Climatic influences on rainfall and runoff variability in the southeast region of the Murray-Darling Basin

The aim of this study is to establish a relationship between 11 correlated climatic indices and rainfall and runoff in the Murray-Darling Basin (MDB). The climatic indices are functions of sea surface temperature (SST) and sea level pressure (SLP) differentials. There are six Pacific Ocean indices,...

Full description

Bibliographic Details
Published in:International Journal of Climatology
Main Authors: Kamruzzaman, M., Beecham, S., Metcalfe, A.
Format: Article in Journal/Newspaper
Language:English
Published: John Wiley & Sons Ltd 2013
Subjects:
regression model
correlation
factor analysis and pre-whitened process
Akaike Information Criterion
Southern Ocean
Pacific
Indian
Soi
Metcalfe
Online Access:http://hdl.handle.net/2440/78959
https://doi.org/10.1002/joc.3422
Description
Summary:The aim of this study is to establish a relationship between 11 correlated climatic indices and rainfall and runoff in the Murray-Darling Basin (MDB). The climatic indices are functions of sea surface temperature (SST) and sea level pressure (SLP) differentials. There are six Pacific Ocean indices, three Atlantic Ocean indices, one Indian Ocean index and one from the Southern Ocean. Trends in the climatic indices are investigated by fitting seasonal trend models using generalized least squares. Relationships between the indices are described by correlation analysis and factor analysis. Correlation analysis of the pre-whitened series is first presented and this is used to guide the choice of climatic indices for the regression models. Regression analyses are then used to investigate the effects of climatic indices on rainfall and runoff at the monthly level during the period 1957–2009. This is undertaken using data from three stations in the southeast region of the MDB, namely Tooma River Basin, Jingellic Catchment and Ovens Catchment. Regression models are fitted using all or a sub-set of the climatic indices and their interactions, and these are compared with regression models based on estimates of latent factors and their interactions. Typical R2 values of 20% were obtained. The Akaike Information Criterion indicated that statistically significant improvement could be obtained from a benchmark model using seasonality, trends and the Southern Oscillation Index (SOI). However, the gain in information is generally modest. This conclusion is specific to the southeast region of the MDB, but the methods used are generally applicable. M. Kamruzzaman, S. Beecham and A. V. Metcalfe

Similar Items