Non-stationary Influences of Large-scale Climate Drivers on Low-flow Characteristics of Streamflows in Southeast Australia.

Abstract:

In the present study, we investigate the potential influence of large-scale climate drivers, viz., the El Niño Southern Oscillation (ENSO), the Indian Ocean Dipole (IOD) and the Southern Annular Mode (SAM) on streamflow in Southeast Australia. We investigate extreme low-flow events, defined by flow volumes below a suitably-chosen threshold, through modelling their intensity, duration and annual frequency. A total of 162 high-quality and undisturbed streamflow gauging sites spread across the state of Victoria are chosen for the purpose. The influences of ENSO, IOD and SAM on the intensity of low-flow events are modelled using non-stationary extreme value distributions that are fitted using the Differential Evolution-based global calibration scheme. Further, Generalised Linear Models are used for modelling duration and annual frequency of the low-flow events. We find that most of the sites that have significant non-stationary influences in the low-flow intensities are located in the energy-limited alpine region known to receive relatively high average annual rainfall. Conversely, in the water-limited regions, where the streams are often ephemeral, the climate drivers are not found to be significant. The duration and frequency of such low-flow events, on the other hand, are found to be less influenced by the climate drivers. We demonstrate that sites that are non-stationary in relation to these climate drivers have a higher probability of experiencing more intense and longer low-flow spells. These findings hold important implications in the context of the future trends and occurrences of such climate teleconnection events, the influence of which will reflect in the low-flow characteristics and drought conditions arising due to these in the region.

Keywords: non-stationarity, ENSO, IOD, SAM, low-flows, streamflow, droughts, climate teleconnections, Extreme Value Theory, Generalised Linear Models, Southeast Australia.