ABSTRACT: Within the context of analyzing daily heavy precipitation events in the Mediterranean under enhanced greenhouse gas forcing in the 21st century, a new method considering non-stationarities in the relationships of large-scale circulation predictors and regional precipitation extremes was applied. The Mediterranean area was split into up to 22 precipitation regions, and analyses were performed separately for 3 different seasons (autumn, winter and spring) and 3 different quantiles (90th, 95th and 99th). Estimations are based on a three-step censored quantile regression. Future estimations are performed by means of 3 model runs of the Max Planck Institute Earth System Model with Low Resolution (MPI-ESM-LR) for representative concentration pathways (RCPs) 4.5 and 8.5. Overall, the Mediterranean is mainly characterized by decreasing quantile values. Especially in the regions in the southeast, declines are significant, with up to 71.7% (-1.65 mm) in the Levante region (autumn) and over 16 mm (-38.2%) on Crete (winter). Increased precipitation quantiles were only assessed for a more or less extended region in the northern parts of the Central Mediterranean (winter and spring), for the northeastern coast of the Iberian Peninsula (autumn) and for northern Spain (spring). Overall, analyses showed that non-stationarities seriously affect precipitation behavior in most parts of the Mediterranean. Results indicated that 2 different regimes (western and eastern) inducing non-stationarities are predominant in the Mediterranean area. In autumn (winter), the western (eastern) regime is limited to the Iberian Peninsula (Levante), whereas in spring, the area of influence of both regimes is of equal size.
KEY WORDS:Statistical Downscaling · Precipitation extremes · Non-stationarities · Mediterranean
Full text in pdf format Supplementary material | Cite this article as: Merkenschlager C, Hertig E
(2020) Seasonal assessments of future precipitation extremes in the Mediterranean area considering nonstationarities in predictor-predictand relationships. Clim Res 80:19-42. https://doi.org/10.3354/cr01590
Export citation Share: Facebook - - linkedIn |
Previous article Next article |