Natural regeneration is used to restock trembling aspen (Populus tremuloides Michx.) cutblocks and factors controlling regeneration are areas of interest and concern to the forest industry. Harvest operations in Manitoba require that coarse woody debris, or slash, be left and distributed in cutblocks. The objective of this study was to investigate the effects of slash loading on soil temperatures and aspen regeneration, and implications for harvest operations in the Duck Mountain area. Early sucker growth, initiation, and soil temperatures were surveyed in six winter and six summer cutblocks under different levels of slash loadings. A growth chamber study, using field temperature data as a guideline, examined the effects of diurnal temperature variation on sucker initiation and production. In winter and summer cutblocks, mean depths to sucker initiation from the parent root were 4.6 + 2.4 cm and 3.4 + 2.1 cm, respectively, and initiation of suckers occurred mainly from parental roots located in the LFH layer. Daily mean soil temperatures during the growing season were significantly lower under higher levels of slash (difference of 3.6 oC during May). Higher amounts of slash also significantly shortened the length of the growing season (89 fewer days above 0 oC in one season) and decreased the number of suckers produced (150 000 ha1 decreased to 14 000 ha-1), sucker volume (decreased by 256 cm3m-2) and leaf area index (decreased by 0.9). There was no difference in sucker production between any diurnal temperature treatments in the growth chamber study. Shallow depth to sucker initiation has important implications for harvest operations using heavy machinery especially those occurring during the summer season. Moderate levels of slash in summer cutblocks, and heavy levels of slash in winter cutblocks limit sucker growth. Although slash decreases diurnal temperature amplitudes, this may not be the reason for the decrease in sucker production associated with increased levels of slash. Both soil temperature and early sucker growth are strongly affected by slash loading; by monitoring harvest operations and the distribution of slash within cutblocks, the negative effect of heavy machine traffic and heavy piles of slash can be reduced and ensure successful forest regeneration.

Natural regeneration is used to restock trembling aspen (Populus tremuloides Michx.) cutblocks and factors controlling regeneration are areas of interest and concern to the forest industry. Harvest operations in Manitoba require that coarse woody debris, or slash, be left and distributed in cutblocks. The objective of this study was to investigate the effects of slash loading on soil temperatures and aspen regeneration, and implications for harvest operations in the Duck Mountain area. Early sucker growth, initiation, and soil temperatures were surveyed in six winter and six summer cutblocks under different levels of slash loadings. A growth chamber study, using field temperature data as a guideline, examined the effects of diurnal temperature variation on sucker initiation and production. In winter and summer cutblocks, mean depths to sucker initiation from the parent root were 4.6 + 2.4 cm and 3.4 + 2.1 cm, respectively, and initiation of suckers occurred mainly from parental roots located in the LFH layer. Daily mean soil temperatures during the growing season were significantly lower under higher levels of slash (difference of 3.6 oC during May). Higher amounts of slash also significantly shortened the length of the growing season (89 fewer days above 0 oC in one season) and decreased the number of suckers produced (150 000 ha1 decreased to 14 000 ha-1), sucker volume (decreased by 256 cm3m-2) and leaf area index (decreased by 0.9). There was no difference in sucker production between any diurnal temperature treatments in the growth chamber study. Shallow depth to sucker initiation has important implications for harvest operations using heavy machinery especially those occurring during the summer season. Moderate levels of slash in summer cutblocks, and heavy levels of slash in winter cutblocks limit sucker growth. Although slash decreases diurnal temperature amplitudes, this may not be the reason for the decrease in sucker production associated with increased levels of slash. Both soil temperature and early sucker growth are strongly affected by slash loading; by monitoring harvest operations and the distribution of slash within cutblocks, the negative effect of heavy machine traffic and heavy piles of slash can be reduced and ensure successful forest regeneration.

