<p>The Okavango river flows from southern Angola, through the Kavango region of Namibia and into the Okavango Delta in Botswana. The recent peace in Angola hopefully marks the end of the intense suffering that the peoples of the river basin have endured, and the beginning of sustainable decision-making in the area. Informed decision-making however requires knowledge; and there is a need for, and a lack of knowledge regarding basin-wide land cover (LC) changes, and their causes, during the Angolan civil war in the basin. Furthermore, there is a need for, and a lack of knowledge on how expanding large-scale agriculture and urban growth along the Angola-Namibia border affects the water quality of the river.</p><p>The aim of this study was therefore to develop a remote sensing method applicable to the basin (with scant ground-truth data availability) to carry out a systematic historic study of LC changes during the Angolan civil war, to apply the method to the basin, to relate these changes to major societal trends in the region, and to analyse potential impacts of expanding large-scale agriculture and urban growth on the water quality of the river along the Angola-Namibia border.</p><p>A range of remote sensing methods to study historic LC changes in the basin were tried and evaluated against reference data collected during a field visit in Namibia in October 2005. Eventually, two methods were selected and applied to pre-processed Landsat MSS and ETM+ satellite image mosaics of 1973 and 2001 respectively: 1. a combined unsupervised classification and pattern-recognition change detection method providing quantified and geographically distributed binary LC class change trajectory information and, 2. an NDVI (Normalised Difference Vegetation Index) change detection method providing quantified and geographically distributed continuous information on degrees of change in vegetation vigour. In addition, available documents and people initiated in the basin conditions were consulted in the pursuit of discerning major societal trends that the basin had undergone during the Angolan civil war. Finally, concentrations of nutrients (total phosphorous & total nitrogen), bacteria (faecal coliforms & faecal streptococci), conductivity, total dissolved solids, dissolved oxygen, pH, temperature and Secchi depth were sampled at 11 locations upstream and downstream of large-scale agricultural facilities and an urban area during the aforementioned field visit.</p><p>The nature, extent and geographical distribution of LC changes in the study area during the Angolan civil war were determined. The study area (150 922 km²) was the Angolan and Namibian parts of the basin. The results indicate that the vegetation vigour is dynamic and has decreased overall in the area, perhaps connected with precipitation differences between the years. However while the vigour decreased in the northwest, it increased in the northeast, and on more local scales the pattern was often more complex. With respect to migration out of Angola into Namibia, the LC changes followed expectations of more intense use in Namibia close to the border (0-5 km), but not at some distance (10-20 km), particularly east of Rundu. With respect to urbanisation, expectations of increased human impact locally were observed in e.g. Rundu, Menongue and Cuito Cuanavale. Road deterioration was also observed with Angolan urbanisation but some infrastructures appeared less damaged by the war. Some villages (e.g. Savitangaiala de Môma) seem to have been abandoned during the war so that the vegetation could regenerate, which was expected. But other villages (e.g. Techipeio) have not undergone the same vegetation regeneration suggesting they were not abandoned. The areal extent of large-scale agriculture increased 59% (26 km²) during the war, perhaps as a consequence of population growth. But the expansion was not nearly at par with the population growth of the Kavango region (320%), suggesting that a smaller proportion of the population relied on the large-scale agriculture for their subsistence in 2001 compared with 1973.</p><p>No significant impacts were found from the large-scale agriculture and urbanisation on the water quality during the dry season of 2005. Total phosphorous concentrations (with range: 0.067-0.095 mg l^-1) did vary significantly between locations (p=0.013) but locations upstream and downstream of large-scale agricultural facilities were not significantly different (p=0.5444). Neither did faecal coliforms (range: 23-63 counts per 100ml) nor faecal streptococci (range: 8-33 counts per 100ml) vary significantly between locations (p=0.332 and p=0.354 respectively). Thus the impact of Rundu and the extensive livestock farming along the border were not significant at this time. The Cuito river on the other hand significantly decreased both the conductivity (range: 27.2-49.7 μS cm^-1, p<0.0001) and the total dissolved solid concentration (range: 12.7-23.4 mg l^-1, p<0.0001) of the mainstream of the Okavango during the dry season.</p><p>Land cover changes during the Angolan civil war, contributing causes and effects on water quality were studied in this research effort. Many of the obtained results can be used directly or with further application as a knowledge base for sustainable decision-making and management in the basin. Wisely used by institutions charged with that objective, the information can contribute to sustainable development and the ending of suffering and poverty for the benefit of the peoples of the Okavango and beyond.</p>

