Stratospheric Variability and Its Relationship to Cold Air Outbreaks in the United States

Only since the early 2000s, has there been a growing body of work linking extreme winter temperature anomalies to the variability of the Northern Hemisphere stratospheric polar vortex (SPV). However, the understanding behind stratosphere-troposphere coupling and cold air outbreaks (CAOs) remain occasionally contradictory in the literature, making it difficult to properly articulate societal risks. Studies often focused on broad stratospheric influences, but this research directly compared stratospheric variability according to their respective tropospheric impacts. This dissertation filled a knowledge gap, as there has yet to be a comprehensive analysis to quantify the differences in stratospheric variability with and without CAOs in the U.S. The first part of the research focused on climatological and statistical analyses of all U.S. CAO events and investigated them in the context of Northern Hemisphere polar stratospheric vortex conditions preceding their occurrence at lead-times of 15-, 30-, and 60-days. Three different types of stratospheric variability were analyzed: weak vortex events (WVE), strong vortex events (SVE), and wave reflection events. Wave reflection events were subdivided using two different metrics for calculating a reflective layer. One was based on the difference of zonal-mean zonal-winds at 60°N between 10-hPa and 2-hPa (DURI) and a second used the difference of regionally averaged meridional heat fluxes between Siberia and Canada (VTRI). Results showed the occurrence of extreme cold CAOs following within 15-days of a WVE was less than expected (90th percent confidence interval), and CAOs that occurred within 30-days of a WVE had shorter durations, (95th percent confidence interval). The frequency of CAOs within 15- and 30-days of a VTRI event were significant (90th percent confidence interval). Next, the analysis partitioned different types of stratospheric variability by whether or not a U.S. CAO occurred within a given window after the onset of a stratospheric event. The first part of this analysis used the quasi-geostrophic height tendency equation to show that the forcings for significant geopotential height features differed between stratospheric event types. Composites of the WVEs with a U.S. CAO occurring in the following 15- and 30- days and VTRI events with a CAO following within 15-days had significant amplification of an Alaskan ridge (> 90th percentile), a known synoptic feature linked to U.S. CAOs. Those VTRI events with a CAO also showed anomalously low geopotential heights at 250-hPa, in a trough centered in northern North America. QG analysis demonstrated that in the cases of WVE with CAOs following in the 15- and 30-day windows, regions near the Alaskan Ridge had the strongest forcing for ridging in the upper troposphere. However, VTRI events with a CAO in 15-days the strongest forcing for height rises in the Alaskan ridge came from differential temperature advection in the lower stratosphere. Downstream over North America, differential temperature advection, maximized in the lower stratosphere and near the tropopause, deepened the trough. To understand if all of the CAOs that occurred with a precursor stratospheric event were indeed driven by stratospheric variability, a cluster analysis revealed WVEs with a CAO in 15-days had a differential thermal advection pattern resembling a Rossby wave train with a second cluster exhibiting stratosphere-troposphere coupling driving the dynamics. WVE with a CAO in 30-days only showed the former. VTRI events with a CAO in the 15-days cluster analysis revealed differential thermal advection patterns consistent with near tropopause and lower-stratosphere QG forcing in both clusters. The final part of the dissertation employed empirical orthogonal function analysis for cross sections of 45-75˚N averaged eddy geopotential height to identify common modes of variability during stratospheric-tropospheric coupled U.S. CAOs. EOF1-4 showed wave structures consistent with the variability of planetary wavenumbers 1 and 2 in the troposphere and stratosphere, while EOF5 and EOF6 were consistent with wave reflections events. Averaged composite magnitudes of WVE showed amplification of both EOF5 and EOF6 occurred in the subset of WVEs with CAOs within 15- and 30-days but did not occur in the group where a CAO did not occur. The result suggested wave reflection concurrent with a WVE was critical for genesis of stratosphere-troposphere coupled CAOs in the U.S. Likewise, the results show that wave reflection events are statistically and dynamically related to stratospheric variability with CAOs, and wave reflection events are common modes of variability in the coupled troposphere-stratosphere system.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.