Thesis (M.A.)--Kent State University, 2008. / Title from PDF t.p. (viewed Dec. 22, 2009). Advisor: Mandy Munro-Stasiuk. Keywords: remote sensing, wetland, wetland identification, GIS. Includes bibliographical references (p. 158-161).

>Magister Scientiae - MSc / Wetlands have been recognised as one of the most intrinsically valuable and threatened ecosystems in the world. Global estimates indicate that wetlands are being lost or transformed at a rapid rate, exacerbated by projected climate change impacts. This has prompted the need to improve wetland mapping to address the conservation and management of these ecosystems effectively. However, this remains a challenge. Current mapping approaches estimates of wetland extent vastly underestimate the true extent. Ancillary data has been acknowledged to improve the accuracy of mapping the distribution of wetlands.

Global trends of transformation and loss of wetlands to other land uses has deleterious effects on surrounding ecosystems, and there is a resultant increasing need for improved mapping of wetlands. This is because wetland conservation and management depends on accurate spatial representation of these systems. Current approaches to mapping wetlands through the classification of satellite imagery typically under-represent actual wetland area, and the importance of ancillary data in improving the accuracy in mapping wetlands is recognized. This study uses likelihood estimates of wetland occurrence in KwaZulu-Natal (KZN), South Africa, using a number of environmental surrogate predictors (such as slope, rainfall, soil properties etc.). Using statistical information from a set of mutually independent environmental variables in known wetland areas, conditional probabilities were derived through a Bayesian network (BN) from which a raster layer of wetland probability was created. The layer represents the likelihood of wetlands occurring in a specific area according to the statistical conditional probability of the wetland determinants. Probability values of 80% and greater also accounted for approximately 6% of the KZN area (5 520 km²), which is substantially more than the previously documented wetland area in KZN (4% of the KZN area or 4 200 km²). Using an independent test dataset, Receiver Operating Characteristic (ROC) curves with the Area Under Curve (AUC) analysis verified that the final model output predicted wetland area well (AUC 0.853). Based on visual comparisons between the probability layer and ground verified wetland systems, it was shown that high wetland probability areas in the final output correlated well with previously highlighted major wetland and wetland-rich areas in KZN. Assessment of the final probability values indicated that the higher the probability values, the higher the accuracy in predicting wetland occurrence in a landscape setting, irrespective of the wetland area. It was concluded that the layer derived from predictor layers in a BN has the potential to improve the accuracy of the KZN wetland layer by serving as valuable ancillary data. Application of the final probability layer could extend into the development of updated spatial freshwater conservation plans, potentially predicting the historical wetland extents, and as input into the land cover classification process. Keywords: ancillary data, Bayesian network, GIS, modelling, probability, wetland mapping. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2014.

Wetlands are vital ecosystems providing essential services to both humans and the environment, yet they face threats from human activities leading to loss and disturbance. This study utilizes remote sensing (RS) methods, including object-based image analysis (OBIA), to map and assess wetland health in Kristianstad’s Vattenrike in the southernmost part of Sweden between 2015 and 2023. Objectives include exploring RS capabilities in detecting wetlands and changes, deriving wetland health indicators, and assessing classification accuracy. The study uses Sentinel-2 imagery, elevation data, and high-resolution aerial images to focus on wetlands along the river Helge å. Detection and classifications were based on Sentinel-2 imagery and elevation data, and the eCognition software was employed. The health assessment was based on the spectral indices Normalized Difference Vegetation Index (NDVI) and Modified Normalized Difference Water Index (mNDWI). Validation was conducted through aerial photo interpretation. The derived classifications demonstrate acceptable accuracy levels and the analysis reveals relatively stable wetland conditions, with an increase in wetland area attributed to the construction of new wetlands. Changes in wetland composition, such as an increase in open meadows and swamp forests, were observed. However, an overall decline in NDVI values across the study area indicates potential degradation, attributed to factors like bare soil exposure and water presence. These findings provide insights into the local changes in wetland extent, composition, and health between the study years.