<p>The purpose of this study is to find potential areas suitable for energy production using renewable sources. For this aim, accurate assessments are necessary. The assessments include geographical suitability, closeness of infrastructure and observing local and regional framework concerning the use of renewable sources together with environmental protection. In addition, economical factor is considered in such an assessment. In this study, the Photovoltaic (PV) production potential for Germany is considered. An accurate and complete data set is necessary in order to achieve reliable results. In addition, a powerful database management and strong analysis tools are required. Geographical Information System (GIS) is a tool for finding suitable sites for the photovoltaic production.Using GIS, energy generation planners are able to visualize solar densities throughout the considered area. In addition, they can find the optimal and most economical sites by the combination of solar potential with the information about land. In this study, data sources consist of meteorological and geographical conditions. Furthermore, all analyses have been performed using Arc GIS Desktop. This study demonstrates the possible places for photovoltaic plants and indicates suitable candidates according to weights and factors in multi criteria analysis. The solar radiation data is from year 1995 to 2005. Land cover data is according to Corine 2000 and the more detailed Raumordnungskataster (Rok) for Weser-Ems. Numerical results are reliable from a comparison point of view. This study demonstrates the sensitivity of the defined criteria with respect to electricity production. In particular, this study is useful to see the capabilities of GIS for site selection regarding photovoltaic plants.The purpose of this study is to find potential areas suitable for energy production using renewable sources. For this aim, accurate assessments are necessary. The assessments include geographical suitability, closeness of infrastructure and observing local and regional framework concerning the use of renewable sources together with environmental protection. In addition, economical factor is considered in such an assessment. In this study, the Photovoltaic (PV) production potential for Germany is considered. An accurate and complete data set is necessary in order to achieve reliable results. In addition, a powerful database management and strong analysis tools are required. Geographical Information System (GIS) is a tool for finding suitable sites for the photovoltaic production.Using GIS, energy generation planners are able to visualize solar densities throughout the considered area. In addition, they can find the optimal and most economical sites by the combination of solar potential with the information about land. In this study, data sources consist of meteorological and geographical conditions. Furthermore, all analyses have been performed using Arc GIS Desktop. This study demonstrates the possible places for photovoltaic plants and indicates suitable candidates according to weights and factors in multi criteria analysis. The solar radiation data is from year 1995 to 2005. Land cover data is according to Corine 2000 and the more detailed Raumordnungskataster (Rok) for Weser-Ems. Numerical results are reliable from a comparison point of view. This study demonstrates the sensitivity of the defined criteria with respect to electricity production. In particular, this study is useful to see the capabilities of GIS for site selection regarding photovoltaic plants.</p>