Quantification of total phosphorus (TP) loads entering a lake or reservoir is important because phosphorus is most often the limiting nutrient in terms of algae growth, thus phosphorus can control the extent of eutrophication. Four methods for assessing the annual tributary phosphorus loads to two different Virginia reservoirs were analyzed, three methods that use tributary monitoring program data and one that uses land-use and rainfall data. In this project, one tributary has been extensively monitored for many years and served as a control on which the other methods were tested. The key difference between this research and previous studies is the inclusion of flow-weighted composite storm sampling instead of simple grab sample analyses of storm flow. Three of the methods employed flow stratification, and the impact of the base flow separation point was examined. It was found that the Regression Method developed in this research was the least sensitive to the base flow separation point, which is a valuable attribute because a wrong choice will not significantly affect the estimate. The Monte Carlo Method was found to underestimate the TP loads. The amount of rainfall impacted the accuracy of the methods, with more error occurring in a year with lower precipitation. / Master of Science

An accurate estimate of the mass of a passenger vehicle is important for several safety systems and environmental aspects. In this thesis, an algorithm for estimating the mass of a passenger vehicle using the recursive least squares methodis presented. The algorithm is based on a physical model of the vehicle and is designed to be able to run in real-time onboard a vehicle and uses the wheel torque signal calculated in the electrical control unit in the engine. Therefore no estimation of the powertrain is needed. This is one contribution that distinguishes this thesis from previous work on the same topic, which has used the engine torque. The benefit of this is that no estimation of the dynamics in the powertrain is needed. The drawback of using this method is that the algorithm is dependenton the accuracy of the estimation done in the engine electrical control unit. Two different versions of the recursive least squares method (RLS) have been developed - one with a single forgetting factor and one with two forgetting factors. The estimation performance of the two versions are compared on several different real-world driving scenarios, which include driving on country roads, highways, and city roads, and different loads in the vehicle. The algorithm with a single forgetting factor estimates the mass with an average error for all tests of 4.42% and the algorithm with multiple forgetting factors estimates the mass with an average error of 4.15 %, which is in line with state-of-the-art algorithms that are presented in other studies. In a sensitivity analysis, it is shown that the algorithms are robust to changes in the drag coefficient. The single forgetting factor algorithm is robust to changes in the rolling resistance coefficient whereas the multiple forgetting factor algorithm needs the rolling resistance coefficient to be estimated with fairly good accuracy. Both versions of the algorithm need to know the wheel radius with an accuracy of 90 %. The results show that the algorithms estimate the mass accurately for all three different driving scenarios and estimate highway roads best with an average error of 2.83 % and 2.69 % for the single forgetting factor algorithm and the multiple forgetting factor algorithm, respectively. The results indicate it is possible to use either algorithm in a real-world scenario, where the choice of which algorithm depends on sought-after robustness.

Recent technological advances have witnessed the rapid encroachment of computing systems into our social spaces. Their acceptance in these social spaces by other occupants, however, might be mostly contingent on their social appropriateness. Notions of social appropriateness might seem vague but even people who don’t act on this commonsense knowledge, and accord to social norms, can sometimes find themselves ostracized from society. It is reflected in behavior that supports a sense of successful engagement and connection. Such behavior communicates a desire to be accepted and a willingness to engage, as opposed to inappropriateness that conveys indifference, rejection or even danger. As social actors, how can systems improve their interactions with us in order to better succeed at their tasks? Perhaps, more interestingly, how might they even improve our communications with each other? In this thesis we describe a framework to identify opportunities to design systems that can begin to act appropriately in social settings, which we call Considerate Systems. It includes a design process and guidelines, which allows an interaction to be viewed from the perspectives of the user, system and task. It also includes an architecture that guides the addition of productive social responses to interactive systems. We demonstrate the utility of this framework by exploring two types of scenarios that impact social interactions in contrasting ways. Remote interactions (such as on a conference call) suffer from an impinging of social cues that people rely on while communicating. On the other hand, situated multitasking interactions (such as texting while driving) can easily overwhelm users and detract from their performance. The framework is applied towards the design of autonomous agents tackling problems endemic to such scenarios. We evaluate their success with respect to specific scenario goals. We conclude by noting that while the challenges of instilling computing systems with a sense of appropriateness seem daunting, our productive use of systems can be enhanced with them.

