Tropical precipitation extremes: Response to SST-induced warming in aquaplanet simulations
Scaling of tropical precipitation extremes in response to warming is studied in aquaplanet experiments using the global Weather Research and Forecasting (WRF) model. We show how the scaling of precipitation extremes is highly sensitive to spatial and temporal averaging: while instantaneous grid point extreme precipitation scales more strongly than the percentage increase (∼7% K^(−1)) predicted by the Clausius‐Clapeyron (CC) relationship, extremes for zonally and temporally averaged precipitation follow a slight sub‐CC scaling, in agreement with results from Climate Model Intercomparison Project (CMIP) models. The scaling depends crucially on the employed convection parameterization. This is particularly true when grid point instantaneous extremes are considered. These results highlight how understanding the response of precipitation extremes to warming requires consideration of dynamic changes in addition to the thermodynamic response. Changes in grid‐scale precipitation, unlike those in convective‐scale precipitation, scale linearly with the resolved flow. Hence, dynamic changes include changes in both large‐scale and convective‐scale motions.
Additional Information© 2017 American Geophysical Union. Received 27 FEB 2017; Accepted 20 MAR 2017; Accepted article online 27 MAR 2017; Published online 14 APR 2017. This work was supported by the Caltech President's and Director's Fund program. Thanks to Kay Sušelj for discussions and helpful suggestions regarding the WRF model. The source code for the model used in this study, the WRF V 3.5.1, is freely available at http://www2.mmm.ucar.edu/wrf/users/download/. The input files necessary to reproduce the experiments with WRF as well as the relevant model output are available from the corresponding author upon request (firstname.lastname@example.org).
Supplemental Material - grl55703-sup-0001-supplementary.pdf