The Okavango river flows from southern Angola, through the Kavango region of Namibia and into the Okavango Delta in Botswana. The recent peace in Angola hopefully marks the end of the intense suffering that the peoples of the river basin have endured, and the beginning of sustainable decision-making in the area. Informed decision-making however requires knowledge; and there is a need for, and a lack of knowledge regarding basin-wide land cover (LC) changes, and their causes, during the Angolan civil war in the basin. Furthermore, there is a need for, and a lack of knowledge on how expanding large-scale agriculture and urban growth along the Angola-Namibia border affects the water quality of the river. The aim of this study was therefore to develop a remote sensing method applicable to the basin (with scant ground-truth data availability) to carry out a systematic historic study of LC changes during the Angolan civil war, to apply the method to the basin, to relate these changes to major societal trends in the region, and to analyse potential impacts of expanding large-scale agriculture and urban growth on the water quality of the river along the Angola-Namibia border. A range of remote sensing methods to study historic LC changes in the basin were tried and evaluated against reference data collected during a field visit in Namibia in October 2005. Eventually, two methods were selected and applied to pre-processed Landsat MSS and ETM+ satellite image mosaics of 1973 and 2001 respectively: 1. a combined unsupervised classification and pattern-recognition change detection method providing quantified and geographically distributed binary LC class change trajectory information and, 2. an NDVI (Normalised Difference Vegetation Index) change detection method providing quantified and geographically distributed continuous information on degrees of change in vegetation vigour. In addition, available documents and people initiated in the basin conditions were consulted in the pursuit of discerning major societal trends that the basin had undergone during the Angolan civil war. Finally, concentrations of nutrients (total phosphorous &amp; total nitrogen), bacteria (faecal coliforms &amp; faecal streptococci), conductivity, total dissolved solids, dissolved oxygen, pH, temperature and Secchi depth were sampled at 11 locations upstream and downstream of large-scale agricultural facilities and an urban area during the aforementioned field visit. The nature, extent and geographical distribution of LC changes in the study area during the Angolan civil war were determined. The study area (150 922 km2) was the Angolan and Namibian parts of the basin. The results indicate that the vegetation vigour is dynamic and has decreased overall in the area, perhaps connected with precipitation differences between the years. However while the vigour decreased in the northwest, it increased in the northeast, and on more local scales the pattern was often more complex. With respect to migration out of Angola into Namibia, the LC changes followed expectations of more intense use in Namibia close to the border (0-5 km), but not at some distance (10-20 km), particularly east of Rundu. With respect to urbanisation, expectations of increased human impact locally were observed in e.g. Rundu, Menongue and Cuito Cuanavale. Road deterioration was also observed with Angolan urbanisation but some infrastructures appeared less damaged by the war. Some villages (e.g. Savitangaiala de Môma) seem to have been abandoned during the war so that the vegetation could regenerate, which was expected. But other villages (e.g. Techipeio) have not undergone the same vegetation regeneration suggesting they were not abandoned. The areal extent of large-scale agriculture increased 59% (26 km2) during the war, perhaps as a consequence of population growth. But the expansion was not nearly at par with the population growth of the Kavango region (320%), suggesting that a smaller proportion of the population relied on the large-scale agriculture for their subsistence in 2001 compared with 1973. No significant impacts were found from the large-scale agriculture and urbanisation on the water quality during the dry season of 2005. Total phosphorous concentrations (with range: 0.067-0.095 mg l-1) did vary significantly between locations (p=0.013) but locations upstream and downstream of large-scale agricultural facilities were not significantly different (p=0.5444). Neither did faecal coliforms (range: 23-63 counts per 100ml) nor faecal streptococci (range: 8-33 counts per 100ml) vary significantly between locations (p=0.332 and p=0.354 respectively). Thus the impact of Rundu and the extensive livestock farming along the border were not significant at this time. The Cuito river on the other hand significantly decreased both the conductivity (range: 27.2-49.7 μS cm-1, p&lt;0.0001) and the total dissolved solid concentration (range: 12.7-23.4 mg l-1, p&lt;0.0001) of the mainstream of the Okavango during the dry season. Land cover changes during the Angolan civil war, contributing causes and effects on water quality were studied in this research effort. Many of the obtained results can be used directly or with further application as a knowledge base for sustainable decision-making and management in the basin. Wisely used by institutions charged with that objective, the information can contribute to sustainable development and the ending of suffering and poverty for the benefit of the peoples of the Okavango and beyond.

