In this thesis, a landcover simulator module is developed to incorporate non-stationary landcover into the hydrological model WATFLOOD. Objectives are to quantify the uncertainty inherent in assuming landcover stationarity in the Winnipeg River basin (WRB), and to improve the projections of future streamflow. Forest fires commonplace in the WRB are modelled through logistic regression and a generalized extreme value distribution for occurrence and extent respectively, fit from historical data. Fire regeneration and natural changes in landcover are modelled though a first order Markov chain, with transition probabilities derived from satellite imagery. Using satellite imagery directly into historical simulations in a sub-basin with substantial forest fire activity improved WATFLOOD results. With climate change, incorporating non-stationary landcover results in lower flows than assuming stationarity, albeit still greater than baseline (1971 - 2000) flows. Projected streamflow uncertainty under climate change also increases as a result of introducing non-stationary landcover in the WRB.

Region called "Sokolovsko" is an area in the Czech Republic in which a coal mining has caused a great interference with the appearance of the landscape. With a subsequent reclamation, the affected areas are recreated into new landscapes, however on the research base, the non-interference approach is applied in order to follow the principles of a natural succession. This diploma thesis examines the influence of the origin, respectively the relief of the dump area on the intensity of the spontaneous vegetation growth, within the example of the Velká podkrušnohorská spoil heap, based on the data collected by remote sensing techniques. The vegetation indices NDVI and SAVI were used to reveal the intensity of the vegetation cover on the area of the interest. It is clear from the results that the vegetation growth is considerably faster in the areas with the original, wavy relief. It was also found that the vegetation growth of the non reclaimed area of Velká podkrušnohorská spoil heap in the parts of which the relief was settled at the time of the origin differs from the non reclaimed area of which the relief was left in the original wavy surface and later over layered with a new material. Finally it was made a comparison between the non reclaimed part of the Velká podkrušnohorská spoil heap whose...

No description available.

The Amazon rainforest, a vital ecosystem of immense biodiversity and global climate significance, faces the ongoing threat of deforestation driven by agricultural expansion. This thesis employs remote sensing techniques, focusing on the Enhanced Vegetation Index (EVI) derived from Landsat satellite imagery, to track land cover dynamics within the Amazon. The study examines historical land cover changes in current plantations in Peru and Brazil, regions where the exact timing of deforestation is uncertain. By analyzing EVI measurements dating back to 1984, inflection points indicative of deforestation events preceding plantation establishment are identified. Statistical modeling techniques, including spline fitting to analyze time series data and Random Forest algorithms for calibration, are employed to enhance the accuracy of EVI measurements. Additionally, predictions for deforestation years derived from ALOS satellite data are compared with those from Landsat imagery, revealing discrepancies and underscoring the need for methodological refinement.