This study investigated the response of the macrophytes in St Lucia Estuary, northern KwaZulu-Natal to drought. The present distribution of vegetation (2010 / 2011) was mapped and changes over time recorded from past aerial photographs. The changes in macrophyte cover in response to environmental factors (sediment and groundwater characteristics) was measured along four transects in 2010 and compared with results from previous years, in 2005 and 2006. In the current study, 1960 images were digitized to illustrate macrophyte distribution and cover of the Narrows, Makakatana and the Eastern Shores during a low rainfall period which started in 1958. The 2001 images were digitized to illustrate vegetation distribution and cover of the same area prior to the current drought which started in June/ July 2002. The 2008 images were digitized to illustrate vegetation distribution and cover of estuarine vegetation within the floodplain during the drought (after 6 years). The sites were visited in June 2011 for verification of the distribution and boundaries of each macrophyte habitat. The area covered by the water column varied over time. In 1960 during a low rainfall period the water was 32705 ha, 33320 ha in 2001 and reduced to 30443 ha in 2008. The area cover of inundated and dry reeds fluctuated with the water level. Under high water levels, low-lying areas such as Fanies Island and Selley‟s Lake were flooded and under low water levels, intertidal sand and mudflats were exposed and colonised by succulent salt marsh (Sarcocornia spp. and Salicornia meyeriana). Similar conditions were observed at Makakatana. Mangroves were observed from the mouth to the Forks. Avicennia marina was the dominant mangrove species and primary colonizer of dredge spoil. The area cover of mangroves in the vicinity of the mouth fluctuated as a result of fluctuating water levels, dredging operations, excavation of the Back Channel and Cyclone Gamede which killed intertidal vegetation. Between 2001 and 2008, mangrove expansion was faster in the Mfolozi Swamps area (± 1.4 ha yr -1) compared to the Narrows (± 0.4 ha yr -1). Long term monitoring transects were set up in 2005, at Makakatana, Charters Creek, Catalina Bay and at Listers Point to document changes in sediment conditions and vegetation cover. These were sampled in July 2005, October 2005, February 2006 and May 2010. Silt was the dominant particle size at Catalina Bay, Charters Creek and Makakatana. At Makakatana, average ground water salinity was 17.2 ± 6.6 ppt, 4.1± 4.9 ppt at Catalina Bay and 32.9 ± 19.3 ppt at Charters Creek. Drought resulted in the accumulation of salt on the surface sediment layer at Listers Point and Charters Creek due to low rainfall. Listers Point, the site with the lowest freshwater input and habitat diversity had the lowest macrophyte species richness with only three species. The dominant species at this site were Sporobolus virginicus and Chenopodium album L. which are highly salt tolerant species. Catalina Bay had the highest species richness (18 to 27); as a result of freshwater input via groundwater seepage from the sand dune aquifers on the Eastern Shores. Along the Eastern Shores, vegetation was dominated by species of Cyperaceae, Juncaceae and Juncaginaceae. Fluctuations in groundwater depth were observed at all sites, Listers Point groundwater depth in February 2006 was 80 cm and the ground water level was not reached during the May 2010 field trip. During the May 2010 field trip, the water column salinity of the St Lucia system was highly variable, Makakatana had the lowest water column salinity of 7.1 ppt, 42.1 ppt at Catalina Bay, 44 ppt at Charters Creek and Listers Point had the highest water column salinity of 95 ppt. An assessment of the changes in macrophyte cover along the transects showed that cover fluctuated in response to rainfall, water level and drought. At Listers Point, there was a continuous decline in the abundance of Sporobolus virginicus over time which was sparsely distributed in the first 40 m of the transect. Sarcocornia natalensis, a succulent and obligate halophyte, was recorded, in areas with high sediment conductivity. In May 2010, bare ground increased to an average percentage cover of 96.5% and was covered with dead organic matter and a salt crust at Listers Point. At Makakatana, there was a significant decrease in bare ground from July 2005 to May 2010 (H = 24.58, N = 197, p<0.001) as bare areas were colonized by salt marsh. Multivariate analysis showed that the abundance of Sporobolus virginicus was positively influenced by sediment moisture content and Paspalum vaginatum abundance was strongly influenced by the water column salinity. At Catalina Bay, low sediment conductivity at the groundwater seepage area resulted in terrestrial vegetation encroaching on estuarine vegetation. Sarcocornia natalensis became more abundant towards the water column. During the period of study, species richness at St Lucia ranged from 2 (Listers Point, May 2010) to 27 (Catalina Bay, February 2006). Salinity and water level fluctuation have a significant impact on the distribution of macrophytes at St Lucia during the drought. In saline areas salt marsh plants have colonized exposed shorelines and along the Eastern Shores groundwater seepage has increased macrophyte species richness. Low water levels have resulted in the exposure and desiccation of submerged macrophytes, which are replaced by macroalgae.