The purpose of this study is to find potential areas suitable for energy production using renewable sources. For this aim, accurate assessments are necessary. The assessments include geographical suitability, closeness of infrastructure and observing local and regional framework concerning the use of renewable sources together with environmental protection. In addition, economical factor is considered in such an assessment. In this study, the Photovoltaic (PV) production potential for Germany is considered. An accurate and complete data set is necessary in order to achieve reliable results. In addition, a powerful database management and strong analysis tools are required. Geographical Information System (GIS) is a tool for finding suitable sites for the photovoltaic production.Using GIS, energy generation planners are able to visualize solar densities throughout the considered area. In addition, they can find the optimal and most economical sites by the combination of solar potential with the information about land. In this study, data sources consist of meteorological and geographical conditions. Furthermore, all analyses have been performed using Arc GIS Desktop. This study demonstrates the possible places for photovoltaic plants and indicates suitable candidates according to weights and factors in multi criteria analysis. The solar radiation data is from year 1995 to 2005. Land cover data is according to Corine 2000 and the more detailed Raumordnungskataster (Rok) for Weser-Ems. Numerical results are reliable from a comparison point of view. This study demonstrates the sensitivity of the defined criteria with respect to electricity production. In particular, this study is useful to see the capabilities of GIS for site selection regarding photovoltaic plants.The purpose of this study is to find potential areas suitable for energy production using renewable sources. For this aim, accurate assessments are necessary. The assessments include geographical suitability, closeness of infrastructure and observing local and regional framework concerning the use of renewable sources together with environmental protection. In addition, economical factor is considered in such an assessment. In this study, the Photovoltaic (PV) production potential for Germany is considered. An accurate and complete data set is necessary in order to achieve reliable results. In addition, a powerful database management and strong analysis tools are required. Geographical Information System (GIS) is a tool for finding suitable sites for the photovoltaic production.Using GIS, energy generation planners are able to visualize solar densities throughout the considered area. In addition, they can find the optimal and most economical sites by the combination of solar potential with the information about land. In this study, data sources consist of meteorological and geographical conditions. Furthermore, all analyses have been performed using Arc GIS Desktop. This study demonstrates the possible places for photovoltaic plants and indicates suitable candidates according to weights and factors in multi criteria analysis. The solar radiation data is from year 1995 to 2005. Land cover data is according to Corine 2000 and the more detailed Raumordnungskataster (Rok) for Weser-Ems. Numerical results are reliable from a comparison point of view. This study demonstrates the sensitivity of the defined criteria with respect to electricity production. In particular, this study is useful to see the capabilities of GIS for site selection regarding photovoltaic plants.

The following study investigates how a sand dune system in the central part of Laholm Bay (Laholmsbukten) in Halland County, Sweden, has evolved over the time period 1947 to 2014. Effort was made to contribute to improved knowledge of how geomorphological variations and anthropogenic activity in the area have been influenced by each other over time. The study is aiming to provide an improved science basis for further development of coastal management in the area. Initial focus points were to investigate the correlation over time between changes in dune system morphology, vegetation distribution and anthropogenic influence. Furthermore connections were made as to how climate has influenced the development and how climate change during the 21st century might affect the area. Methods used during the course of this study entailed fieldwork and remote sensing of aerial photographs. Changes in dune system dynamics, land cover and human impact on the area over time were analyzed. The result demonstrates that the area has undergone dynamic changes, affected by climatological aspects, human activities as well as vegetation changes. Decreasing topographic variations in the southern part of the dune system compared to the north is identified to be dependent on variations in soil fractions. This combined with the identified changes in vegetation distribution over time has an affect on erosion and deposition processes within the area. Future climate change during this century may further increase the dynamic behavior of the dune system, an important aspect to consider within local coastal management. / Följande studie utreder hur ett sanddynsområde i de centrala delarna av Laholmsbukten i Hallands län utvecklats under tidsperioden 1947-2014. Syftet var att skapa en uppdaterad kunskapsbild av områdets geomorfologiska utveckling och hur den antropogena aktiviteten i området har påverkat denna, för att, om möjligt bidra med underlag till en utveckling av förvaltningsarbetet i kustområdet. Fokus låg på att utreda sambanden mellan förändringar av landskapets morfologi, vegetationens utbredning och antropogen påverkan. Vidare undersöks hur klimatet under tidsperioden kan ha bidragit till den geomorfologiska utvecklingen och hur området kan komma att förändras fram till sekelskiftet år 2100. Arbetet innefattade fältarbete och fjärranalys av flygbilder. Faktorer såsom dynsystemets dynamik, areella förändringar i marktäcke och mänsklig påverkan på området analyserades. Resultatet påvisar att sanddynsområdet genomgått dynamiska förändringar över tid som kan antas bero på klimatologiska aspekter i kombination med mänsklig aktivitet och vegetationsförändringar. Dynområdets minskande relief i nord-sydlig riktning och variationer i dynkantens förskjutning över tid är beroende av det dynbildande materialets sammansättning. I kombination med identifierade variationer i vegetationens utbredning påverkar detta erosions- och ackumulationsförutsättningarna i området. Klimatförändringar under detta sekel kan komma att påverka de faktorer som reglerar dynsystemets uppbyggnad vilket kan öka dynamiken i systemet ytterligare. En aspekt som är viktig att ta hänsyn till i förvaltningen av dynområdet.

