Non-stationary covariance function modelling in 2D least-squares collocation

Standard least-squares collocation (LSC) assumes 2D stationarity and 3D isotropy, and relies on a covariance function to account for spatial dependence in the ob-served data. However, the assumption that the spatial dependence is constant through-out the region of interest may sometimes be violated....

Full description

Bibliographic Details
Main Authors: Darbeheshti, Neda, Featherstone, Will
Format: Journal Article
Published: Springer - Verlag 2009
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/42464
_version_ 1848756426822909952
author Darbeheshti, Neda
Featherstone, Will
author_facet Darbeheshti, Neda
Featherstone, Will
author_sort Darbeheshti, Neda
building Curtin Institutional Repository
collection Online Access
description Standard least-squares collocation (LSC) assumes 2D stationarity and 3D isotropy, and relies on a covariance function to account for spatial dependence in the ob-served data. However, the assumption that the spatial dependence is constant through-out the region of interest may sometimes be violated. Assuming a stationary covariance structure can result in over-smoothing of, e.g., the gravity field in mountains and under-smoothing in great plains. We introduce the kernel convolution method from spatial statistics for non-stationary covariance structures, and demonstrate its advantage fordealing with non-stationarity in geodetic data. We then compared stationary and non-stationary covariance functions in 2D LSC to the empirical example of gravity anomaly interpolation near the Darling Fault, Western Australia, where the field is anisotropic and non-stationary. The results with non-stationary covariance functions are better than standard LSC in terms of formal errors and cross-validation against data not used in the interpolation, demonstrating that the use of non-stationary covariance functions can improve upon standard (stationary) LSC.
first_indexed 2025-11-14T09:12:01Z
format Journal Article
id curtin-20.500.11937-42464
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T09:12:01Z
publishDate 2009
publisher Springer - Verlag
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-424642017-09-13T16:05:51Z Non-stationary covariance function modelling in 2D least-squares collocation Darbeheshti, Neda Featherstone, Will gravity field interpolation elliptical kernel convolution non-stationary covariance function Darling Fault Australia modelling Least squares collocation (LSC) Standard least-squares collocation (LSC) assumes 2D stationarity and 3D isotropy, and relies on a covariance function to account for spatial dependence in the ob-served data. However, the assumption that the spatial dependence is constant through-out the region of interest may sometimes be violated. Assuming a stationary covariance structure can result in over-smoothing of, e.g., the gravity field in mountains and under-smoothing in great plains. We introduce the kernel convolution method from spatial statistics for non-stationary covariance structures, and demonstrate its advantage fordealing with non-stationarity in geodetic data. We then compared stationary and non-stationary covariance functions in 2D LSC to the empirical example of gravity anomaly interpolation near the Darling Fault, Western Australia, where the field is anisotropic and non-stationary. The results with non-stationary covariance functions are better than standard LSC in terms of formal errors and cross-validation against data not used in the interpolation, demonstrating that the use of non-stationary covariance functions can improve upon standard (stationary) LSC. 2009 Journal Article http://hdl.handle.net/20.500.11937/42464 10.1007/s00190-008-0267-0 Springer - Verlag fulltext
spellingShingle gravity field interpolation
elliptical kernel convolution
non-stationary covariance function
Darling Fault
Australia
modelling
Least squares collocation (LSC)
Darbeheshti, Neda
Featherstone, Will
Non-stationary covariance function modelling in 2D least-squares collocation
title Non-stationary covariance function modelling in 2D least-squares collocation
title_full Non-stationary covariance function modelling in 2D least-squares collocation
title_fullStr Non-stationary covariance function modelling in 2D least-squares collocation
title_full_unstemmed Non-stationary covariance function modelling in 2D least-squares collocation
title_short Non-stationary covariance function modelling in 2D least-squares collocation
title_sort non-stationary covariance function modelling in 2d least-squares collocation
topic gravity field interpolation
elliptical kernel convolution
non-stationary covariance function
Darling Fault
Australia
modelling
Least squares collocation (LSC)
url http://hdl.handle.net/20.500.11937/42464