The Grassland biome of South Africa has been identified as critically endangered and the biome in South Africa most requiring conservation attention through the implementation of efficient, sustainable systematic conservation plans. The ability to predict where land-cover transformation as well as information on the occurrence and severity of current land cover transformation activities, as threats to biodiversity, is required as part of the systematic conservation planning process. Neke & du Plessis (2004) predicted land cover transformation and the severity of the impact on biodiversity in the Grassland biome. This model was based on potential land use suitability models and land cover information for the 1994/5 season extracted from the National Land Cover database (NLC1994). These predictions were tested by assessing actual land cover change in the Grassland biome using observed differences in grassland land cover between the NLC1994 and NLC2000 databases. Methodology Because of differences in format and land-cover classification between the original datasets, both NLC1994 and NLC2000 had to be modified before any analyses could be carried out. These differences exist because different techniques were used to collate the respective datasets, thus introducing the potential for significant mapping error in the original datasets and more significantly erroneous results with respect to landcover change detection. The implications of this were presented in the discussion. Both datasets were spatially resampled and class-standardised and it was felt that this would significantly reduce any the impact of any such existing errords in the original datasets. Thereafter landcover information for the Grassland biome was be extracted and the comparative landcover analyses executed. The analyses carried out included: • Landcover change per landcover class within the Grassland biome with emphasis on the Grassland landclass losses and gains • An assessment and comparison of the relative fragmentation of the remaining grassland patches in both datasets • An assessment of current grassland habitat degradation • The comparison of the predicted land cover change as given by Neke & du Plessis (2004) against the observed grassland changes • The creation of a new Grassland Transformation threat map reflecting current land cover change threats, and including information pertaining to the threats to Grassland biodiversity posed by invasive alien plants, road effects, urban areas and soil erosion hazards. Results and Discussion 25% of the remaining grassland patches underwent transformation to other land classes. Grassland clearing for cultivation, bush encroachment and bushland vegetation regeneration were the main causal factors behind the observed grassland losses. However, grassland vegetation regeneration on formerly cultivated land, bush clearing and reclassification of degraded lands as grasslands in the NLC2000 dataset contributed to a net 2% gain in area of the grassland land class. The remaining grassland patches are more fragmented than they were in NLC1994, the average patch size (NLC2000) is three times smaller and the total number of grassland patches has increased (also by a factor of 3) and the remaining grassland patches are more isolated. The largest, least fragmented grassland patches occur along and to the west of the Great Escarpment as it traverses the Grassland biome. Most of the predictions of grassland transformation were realised, however the model used by Neke & du Plessis (2004) consistently underestimated and in some cases failed to predict the occurrence of grassland transformation in the central interior of the Grassland biome. Current, measurable human activities that act as grassland transformation agents were incorporated to create a threat map showing the extent and severity of land-cover transformation activities within the biome; grassland bird species richness information was then incorporated into this map to create biodiversity transformation threat map. This map was used to show the location and severity of the impacts of human transformation activities on grassland biodiversity. Both transformation threat map reflect the current situation across the biome today and were compared against the Potential transformation threat map produced by Neke & du Plessis (2004). The human transformation threat map confirmed the inability of the Neke & du Plessis model to make correct predictions of land cover change away from the eastern, 7 high altitude boundary of the biome. Given that the biome is defined by its climatic characteristics, the incorporation of global climate change effects would further refine the results gained, and perhaps provide more accurate predictions. As aforementioned, there are however factors existing within the original datasets used in this analysis that may have affected the accuracy of the landcover change analyses. These factors are centred on the potential effects of mapping errors within either of the NLC datasets. The delineation of landclass boundaries in the NLC1994 dataset is one such factor- placing a line over what is in reality a gradient of changing vegetation, is a subjective exercise and depends entirely on the technician involved this in itself may have introduced a fair amount of error in the mapping process. When coupled with the automated classification techniques used, for the most part, for the NLC2000 dataset, it becomes apparent that it is highly unlikely that even in the absence of actual landcover change the same boundaries would be drawn between two landclasses in the same area. This would provide false positive results for landcover change where in fact this is as a result of mapping errors. This is acknowledged and included in the interpretation of the results and it is felt that in spite of this, all possible steps were taken to minimize the impact of these effects on the reslults. The analysis allowed the identification of the current land cover transformations leading to grassland loss. However, land-cover change is only the physical expression of the complex interactions between socio-economic factors. To create effective and sustainable conservation plan for the Grassland biome, with an aim to reducing habitat loss requires an action plan to address these factors as the ultimate drivers of land cover change.

The connectivity of fluvial processes or hydrological connectivity are terms often used to describe internal linkages in landscape that intensify/reduce water and sediment fluxes as well as the substances they contain. It is based on the concept of landscape connectivity defined, in particular, by fragmentation of landscape and barriers that obstruct the movement of material. While researches focusing on water erosion or runoff focus on the volume of different fluxes, connectivity reveals new information on catchment functioning as well as reactions to various types of stimuli, and thus represents a more complex phenomenon. This master thesis informs about the topic as a whole, it describes the main forms of its modelling and specific influences that have major impact on presented results. Three approaches of modelling were tested. The main concept lies in evaluation of the index of connectivity, firstly, with approach proposed by Borselli et al. (2008) and implemented in software ArcGIS and secondly, in software SedInConnect 2.0. The third approach is modelling connectivity in software LAPSUS 5.0 that belongs to landscape evolution models. Modelling in LAPSUS proved that using even this indirect method for evaluation of hydrological connectivity is a suitable approach. Selected area for modelling...