In recent years the wide availability of high-resolution satellite images has made the remote monitoring of water resources all over the world possible. While the detection of open water from satellite images is relatively easy, a significant percentage of the water extent of wetlands is covered by vegetation. Convolutional Neural Networks have shown great success in the task of detecting wetlands in satellite images. However, these models require large amounts of manually annotated satellite images, which are slow and expensive to produce. In this paper we use self-supervised training methods to train a Convolutional Neural Network to detect water from satellite images without the use of annotated data. We use a combination of deep clustering and negative sampling based on the paper ”Unsupervised Single-Scene Semantic Segmentation for Earth Observation”, and we expand the paper by changing the clustering loss, the model architecture and implementing an ensemble model. Our final ensemble of self-supervised models outperforms a single supervised model, showing the power of self-supervision. / Under de senaste åren har den breda tillgången på högupplösta satellitbilder möjliggjort fjärrövervakning av vattenresurser över hela världen. Även om det är relativt enkelt att upptäcka öppet vatten från satellitbilder, täcks en betydande andel av våtmarkernas vattenutbredning av vegetation. Lyckligtvis kan radarsignaler tränga igenom vegetation, vilket gör det möjligt för oss att upptäcka vatten gömt under vegetation från satellitradarbilder. Under de senaste åren har Convolutional Neural Networks visat stor framgång i denna uppgift. Tyvärr kräver dessa modeller stora mängder manuellt annoterade satellitbilder, vilket är långsamt och dyrt att producera. Självövervakad inlärning är ett område inom maskininlärning som syftar till att träna modeller utan användning av annoterade data. I den här artikeln använder vi självövervakad träningsmetoder för att träna en Convolutional Neural Network-baserad modell för att detektera vatten från satellitbilder utan användning av annoterade data. Vi använder en kombination av djup klustring och kontrastivt lärande baserat på artikeln ”Unsupervised Single-Scene Semantic Segmentation for Earth Observation”. Dessutom utökar vi uppsatsen genom att modifiera klustringsförlusten och modellarkitekturen som används. Efter att ha observerat hög varians i våra modellers prestanda implementerade vi också en ensemblevariant av vår modell för att få mer konsekventa resultat. Vår slutliga ensemble av självövervakade modeller överträffar en enda övervakad modell, vilket visar kraften i självövervakning.

Wetlands are fragile ecosystems that deliver a wide range of environmental and ecosystem services that contribute to human well-being. This fragility has seen a number of wetlands succumb to degradations and loss arising from different causes. The positive contribution of wetlands to various ecosystems and the increasing rate of their degradation and loss has been a cause for concern to different stakeholders for a long time. One of the landmark indications of this concern is the constitution of the Ramsar Convention that was held in Iran in 1971 with the objective of addressing concerns regarding to wetland degradation and loss. Among the competing land uses resulting in wetlands loss and degradation could be listed urban agriculture, industrial development and housing. The research sought to assess the extent of wetland degradation with respect to housing developments in Harare, Zimbabwe. The dilemma as to why and how wetlands were being degraded was an issue of concern. The research sought to determine whether abatement or mitigation measures were in place to address the challenges Associated with wetlands loss and degradation. The subsequent growth of developments, their associated activities and possible intervention measures had to be thoroughly assessed. The administration of questionnaires and conducting of interviews was done to solicit data from the respondents. Collected data was analyzed through the Statistical Package for Social Scientists. Further to this, Google satellite imagery was used in the mapping of the Belvedere North and Epworth suburbs. Disturbed and undisturbed portions of the wetlands were analyzed by Geographical Information System software. Direct observation of the wetland area was used to examine existing natural features. To determine extent of pollution, water samples were collected in the wetlands and submitted to the laboratory for analysis. Analysis of selected chemical and physical parameters was carried out in the laboratory using selected methods. The research made two important and broad findings. First, both the Belvedere North and Epworth wetlands have been severely degraded by anthropogenic activities with housing developments and urban agriculture as the major contributors of this degradation. The development of residential properties in the wetlands is a result of a high demand for residential space in Harare. Whereas housing developments in Belvedere are formal and planned, developments in Epworth were informal and haphazard. With no formal planning, the Epworth wetlands have been severely impacted by water and sand extraction as well as contamination of underground water by pit-latrines that are common in these informal settlements. Second, the research also found that wetland degradation was closely associated with the lack of clear wetland policies both at national and local levels. The laboratory results showed that the water in Epworth was more polluted than the water in Belvedere. The measure of pollutants in Epworth was higher due to the activities being carried out as compared to Belvedere. Measures of Dissolved Oxygen, Total Dissolved Substances and conductivity increased from October to March in Epworth. Of note is the measure of Dissolved Oxygen that increased between October and March with a mean of 7.9 and 2.5 respectively. Total nitrogen and total phosphate also increased significantly in midstream and downstream sections. In essence, averages of 0.01 in October and 0.04 in March were recorded respectively for total nitrogen. The study concludes that an integrated land use approach has the potential of minimizing wetland loss and degradation. This is possible through the formulation and implementation of a comprehensive policy involving all relevant stakeholders. This will ensure a coherent decision making process. In as much as integrated land use planning and policy formulation are potential approaches to wetland sustainability, the drive to implement policy by policy makers concerning wetlands has to be explored. As ignorance concerning the ecological importance of wetlands exists in some cases, it is critical to consider wetland benefits above other competing developments such as housing and urban agriculture. This can be achieved through merging interrelated disciplines with the idea of integrating wetland information. For instance, deriving engineering solutions could promote development yet preserving wetlands. It is also suggested that environmental impact assessments, as planning tools should be carried out simultaneously with the town planning aspect. If given to proper planning, wetlands have a huge chance of survival. / Environmental Sciences / M.Sc. (Environmental Management)

