Thesis (Ph. D.)--University of Hawaii at Manoa, 2005. / Includes bibliographical references (leaves 197-202).

Thesis--University of Florida. / Description based on print version record. Typescript. Vita. Includes bibliographical references (leaves 220-222).

Thesis (M. Eng.)--Carleton University, 1986. / Also available in electronic format on the Internet.

This thesis discusses the convergence of nets through a series of theorems and proofs.

Exploitation of Skeena River sockeye salmon (Oncorhynchus nerka) has been conducted almost solely by drift gillnets since inception of the commercial fishery in 1877. Selective action of gillnets is a factor which may have contributed to a long term decline in sockeye production and to other features of the population biology. This study was designed to determine the selective properties of nylon gillnets presently in use; to compare these with properties of linen nets used prior to 1955; to adjust age composition estimates of escapement levels prior to 1946; and to re-examine brood year production. In addition, the selective action was examined of the fishery as a whole on sockeye and pink salmon (O. gorbuscha) in 1968. The selective properties of six nylon gillnets ranging in mesh size between 4-5/8 inches and 5-5/8 inches, which corresponded with sizes normally used in the commercial fishery, were to be determined through a simultaneous fishing experiment, and selectivity curves were computed by Holt's (1963) normal probability technique. Unique selectivity curves for each mesh size could not be determined from the sockeye data. Mean size of age class l.2 sockeye (representing about 12 per cent of the sample) increased with mesh size but mean size of age class 1.3 sockeye (representing about 82 per cent of the sample) demonstrated no trend. Age class 1.3 sockeye were among the largest on record and it was postulated that fish of this age class were too large to gill properly in all mesh sizes used. A comparison of the predominant mesh of nylon gillnet (5-1/4 inch) with linen nets of the mesh used in the historic fishery (5-5/8 inch) was also influenced by the large size of age class 1.3 sockeye. Nylon nets were 2.5 and 2.7 times as efficient as linen for sockeye, and 8.0 and 9.0 times as effective for pink salmon. Nylon gillnets, although smaller, caught larger sockeye and pink salmon than did the linen gillnets. Variances about mean size of sockeye and pink salmon were also greater for catches in nylon as opposed to linen gillnets. In total, the selective properties of the commercial fishery reflected the results obtained in the nylon gillnet multi-mesh experiment. An overall selectivity curve computed for sockeye salmon by the direct method of Regier and Robson (1966) approximated a skew-normal distribution. Examination of the age-sex class components of the stock indicated that selection increased linearly over the length range of age 1.2 sockeye of both sexes, and decreased linearly over the length range of age class 1.3 sockeye. A selectivity curve was also computed for pink salmon and this curve, in total, also assumed a normal shape in spite of the extremely small size of pink salmon in 1968. The length-girth relationships of the two species were shown to differ and this accounted for most of the difference between the selectivity curves. The relationship between maximum efficiency and the maximum girth to perimeter mesh measure closely approximated the value of 1.2: 1.0 reported by McCombie and Berst (1969) for other species. Retention by gillnets declined once the girth/mesh ratio exceeded 1.2 for sockeye. For pink salmon, no females were of a size to equal this ratio; the descending limb of the selectivity curve was due solely to males as the retention rates declined once girth/mesh ratio exceeded 1.0. These findings suggest that in most years the gillnet fishery on the Skeena River would tend to select relatively larger sockeye salmon. In years such as 1968, however, selection would be against smaller fish. This frequent reversal combined with the intense modern fishery, which tends to remove virtually all fish during a short period and allows almost complete escapement in periods between fishing, suggested that selective fishing has probably not been a significant factor in decreased production. / Science, Faculty of / Zoology, Department of / Graduate

No description available.

No description available.

An outbreak of an infectious illness can have a devastating impact on a population. Once confirmed, local health care organizations will attempt to reduce the spread of the disease by adopting a set of pre-defined guidelines. Modelling such a system presents a number of unique challenges: timing and probability constraints must be captured, scaling must be seamless and methods for analysis must be robust and efficient. To satisfy these requirements, an augmented form of Petri net known as a choice-point net is introduced in this thesis. In this data structure, timing is associated with event-based transitions that may fire multiple times to simulate the same event occurring several times in parallel. Events may result in several possible outcomes, or choices, each of which is given a probability of occurrence. A choice-point net may be scaled without requiring structural changes to the model and may be analyzed by unravelling it into a finite-state automaton representing (perhaps portions of) its behaviour. By translating questions about the protocol into the mathematical language of the net, recursive algorithms may then be employed to provide health-care professionals with answers to their questions. To demonstrate the expressiveness of choice-point nets, an actual, in-use protocol to control respiratory infection outbreaks in long-term care homes is modelled. Three similar abridged scenarios set in a small long-term care home are also modelled, analyzed and compared. / Thesis (Ph.D, Computing) -- Queen's University, 2011-09-27 21:23:16.13

No description available.

No description available.