Northern bobwhite (Colinus virginianus) populations have continually declined across much of their native range due to changes in land use and habitat. This notable decline causes concern for the future of this important game bird species and highlights the need for research on how habitat influences bobwhite vital rates. My research, conducted during 2018-2020 at Burning Star State Fish and Wildlife Area, addressed the following two objectives to provide information on how landcover characteristics influence bobwhite populations: (1) investigate how landcover characteristics influence annual bobwhite home range size, and (2) investigate how landcover characteristics influence bobwhite breeding season survival. For Objective 1, I radio-marked 113 individuals of which 50 bobwhites (31 males and 19 females) had ≥25 independent tracking locations and were used in my analyses. I modeled landcover types related to bobwhite habitat selection, habitat avoidance, foraging habitat, and protective cover to annual bobwhite home range size. Landcover types related to foraging habitat formed the most parsimonious model, with the proportion of grassland having the greatest negative influence on bobwhite home range size. As expected, landcover types that are thought to be beneficial to bobwhite fitness were negatively related to bobwhite home range size, suggesting those landcover types provide necessary resources for bobwhites throughout their full annual cycle. For Objective 2, I radio-marked 113 individuals of which 50 bobwhites (31 males and 19 females) had ≥25 independent tracking locations and were used in my analyses. I used a known fate model of the influence of landcover types, in a stepwise fashion, on bobwhite breeding season survival at two scales, home range and landscape. At the home range scale, the univariate model including early successional forest was the most parsimonious model, wherein the proportion of early successional forest negatively influenced bobwhite breeding season survival. Other competitive models included agriculture and early successional forest + scrub shrub where agriculture positively influenced breeding season survival while early successional forest and scrub shrub negatively influenced breeding season survival. The negative influence of landcover types thought to provide necessary protective cover for bobwhites on breeding season survival is inconsistent with the current paradigm that woody cover is an important habitat component that provides protective and escape cover, thermoregulatory functions, and midday loafing sites for bobwhites. Early successional forest provides visually appealing woody protective cover but appears to expose bobwhites to a higher predation risk because trees provide perch habitat for avian predators. Agriculture positively influenced breeding season survival because during the breeding season, agriculture provides overhead cover and bare ground that provides protective cover, foraging opportunities, and enables ease of movement. At the landscape scale, the null model was the most parsimonious model, but there was weak evidence that the proportion of grassland positively influenced breeding season survival. My study demonstrates how landcover composition influences annual home range size and breeding season survival at different scales.

The Swedish forest has been transformed during the last hundred years, from semi-open forest of multiple tree species to dense production forest of mainly spruce and pine. This has led to alterations of species composition in the forest floor, reducing coverage of dwarf shrubs like Vaccinium myrtillus (European blueberry). Bumblebees (Apidae: Bombus spp.) forage from V. myrtillus flowers and it forms together with other early flowering plant species a foundation for bumblebee colony establishment in spring. Against this background I examined how resources of V. myrtillus in the landscape affect bumblebees. I also studied the influence of season on preferred bumblebee forage habitat, comparing forests with road verges in the open landscape. The research was conducted in spring and summer of 2021 in the county of Södermanland in Sweden. Bumblebees were sampled in forests and road verges in 20 study landscapes, dominated by a forest-farmland mosaic. Bumblebee abundance and species richness was found to increase with coverage of V. myrtillus shrubs in forests during flowering of V. myrtillus in spring and with coverage of flowers in forests and road verges in summer. Bumblebees were also found to mainly forage in forests in spring and in road verges in summer and approximately the same bumblebee species occurred in both habitats. I conclude that bumblebees moves in between the forest and the open landscape in seasons, controlled by availability of flowers in the habitats. It is further concluded that V. myrtillus flowers is an important forage resource for bumblebees in spring. This calls for a landscape perspective managing bumblebees in forest-farmland landscapes in Sweden, recognizing the importance of forest forage resources as well as forage resources in the open landscape. A forest management that acknowledge the need for forest floor conditions supporting dwarf shrubs like V. myrtillus and summer flowering species is further recommended in order to preserve bumblebees in Swedish forest-farmland landscape.

Recent developments in commercial satellite products have resulted in a broader range of high quality image data, enabling detailed analysis. Transportation features have historically been difficult to accurately identify and structure into coherent networks; prior analyses have demonstrated problems in locating smaller features. One problem is that roadways in urban environments are often partly obscured by proximity to land cover or impervious objects. Ongoing research has focused on object-based methods for classification and different segmentation techniques key to this approach. For this application, software packages such as eCognition have shown encouraging results in assessing spatial and spectral patterns at varied scales in intelligent classification of aerial and satellite imagery. In this study 2.44m QuickBird and 4m Ikonos multispectral imagery for a 7.5' quad near the Mississippi Gulf Coast are examined. Challenges in analysis include intricate networks of smaller roads in residential zones and regions of tall/dense tree cover. Both spectral and object-based approaches are implemented for pre-classification, and road features are extracted using various techniques, after which the results are compared based on a ?Raster Completeness? model developed.

Mesoscale networks, also known as Mesonets, are advanced environmental monitoring systems that consist of multiple high-quality weather stations that collect meteorological data at frequent intervals (5 minutes or less). Tennessee currently lacks a Mesonet system, and this study identifies suitable and feasible locations for stations in all 95 counties. Layers needed to develop the suitability model included land cover, road centerlines, elevation, flood zones, building footprints, and hydrologic soil types. Geospatial analysis techniques were used to process and combine these layers to identify suitable areas. Once suitable areas were determined, a feasibility assessment that examined land ownership, accessibility, and cell coverage was conducted and multiple potential station locations were identified in each county. The establishment of the Tennessee Mesonet is expected to enhance weather monitoring and provide reliable environmental data for multiple sectors, including emergency management, agriculture, transportation, conservation, health, and many more.