No description available.

We have used broad-band near infrared photometry in combination with optical Johnson-Cousins photometry to study the dust properties in the line of sight to the Galactic globular cluster M4. These data have been used to investigate the reddening effects in terms of absolute strength, distribution and variations across the cluster field, as well as the shape of the reddening law defined by the type of dust. All three aspects were poorly defined for this system and therefore there has been controversy about the absolute distance to the globular cluster which is closest to the sun. Here, we introduce a new method to determine the ratio of absolute to selective extinction (RV ) in the line of sight toward resolved stellar populations, which is known to be a useful indicator for the type of dust and therefore characterizes the applicable reddening law. This method is independent of age assumptions and appears to be significantly more precise and accurate than existing approaches. In a first application, we determine AV /E(B − V ) = 3.76 ± 0.07 (random error) for the dust in the line of sight to M4 for our set of filters. That corresponds to a dust-type parameter RV = 3.62 ± 0.07 in the Cardelli, Clayton & Mathis (1989) reddening law. With this value, the distance to M4 is found to be d = 1.80 ± 0.05 kpc, corresponding to a true distance modulus of (m − M)0 = 11.28 ± 0.06. These uncertainties do not include possible systematic errors in the theoretical isochrones. A reddening map for M4 has been created which reveals a spatial differential reddening of δE(B − V ) ≥ 0.2 mag across the field within 10′ around the cluster centre; this is about 50% of the total mean reddening, which has been determined to be E(B − V ) = 0.37 ± 0.01. In order to provide accurate zero points for the extinction coefficients of our photometric filters, a computer code has been written to investigate the impact of stellar parameters such as temperature, surface gravity and metallicity on the extinction properties and the necessary corrections in different bandpasses. Using both synthetic ATLAS9 spectra and observed spectral energy distributions, we found similar sized effects for the range of temperature and surface gravity typical of globular cluster stars: both cause a change of about 3% in the necessary correction factor for each filter combination. Interestingly, variations in the metallicity cause effects of the same order when the assumed value is changed from the solar metallicity ([Fe/H] = 0.0) to [Fe/H]=-2.5. Our analysis showed that the systematic differences between the flux of a typical main-sequence turnoff star in a metal poor globular cluster and a Vega-like star are even stronger (∼ 5%). We compared the results from synthetic spectra to those obtained with observed spectral energy distributions and found significant differences in detail for temperatures lower than 5 000 K. We have attributed these discrepancies to the inadequate treatment of molecular bands in the B filter within the ATLAS9 models. Accordingly, for those cooler temperatures we obtained corrections for temperature, gravity and metallicity primarily from the observed spectra. Fortunately, these differences do not affect our principal astrophysical conclusions in this study, which are based on stars hotter than 5 000 K. / Graduate

Data from a new, wide field, coincident optical and X-ray survey, the X-ray Dark Cluster Survey (XDCS) are presented. The aim was to conduct simultaneous and independent searches for clusters of galaxies in the optical and X-ray passbands. Optical cluster detection algorithms implemented on the data are detailed. This resulted in catalogues of 185 I-band selected, 290 colour selected and 15 X-ray selected systems, residing in of optical + X-ray imaging. The relationship between optical (L(_E)) and X-ray luminosity (L(_x) ) was examined and found to exhibit significant scatter. This study highlights the higher efficiency and resolution of optical colour selection compared with other cluster detection methods. Spectroscopic redshifts confirmed the reality of a subsample of systems which were found with the optical algorithms, but required to have no detectable X-ray emission. These systems show comparable optical luminosity to the most X-ray luminous clusters, but orders of magnitude lower X-ray emission. This is consistent with the large scatter seen in the L(_x)-L(_e) relation. A near-infrared multicolour technique was explored and extended to search for high redshift (z>l) clusters. Finally, application of such techniques to forthcoming wide field near-infrared surveys was discussed and predictions for cluster finding in such surveys made.

Single nucleotide polymorphisms (SNPs) have been increasingly popular for a wide range of genetic studies. A high-throughput genotyping technologies usually involves a statistical genotype calling algorithm. Most calling algorithms in the literature, using methods such as k-means and mixturemodels, rely on elliptical structures of the genotyping data; they may fail when the minor allele homozygous cluster is small or absent, or when the data have extreme tails or linear patterns. We propose an automatic genotype calling algorithm by further developing a linear grouping algorithm (Van Aelst et al., 2006). The proposed algorithm clusters unnormalized data points around lines as against around centroids. In addition, we associate a quality value, silhouette width, with each DNA sample and a whole plate as well. This algorithm shows promise for genotyping data generated from TaqMan technology (Applied Biosystems). A key feature of the proposed algorithm is that it applies to unnormalized fluorescent signals when the TaqMan SNP assay is used. The algorithm could also be potentially adapted to other fluorescence-based SNP genotyping technologies such as Invader Assay. Motivated by the SNP genotyping problem, we propose a partial likelihood approach to linear clustering which explores potential linear clusters in a data set. Instead of fully modelling the data, we assume only the signed orthogonal distance from each data point to a hyperplane is normally distributed. Its relationships with several existing clustering methods are discussed. Some existing methods to determine the number of components in a data set are adapted to this linear clustering setting. Several simulated and real data sets are analyzed for comparison and illustration purpose. We also investigate some asymptotic properties of the partial likelihood approach. A Bayesian version of this methodology is helpful if some clusters are sparse but there is strong prior information about their approximate locations or properties. We propose a Bayesian hierarchical approach which is particularly appropriate for identifying sparse linear clusters. We show that the sparse cluster in SNP genotyping datasets can be successfully identified after a careful specification of the prior distributions.