The probability and impact of flooding is projected to increase in the future. This is due to climate and land-use changes (e.g. urbanization) in addition to the ongoing socioeconomic development of many floodplain areas. Exploiting the increasing availability of satellite data for flood inundation modelling will allow mapping floods in remote, data-poor areas to lower costs, and thereby make it possible to estimate flood risks in areas that today lack the economic resources needed for supporting risk assessment. In this context, this study has investigated the potentials and limitations of using low-cost, global remote sensing data (i.e. SRTM) to support flood inundation modelling. To this end, a case study of a river reach along the Mississippi was exploited. In particular, two flood inundation models were built by using the same 2D hydraulic model code (LISFLOOD-FP), but with two different topographical inputs, i.e. high quality/accuracy LiDAR topography data and the freely available SRTM topography data. The LiDAR data was lowered to the same resolution as the SRTM data and the two models were run with the resolution of 83x83 m2 . Thereafter, the models were compared by simulating two historical flood events of different magnitude. The comparison of the two models showed that flood inundation modelling with satellite data is more accurate (closer to the reference model, i.e. LiDAR-based model) for the higher magnitude flood event than for the lower magnitude flood event. This was attributed to the relatively reduced importance of micro topography during bigger flood events. An area-based performance measure gave a value of the correspondence (i.e. the fit) between the predicted flood extents for the two models. The areas/pixels were reclassified in ARC GIS to flooded or dry. Thereafter, areas flooded in both the LiDAR and the SRTM simulations were divided by the sum of the areas flooded in both or in one of the simulations (LiDAR or SRTM). From this procedure the fit could be determined, where a fit of 100 % would mean that the simulations had predicted the same flood extents. For the high magnitude flood event simulated in this study, the fit in terms of flood extent between the LiDAR-based and the SRTM-based model was 72 %, while the fit for the smaller flood was only 38 %. In this study, model calibration was preformed manually because of limited availability of time and computational power. However, this is not considered a major limitation as the work does not aim to make a faultless model of this river reach of the Mississippi, but rather to determine the potentials and limitations of SRTM topography data in supporting flood inundation modelling. Additional studies of rivers systems with different properties, flood magnitudes, vegetation covers and river scales should be conducted, to further validate the usability of remote sensing data for flood inundation modelling. / Stora områden runt om i världen har problem med översvämningar, som står för 40 % av alla dödsfall orsakade av naturkatastrofer. Det är troligt att risken för översvämningar kommer att öka i framtiden på grund av klimatförändringar och ändrad landanvändning, som till exempel urbanisering. Ett problem är att det ofta är dyrt att göra kartor som beskriver översvämningsrisker och därför finns det många områden där kunskap om riskerna saknas. I denna studie har det undersökts huruvida det är möjligt att använda globala fjärranalysdata (data från satelliter) för översvämningsmodellering. Detta skulle möjliggöra framställandet av kartor över översvämningsrisker till en låg kostnad, och därmed nå ut till områden där idag inte finns ekonomiska resurser nog för detta. En fallstudie har gjorts av en sträcka utmed Mississippifloden (USA) och två översvämningsmodeller har byggts genom att använda samma hydrauliska modelleringskod (LISFLOOD-FP). Skillnaden mellan modellerna var att den ena modellen byggdes med hjälp av LiDAR-topografidata, medan den andra modellen baserades på gratis SRTMtopografidata. LiDAR-data är högkvalitativt och högupplöst data (1 meter upplösning) insamlat från flygplan med hjälp av laser. SRTM-data har endast 30-90 meters upplösning (83 meter inom fallstudieområdet) och är insamlat av satelliter. Upplösningen av LiDAR-datat ändades till samma upplösning som för SRTM-datat och båda modellerna kördes med en upplösning av 83x83 m2 . De två modellerna jämfördes genom att två historiska översvämningar, en liten år 2008 och en mycket stor år 1993, simulerades. Jämförelsen av de två modellerna visade på att modellering med hjälp av satellitdata är mer precist och närmare referensmodellen, det vill säga den LiDAR-baserade modellen, för större översvämningar än för mindre översvämningar. Förklaringen till detta tillskrevs den relativt reducerade betydelsen av mikrotopografi för större översvämningar. Överrensstämmelsen mellan modellresultaten räknades ut genom att områdena/pixlarna först blev omklassificerade i ARC GIS som översvämmande eller icke översvämmade. Därefter delades antalet områden som svämmades över i båda simuleringarna med antalet områden som svämmades över i båda simuleringarna eller i den ena av simuleringarna. På detta sett kunde en faktor för överensstämmande bestämmas, där en faktor på 100 % innebar att modellerna förutspådde lika stora översvämningar. För den större översvämningen som simulerades överensstämde, i fråga om utbredning, de två modellerna (LiDAR och SRTM) till 72 %, medan modellerna för den mindre översvämningen endast överensstämde till 38 %. I denna studie gjordes kalibreringen manuellt då den tillgängliga tiden och datorkapaciteten var begränsad. Dock så anses inte detta vara en stor begränsning eftersom studien inte syftade till att göra en felfri modell av översvämningsriskerna utmed en sträcka av Mississippifloden, utan till att undersöka användbarheten och begränsningarna av satellitdata för översvämningsmodellering. Denna studie stödjer tidigare teorier om att globala satellitdata har stort användningsområde för att simulera översvämningsrisker. Dock behövs fler studier av flodsystem med olika egenskaper, storlek på översvämningar och vegetation göras för att ytterligare validera detta.

