 |
| Titel |
Linear and nonlinear spatio-temporal dependences of precipitation and river runoffs in a catchment |
| VerfasserIn |
Reik Donner, Maria Koch |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250057943
|
|
|
|
|
|
| Zusammenfassung |
| Hydro-meteorological processes are characterised by spatio-temporal patterns with distinct
correlations in both time and space. Thus, time series of relevant observables such as
precipitation and runoff – recorded at different locations – show relevant correlations if these
locations are influenced by the same variability patterns (i.e. the same atmospheric
regimes) and/or belong to the same catchment. The actual strength and temporal as
well as spatial extensions of these correlations depend crucially on the considered
variable.
Here, we investigate the significance of the corresponding interrelationships and their
possible temporal changes in the presence of changing environmental conditions. For
this purpose, a variety of linear as well as nonlinear measures for the statistical
dependence between univariate time series is applied to spatial fields of runoff and
precipitation records taken from an intermediate-size (about 4.200 km2) river catchment in
Southern Germany [1]. The qualitative behaviour of spatial correlations during
extreme weather events is characterised by a variant of the LVD dimension density, a
complexity measure for multivariate (or spatially distributed univariate) time series
[2,3].
As a further research question, we investigate the mutual interplay between precipitation and
runoff for the same catchment. We demonstrate that at the different gauges under
consideration, runoffs react to precipitation patterns in the source region with a time delay of
1-3 days, depending on the gauge position and the considered measure of interdependence.
Specifically, we find that Hilbert transform based phase synchronisation measures show a
clear tendency towards shorter delays, whereas mutual information and event synchronisation
[4,5] indicate the strongest interrelationships at somewhat larger delays. The latter
observation points to different time scales relevant for weak/moderate and strong
precipitation scenarios, respectively, and a temporal clustering of heavy rainfall events. In
addition to the obvious dependence on the gauge position, we discuss the relationship of the
observed spatio-temporal correlations on further relevant factors such as soil properties and
hillslope.
[1] R. Donner, Spatial Correlations of River Runoffs in a Catchment. In: J. Kropp and H.-J.
Schellnhuber (eds): In Extremis - Disruptive Events and Trends in Climate and Hydrology
(Springer, Berlin, 2011), pp. 286-313
[2] R. Donner and A. Witt, Characterisation of long-term climate change by dimension
estimates of multivariate palaeoclimatic proxy data. Nonlin. Processes Geophys. 13, 485-497
(2006)
[3] R. Donner, T. Sakamoto, and N. Tanizuka, Complexity of Spatio-Temporal Correlations
in Japanese Air Temperature Records. In: R.V. Donner and S.M. Barbosa (eds.): Nonlinear
Time Series Analysis in the Geosciences - Applications in Climatology, Geodynamics, and
Solar-Terrestrial Physics (Springer, Berlin, 2008), pp. 125-154
[4] R. Quian Quiroga, T. Kreuz, and P. Grassberger, Event synchronization: A simple and fast
method to measure synchronicity and time delay patterns. Phys. Rev. E 66, 041904
(2002)
[5] N. Malik, N. Marwan, and J. Kurths, Spatial structures and directionalities in Monsoonal
precipitation over South Asia. Nonlin. Processes Geophys. 17, 371-381 (2010) |
|
|
|
|
|
|