Appropriately constructed pollutant export models can help set management priorities for catchments, identify critical pollutant source areas, and are important tools for developing and evaluating economically viable ways of minimising surface water pollution.¶ This thesis presents a comparison, an evaluation and an integration of models for predicting the export of environmental pollutants, in particular sediment, through river systems. A review of the capabilities and limitations of current water quality modelling approaches is made. Several water quality and quantity modelling approaches are applied and evaluated in the catchment of the upper Murrumbidgee River.¶ The IHACRES rainfall-runoff model and a simple hydrologic routing model are applied with the aim of developing a capacity to predict streamflow at various catchment scales and to enable integration with other pollutant load estimation techniques. Methods for calculating pollutant loads from observed pollutant concentration and modelled streamflow data are also investigated. Sediment export is estimated using these methods over a 10-year period for two case study subcatchments. Approaches for water quality sampling are discussed and a novel monitoring program using rising stage siphon samplers is presented. Results from a refinement of the Sediment River Network model in the upper Murrumbidgee catchment (SedNet-UM) are presented. The model provides a capacity to quantify sediment source, transport and to simulate the effects of management change in the catchment. The investigation of the model includes rigorous examination of the behaviour of the model through sensitivity assessment and comparison with other sediment modelling studies. The major conclusion reached through sensitivity assessment was that the outputs of the model are most sensitive to perturbation of the hydrologic parameters of the model.¶ The SedNet-UM application demonstrates that it is possible to construct stream pollutant models that assist in prioritising management across catchment scales. It can be concluded that SedNet and similar variants have much potential to address common resource management issues requiring the identification of the source, propagation and fate of environmental pollutants. In addition, incorporating the strengths of a conceptual rainfall-runoff model and the semi-distributed SedNet model has been identified as very useful for the future prediction of environmental pollutant export.

In electric power distribution systems, the major determinant in electricity supply strategy is the quantity of demand. Customers need to be accurately represented using updated nodal load information as a requirement for efficient control and operation of the distribution network. In Distribution Load Estimation (DLE), two major categories of data are utilized: historical data and direct real-time measured data. In this thesis, a comprehensive survey on the state-of-the-art methods for estimating loads in distribution networks is presented. Then, a novel method for representing historical data in the form of Representative Load Curves (RLCs) for use in realtime DLE is also described. Adaptive Neuro-Fuzzy Inference Systems (ANFIS) is used in this regard to determine RLCs. An RLC is a curve that represents the behavior of the load during a specified time span; typically daily, weekly or monthly based on historical data. Although RLCs provide insight about the variation of load, it is not accurate enough for estimating real-time load. This therefore, should be used along with real-time measurements to estimate the load more accurately. It is notable that more accurate RLCs lead to better real-time load estimation in distribution networks. This thesis addresses the need to obtain accurate RLCs to assist in the decision-making process pertaining to Radial Distribution Networks (RDNs).This thesis proposes a method based on Adaptive Neuro-Fuzzy Inference Systems (ANFIS) architecture to estimate the RLCs for Distribution Networks. The performance of the method is demonstrated and simulated, on a test 11kV Radial Distribution Network using the MATLAB software. The Mean Absolute Percent Error (MAPE) criterion is used to justify the accuracy of the RLCs.

Reservoir sedimentation is a major issue for water resource management around the world. It has serious economic, environmental, and social consequences, such as reduced water storage capacity, increased flooding risk, decreased hydropower generation, and deteriorated water quality. Increased rainfall intensity, higher temperatures, and more extreme weather events due to climate change are expected to exacerbate the problem of reservoir sedimentation. As a result, sedimentation must be managed to ensure the long-term viability of reservoirs and their associated infrastructure. Effective reservoir sedimentation management in the face of climate change necessitates an understanding of the sedimentation process and the factors that influence it, such as land use practices, erosion, and climate. Monitoring and modelling sedimentation rates are also useful tools for forecasting future impacts and making management decisions. The goal of this research is to create long-term reservoir management strategies in the face of climate change by simulating