Aquatic weeds transform and degrade the ecosystems which they invade, impacting various aspects of their surroundings ranging from the community level to disrupting important processes affecting ecosystem services. All of the major aquatic weeds of South Africa are found in the Cape Town Metropolitan Area. Landowners, whether private or public, are legally obliged to manage the listed invasive species through applying environmentally acceptable methodologies. This thesis provides an overview of the strategic management options, prevention, early detection, rapid response and eradication of new invasions, and containment and control species of established species. It discusses the different control methods available for managing aquatic weeds, namely mechanical, manual, chemical and biological, and the integration of different methods to improve their effectiveness. Although various studies have shown that biological control is the most cost–effective, environmentally-friendly and sustainable method, it is not yet fully integrated into weed management programmes in South Africa. In addition, the successes achieved in other parts of the world with the control of water hyacinth through biological control have not been repeated in the urban environment, despite the fact that South Africa has the highest number of biological control agents available for the weed. Urbanisation puts pressure on the natural environment and ecosystem functioning. Nutrient-enriched waters support aquatic weed growth and pose a challenge to the management thereof, in particular with regard to integrating biological control into management programmes. The aims of this study were to determine the reasons for the lack of integration of biological control into weed management programmes in South Africa, to determine the feasibility of integrating biological control in aquatic weed management programmes in a complex urban environmental and socio-political landscape by means of three case studies in the Cape Town Metropolitan Area, which showed that biological control is feasible in urban environments and should be considered. Two surveys were conducted to determine the reasons for the lack of integration of biological control into weed management programmes. The surveys showed that there is a gap between research and implementation as a result of poor communication, non-supporting institutional arrangements and a lack of appropriate capacity and skills at the implementation level. Recommendations were offered to address these issues.

The objectives of this project were to measure changes in frequency of occurrence of submerged macrophytes over the first two growing seasons following stocking with triploid grass carp at two fish per acre and to measure differences in macrophyte biomass between areas excluded from herbivory and adjacent control sites after 16 months following establishment of exclosures. The project also seeks to measure concentrations of fluridone following the herbicide treatment, and to compare two methods of aquatic vegetation sampling.

No description available.

No description available.

In this dissertation, methods are designed and validated for the utilization of hyperspectral hypertemporal remotely sensed data in target detection applications. Two new classes of methods are designed to optimize the selection of target detection features from spectro-temporal space data. The first method is based on the consideration that all the elements of the spectro-temporal map are independent of each other. The second method is based on the consideration that the elements of the spectro-temporal map have some vicinal dependency among them. Methods designed for these two approaches include various stepwise selection methods, windowing approaches, and clustering techniques. These techniques are compared to more traditional feature extraction methods such as Normalized Difference Vegetation Index (NDVI), spectral analysis, and Principal Component Analysis (PCA). The efficacies of the new methods are demonstrated within an aquatic invasive species detection application, namely discriminating waterhyacinth from other aquatic vegetation such as American lotus. These two aquatic plant species are chosen for testing the proposed methods as they have very similar physical characteristics and they represent a practical life target detection problem. It is observed from the overall classification accuracy estimates that the proposed feature extraction methods show a marked improvement over conventional methods. Along with improving the accuracy estimates, these methods demonstrate a capability to drastically reduce the dimensionality while retaining the desired hyperspectral hypertemporal features. Furthermore, the feature set extracted using the newly developed methods provide information about the optimum subset of the hyperspectral hypertemporal data for a specific target detection application, which makes these methods serve as tools to strategize more intelligent data collection plans.

No description available.

No description available.

No description available.

Doctor of Philosophy / Department of Geography / Melinda Daniels / Geomorphic properties of streams are linked to ecosystem function through processes related to storage, transport, and other drivers regulating biogeochemical conditions. Disturbances altering the physical template of a stream are associated with cascading impacts on ecosystem function. However, few disturbances are studied at long time scales and so the legacy of such events and the implications for ecosystem structure and function are not well understood. This research investigates the role of historic tie-driving, a channel disturbance legacy, in shaping present-day stream channel conditions in the Rocky Mountain region and the associated implications for organic matter dynamics. Using a combination of geomorphic and riparian surveys, organic matter and vegetation sampling, and modeling, I show that components of mountain stream ecosystems have recovered from tie-driving at varying rates. First, I addressed how tie-driving has altered channel morphology and wood loading. Tie-driven streams are narrower, shallower, less rough, and have less wood than non-driven reference reaches. In a second study, I focused on differences in carbon storage within the stream and riparian area between tie-driven and non-driven streams. Carbon stored on the landscape represents a long-term component of the terrestrial carbon cycle and some, but not all, components have been impacted by tie-driving. Large instream wood, coarse downed wood, and fine downed wood were identified as carbon storage components that were significantly smaller in tie-driven stream-riparian corridors. Finally, I modeled whole stream ecosystem metabolism and tested whether abiotic drivers influenced variations in rates of gross primary productivity (GPP), ecosystem respiration (ER), and net ecosystem productivity (NEP). Results from this work suggest that rates of GPP were significantly different between tie-driven and non-driven streams and were partially explained by variations in light related to canopy structure. However, variations in ER and NEP were not significantly different between tie-driven and non-driven sites. Taken as a whole, this work shows that ecosystems bear the imprint of historic disturbances but individual ecosystem components recover at differing rates. Additionally, integrating stream hydro-geomorphic and ecological dynamics is an effective approach to understanding the impact of channel disturbances in shaping ecosystem function at a variety of spatial and temporal scales.

Freshwater is a resource under threat due to anthropogenic actions. Stream restoration is a common method for mitigating disturbance. Inconsistent methodologies used for evaluating restorations have drawn criticism. Limited use of baseline data for guiding stream restoration activities is of particular concern. This study was developed to elucidate metrics that differentiate reference and disturbed sites in Upper Coastal Plain streams. This information could improve resource use and successes of restorations. Structural and functional variables were examined in 10 reference and 10 streams that meet the traditional definition of disturbance and would be restoration priorities. Disturbed streams were classified into two regimes, temporal, based on time since disturbance, and categorical, based on disturbance cause. Some metrics of geomorphology, water chemistry and macroinvertebrates differentiated reference from disturbed regimes and while other metrics separated streams within disturbance regimes. Surprisingly, leaf decay rate was not an effective metric for determining disturbance. However, macroinvertebrate leaf pack colonizers were found to be useful for differentiating reference sites and disturbance regimes. Of the 10 disturbed streams this study examined, my data suggests that only three are in immediate need of restoration. This study emphasizes the importance of baseline data and its potential benefits for guiding stream restoration.