Weather and Climate Systems

Understanding the Features and Mechanisms of the Great Plains Low-Level Jet in CMIP5 Models

James Danco

School of Meteorology

13 April 2016, 3:00 PM

National Weather Center, Room 5600
120 David L. Boren Blvd.
University of Oklahoma
Norman, OK

The Great Plains low-level jet (GPLLJ) is a nocturnal southerly flow of air from 850 hPa to just above the surface that has been shown to significantly increase nighttime convection and severe weather outbreaks over the U.S. Great Plains during the spring and summer months. Therefore, it is extremely important that the various features of the GPLLJ, including the main atmospheric and oceanic mechanisms controlling its variations, are understood and simulated by models accurately so Great Plains precipitation can be better predicted. While past studies have discovered some of these features and mechanisms in observations/reanalysis, less focus has been given on how they are represented in climate models. This study uses HadISST1 and ERSST sea surface temperature (SST) observations, meridional wind data from three reanalyzes (20th Century Reanalysis, ERA-Interim, and NCEP CFSR), and historical simulations from 40 models from phase 5 of the Coupled Model Intercomparison Project (CMIP5) to examine some of the features and mechanisms of the GPLLJ in the spring and summer in the models compared to reanalysis, as well as reasons why they may not be sufficiently captured in the models.

Analysis of various features of the GPLLJ itself shows that most CMIP5 models simulate a GPLLJ in spring that is significantly weaker than the reanalysis, extend its peak too late into the summer, and place the peak wind at a level too low in the troposphere. Results from the reanalyzes also find that El Niño Southern Oscillation (ENSO) from the previous December, January, February (DJF) has a significant negative correlation with the GPLLJ in March, April, and May (MAM) and a significant positive correlation with the GPLLJ in July, August, and September (JAS), agreeing with past studies. In contrast, while a majority of the models exhibit significant negative correlations in the spring, all of them have minimal or negative correlations in the summer. The influence of ENSO on the GPLLJ is also found to be highest following strong DJF ENSO events and weak to nonexistent following weak DJF ENSO events. Besides ENSO, the reanalyzes and models indicate an SST dipole between the Caribbean and Gulf of Mexico and a negative Pacific Decadal Oscillation (PDO) is associated with a stronger GPLLJ in MAM, but models fail to capture the flip to a positive correlation between the PDO and GPLLJ in JAS seen in the reanalyzes. Lastly, it is shown that the models that do the poorest job simulating the negative MAM GPLLJ – DJF ENSO correlation also simulate ENSO events that are much weaker on average than the other models and observations, indicating that this may be playing a role in their inability to capture the correct relationship.