Journal article

Model averaging methods to merge operational statistical and dynamic seasonal streamflow forecasts in Australia

A Schepen, QJ Wang

Water Resources Research | AMER GEOPHYSICAL UNION | Published : 2015

Abstract

The Australian Bureau of Meteorology produces statistical and dynamic seasonal streamflow forecasts. The statistical and dynamic forecasts are similarly reliable in ensemble spread; however, skill varies by catchment and season. Therefore, it may be possible to optimize forecasting skill by weighting and merging statistical and dynamic forecasts. Two model averaging methods are evaluated for merging forecasts for 12 locations. The first method, Bayesian model averaging (BMA), applies averaging to forecast probability densities (and thus cumulative probabilities) for a given forecast variable value. The second method, quantile model averaging (QMA), applies averaging to forecast variable valu..

View full abstract

University of Melbourne Researchers

Grants

Funding Acknowledgements

This work was funded by the Water Information Research and Development Alliance, a partnership between the CSIRO and the Bureau of Meteorology. The main author is formerly of the Australian Bureau of Meteorology and so we would like to thank the Bureau and employees of the extended hydrological prediction section for contributing resources to this work. We thank Daehyok Shin and Julien Lerat for providing dynamic seasonal streamflow forecasts. We thank Bat Le and Senlin Zhou for providing statistical seasonal streamflow forecasts. We would also like to thank Senlin and Julien for their constructive reviews of the initial manuscript. We thank four anonymous reviewers for their reviews of an earlier version of the manuscript. Study data are available by contacting the authors.