Abstract
We explore use of the familiar tercile framework of seasonal forecasting for the characterization of 21st-century precipitation projections over North America. Consistent with direct analyses of modeled precipitation change, in a superensemble of CMIP5 simulations an unambiguous pattern of shifted tercile population statistics develops as the globe warms. Expressed categorically, frequencies for the low (i.e., dry) tercile increase in the southwestern United States and southward into Mexico and decrease across the northern tier of North America, while counts for the high tercile shift in the opposite sense. We show that as the 21st-century proceeds, changes become statistically significant over wide regions in the pointwise sense, and also when considered as projections on model-specific climate change “fingerprints”. Background noise in the superensemble, against which significance is established, comprises both structural model uncertainty and natural climate variability. The robustness of these findings makes a compelling case for long-range planning for a dryer future in the American Southwest and southward, and wetter one to the north and especially northeast, while communication is facilitated by widespread user familiarity with the tercile format.
Highlights
We utilize all ensemble members from 29 global climate models participating in the Coupled Model Intercomparison Project, Phase 5 (CMIP5) having both 20th-century (“historical”) and Representative Concentration Pathway 8.5 (“RCP8.5”) simulations. (A complete listing is provided as Supplementary Table S1.) These span the historical and future (2011–2080) periods, respectively
We show results based on weighting all ensemble members, which is equivalent to weighting models differentially depending on ensemble size
Computations based on weighting the models, irrespective of the number of ensemble members, result in somewhat greater uncertainty bounds for establishing pointwise significance, but differences are small and do not affect the conclusions presented. (See Supplementary Fig. S1 and associated discussion.)
Summary
From each model we obtained variable “pr” from all available ensemble members, for both the historical and RCP8.5 experiments[24]. Historical runs for 1901–2005 for each model were pooled in order to compute climatological tercile boundaries; RCP8.5 runs for each model were pooled to compute future tercile occupation statistics, based on those climatological boundaries. Probabilities were subsequently averaged over models in order to generate the maps of Fig
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
