Disaggregation of Climate-Projected Rainfall Using an Empirical Approach for Design Rainfall Estimation

The robust regional and seasonal variability exhibited in Australian rainfall patterns, superimposed by the large-scale-continental-climate-volatility, is expected to further intensify under climate change impacts, altering the recurrence and austerity of extreme rainfall intensity event(s) prevalence. This needs to be conscientiously addressed while developing Intensity-Frequency-Duration (IFD) curves for employment in the design of flood-mitigation-infrastructure. Current Australian IFD practices are developed based upon the temporal-stationarity-concept, thereby calling for updated IFD practices based upon non-stationarity approaches for future flood mitigation/planning. However, a major obstacle in the adaptation of this approach is centered around the unavailability of projected future rainfall data records at sub-hourly/sub-daily timescales, crucial for developing IFD curves of any sort. This has led to extensive research on various rainfall disaggregation techniques, using both statistical and empirical methods. This paper proposes the novel application of one such empirical method, a reduction formula used by the Ethiopian Road Authority, dubbed as the ERA Formula, for disaggregating projected daily rainfall data into sub-daily/sub-hourly timescales. The proposed method is attested on an Eastern Melbourne urban catchment, Gardiners Creek, with good-quality observed rainfall data. The original ERA equation, is calibrated to befit Australian climatic and geographical conditions, following which it is applied and evaluated. The results highlight that the application of the ERA approach exhibited supremacy in the accurate replication of the observed temporal variability in the annual maxima rainfall timeseries at the sub-daily/sub-hourly timesteps, with high estimation accuracy (R<sup>2</sup> = 88-92% & NSE = 0.89-0.9) and minimum error magnitude (MAE = 0.85mm & RMSE =1.37 mm), thereby highlighting the efficacy of potentially adopting this approach for disaggregation of the projected rainfall.