Publication: Análisis de series temporales en estaciones permanentes GPS
Loading...
Official URL
Full text at PDC
Publication Date
2016-04-15
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Complutense de Madrid
Citations
Exportar
Abstract
En los ultimos años el número de estaciones permanentes GPS ha aumentado considerablemente para diferentes aplicaciones geodésicas y se comienza a disponer de series de coordenadas suficientemente largas como para poder aplicar ánalisis de series temporales. El principal objetivo de esta tesis es estudiar la influencia de la red de estaciones utilizada, marco de referencia y/o estrategia de procesado en las series temporales de coordenadas de estaciones CGPS y todos aquellos parámetros que se obtienen a partir de ellas, como la velocidad o deformaciones. Para conseguirlo se han utilizado varios conjuntos de series temporales de estaciones CGPS, todas ellas en las Islas Canarias y sus alrededores, obtenidos a partir de diferentes procesados. Se han empleado dos estrategias de procesado diferentes, una de Dobles Diferencias realizada con Bernese v4.2 (DD4.2) y otra de Posicionamiento Puntual Preciso (PPP) con Bernese v5.0 para las series del 2002 al 2006. Posteriormente, considerando una red con un mayor número de estaciones, se ha realizado un reprocesado dobles diferencias con Bernese v5.0, del 2002 al 2008. Estas series están afectadas por el cambio de ITRF2000 a ITRF2005 y modelo de calibración del centro de fase de antenas en noviembre del 2006. Para abordar este cambio se han planteado dos soluciones diferentes: una de ellas consiste en utilizar los parámetros de transformación entre los sistemas de referencia para obtener las series en ITRF2005 y en la otra se ha realizado un reprocesado. A los conjuntos de series obtenidos se les han aplicado diferentes análisis de series temporales para conocer sus características y obtener modelos que permitan estimar la velocidad con mayor exactitud y una precisión más realista. Mediante el cálculo de la función de autocorrelación y autocorrelación parcial se ha comprobado la existencia de variaciones estacionales en todos los procesados y componentes. El análisis espectral ha permitido calcular la amplitud de las variaciones estacionales, siendo mayores en los procesados PPP y DD4.2 que en los restantes, y se ha calculado el índice espectral comprobando la existencia de ruido correlado...
The number of permanent GPS stations has increased signi cantly in recent years for different geodetic applications and we begin to have coordinates time series long enough to apply time series analyses. The main goal of this thesis is to study the in uence of the network, framework and/or processing strategy in the coordinate time series from CGPS stations and those parameters obtained from them, such as velocities or deformations. For this aim, several sets of time series of CGPS stations have been used, obtained from different processing methods, all in the Canary Islands and the surrounding area. Two different processing strategies have been used for the series from 2002 to 2006, a double differences with Bernese v4.2 (DD4.2) and other Precise Point Positioning (PPP) with Bernese v5.0. Later, a double differences reprocessing with Bernese v5.0 has been made for the series from 2002 to 2008, considering a network with a greater number of stations. These series are affected by the change from ITRF2000 to ITRF2005 and the antenna phase centre model calibration in November 2006. To deal with this change two different solutions have been performed: one is to use the transformation parameters between the reference systems to obtain series in ITRF2005 and the other is to reprocesses the series. Different time series analyses have been applied to the coordinate sets obtained in order to identify their characteristics and obtain models to allow velocities to be estimated with greater accuracy and with more realistic uncertainties. The existence of seasonal variations in all processing methods and components has been con rm by computing the autocorrelation and partial autocorrelation functions. The spectral analysis has been able to estimate the magnitude of the seasonal variations, which are higher in PPP and DD4.2 processing than in the others. The spectral index has also been computed to prove the existence of correlated noise...
The number of permanent GPS stations has increased signi cantly in recent years for different geodetic applications and we begin to have coordinates time series long enough to apply time series analyses. The main goal of this thesis is to study the in uence of the network, framework and/or processing strategy in the coordinate time series from CGPS stations and those parameters obtained from them, such as velocities or deformations. For this aim, several sets of time series of CGPS stations have been used, obtained from different processing methods, all in the Canary Islands and the surrounding area. Two different processing strategies have been used for the series from 2002 to 2006, a double differences with Bernese v4.2 (DD4.2) and other Precise Point Positioning (PPP) with Bernese v5.0. Later, a double differences reprocessing with Bernese v5.0 has been made for the series from 2002 to 2008, considering a network with a greater number of stations. These series are affected by the change from ITRF2000 to ITRF2005 and the antenna phase centre model calibration in November 2006. To deal with this change two different solutions have been performed: one is to use the transformation parameters between the reference systems to obtain series in ITRF2005 and the other is to reprocesses the series. Different time series analyses have been applied to the coordinate sets obtained in order to identify their characteristics and obtain models to allow velocities to be estimated with greater accuracy and with more realistic uncertainties. The existence of seasonal variations in all processing methods and components has been con rm by computing the autocorrelation and partial autocorrelation functions. The spectral analysis has been able to estimate the magnitude of the seasonal variations, which are higher in PPP and DD4.2 processing than in the others. The spectral index has also been computed to prove the existence of correlated noise...
Description
Tesis inédita de la Universidad Complutense de Madrid, Facultad de Ciencias Matemáticas, Departamento de Física de la Tierra, Astronomía y Astrofísica I, leída el 30-10-2015