Assessment of Regional Climate Model simulations at reproducing the link between PDO and Climate Extreme Precipitation indices in Mexico

The primary objective of this study is to evaluate whether Regional Climate Models (RCM)-simulations and their driving dataset (ERA-Interim) can reproduce the spatial pattern of the temporal correlation between the Pacific Decadal Oscillation (PDO) and four Climate Extreme Indices –CDD, CWD, Rx1, and Rx5– over Mexico from 1980 to 2012. First the ERA-Interim dataset’s ability to reproduce the PDO index time series were assessed. Next, both ERA-Interim and four ERA-Interin-driven RCM simulations (CRCM5-0.22, CRCM5-0.44, RCA4-0.5, and RegCM4-7-0.25) were evaluated for their representation of the average monthly accumulated precipitation and the seasonal spatial patterns of the climate extreme indices. Finally, the study assessed how well ERA-Interim and RCM simulations reproduced the spatial pattern of the temporal correlation between the PDO and climate extreme indices. The findings indicate that ERA-Interim effectively reproduces the PDO. Both ERA-Interim and RCM simulations capture the main features of monthly accumulated precipitation, although some overestimation occurs in summer and spring. The CRCM5 and RCA simulations perform better in reproducing the seasonal spatial patterns of the climate extreme indices. Spatial correlation patterns, with statistical significance, were found during the two driest seasons in Mexico, winter and spring, and mainly relate to moisture conditions in the north-west part of the country (spring) and in the northern central part of the country (winter). Given the location of the Sierra Madre Occidental Mountain range, the spring moisture conditions could also be associated with orography. The ERA-Interim dataset as well as the two CRCM5 simulations generally perform best at reproducing the observed correlations.