Assessing Physical Relationships Between Forcing Mechanisms and Boundary Layer Variability at McMurdo Station, Antarctica

Radiosonde and surface meteorology and flux observations at McMurdo Station Antarctica from
23 November 2015 through 5 January 2017 were used to analyze the boundary layer variability present
annually and seasonally. These measurements were taken during the ARM (Atmospheric Radiation
Measurement) West Antarctic Radiation Experiment (AWARE), which was conducted with the aims of
explaining the dynamical mechanisms of climate change in West Antarctica, one of the most rapidly
warming places on Earth. Radiosondes were launched twice per day during AWARE, and the hundreds of resulting
potential temperature profiles were analyzed using a self-organizing map neural network to identify the
range of potential temperature profiles present in each season of observations as well as annually. The
goal of the self-organizing map analysis is to view the differences in temperature profiles in the lowest
500 meters of the atmosphere in a useful and compact way while still capturing all of the observations
during the study period. Instead of viewing hundreds of individual profiles, these data are organized by
the self-organizing map into patterns of similar potential temperature profiles. It was found that in the
winter (MJJA), strongly stable profiles dominate the boundary layer structure, and weakly stable profiles
dominate in the summer (DJ). Flux and surface meteorological data, such as wind speed, can also be
viewed in this way by calculating the averages of these data taken at the time of the radiosonde launches
that correspond to each self-organizing map pattern, a technique called compositing. For example, this
technique was used to calculate the average wind speed at the time of launch for each self-organizing map
pattern, and it was found that weak winds correspond with stronger inversions, and stronger winds
correspond to a well-mixed, weakly stable profile. Profiles of bulk Richardson numbers were calculated
for each sounding to evaluate the layers of the atmosphere that are turbulent, and how this varies across
the identified potential temperature profiles in the self-organizing map. Areas of enhanced stability in the
boundary layer, such as surface inversions, and/or areas of especially weak wind shear reflect more
positive bulk Richardson numbers compared to areas of weak stability and/or areas of strong wind shear.
Observations of surface radiative fluxes, turbulent fluxes, relative humidity, and temperature have been
utilized in this analysis to show how these observations relate to changes in boundary layer stability. For
example, it was found in this study that the average sensible heat flux in the summer (DJ) was generally
less positive for weakly stable profiles compared to strongly stable profiles. Additionally, averages of the
downwelling longwave radiation in each self-organizing map pattern revealed that downwelling longwave
radiation was larger for weakly stable profiles and smaller for strongly stable profiles in the winter
(MJJA). This is expected because reduced downwelling longwave radiation in weakly stable winter
patterns is consistent with the negative radiation budget which favors the formation of surface inversions.
This presentation will further discuss these findings and the physical relationships between the
composited variables and the self-organizing map potential temperature patterns.