Joint assembly AGU-GAC-MAC-CGU
Year of publication: 
Air masses are a fundamental quantity in both weather and climate. The study of air masses largely originated with the Norwegian Cyclone Model led by Vilhelm and Jacob Bjerknes, who put emphasis on both the dynamics and thermodynamics of air masses. Since the classic works of Charney and Eady, modern use of the concept of air masses has revolved around their dynamics. In a recent essay, Kerry Emanuel suggests that it is time to re-visit the concept of air masses in both a dynamic and thermodynamic sense. Here, equivalent potential temperature (Θe) is used as an air mass tracer as it takes into account both temperature and moisture, and because it is a conserved thermodynamic property. A monthly climatology of 850 hPa Θe from 1979-2011 using the NCEP (National Centers for Environmental Prediction) Reanalysis 2 is presented for North America. Our statistical analysis shows that while the variance of Θe has not changed during the 33 years, there is a statistically significant increase in the mean Θe. Extreme events were selected by applying a 3-day running mean to the Θe data and ranking them from high to low by month. The top 10 (representing roughly the 99th percentile) Θe events were selected for each month, totaling 120 extreme events. Monthly composites show that these anomalously high Θe events had contributions from both high values of potential temperature and moisture content. Back trajectories were created using the NOAA (National Oceanic and Atmospheric Administration) Air Resources Laboratory HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory) model to determine the source regions of the parcels contributing to the extremes. Results showed that most of the parcels had origins in the Gulf of Mexico, subtropical Atlantic Ocean, and US Midwest region. The HYSPLIT results showed that some of the parcels descended from the mid-troposphere to 850 hPa. Data from the HYSPLIT model runs was used to calculate theta-e along the parcel trajectories. These results were used to separate the conservative theta-e events from those of non-conservation. Such a partitioning facilitates further analysis of the events’ dynamics and thermodynamics.