Welcome to the NOAA Institutional Repository |
Stacks Logo
Advanced Search
Select up to three search categories and corresponding keywords using the fields to the right. Refer to the Help section for more detailed instructions.
Clear All Simple Search
Advanced Search
Regional climate processes and projections for North America: CMIP3/CMIP5 differences, attribution and outstanding issues
Filetype[PDF-17.31 MB]

This document cannot be previewed automatically as it exceeds 5 MB
Please click the thumbnail image to view the document.
Regional climate processes and projections for North America: CMIP3/CMIP5 differences, attribution and outstanding issues
  • Description:
    The Coupled Model Intercomparison Project, phase 5 (CMIP5) provides an unprecedented set of climate model data from coordinated experiments that can be used to address a variety of questions related to climate change and climate variability. The CMIP5 builds on the previous model intercomparison phase (CMIP3) in several ways, including a larger number of modeling centers and models, the use of generally moderately higher resolution models, and the inclusion of more complex and complete representation of Earth system processes. A key question is whether the CMIP5 results have improved since CMIP3 in terms of the representation of observed climate features and processes, and whether the future projected changes are more robust, and why. This report addresses these questions for a suite of climate variables and regional processes for North America and provides recommendations for future analyses and experiments to resolve some of the ongoing issues. Overall, the multi-model ensemble (MME) mean performance has not improved substantially in CMIP5 relative to CMIP3 for climatological variables (precipitation, sea surface temperature - SST), except for a slight improvement for near surface air temperature over land. CMIP5 models tend to underestimate the frequency of heavy and extreme daily precipitation events, despite a slight improvement over CMIP3, especially in the southeastern US. Projected increases in moderate to extreme precipitation events are similar to CMIP3. Generally, the CMIP5 models show better skill for basic attributes of El Niño and the Southern Oscillation (ENSO) with performance related to the mean SST state. It is unclear whether the representation of teleconnections with North American climate has improved. It is likely that the structure of Pacific Decadal Variability (PDV) as indexed by the PDO is slightly better simulated and teleconnections for precipitation also are improved slightly but remain poor overall. Atlantic Multidecadal Variability (AMV) as represented by the AMO is better represented in CMIP5 models in terms of decadal variability and persistence than CMIP3 models, but its SST footprint and teleconnections with North American climate are still poorly represented. Regionally, projections of changes in precipitation from CMIP5 for the sub-tropics tend to be more robust overall than CMIP3, in particular for summer drying in the Caribbean and southwest Mexico. The boundary between winter wetting and drying in the Southwest US is projected to move southward in CMIP5 relative to CMIP3 results, although the changes are highly dependent on the region and season. The CMIP5 models project a more significant decrease in extra-tropical storm track activity than CMIP3, which may be related to a larger projected decrease in the temperature gradient between lower and higher latitudes. doi:10.7289/V5DB7ZRC (http://dx.doi.org/10.7289/V5DB7ZRC)
  • Document Type:
  • Main Document Checksum:
  • Supporting Files:
    No Additional Files
No Related Documents.