Long-term variations of hydrological and climate time series from the German part of the Elbe River basin

Abstract: Long-term variations in the structure of mean monthly hydro-meteorological time series from the German part of the Elbe River Basin are analyzed. Statistically significant correlations between the 2-15 yr. scale av-eraged wavelet spectra of the mean monthly climate variables and the North...

Full description

Bibliographic Details
Main Authors: D. Markovic, M. Koch, H. Lange
Other Authors: The Pennsylvania State University CiteSeerX Archives
Format: Text
Language:English
Published: 2008
Subjects:
Singular Spectrum Analysis
Precipitation
Discharge
Persis- tence
Low-frequency variability
Forecast
Elbe River Basin
NAO
AO 1
Arctic
Gage
North Atlantic
North Atlantic oscillation
Online Access:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.513.1224
http://www.uni-kassel.de/fb14/geohydraulik/koch/paper/2007/Elbe_SSA_Modes.pdf
Description
Summary:Abstract: Long-term variations in the structure of mean monthly hydro-meteorological time series from the German part of the Elbe River Basin are analyzed. Statistically significant correlations between the 2-15 yr. scale av-eraged wavelet spectra of the mean monthly climate variables and the North Atlantic Oscillation (NAO)- and Arctic Oscillation (AO)- Index are found, which provides evidence that such long-term patterns in climate time series are externally forced. Application of Singular Spectrum Analysis (SSA) re-sults in major low-frequency modes for the basin precipitation of the Striegis, Ohre and the Elbe River that coincide with those detected in the discharge time series. The percentage of the variance explained by the annual cycle and low frequency components is clearly larger for the discharge than for precipi-tation. This manifests itself also through higher DFA (Detrended Fluctuation Analysis) Hurst parameter (H) estimates for discharge than for precipitation. Upon subtraction of the annual- and the major low frequency SSA- signal from the raw time series data, the DFA H parameter estimates suggest a short-range memory structure of the residuals for both precipitation and dis-charge. For the Este-River flows at gage Emmen we show additionally that the low frequency variability modes can be predicted by the SSA recurrent algorithm.

Similar Items