A review of flow estimation by runoff routing in Australia – and the way forward

ABSTRACT Flood hydrograph estimation in Australia is mostly undertakenby a class of models known as runoff routing models. These models have in common the sub-division of the catchment into a number of sub-catchments, with the application of excess rainfall to each sub-catchment and the routing through a series of nonlinear storages representing the channel flow through the catchment. The first Australian runoff routing model was developed in the 1960s, and there have been numerous models developed that contain the same basic structure, with enhancements. With the release of Australian Rainfall & Runoff there were significant changes in the way design floods are estimated in Australia, with flood estimation by runoff routing beingmodified to include a joint probability approach rather than a design flood approach, so that the variability and interaction of flood-producing factors are better allowed for. This paper reviews the development of runoff routing within Australia, from the original 1960s model to joint probability, and identifies issues associated with the models, including self-consistency and the limitation on the runoff processes that can be modelled. The required features of a model that address the limitations of current models are described.