Publication: Journal of Geophysical Research-Space Physics; Authors: S. Datta-Barua, L. Nutter, N. Pedatella, V. L. Harvey, and N. Wang

Left: Specified Dynamics Whole Atmosphere Community Climate Model with thermosphere-ionosphere eXtension (SD-WACCMX) Lagrangian coherent structures (LCSs) for a space shuttle launch during (a) spring and (b) summer. Right: Observations of GUVI Lyman-alpha column emission rates in Rayleighs reprinted from Meier et al. (2011) and annotated with the LCS locations.

Left: Specified Dynamics Whole Atmosphere Community Climate Model with thermosphere-ionosphere eXtension (SD-WACCMX) Lagrangian coherent structures (LCSs) for a space shuttle launch during (a) spring and (b) summer. Tracers Ao and Bo mark the initial location of space shuttle water vapor deposition, and are plotted tri-hourly as A (white) and B (red) advect, ending at Af and Bf. Right: Observations of GUVI Lyman-alpha column emission rates in Rayleighs reprinted from Meier et al. (2011) and annotated with the LCS locations. The results illustrate that the LCSs tend to provide a barrier for the water vapor plume generated by the space shuttle.

Lagrangian coherent structures (LCSs), indicating regions of material transport, are derived from models of the lower thermosphere for five space shuttle water vapor plume events. LCSs defined using flow fields from the Specified Dynamics version of the Whole Atmosphere Community Climate Model with thermosphere eXtension (SD-WACCMX) are compared to global ultraviolet imager (GUVI) observations of water vapor documented in the literature. Tracer distributions and LCS locations largely agree with GUVI data for spring, summer, and winter. For a fall season event, SD-WACCMX predicts an LCS ridge to produce spreading not found in the observations. To understand the dependence of the location of LCSs on the method for constraining the model meteorology, for a shuttle plume event for which SD-WACCMX and data assimilative WACCM-X (WACCMX-DART) model outputs are both available, the global LCSs are compared. There are significant differences in the modeled LCSs at a geometric altitude of 109 km. Low-pass filtering the zonal and meridional winds with a Fourier series of up to zonal wavenumber 6 smooths out differences between models but does not eliminate the most significant LCS differences. There is improved inter-model agreement at a lower height of 60 km. The LCS method represents a novel approach in how to analyze material transport in the atmosphere, but should be used carefully with models of the lower thermosphere.