No description available.

No description available.

It has been a practice to use seawater for preparing acid in offshore operations where fresh water is relatively expensive or logistically impossible to use. However, hydrochloric acid will release calcium ion into solution, which will combine with sulfate ion in seawater (greater than 3000 ppm) and calcium sulfate will precipitate once it exceeds its critical scaling tendency. A few studies have provided evidence for this problem and how to address this problem has not been fully examined. Core flood tests were conducted using Austin Chalks cores (1.5 in. x 6 in. and 1.5 in. x 20 in.) with permeability 5 md to investigate the effectiveness of scale inhibitor. A synthetic seawater was prepared according to the composition of seawater in the Arabian Gulf. Calcium, sulfate ions, and scale inhibitor concentrations were analyzed in the core effluent samples. Solids collected in the core effluent samples were analyzed using X-ray photoelectron spectroscopy (XPS) technique and thermodynamic calculation using OLI Analyzer software were conducted to identify the critical scaling tendency of calcium sulfate at different temperatures. Results showed that calcium sulfate precipitation occurred when seawater was used in any stage during matrix acidizing including preflush, post-flush, or in the main stage. Injection rate was the most important parameter that affected calcium sulfate precipitation; permeability reduction was significant at low flow rates, while at high rates wormhole breakthrough reduced the severity of the problem. More CaSO4 precipitated at high temperatures, accounting for more significant permeability reduction in the cores. The values of critical scaling tendency at various temperatures calculated by OLI ScaleChem 4.0.3 were believed to be 2.1, 2.0, and 1.2 respectively. A scale inhibitor (a sulfonated terpolymer) was found to be compatible with hydrochloric acid systems and can tolerate high concentration of calcium (30,000 mg/l). Analysis of core effluent indicated that the new treatment successfully eliminated calcium sulfate scale deposition. The concentration of scale inhibitor ranged from 20 to 250 ppm, depending on the scaling tendencies of calcium sulfate. This work confirms the damaging effect of preparing hydrochloric acid solutions using seawater on the permeability of carbonate cores. Therefore, it is recommended to use fresh water instead of seawater to prepare HCl acids whenever possible. If fresh water is not available, then a proper scale inhibitor should be added to the acids to avoid calcium sulfate precipitation.

No description available.

No description available.

Statistical machine learning methods provide us with a principled framework for extracting meaningful information from noisy high-dimensional data sets. A significant feature of such procedures is that the inferences made are statistically significant, computationally efficient and scientifically meaningful. In this thesis we make several contributions to such statistical procedures. Our contributions are two-fold. We first address prediction and estimation problems in non-standard situations. We show that even when given no access to labeled samples, one can still consistently estimate error rate of predictors and train predictors with respect to a given (convex) loss function. We next propose an efficient procedure for predicting with large output spaces, that scales logarithmically in the dimensionality of the output space. We further propose an asymptotically optimal procedure for sparse multi-task learning when the tasks share a joint support. We show consistency of the proposed method and derive rates of convergence. We next address the problem of learning meaningful representations of data. We propose a method for learning sparse representations that takes into account the structure of the data space and demonstrate how it enables one to obtain meaningful features. We establish sample complexity results for the proposed approach. We then propose a model-free feature selection procedure and establish its sure-screening property in the high dimensional regime. Furthermore we show that with a slight modification, the approach previously proposed for sparse multi-task learning enables one to obtain sparse representations for multiple related tasks simultaneously.

By counting 1's in the "right half" of 2w consecutive rows, we locate the main diagonal of any doubly infinite permutation matrix with bandwidth w. Then the matrix can be correctly centered and factored into block-diagonal permutation matrices. Part II of the paper discusses the same questions for the much larger class of band-dominated matrices. The main diagonal is determined by the Fredholm index of a singly infinite submatrix. Thus the main diagonal is determined "at infinity" in general, but from only 2w rows for banded permutations.

In this thesis we consider the N=1 super Yang-Mills theory on S5 with a single hypermultiplet in the adjoint representation. We argue that there is a critical value of the hypermultiplet mass M=3/2r, where r is the radius of S5, for which the free energy vanishes, and we study the model in the proximity of this value. For large N we provide analytical results for the free energy and the eigenvalue density in the weak and strong coupling limits, and in one case we solve the saddle point equation using a technique introduced by Hoppe. We present numerical results to show where each approximation is justified, and to explore the regimes where the model cannot be solved analytically. Based on the numerical results, we argue that in most cases the behaviour of the model is better understood in terms of an effective coupling constant λ'=λM. For small M the model simplifies to one whose kernel is non-singular. This simplified model shows a peculiar peak structure in the eigenvalue distribution, with the number of peaks growing as the effective coupling is increased. We interpret this as a series of phase transitions as M approaches 3/2r.

Finding a common language in describing and interpreting multivariate data associated with rehabilitation and disturbance ecology, has became a major challenge. The main objective of this study is to find and evaluate mathematical models to describe ecosystem change based on selected indicators of change. Existing data from a previous rehabilitation project on Hendrina Power Station (Mpumalanga, South Africa) was used as a database for this study and this study aims to report on the development of models concentrating on radar graphs and a model based on matrix mathematics. The main groups of organisms selected for the construction of models, were vegetation, soil mesofauna and ant species. The datasets were limited to some indicative species and their mean abundances were determined. The grids that were used were randomly chosen and the models were constructed. Radar graphs were constructed to model the suite of species identified, through a sensitivity analysis, to indicate possible rehabilitation success over time and was applied to the different rehabilitation ages. The surface areas under the radar graphs were determined and compared for the different rehabilitation ages in the same year of survey. Correlation graphs were drawn between the surface area and the rehabilitation ages. These graphs did not indicate much relevance in indicating rehabilitation success, but the radar graphs proved to be good indicators of change in abundance of the selected species over time. iv The vegetation species, Eragrostis curvula, was the only species that showed a strong significant positive relationship with rehabilitation age and could be considered a good rehabilitation species and indicator of rehabilitation success. After the evaluation of this model, Eragrostis curvula, and two additional ant species, Tetramorium setigerum and Lepisiota laevis, were added. These species that were added, showed an increase in abundance over time, as found in a previous study. These radar graphs also did not indicate much relevance and it can be concluded that the radar graphs can only be used for a visual representation of the changes in abundance of the relevant species over time. This study also refers to a matrix model. This model focused on the interactions between the different variables selected. The percentage carbon in the soil were also added to the list of species. Model fitting graphs were constructed and correlations were drawn between the species that had significant values in the interaction table. This model could be useful for future studies, but more data and replication is necessary, over a longer period of time. This will serve to eliminate possible shortcomings of the model. / Thesis (M. Environmental Science (Biodiversity and Conservation Biology))--North-West University, Potchefstroom Campus, 2007.

Avian point counts were conducted in tall-grass prairie fragments and adjacent grassland and agricultural matrix habitat in southern Manitoba. Bird density/abundance was compared between habitat types, while variables within prairie at local, patch or landscape level were modeled to determin avian habitat selection. Prairies and matrix grassland habitat supported the same number of species in both years, and densities of all focal bird species were the same in non-native grasslands as compared with native tall-grass prairies. Overall species richness in tall-grass prairies was mainly driven by vegetation variables. Variable responses to habitat structure and composition between avian species indicate that managing grasslands to promote heterogeneity is important to sustain a diverse assemblage of avian species. As individual species were affected most strongly by vegetation structure and richness, it follows that management of prairie vegetation through techniques such as grazing and prescribed burning could optimize habitat usability for birds.