Understanding the impacts of climate change on the hydrological cycle is critical, for example, for ensuring sustainable water resources management, agricultural production or infrastructure development. Hydrological impact studies commonly use hydrological models forced with corrected climate inputs from general circulation models (GCMs) that simulate the future climate under a range of greenhouse gas concentration pathways. Climate models are typically run at a relatively coarse resolution – coarser than what is required to force hydrological models. Additionally, approximations for small-scale processes can lead to biases in some variables or processes. Therefore, various bias-correction and downscaling (BCDS) methods have been developed to remove systemic biases in GCM outputs and to increase the spatial resolution to match the resolution required by impact models.

The Bureau of Meteorology is currently developing a National Hydrological Projections Service that will provide data on future climate change impacts on Australian water resources, such as changes in soil moisture, runoff or hydrological extremes. Four statistical BCDS methods were applied to climate model outputs to produce bias-corrected and downscaled climate projections data. The target dataset for the BCDS methods is the Australian Water Availability Project data (AWAP; Jones et al., 2009), a gridded dataset that contains climate observations (including precipitation, temperature) at 0.5 km grid resolution. For the evaluation, we forced AWRA-L (Frost et al., 2018) – a gridded land surface water balance model – with the corrected climate data, and produced hydrological simulations for a 30-year validation period (1976-2005).

In this talk, we present the evaluation of the bias-corrected and downscaled climate inputs and simulated hydrological variables. The evaluation includes assessments of mean biases as well as biases in temporal variability and extremes – at seasonal, annual and multi-annual time scale. We discuss implications of our findings for hydrological impact assessments and outline potential uses of these methods.