2016 JISAO Summer Internship Program
Analysis of the North American Monsoon Retreat
My name is Jeremy Sousa and I’m a rising senior in the Atmospheric Science and Mathematics departments at Lyndon State College in Lyndonville, Vermont. This summer, I worked with Dr. Yolande Serra to investigate the accuracy of the CMIP5 (Coupled Model Intercomparison Project Phase 5) models’ depiction of the retreat of the NAM (North American Monsoon) season. The onset of the NAM is relatively rapid and distinct, but the demise is harder to pinpoint due to the lingering influence of tropical systems that may advect moisture into the region late in the season. Previous analogous studies have shown that climate models for historical periods have difficulty accurately depicting the retreat of the NAM, so the motivation for this project was to see if models for future periods would perform better.
This analysis was completed using 21 models’ data for the historical period (1979-2005) and 17 models’ RCP 8.5 (representative concentration pathway of 8.5 W/m2) data for the 2070-2099 period. We used TRMM satellite precipitation data over the 1998-2010 period to form daily precipitation climatologies and plotted this against the top 3 and bottom 3 performing model composites for the historical and future periods in order to quantify bias in the models and to highlight change in precipitation over the observed region with respect to time. Besides analyzing precipitation data, we also used MATLAB to calculate moisture flux convergence over the observed region during the month of October. The purpose of this was to detect any possible changes in moisture advection over the monsoon region that may occur as anthropogenic warming continues.
Our results showed that while moisture advection increased in the future period during the demise of the monsoon, overall precipitation during the monsoon decreased. There was also large variability in modeled precipitation output during the retreat of the monsoon. This led us to conclude that like the historical model data, the future model data in CMIP5 doesn’t accurately depict the retreat of the NAM.
Although I was mostly focused on my work while in Seattle, I did have some time to explore. I enjoyed the variety in cuisine that the city has to offer. As an avid hiker, however, the highlight of my summer was hiking Mt. Rainier with the other interns. I also enjoyed checking out the city on the weekends. Overall, interning this summer at JISAO was a fantastic experience where I gained invaluable research experience in my field of study and formed bonds and memories with the other interns.
Click on image for full size PDF