Nanoclusters have special importance in nanotechnology because of their low dimensionality, which provides electronic, chemical, and magnetic properties that differ from those of the equivalent bulk materials. Suitably controllable self-assembly methods are required in order to incorporate nanoclusters into useful devices. The self-assembly method used in this study employs V-grooves as a template element for nanocluster device fabrication. The V-grooves are fabricated by optical lithography on SiO2/Si wafers and KOH wet etching. Bi clusters deposited on a V-groove form a self-assembled conducting wire. The clusters are produced using an inert gas aggregation source inside an ultra high vacuum compatible system. In order to characterise the assembly process, Bi clusters with different average sizes and velocities are deposited on V-grooves with different widths. The cluster bouncing was found to be the main process in forming the cluster wires. The bouncing angles were smaller than the incident angle, and they are dependent on the cluster size and velocity. For a certain bouncing angle, the wire width reflects the V-groove width because of the fixed bouncing angle. Nanocluster devices were fabricated by depositing the clusters on V-grooves with pre-formed Au/NiCr electrical contacts. The amount of the deposited material required to form an electrically conducting wire was found to be a function of the V-groove width and the wire length. The two point I(V) measurements in the voltage range between -1 and +1V showed linear characteristics for low resistance wires (kΩ), and non-linear characteristics for the high resistance ones (MΩ). The silicon substrate was used as a back gate. Applying a voltage to the gate was found to modify the electrical conduction of the cluster wire. The temperature dependence of the resistance of the nanocluster wires was studied in the temperature range of 4.2-473K, and all of the measured wires showed a negative temperature coefficient of resistance. These measurements allowed a detailed study of the conduction mechanisms through the cluster wires. The study showed that Bi clusters can be used for device fabrication. To size select the clusters prior to using them for the device fabrication, a high transmission mass filter is required. This transmission can be obtained using the von Issendorff and Palmer mass filter if it is operated using the optimum operation conditions. The mass filter consists of two pairs of parallel plates with horizontal openings in Plates 1 and 2, and it operates on the time of flight principle. During this project, the operation conditions of this mass filter were studied using both experiment and simulation. The study showed that the beam deflection angle is a critical factor in optimising the mass filter transmission efficiency. This angle is dependent on the accelerating voltage, ion mass, and the horizontal velocity of the ions. The optimum operation conditions for the mass filter were found and used to study the mass distribution of Pd ions produced by a magnetron sputtering source with variable cluster aggregation length.

Clustering is the search for those partitions that reflect the structure of an object set. Traditional clustering algorithms search only a small sub-set of all possible clusterings (the solution space) and consequently, there is no guarantee that the solution found will be optimal. We report here on the application of Genetic Algorithms (GAs) -- stochastic search algorithms touted as effective search methods for large and complex spaces -- to the problem of clustering. GAs which have been made applicable to the problem of clustering (by adapting the representation, fitness function, and developing suitable evolutionary operators) are known as Genetic Clustering Algorithms (GCAs). There are two parts to our investigation of GCAs: first we look at clustering into a given number of clusters. The performance of GCAs on three generated data sets, analysed using 4320 differing combinations of adaptions, establishes their efficacy. Choice of adaptions and parameter settings is data set dependent, but comparison between results using generated and real data sets indicate that performance is consistent for similar data sets with the same number of objects, clusters, attributes, and a similar distribution of objects. Generally, group-number representations are better suited to the clustering problem, as are dynamic scaling, elite selection and high mutation rates. Independent generalised models fitted to the correctness and timing results for each of the generated data sets produced accurate predictions of the performance of GCAs on similar real data sets. While GCAs can be successfully adapted to clustering, and the method produces results as accurate and correct as traditional methods, our findings indicate that, given a criterion based on simple distance metrics, GCAs provide no advantages over traditional methods. Second, we investigate the potential of genetic algorithms for the more general clustering problem, where the number of clusters is unknown. We show that only simple modifications to the adapted GCAs are needed. We have developed a merging operator, which with elite selection, is employed to evolve an initial population with a large number of clusters toward better clusterings. With regards to accuracy and correctness, these GCAs are more successful than optimisation methods such as simulated annealing. However, such GCAs can become trapped in local minima in the same manner as traditional hierarchical methods. Such trapping is characterised by the situation where good (k-1)-clusterings do not result from our merge operator acting on good k-clusterings. A marked improvement in the algorithm is observed with the addition of a local heuristic.

Thesis (M.S.)--West Virginia University, 2007. / Title from document title page. Document formatted into pages; contains vi, 59 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 57-59).

Zugl.: Braunschweig, Techn. Univ., Diss., 2005

Zugl.: Karlsruhe, Univ., Diss., 2005

Zugl.: Cottbus, Techn. University, Diss., 2005.