound- ary proximity heatmaps and semantic segmentations. A new neural network is designed with the novel feature of supplying the predictions from one task to another with the goal of improving performance. A number of strategies to ensure the model generalizes to un- seen data are employed. The model is trained using satellite and aerial imagery from a variety of cities on the planet. The model is meticulously evaluated by using four common performance metrics. For datasets with no ground truth data, the results were assessed visually. This thesis concludes that it is possible to create a deep learning network capa- ble of making predictions for the three tasks with varying success, performing best for heightmaps and worst for semantic segmentation. It was observed that supplying estima- tions from one task to another can both improve and decrease performance. Analysis into what features in an image is important for the three tasks was clear in some images, unclear in others. Lastly, validation proved that a number of random transformations during the training process helped the model generalize to unseen data. / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

Change detection (CD) has, in recent years, shown promising results in remote sensing (RS). The development of deep learning CD (DLCD) has, in even more recent years, taken change detection to another level and it has become more widely researched. However, the research depends on publicly available datasets that have been manually annotated for the task of CD. This method is cumbersome and the resulting datasets do not often include all types of change. In this thesis, the generalizability to different areas and different change types of a model trained on a widely used public dataset is analyzed. Also, the thesis investigates how 3D information from Maxar Technologies 3D models can be used to automatically create new more general datasets for CD with both binary or non-binary outputs. The access to large amounts of satellite images together with 3D information enables the creation of more general datasets that can capture more types of change.The thesis concludes that a model trained on the publicly available dataset does not generalize to other areas or other types of change. Models trained on the automatically generated datasets yield relatively good results which indicates that using 3D information to automatically create large datasets is a valid method for CD. Even non-binary approaches show promising results which enable using to gain more practical information on the change of an area. While the thesis presents encouraging results, work can definitely be done to further improve the generalization of the models and improve the dataset generation.

Analysis of overhead imagery has a great potential to produce real-time data cost-effectively. This can be an important foundation for decision-making for businesses and politics. Every day a massive amount of new satellite imagery is produced. To fully take advantage of these data volumes a computationally efficient pipeline is required for the analysis. This thesis proposes a pipeline which outperforms the Segment Before you Detect network [6] and different types of fast region based convolutional neural networks [61] with a large margin in a fraction of the time. The model obtains a prediction error for counting cars of 1.67% on the Potsdam dataset and increases the vehiclewise F1 score on the VEDAI dataset from 0.305 reported by [61] to 0.542. This thesis also shows that it is possible to outperform the Segment Before you Detect network in less than 1% of the time on car counting and vehicle detection while also using less than half of the resolution. This makes the proposed model a viable solution for large-scale satellite imagery analysis. / Analys av flyg- och satellitbilder har stor potential att kostnadseffektivt producera data i realtid för beslutsfattande för företag och politik. Varje dag produceras massiva mängder nya satellitbilder. För att fullt kunna utnyttja dessa datamängder krävs ett beräkningseffektivt nätverk för analysen. Denna avhandling föreslår ett nätverk som överträffar Segment Before you Detect-nätverket [6] och olika typer av snabbt regionsbaserade faltningsnätverk [61]  med en stor marginal på en bråkdel av tiden. Den föreslagna modellen erhåller ett prediktionsfel för att räkna bilar på 1,67% på Potsdam-datasetet och ökar F1- poängen for fordons detektion på VEDAI-datasetet från 0.305 rapporterat av [61]  till 0.542. Denna avhandling visar också att det är möjligt att överträffa Segment Before you Detect-nätverket på mindre än 1% av tiden på bilräkning och fordonsdetektering samtidigt som den föreslagna modellen använder mindre än hälften av upplösningen. Detta gör den föreslagna modellen till en attraktiv lösning för storskalig satellitbildanalys.

This thesis investigates the seasonal predictive capabilities of Neural Radiance Fields (NeRF) applied to satellite images. Focusing on the utilization of satellite data, the study explores how Sat-NeRF, a novel approach in computer vision, per- forms in predicting seasonal variations across different months. Through compre- hensive analysis and visualization, the study examines the model’s ability to cap- ture and predict seasonal changes, highlighting specific challenges and strengths. Results showcase the impact of the sun on predictions, revealing nuanced details in seasonal transitions, such as snow cover, color accuracy, and texture represen- tation in different landscapes. The research introduces modifications to the Sat- NeRF network. The implemented versions of the network include geometrically rendered shadows, a signed distance function, and a month embedding vector, where the last version mentioned resulted in Planet-NeRF. Comparative evalua- tions reveal that Planet-NeRF outperforms prior models, particularly in refining seasonal predictions. This advancement contributes to the field by presenting a more effective approach for seasonal representation in satellite imagery analysis, offering promising avenues for future research in this domain.

This Master’s thesis investigates the application of dynamically generated procedural terrain textures for texturing 3D representations of the Earth’s surface. The study explores techniques to overcome limitations of the currently most common method – projecting satellite imagery onto the mesh – such as insufficient resolution for close-up views and challenges in accommodating external lighting models. Textures for sand, rock and grass were generated procedurally on the GPU. Aliasing was prevented using a clamping technique, dynamically changing the level of detail when freely navigating across diverse landscapes. The general color of each terrain type was extracted from the satellite images, guided by land cover rasters, in a process where shadows were eliminated using HSV color space conversion and filtering. The procedurally generated textures provide significantly more details than the satellite images in close-up views, while missing some information in medium- to far-distance views, due to the satellite images containing information lacking in the 3D mesh. A qualitative analysis spanning six data sets from diverse global locations demonstrates that the proposed methods are applicable across a range of landscapes and climates.

This study aims to conduct an assessment of the land cover change of the Mississippi and Alabama coastal region, an integral part of the Gulf Coast ecological makeup. Landsat satellite data were used to perform a supervised classification using the imagery captured by Landsat sensors including Landsat 1-2 Multispectral Scanner (MSS), Landsat 4-5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper (ETM+), and Landsat 8 Operational Land Imager (OLI) from 1973 to 2015. The objective of this study is to build a long-term assessment of urban development and land cover change over the past four decades for the Alabama and Mississippi Gulf Coast and to characterize these changes using Landscape Metrics (LM). The findings of this study indicate that the urban land cover doubled in size between 1973 and 2015. This expansion was accompanied by a high degree of urban fragmentation during the first half of the study period and then a gradual leveling off. Local, state, and federal authorities can use the results of this study to build mitigation plans, coastal development planning, and serve as the primary evaluation of the current urban development for city planners, environmental advocates, and community leaders to reduce degradation for this environmentally sensitive coastal region.