 |
| Titel |
An extended approach for spatio-temporal gap filling: dealing with large and systematic gaps in geoscientific datasets |
| VerfasserIn |
J. von Buttlar, J. Zscheischler, M. D. Mahecha |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250060035
|
|
|
|
|
|
| Zusammenfassung |
| Geo-scientific datasets often contain numerous and possibly systematically distributed gaps.
This data fragmentation which may be due to instrument failures, sparse measurement
protocols or unfavorable conditions (e.g. clouds or vegetation thickness in case of remote
sensing data). It affects and often inhibits most statistical analysis which often require
continuously sampled data points. Hence, gap filling is an undesired but often necessary task
in geo-sciences. In cases where multivariate relationships are investigated they are often
biased similar relationships which are used by the preceeding gap filling algorithm. In these
cases univariate methods are needed.
Kondrashov and Ghil (2006) proposed a gap filling approach which exploits the temporal
(possibly multidimensional) patterns as identified by Singular Spectrum Analysis (SSA).
Here we propose an univariate extension of this method in order to additionally consider the
spatial processes and patterns underlying most geo-scientific data sets. The latter has been
made possible by including a novel 2D-SSA approach recently introduced by Golyandina and
Usevich (2010).
Using both artificial and real-world test data we show that considering spatial
and temporal patterns simultaneously improves the gapfilling substantially. We
outperform the conventional approach particularly for large and systematically recurring
gaps. Our method is fast, can be applied with a minimum of a priori assumptions
on the data structure and is implemented ready-to-use in an open source software
package.
N. E. Golyandina and K. D. Usevich. 2D-extension of singular spectrum analysis:
algorithm and elements of theory. In Matrix Methods: Theory, Algorithms, Applications,
pages 449–474. World Scientific, 2010.
D. Kondrashov and M. Ghil. Spatio-temporal filling of missing points in geophysical data
sets. Nonlinear Processes in Geophysics, 13:151–159, 2006. |
|
|
|
|
|
|