Uppmärksamheten och intresset för LIDAR data har blivit stort under 2000-talet. Som en väletablerad fjärranalysmetod uppvisar den resultat att studera och den kritisk frågan är ”Hur bra är egentligen resultatet att studera?”. I flera tidigare publiceringar infinner sig ofta en okritisk och positiv inställning till tekniken. Fördelaktiga resultat framhävs istället för helhets kritiska granskningar. Tidigt i processen finns det en rad olika aspekter, som kan utgöra störningsmoment. Hur inskanningen av LIDAR gemoförs, samt efterbehandlingens process. Den valda fornlämningstypen fossil åkermark, kan bli problematisk. Den fossila åkermarkens komplexa systemen kan med felaktiga visualisering döljas. Att kombinera höjdmodeller utifrån Hillshade och Slope förhindrar att den fossila åkermarkens system döljs. En annan aspekt att ta i beräkning är vegetationen. I vår frågeställning vill vi undersöka hur vanliga marktyper påverkar synligheten. Vi valde att utgå från Naturvårdsverkets marktäckesdata och har där tittat på tre olika marktyper. Granskog (utanför våtmark), Ädellövskog (utanför våtmark) och Temporärt ej skog (utanför våtmark). Av de tre marktyperna uppvisar Temporärt ej skog (utanför våtmark) svårigheter i samband med visualiseringen, kvalitén för markytan i höjdmodellaren skiftar avsevärt. Granskog (utanför våtmark) har möjligheter, däremot kan resultaten vara skiftande. Ädellövskog (utanför våtmark) har enklast material att tolka, med tydliga lämningar som tidigare inte vart registrerade hos FMIS. / The attention and interest in LIDAR data has grown considerably during the 2000s. As a well-established distance analysis method, it presents results to study and the critical question is "How good is the result to study?". I several previous publications there often is an uncritical and positive attitude to the technology. Beneficial results are emphasized instead of critical reviews of the whole technology. Early in the process, there are several different aspects, which can be disruptive moments. How the scan of LIDAR is organized, as well as the post-treatment process. The chosen ancient type of fossil arable land can be problematic. The complex systems of the fossil farmland can be hidden with incorrect visualization. Combining altitude models based on Hillshade and Slope prevents the fossil farmland system from being concealed. Another aspect to consider is vegetation. In our question we want to investigate how common types of vegetation affect visibility. We chose to use the Swedish Environmental Protection Agency's ground cover data and have looked at three different vegetation types. Spruce forest (outside wetland), Nobel deciduous forest (outside wetland) and Temporary non forest (outside wetland). Of the three vegetation types, Temporary non forest (outside wetland) presents difficulties in visualization, the quality of the ground surface in the height model varies considerably. Spruce forest (outside wetlands) has opportunities, however the results can be varied. Noble deciduous forest (outside wetland) has the simplest material for interpreting, with remains that have not previously been registered with FMIS.