Wetlands are fragile ecosystems that deliver a wide range of environmental and ecosystem services that contribute to human well-being. This fragility has seen a number of wetlands succumb to degradations and loss arising from different causes. The positive contribution of wetlands to various ecosystems and the increasing rate of their degradation and loss has been a cause for concern to different stakeholders for a long time. One of the landmark indications of this concern is the constitution of the Ramsar Convention that was held in Iran in 1971 with the objective of addressing concerns regarding to wetland degradation and loss. Among the competing land uses resulting in wetlands loss and degradation could be listed urban agriculture, industrial development and housing. The research sought to assess the extent of wetland degradation with respect to housing developments in Harare, Zimbabwe. The dilemma as to why and how wetlands were being degraded was an issue of concern. The research sought to determine whether abatement or mitigation measures were in place to address the challenges Associated with wetlands loss and degradation. The subsequent growth of developments, their associated activities and possible intervention measures had to be thoroughly assessed. The administration of questionnaires and conducting of interviews was done to solicit data from the respondents. Collected data was analyzed through the Statistical Package for Social Scientists. Further to this, Google satellite imagery was used in the mapping of the Belvedere North and Epworth suburbs. Disturbed and undisturbed portions of the wetlands were analyzed by Geographical Information System software. Direct observation of the wetland area was used to examine existing natural features. To determine extent of pollution, water samples were collected in the wetlands and submitted to the laboratory for analysis. Analysis of selected chemical and physical parameters was carried out in the laboratory using selected methods. The research made two important and broad findings. First, both the Belvedere North and Epworth wetlands have been severely degraded by anthropogenic activities with housing developments and urban agriculture as the major contributors of this degradation. The development of residential properties in the wetlands is a result of a high demand for residential space in Harare. Whereas housing developments in Belvedere are formal and planned, developments in Epworth were informal and haphazard. With no formal planning, the Epworth wetlands have been severely impacted by water and sand extraction as well as contamination of underground water by pit-latrines that are common in these informal settlements. Second, the research also found that wetland degradation was closely associated with the lack of clear wetland policies both at national and local levels. The laboratory results showed that the water in Epworth was more polluted than the water in Belvedere. The measure of pollutants in Epworth was higher due to the activities being carried out as compared to Belvedere. Measures of Dissolved Oxygen, Total Dissolved Substances and conductivity increased from October to March in Epworth. Of note is the measure of Dissolved Oxygen that increased between October and March with a mean of 7.9 and 2.5 respectively. Total nitrogen and total phosphate also increased significantly in midstream and downstream sections. In essence, averages of 0.01 in October and 0.04 in March were recorded respectively for total nitrogen. The study concludes that an integrated land use approach has the potential of minimizing wetland loss and degradation. This is possible through the formulation and implementation of a comprehensive policy involving all relevant stakeholders. This will ensure a coherent decision making process. In as much as integrated land use planning and policy formulation are potential approaches to wetland sustainability, the drive to implement policy by policy makers concerning wetlands has to be explored. As ignorance concerning the ecological importance of wetlands exists in some cases, it is critical to consider wetland benefits above other competing developments such as housing and urban agriculture. This can be achieved through merging interrelated disciplines with the idea of integrating wetland information. For instance, deriving engineering solutions could promote development yet preserving wetlands. It is also suggested that environmental impact assessments, as planning tools should be carried out simultaneously with the town planning aspect. If given to proper planning, wetlands have a huge chance of survival. / Environmental Sciences / M. Sc. (Environmental Management)