Abstract

Climate change GCM-based projections and their uncertainty are relevant to study hydrological impacts, and to analyze, operate and design water infrastructure. Impact studies and uncertainty evaluations commonly use several downscaled and/or bias-corrected GCM projections to map raw GCMs’ changes to local stations. The preservation of GCMs statistical attributes is by no means guaranteed, and thus alternative methods to cope with this issue are needed. We developed an ensemble technique for the unbiased mapping of GCM changes to local stations, which preserves local climate variability and GCMs’ statistics. In this approach, trend percentiles are extracted from GCMs to represent the future long-term climate range to which local climatic variability is added. We compare the method’s ability to preserve GCM and local statistics against traditional approaches such as delta change, bias correction and a selection of a subset of GCMs. The approach is also compared against using each GCM individually to build future climatic scenarios from which percentiles are computed. The comparison considered studying precipitation conditions in three Chilean basins under future projections based on 45 GCM runs under RCP 8.5. Overall, the proposed approaches produce acceptable results, even when using a few trend percentiles. In fact, using 5 to 10 percentiles produces a mean absolute difference of 0.4% in the estimation of the probabilities of consecutive years under different precipitation thresholds, which is ~60% less than the error obtained using the median trend. Thus, the approach successfully preserves the GCM’s statistical attributes while incorporating the range of projected climates.