the effects of various reservoir-operating strategies on reservoir sedimentation and sediment delta movement at Mangla Reservoir in Pakistan (the second-largest dam in the country). In order to assess the impact of the Mangla Reservoir's sedimentation and reservoir life, a framework was developed. This framework incorporates both hydrological and morphodynamic models and various soft computing models. In addition to taking climate change uncertainty into consideration, the proposed framework also incorporates sediment source, sediment delivery, and reservoir morphology changes. Furthermore, the purpose of this study is to provide a practical methodology based on the limited data available. In the first phase of this study, it was investigated how to accurately quantify the missing suspended sediment load (SSL) data in rivers by utilizing various techniques, such as sediment rating curves (SRC) and soft computing models (SCMs), including local linear regression (LLR), artificial neural networks (ANN) and wavelet-cum-ANN (WANN). Further, the Gamma and M-test were performed to select the best-input variables and appropriate data length for SCMs development. Based on an evaluation of the outcomes of all leading models for SSL estimation, it can be concluded that SCMs are more effective than SRC approaches. Additionally, the results also indicated that the WANN model was the most accurate model for reconstructing the SSL time series because it is capable of identifying the salient characteristics in a data series. The second phase of this study examined the feasibility of using four satellite precipitation datasets (SPDs) which included GPM, PERSIANN_CDR, CHIRPS, and CMORPH to predict streamflow and sediment loads (SL) within a poorly gauged mountainous catchment, by employing the SWAT hydrological model as well as SWAT coupled soft computing models (SCMs) such as artificial neural networks (SWAT-ANN), random forests (SWAT-RF), and support vector regression (SWAT-SVR). SCMs were developed using the outputs of un-calibrated SWAT hydrological models to improve the predictions. The results indicate that during the entire simulation, the GPM shows the best performance in both schemes, while PERSIAN_CDR and CHIRPS also perform well, whereas CMORPH predicts streamflow for the Upper Jhelum River Basin (UJRB) with relatively poor performance. Among the best GPM-based models, SWAT-RF offered the best performance to simulate the entire streamflow, while SWAT-ANN excelled at simulating the SL. Hence, hydrological coupled SCMs based on SPDs could be an effective technique for simulating streamflow and SL, particularly in complex terrain where gauge network density is low or uneven. The third and last phase of this study investigated the impact of different reservoir operating strategies on Mangla reservoir sedimentation using a 1D sediment transport model. To improve the accuracy of the model, more accurate boundary conditions for flow and sediment load were incorporated into the numerical model (derived from the first and second phases of this study) so that the successive morphodynamic model could precisely predict bed level changes under given climate conditions. Further, in order to assess the long-term effect of a changing climate, a Global Climate Model (GCM) under Representative Concentration Pathways (RCP) scenarios 4.5 and 8.5 for the 21st century is used. The long-term modelling results showed that a gradual increase in the reservoir minimum operating level (MOL) slows down the delta movement rate and the bed level close to the dam. However, it may compromise the downstream irrigation demand during periods of high water demand. The findings may help the reservoir managers to improve the reservoir operation rules and ultimately support the objective of sustainable reservoir use for societal benefit. In summary, this study provides comprehensive insights into reservoir sedimentation phenomena and recommends an operational strategy that is both feasible and sustainable over the long term under the impact of climate change, especially in cases where a lack of data exists. Basically, it is very important to improve the accuracy of sediment load estimates, which are essential in the design and operation of reservoir structures and operating plans in response to incoming sediment loads, ensuring accurate reservoir lifespan predictions. Furthermore, the production of highly accurate streamflow forecasts, particularly when on-site data is limited, is important and can be achieved by the use of satellite-based precipitation data in conjunction with hydrological and soft computing models. Ultimately, the use of soft computing methods produces significantly improved input data for sediment load and discharge, enabling the application of one-dimensional hydro-morphodynamic numerical models to evaluate sediment dynamics and reservoir useful life under the influence of climate change at various operating conditions in a way that is adequate for evaluating sediment dynamics.:Chapter 1: Introduction Chapter 2:Reconstruction of Sediment Load Data in Rivers Chapter 3:Assessment of The Hydrological and Coupled Soft Computing Models, Based on Different Satellite Precipitation Datasets, To Simulate Streamflow and Sediment Load in A Mountainous Catchment Chapter 4:Simulating the Impact of Climate Change with Different Reservoir Operating Strategies on Sedimentation of the Mangla Reservoir, Northern Pakistan Chapter 5:Conclusions and Recommendations

