Answers are grounded in this paper and cite their sources.
Bias correction of rainfall data is often required for the production of hydroclimate projections due to the misestimation of important characteristics of the precipitation process in global and/or regional climate models. Most bias correction methods ignore rainfall sequencing and correct the daily distribution only, which can lead to biases in runoff characteristics when corrected rainfall is used as an input into hydrological models. A bias correction method (RhoQM) is developed in which rainfall sequencing is corrected using estimated autocorrelation functions from autoregressive-moving average (ARMA) models. We correct historical and future rainfall from an ensemble of regional climate models for south-eastern Australia and compare the results to standard quantile-quantile mapping. Correctly specifying the autocorrelation function complexity can reduce residual cross-validated bias in rainfall and runoff metrics. There is little difference in future change signal in metrics from different model complexities, which suggests that the method is robust for hydroclimate projections.
No takes yet. Share an insight, caveat, or question.
Potter et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: