Evaluating Different Techniques for Constraining Lower Atmospheric Variability in an Upper Atmosphere General Circulation Model: A Case Study During the 2010 Sudden Stratospheric Warming. (15th December 2018)

• Record Type: Journal Article
• Authors: Jones, McArthur; Drob, Douglas P.; Siskind, David E.; McCormack, John P.; Maute, Astrid; McDonald, Sarah E.; Dymond, Kenneth F.
• Is Part Of: Journal of advances in modeling earth systems. Volume 10:Number 12(2018)
• Page Start: 3076
• Page End: 3102
• Publication Date: 2018-12-15
• Subjects: sudden stratospheric warming -- mesosphere lower thermosphere -- nudging -- ionosphere -- atmospheric tides; Geological modeling -- Periodicals; Climatology -- Periodicals; Geochemical modeling -- Periodicals; 551.5011
• DOI: 10.1029/2018MS001440
• Languages: English
• ISSNs: 1942-2466

Abstract

Abstract: We analyze the effects specified dynamics (SD) and 4D Tendency nudging have on accurately reproducing the middle and upper atmospheric variability induced by the 2010 sudden stratospheric warming (SSW) event in the National Center for Atmospheric Research thermosphere‐ionosphere‐mesosphere‐electrodynamics general circulation model (TIME‐GCM). TIME‐GCM numerical experiments were performed using constrained middle atmospheric winds and temperatures from a high‐altitude version of the Navy Global Environmental Model to compare the previously implemented SD scheme, with the newly implemented 4D Tendency scheme. Model comparisons focused on zonal mean winds, composition, planetary waves, and tides in the thermosphere‐ionosphere system. Through 4D Tendency nudging we reveal that coupling coefficients of the one‐way SD coupling approach between the TIME‐GCM and observed SSW conditions were too strong. Prior implementations produced unusually strong vertical shears in the zonal mean winds in the mesosphere and lower thermosphere (MLT), where the model is free running. Differences in zonal mean MLT winds between SD and 4D Tendency nudging simulations resulted in migrating diurnal (DW1) and semidiurnal (SW2) tidal amplitude differences at lower thermospheric altitudes. The consequences of simulating different MLT dynamics using SD and 4D Tendency nudging in the overlaying ionosphere are reported and validated using electron density data from the Constellation Observing System for Meteorology, Ionosphere, and Climate satellites. Although we demonstrate that SD and 4D Tendency nudging techniques are approximately equivalent, results presented herein establish that 4D Tendency nudging has the added potential to identify physical model parameters that contribute to data‐model differences during the 2010 SSW. Plain Language Summary: Understanding the meteorology of Earth's upper atmosphere on the time scales necessary to forecast space weather requires knowledge of the sources and propagation conditions of waves originating in the lower atmosphere, in addition to solar forcing. This principle underlies the ongoing effort in the middle and upper atmospheric modeling communities to replicate short‐term variability of the mesosphere, thermosphere, and ionosphere associated with lower atmospheric phenomena that couple Earth's lower and upper atmosphere. Using a commonly used upper atmosphere general circulation model, this study demonstrates that the same model will produce different results in Earth's thermosphere and ionosphere depending on the technique employed to constrain model fields in the underlying middle atmosphere. Simulating different model ionospheres (and thermospheres) could have an important impact on the uncertainty associated with ionospheric forecasts used by communication and navigation users (e.g., Global Positioning System). We therefore evaluate two leading methods for constraining lower atmospheric variability in order to better understand the underlying assumptions associated with each. We show how these techniques, previously thought to be mathematically distinct, can be reconciled provided certain considerations are made about the strength of and the time constants associated with constraining an upper atmospheric general circulation model to reanalysis data. Key Points: We compare TIME‐GCM simulations of the 2010 sudden stratospheric warming (SSW) using specified dynamics (SD) and 4D Tendency nudging The 4D Tendency nudging is different than SD nudging as data‐model differences are represented by a source term in the prognostic equations Selecting appropriate nudging parameters is important to accurately reproducing SSW induced thermosphere‐ionosphere variability in TIME‐GCM