Para implantação de diversas funções de controle e operação em tempo real em Sistemas de Distribuição (SDs), como, por exemplo, restabelecimento de energia, é necessário um procedimento para representar a carga em tempo real. Ou seja, uma metodologia que possibilite a estimação em tempo real das demandas dos transformadores de distribuição que em geral não são monitoradas de forma direta. Para esse fim propõe-se, neste trabalho, um Estimador de Demanda em Tempo Real (EDTR) baseado em: informações off-line (consumo mensal dos consumidores e curvas de carga típicas); um algoritmo computacionalmente eficiente para cálculo de fluxo de potência baseado na estrutura de dados denominada Representação Nó-Profundidade (RNP); e nas poucas medidas disponíveis em tempo real nos SDs. O EDTR proposto opera em dois estágios: (1) Estimação Off-line das Demandas; e (2) Refinamento em Tempo Real das Demandas, executados em instantes diferentes (um de maneira off-line e outro em tempo real), de forma a prover uma estimativa das demandas dos transformadores de distribuição. Considerando somente as informações off-line, o EDTR proposto permite a estimação das demandas dos transformadores de distribuição com uma medida da incerteza da estimativa. Através do processamento das medidas disponíveis em tempo real, via um algoritmo eficiente para cálculo de fluxo de potência, o EDTR proposto permite o refinamento das estimativas off-line. Neste trabalho serão apresentados resultados de diversas simulações computacionais demonstrando a eficiência do EDTR proposto. Alguns parâmetros são avaliados quanto à influência nas estimativas do EDTR proposto, como a presença de erros grosseiros nas medidas disponíveis em tempo real e alimentadores somente com medidas de magnitude de corrente. Além disto, destaca-se a influência da qualidade das estimativas iniciais obtidas pelo Estágio (1), e a importância das hipóteses estatísticas utilizadas nesse estágio para o processo de estimação. Apresenta-se, ainda, a aplicação do EDTR proposto em um SD real brasileiro. Um teste de validação foi realizado através de uma campanha de medição em um alimentador real, que consistiu na instalação de medidores de demanda em três transformadores de distribuição para aferir a qualidade das estimativas obtidas pelo EDTR proposto. Finalmente, o EDTR proposto é aplicado em um SD real de larga escala, para aferir o desempenho computacional da metodologia implantada e as dificuldades de implantação. Vale ressaltar que sua implantação é condizente com ferramentas consolidadas nos Centros de Operação da Distribuição, como o uso do processo de agregação de cargas e o cálculo de fluxo de potência, e poucas rotinas precisam ser adicionadas para integração do EDTR. / Several real time control and operation applications for Distribution Systems (DS), such as, service restoration, require a procedure for real time load modeling. That is, a methodology for real time estimation of the distribution transformers loading which are generally not monitored. For this purpose, in this dissertation, a Real Time Load Estimator (RTLE) is proposed based on: off-line information (monthly consumption and typical load curves); a computationally efficient algorithm for power flow calculation based on the data structure called Node-Depth Encoding; and on the few available real time measurements on the distribution system. The proposed RTLE operates in two stages: (1) Off-line Load Estimation and (2) Real Time Load Refinement, performed in different moments (one off-line and the other in real time), providing the distribution transformers load estimates. Using only the offline information, the proposed RTLE allows the estimation of the loads of the distribution transformers with a measure of uncertainty. By processing the available real time measurements, using an efficient power flow calculation algorithm, the proposed RTLE refines these off-line estimates. This dissertation presents several simulations showing the efficiency of the proposed RTLE. Some parameters are evaluated and their influence on the RTLE load estimates, such as gross errors in the available real time measurements and feeders with only current magnitude measurements. Besides, it is emphasized the influence of the initial load estimates obtained from Stage (1), and the importance of the statistical hypothesis used in this stage in the load estimation process. Also, this work presents the application of the proposed RTLE in a real Brazilian DS. A validation test was performed through in-field verification in a real distribution feeder, which was executed via load meters installation in three distribution transformers to evaluate the quality of the load estimates provided by the RTLE. Finally, the proposed RTLE was tested in a real large scale DS to evaluate its computational performance and the difficult level of its implementation. It is noteworthy that its implementation is straightforward with other Distribution Operation Center tools, such as load aggregation and load flow calculation, and few routines must be added for integrating the RTLE.