Humanitarian help organisations today may benefit on improving their location analysis when placing field hospitals in countries hit by a disasters or catastrophe. The main objective of this thesis is to develop and evaluate a spatial decision support method for strategic placing of field hospitals for two time perspectives, long term (months) and short term (weeks). Specifically, the possibility of combining existing infrastructure and satellite data is examined to derive a suitability map for placing field hospitals. Haut-Katanga in Congo is used as test area where exists a large variety of ground features and has been visited by aid organisations in the past due to epidemics and warzones. The method consists of several steps including remote sensing for estimation of population density, a Multi Criteria Evaluation (MCE) for analysis of suitability, and visualization in a webmap. The Population density is used as a parameter for an MCE operation to create a decision support map for locating field hospitals. Other related information such as road network, water source and landuse is also taken into consideration in MCE. The method can generate a thematic map that highlights the suitability value of different areas for field hospitals. By using webmap related technologies, these suitability maps are also dynamic and accessible through the Internet. This new approach using the technology of dasymetric mapping for population deprival together with an MCE process, yielded a method with the result being both a standalone population distribution and a suitability map for placing field hospitals with the population distribution taken into consideration. The use of dasymetric mapping accounted for higher resolution and the ability to derive new population distributions on demand due to changing conditions rather than using pre-existing methods with coarser resolution and a more seldom update rate. How this method can be used in other areas is also analysed. The result of the study shows that the created maps are reasonable and can be used to support the locating of field hospitals by narrowing down the available areas to be considered. The results from MCE are compared to a real field hospital scenario, and it is shown that the proposed method narrows down the localisation options and shortens the time required for planning an operation. The method is meant to be used together with other decision methods which involves non spatial factors that are beyond the scope of this thesis.

Deep neural networks (DNN) have been widely demonstrated to be more powerful than their shallower counterparts in a variety of computer vision tasks and remote sensing applications. However, as many techniques are based on trial-and-error experiments as opposed to systematic evaluation, scientific evidence for the superiority of DNN needs more theoretical and experimental foundations. Recent work has shown that the neural network expressivity, measured by the number of linear regions, is independent of the network structure, suggesting that the success of deep neural networks is attributed to its ease of training. Inspired by this, this project aims to investigate novel approaches to train neural networks and obtain desired properties of the regional properties of linear regions. In particular, it highlights the regional structure of linear regions in different decision regions and seeks to initialize the network in a better position that makes it easier to have this regional structure. By counting the total number of linear regions in the input space, we validated that the shallow wider networks and the deep narrow networks share the same upper-bound expressivity in different synthetic datasets. We also discovered that the linear regions along the decision boundary are larger in shape and fewer in number, while being denser and fitted to the data manifold when close to the data. Our experiments indicate that the proposed initialization method can generate more linear regions at initialization, make the training converge faster, and finally generate linear regions that better fit the data manifold on synthetic data. On the EuroSAT satellite dataset, the proposed initialization method does not facilitate the convergence of ResNet-18, but achieves better performance with an average increase of 0.14% on accuracy compared to pre-trained weights and 0.19% compared to He uniform initialization. / Djupa neurala nätverk (Deep Neural Networks, DNN) har i stor utsträckning visat sig vara mer kraftfulla än sina grunda motsvarigheter i en mängd olika datorseendeuppgifter och fjärranalystillämpningar. Många tekniker är dock baserade på försök och misstag snarare än systematisk utvärdering, och vetenskapliga bevis för DNN:s överlägsenhet behöver mer teoretiska och experimentella grunder. Nyligen utförda arbeten har visat att det neurala nätverkets uttrycksförmåga, mätt som antalet linjära regioner, är oberoende av nätverksstrukturen, vilket tyder på att framgången för djupa neurala nätverk beror på att de är lätta att träna. Inspirerat av detta syftar detta projekt till att undersöka nya metoder för att träna neurala nätverk och få önskade egenskaper hos de regionala egenskaperna hos linjära regioner. I synnerhet belyser det den regionala strukturen hos linjära regioner i olika beslutsregioner och försöker initiera nätverket i ett bättre läge som gör det lättare att få denna regionala struktur. Genom att räkna det totala antalet linjära regioner i ingångsutrymmet validerade vi att de grunda bredare nätverken och de djupa smala nätverken har samma övre gräns för uttrycklighet i olika syntetiska dataset. Vi upptäckte också att de linjära regionerna längs beslutsgränsen är större till formen och färre till antalet, samtidigt som de är tätare och anpassade till datamångfalden när de ligger nära data. Våra experiment visar att den föreslagna initialiseringsmetoden kan generera fler linjära regioner vid initialiseringen, få träningen att konvergera snabbare och slutligen generera linjära regioner som bättre passar datamångfalden på syntetiska data. På EuroSAT-satellitdatamängden underlättar den föreslagna initialiseringsmetoden inte konvergensen för ResNet-18, men uppnår bättre prestanda med en genomsnittlig ökning av noggrannheten med 0,14% jämfört med förtränade vikter och 0,19% jämfört med He uniform initialisering.

