Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl

Extensive time series of measurements are often essential to evaluate long term changes and averages such as tidal datums and sea level rises. As such, gaps in time series data restrict the type and extent of modeling and research which may be accomplished. The Texas A&M University Corpus Christ...

Full description

Main Authors: Mostella, Aimee, Sadovski, Alexey, Duff, Scott, Michaud, Patrick, Tissot, Philippe E. Tissot, Steidley, Carl W.
Format: Artículo
Language: Español
Published: 2015
Online Access: http://revistas.ucr.ac.cr/index.php/matematica/article/view/260
http://hdl.handle.net/10669/12906
id RepoKERWA12906
recordtype dspace
spelling RepoKERWA129062017-08-08T18:50:21Z Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl Mostella, Aimee Sadovski, Alexey Duff, Scott Michaud, Patrick Tissot, Philippe E. Tissot Steidley, Carl W. Extensive time series of measurements are often essential to evaluate long term changes and averages such as tidal datums and sea level rises. As such, gaps in time series data restrict the type and extent of modeling and research which may be accomplished. The Texas A&M University Corpus Christi Division of Nearshore Research (TAMUCC-DNR) has developed and compared various methods based on forward and backward linear regression to interpolate gaps in time series of water level data. We have developed a software system that retrieves actual and harmonic water level data based upon user provided parameters. The actual water level data is searched for missing data points and the location of these gaps are recorded. Forward and backward linear regression are applied in relation to the location of missing data or gaps in the remaining data. After this process is complete, one of three combinations of the forward and backward regression is used to fit the results. Finally, the harmonic component is added back into the newly supplemented time series and the results are graphed. The software created to implement this process of linear regression is written in Perl along with a Perl module called PDL (Perl Data Language). Generally, this process has demonstrated excellent results in filling gaps in our water level time series. The program was tested on existing data under three typesof typical weather conditions: calm summers, frontal passages and extreme weather conditions, such as hurricanes. The parameters varied in order to test the accuracy of the methodology included the number of coefficients utilized in the linear regression processes as well as the size of the gaps to be filled. Results are presented for the different weather conditions and the different gap size and coefficient combinations. Serie de tiempo extensivas de medidas a menudo con esenciales para evaluar cambios a largo plazo y promedios como los datos de mareas y crecidas del nivel de agua. As?, huecos en los datos de series de tiempo restringe el tipo y extensi?n del modelamiento e investigaci?n que pueda hacerse. La Divisi?n de Investigaci?n de la Costa de la Universidad de Texas A&M en Corpus Christi (TAMUCC-DNR, por sus siglas en ingl?s) ha desarrollado y comparado varios m?todos basados en regresi?n lineal hacia adelante y hacia atr?s, para interpolar los huecos en las series de tiempo de datos del nivel de agua. Hemos desarrollado un sistema inform?tico que recupera datos reales y arm?nicos de nivel de agua, basado en par?metros dados por el usuario. Los datos reales de nivel de agua se buscan para puntos con datos faltantes y la localizaci?n de estos huecos es registrada. Se aplica regresi?n lineal hacia adelante y hacia atr?s en relaci?n con la localizaci?n de datos faltantes o huecos en los datos restantes. Despu?s de que este proceso se completa, se usa una de tres combinaciones de la regresi?n hacia adelante y hacia atr?s para ajustar los resultados. Finalmente, se a?ade la componente arm?nica para las nuevas series de tiempo suplementarias, y se grafican los resultados. El paquete inform?tico creado para implementar este proceso de regresi?n lineal est? escrito en Perl con un m?dulo llamado PDL (Perl Data Language). Generalmente, este proceso ha demostrado excelentes resultados en llenar huecos en nuestras series de tiempo sobre nivel de agua. El programa ha sido probado sobre datos existentes bajo tres tipos de condiciones clim?ticas: veranos calmos, pasos frontales y condiciones de clima extremas, como huracanes. Se variaron los par?metros con el fin de probar la precisi?n del m?todo, como por ejemplo el n?mero de par?metros usados en las regresiones lineales as? como el tama?o de los huecos a llenar. Se presentan resultados para las diferentes condiciones clim?ticas y los distintos tama?os de los huecos y las combinaciones de coeficientes. 2015-05-19T18:43:02Z 2015-05-19T18:43:02Z 2012-03-02 00:00:00 2015-05-19T18:43:02Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion http://revistas.ucr.ac.cr/index.php/matematica/article/view/260 http://hdl.handle.net/10669/12906 10.15517/rmta.v12i1-2.260 es Revista de Matem?tica: Teor?a y Aplicaciones Vol. 12 N?m. 1-2 2012 157-164 application/pdf
institution Universidad de Costa Rica
collection Repositorio KERWA
language Español
description Extensive time series of measurements are often essential to evaluate long term changes and averages such as tidal datums and sea level rises. As such, gaps in time series data restrict the type and extent of modeling and research which may be accomplished. The Texas A&M University Corpus Christi Division of Nearshore Research (TAMUCC-DNR) has developed and compared various methods based on forward and backward linear regression to interpolate gaps in time series of water level data. We have developed a software system that retrieves actual and harmonic water level data based upon user provided parameters. The actual water level data is searched for missing data points and the location of these gaps are recorded. Forward and backward linear regression are applied in relation to the location of missing data or gaps in the remaining data. After this process is complete, one of three combinations of the forward and backward regression is used to fit the results. Finally, the harmonic component is added back into the newly supplemented time series and the results are graphed. The software created to implement this process of linear regression is written in Perl along with a Perl module called PDL (Perl Data Language). Generally, this process has demonstrated excellent results in filling gaps in our water level time series. The program was tested on existing data under three typesof typical weather conditions: calm summers, frontal passages and extreme weather conditions, such as hurricanes. The parameters varied in order to test the accuracy of the methodology included the number of coefficients utilized in the linear regression processes as well as the size of the gaps to be filled. Results are presented for the different weather conditions and the different gap size and coefficient combinations.
format Artículo
author Mostella, Aimee
Sadovski, Alexey
Duff, Scott
Michaud, Patrick
Tissot, Philippe E. Tissot
Steidley, Carl W.
spellingShingle Mostella, Aimee
Sadovski, Alexey
Duff, Scott
Michaud, Patrick
Tissot, Philippe E. Tissot
Steidley, Carl W.
Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
author_sort Mostella, Aimee
title Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
title_short Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
title_full Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
title_fullStr Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
title_full_unstemmed Comparison of Gap Interpolation Methodologies for water Level time Series using perl/pdl
title_sort comparison of gap interpolation methodologies for water level time series using perl/pdl
publishDate 2015
url http://revistas.ucr.ac.cr/index.php/matematica/article/view/260
http://hdl.handle.net/10669/12906
_version_ 1650735690879598592
score 11.771365