Para implantação de diversas funções de controle e operação em tempo real em Sistemas de Distribuição (SDs), como, por exemplo, restabelecimento de energia, é necessário um procedimento para representar a carga em tempo real. Ou seja, uma metodologia que possibilite a estimação em tempo real das demandas dos transformadores de distribuição que em geral não são monitoradas de forma direta. Para esse fim propõe-se, neste trabalho, um Estimador de Demanda em Tempo Real (EDTR) baseado em: informações off-line (consumo mensal dos consumidores e curvas de carga típicas); um algoritmo computacionalmente eficiente para cálculo de fluxo de potência baseado na estrutura de dados denominada Representação Nó-Profundidade (RNP); e nas poucas medidas disponíveis em tempo real nos SDs. O EDTR proposto opera em dois estágios: (1) Estimação Off-line das Demandas; e (2) Refinamento em Tempo Real das Demandas, executados em instantes diferentes (um de maneira off-line e outro em tempo real), de forma a prover uma estimativa das demandas dos transformadores de distribuição. Considerando somente as informações off-line, o EDTR proposto permite a estimação das demandas dos transformadores de distribuição com uma medida da incerteza da estimativa. Através do processamento das medidas disponíveis em tempo real, via um algoritmo eficiente para cálculo de fluxo de potência, o EDTR proposto permite o refinamento das estimativas off-line. Neste trabalho serão apresentados resultados de diversas simulações computacionais demonstrando a eficiência do EDTR proposto. Alguns parâmetros são avaliados quanto à influência nas estimativas do EDTR proposto, como a presença de erros grosseiros nas medidas disponíveis em tempo real e alimentadores somente com medidas de magnitude de corrente. Além disto, destaca-se a influência da qualidade das estimativas iniciais obtidas pelo Estágio (1), e a importância das hipóteses estatísticas utilizadas nesse estágio para o processo de estimação. Apresenta-se, ainda, a aplicação do EDTR proposto em um SD real brasileiro. Um teste de validação foi realizado através de uma campanha de medição em um alimentador real, que consistiu na instalação de medidores de demanda em três transformadores de distribuição para aferir a qualidade das estimativas obtidas pelo EDTR proposto. Finalmente, o EDTR proposto é aplicado em um SD real de larga escala, para aferir o desempenho computacional da metodologia implantada e as dificuldades de implantação. Vale ressaltar que sua implantação é condizente com ferramentas consolidadas nos Centros de Operação da Distribuição, como o uso do processo de agregação de cargas e o cálculo de fluxo de potência, e poucas rotinas precisam ser adicionadas para integração do EDTR. / Several real time control and operation applications for Distribution Systems (DS), such as, service restoration, require a procedure for real time load modeling. That is, a methodology for real time estimation of the distribution transformers loading which are generally not monitored. For this purpose, in this dissertation, a Real Time Load Estimator (RTLE) is proposed based on: off-line information (monthly consumption and typical load curves); a computationally efficient algorithm for power flow calculation based on the data structure called Node-Depth Encoding; and on the few available real time measurements on the distribution system. The proposed RTLE operates in two stages: (1) Off-line Load Estimation and (2) Real Time Load Refinement, performed in different moments (one off-line and the other in real time), providing the distribution transformers load estimates. Using only the offline information, the proposed RTLE allows the estimation of the loads of the distribution transformers with a measure of uncertainty. By processing the available real time measurements, using an efficient power flow calculation algorithm, the proposed RTLE refines these off-line estimates. This dissertation presents several simulations showing the efficiency of the proposed RTLE. Some parameters are evaluated and their influence on the RTLE load estimates, such as gross errors in the available real time measurements and feeders with only current magnitude measurements. Besides, it is emphasized the influence of the initial load estimates obtained from Stage (1), and the importance of the statistical hypothesis used in this stage in the load estimation process. Also, this work presents the application of the proposed RTLE in a real Brazilian DS. A validation test was performed through in-field verification in a real distribution feeder, which was executed via load meters installation in three distribution transformers to evaluate the quality of the load estimates provided by the RTLE. Finally, the proposed RTLE was tested in a real large scale DS to evaluate its computational performance and the difficult level of its implementation. It is noteworthy that its implementation is straightforward with other Distribution Operation Center tools, such as load aggregation and load flow calculation, and few routines must be added for integrating the RTLE.