Cloud detection on satellite imagery is an essential pre-processing step for several remote sensing applications. In general, machine learning based methods for cloud detection perform well, especially the ones based on deep learning as they consider both spatial and spectral features of the input image. However, false alarms become a major issue in winter images, wherein bright objects like snow/ice are also detected as cloud. This affects further image analysis like urban change detection, weather forecast, disaster risk management. In this thesis, we consider optical remote sensing images from small satellites constellation of PlanetScope. These have limited multispectral capacity of four bands: Red, Green, Blue (RGB) and Near-Infrared (NIR) bands. Detection algorithms tend to be more efficient when considering information from more than one spectral band to perform the detection. This study explores the data diversity provided by NIR band to RGB band images in terms of improvement in cloud detection accuracy. Two deep learning algorithms based on convolutional neural networks with different architectures are trained on RGB, NIR and RGB+NIR image data, resulting in six trained models. Each of these networks is tested with winter images of varying amounts of clouds and land covered with snow and ice. The evaluation is done based on performance metrics for accuracy and Intersection-over-Union (IoU) scores, as well as visual inspection. A total of eighteen experiments are performed, and it is observed that NIR band provides significant data diversity when combined with RGB bands, by reducing the false alarms and improving the accuracy. In terms of processing time, there is no significant increase for the algorithms evaluated, therefore better cloud detection can be achieved without significantly increasing the computational costs. Based on this analysis, Unibap iX10-100 embedded system is a possible choice for implementing these algorithms as it is suitable for AI applications. / Detektering av moln på satellitbilder är ett viktigt bearbetningssteg för flera fjärr analysapplikationer. I allmänhet fungerar maskininlärningsbaserade metoder för molndetektering bra, särskilt de som är baserade på djupinlärning eftersom de tar hänsyn till både spatiala och spektrala egenskaper i input bilder. Men falsklarm blir ett stort problem i vinterbilder, där medbringande föremål som snö/is också upptäcks som moln. Detta påverkar ytterligare bildanalyser som upptäckt av stadsförändringar, väderprognos, katastrofrisk-hantering. I denna avhandling tar vi hänsyn till optiska fjärranalysbilder från små satellitkonstellationer PlanetScope. Dessa har begränsad multispektral kapacitet på fyra band: röda, gröna, blå (RGB) och near-infrared (NIR) band. Detektionsalgoritmer tenderar att vara mer effektiva när man överväger information från mer än ett spektralband för att utföra detekteringen. Denna studie utforskar datadiversiteten som tillhandahålls av NIR-band till RGB-bandbilder när det gäller förbättring av molndetekteringsnoggrannheten. Två djupinlärningsalgoritmer baserade på konvolutionella neurala nätverk med olika arkitekturer tränas på RGB-, NIR- och RGB+NIR-bilddata, vilket resulterar i sex tränade modeller. Vart och ett av dessa nätverk testas med vinterbilder av varierande mängder moln och land täckt med snö och is. Utvärderingen görs baserat på prestandamått för noggrannhet och Intersection-over-Union (IoU) poäng, samt visuell inspektion. Totalt arton experiment utförs, och det observeras att NIR-bandet ger betydande datadiversitet när det kombineras med RGB-band, genom att minska de falska larmen och förbättra noggrannheten. När det gäller bearbetningstid finns det ingen signifikant ökning av den för de utvärderade algoritmerna, därför kan bättre molndetektering uppnås utan att nämnvärt öka beräkningskostnaderna. Baserat på denna analys är Unibap iX10-100 inbyggt system ett möjligt val för implementera dessa algoritmer eftersom det är lämpligt för AI-tillämpningar.