Satellite-based navigation systems make it possible to determine the relative positions of points on the earth with centimetre or even millimetre level accuracy over baselines of up to several thousand kilometres. The highest possible accuracy can only be achieved if the carrier phase integer ambiguities can be resolved. In order to resolve the L1 and L2 integer ambiguities over long GPS baselines, the double difference residual ionospheric errors must be estimated for every satellite, every epoch. The resulting number of parameters is usually too large for estimation using ordinary least squares to be practical due to the time or computing resources needed for the processing. The technique currently used to efficiently estimate the parameters is known as pre-elimination. Pre-elimination divides the unknowns into parameters of interest (the coordinates and ambiguities) and nuisance parameters (the ionospheric path delays). The nuisance parameters are treated as stochastic variables and modelled as process noise, avoiding the need for them to be explicitly estimated. Whilst this approach is highly efficient, it makes assumptions about the stochastic behaviour of the residual ionospheric error that are not necessarily valid. The effectiveness of preelimination can be increased through the use of a deterministic model of the ionosphere. It is the hypothesis of this research that the ionospheric error can be more effectively estimated than is possible with pre-elimination, leading to more reliable ambiguity resolution for long baseline precise positioning. (For complete abstract open document)

Red deer (Cervus elaphus) is one of at least 31 herbivorous exotic mammals existing in New Zealand. All of these species have the potential to affect environmentral and production values. Reducing their impacts on their values, strengthening effective managements are important issues to a variety of agencies within New Zealand including the Department of Conservation (DOC), Ministry of Agriculture and Forestry (MAF), local and regional governments. This research studied animal movement pattern and habitat use of 2 GPScollared red deer in the Canterbury high country and found (1) deer movement was affected by climatic variables such as rainfall and temperature, which had positive or negative effect on it, and had seasonal variation; (2) deer had dominant landcover use categories, depending on climate, season, and individual characteristics (3) deer had different movement patterns in terms of hourly distances. The purpose of the study is to draw accurate inferences from spatially explicit data for biosecurity managers and policy-makers through: (1) using global positioning system (GPS) as a tool to elucidate the application of GPS on red deer in wildlife management; (2) Animal Movement Analysis Arc View® 3.2 Extension under Arc View® Geographic Information System (GIS); (3) Animal movement analysis which used Generalized Additive Models (GAMs) to show how the movement of red deer was affected by different periods of time, seasons months and climatic variables (for example, rainfall and temperature).

Zugl.: Göttingen, Universiẗat, Diss., 2008.

Thèse (M.Sc.)--Université Laval, 2006. / Titre de l'écran-titre (visionné le 13 févr. 2008). Bibliogr.

Thesis (M.S.)--Ohio State University, 2008. / Title from first page of PDF file. Includes bibliographical references.

Thèses (M.Sc.)--Université de Sherbrooke (Canada), 2004. / Titre de l'écran-titre (visionné le 20 juin 2006). Publié aussi en version papier.

Thesis (Ph.D.)--Ohio University, August, 2004. / Includes bibliographical references (p. 177-183)

Zugl.: Stuttgart, Univ., Diss., 2003.

Zugl.: Giessen, Univ., Univ. Diss., 2009

Zugl.: Cottbus, Techn. Univ., Diss., 2008