The Response of Future Projections of the North American Monsoon When Combining Dynamical Downscaling and Bias Correction of CCSM4 Output

Presenter Information

Jonathan Meyer

Location

USU Eccles Conference Center

Event Website

http://water.usu.edu

Start Date

4-5-2016 5:42 PM

End Date

4-5-2016 5:45 PM

Description

A 20-km regional climate model (RCM) dynamically downscaled the Community Climate System Model version 4 (CCSM4) to compare 32-year historical and future “end-of-the-century” climatologies of the North American Monsoon (NAM). CCSM4 and other phase 5 Coupled Model Intercomparison Project (CMIP5) models have indicated a delayed NAM and overall general drying trend. Here, we test the suggested mechanism for this drier NAM where increasing atmospheric static stability and reduced early-season evapotranspiration under global warming will limit early-season convection and compress the mature-season of the NAM. Through our higher resolution RCM, we found the role of increased evaporation under a warmer climate is likely understated in coarse resolution models such as CCSM4. Improving the representation of mesoscale interactions associated with the Gulf of California and surrounding topography produced additional surface evaporation, which overwhelmed the convection-suppressing effects of a warmer troposphere. Furthermore, the improved land-sea temperature gradient helped drive stronger northerly winds and greater moisture transport. Finally, we addressed limitations from inherent CCSM4 biases through a form of mean bias correction, which resulted in a more accurate seasonality of the atmospheric thermodynamic profile. After bias correction, greater surface evaporation from average peak GoC SSTs of 32 °C compared to 29 °C from the original CCSM4 led to roughly 50 percent larger changes to low-level moist static energy compared to that produced by the downscaled original CCSM4. The increasing destabilization of the NAM environment produced onset dates that were one to two weeks earlier in the core of the NAM and northern extent, respectively. Furthermore, a significantly more vigorous NAM signal was produced after bias correction, with greater than 50 mm month-1 increases the June-September precipitation found along east and west coasts of Mexico and into parts of Texas. A shift towards more extreme daily precipitation was found in both downscaled climatologies, with the bias-corrected climatology containing a much more apparent and extreme shift.

Comments

A poster by Jonathan Meyer, who is with Utah State University, Plants, Soils and Climate

This document is currently not available here.

Share

COinS
 
Apr 5th, 5:42 PM Apr 5th, 5:45 PM

The Response of Future Projections of the North American Monsoon When Combining Dynamical Downscaling and Bias Correction of CCSM4 Output

USU Eccles Conference Center

A 20-km regional climate model (RCM) dynamically downscaled the Community Climate System Model version 4 (CCSM4) to compare 32-year historical and future “end-of-the-century” climatologies of the North American Monsoon (NAM). CCSM4 and other phase 5 Coupled Model Intercomparison Project (CMIP5) models have indicated a delayed NAM and overall general drying trend. Here, we test the suggested mechanism for this drier NAM where increasing atmospheric static stability and reduced early-season evapotranspiration under global warming will limit early-season convection and compress the mature-season of the NAM. Through our higher resolution RCM, we found the role of increased evaporation under a warmer climate is likely understated in coarse resolution models such as CCSM4. Improving the representation of mesoscale interactions associated with the Gulf of California and surrounding topography produced additional surface evaporation, which overwhelmed the convection-suppressing effects of a warmer troposphere. Furthermore, the improved land-sea temperature gradient helped drive stronger northerly winds and greater moisture transport. Finally, we addressed limitations from inherent CCSM4 biases through a form of mean bias correction, which resulted in a more accurate seasonality of the atmospheric thermodynamic profile. After bias correction, greater surface evaporation from average peak GoC SSTs of 32 °C compared to 29 °C from the original CCSM4 led to roughly 50 percent larger changes to low-level moist static energy compared to that produced by the downscaled original CCSM4. The increasing destabilization of the NAM environment produced onset dates that were one to two weeks earlier in the core of the NAM and northern extent, respectively. Furthermore, a significantly more vigorous NAM signal was produced after bias correction, with greater than 50 mm month-1 increases the June-September precipitation found along east and west coasts of Mexico and into parts of Texas. A shift towards more extreme daily precipitation was found in both downscaled climatologies, with the bias-corrected climatology containing a much more apparent and extreme shift.

https://digitalcommons.usu.edu/runoff/2016/2016Posters/25