Understanding Strong Neutral Vertical Winds and Ionospheric Responses to the 2015 St. Patrick's Day Storm Using TIEGCM Driven by Data‐Assimilated Aurora and Electric Fields. Issue 2 (13th February 2023)

• Record Type: Journal Article
• Authors: Lu, Xian; Wu, Haonan; Kaeppler, Stephen; Meriwether, John; Nishimura, Yukitoshi; Wang, Wenbin; Li, Jintai; Shi, Xueling
• Is Part Of: Space weather. Volume 21:Issue 2(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-02-13
• Subjects: space weather modeling -- data assimilation -- aurora and electric fields -- TIEGCM -- PFISR and SuperDARN -- THEMIS ASIs; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2022SW003308
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: As one of the strongest geomagnetic storms in Solar Cycle 24, the 2015 St. Patrick's Day storm has attracted significant attention. We revisit this event by taking advantage of simultaneous observations of high‐latitude forcings (aurora and electric fields) and ionosphere‐thermosphere (I‐T) responses. The forcing terms are assimilated to drive the Thermosphere Ionosphere Electrodynamics General Circulation Model (TIEGCM) using a newly adopted Lattice Kriging method (Wu & Lu, 2022, https://doi.org/10.1029/2021SW002880 ; Wu et al., 2022, https://doi.org/10.1029/2022SW003146 ). Compared to the default run, the TIEGCM simulation with assimilation captures: (a) secondary E‐region electron density peak due to aurora intensification; (b) strongly elevated ion temperatures (up to ∼3000 K) accompanied by a strong northward electric field (∼80 mV/m) and associated ion frictional heating; (c) elevation of electron temperatures; and (d) substantially enhanced neutral vertical winds (order of 50 m/s). Root‐mean‐square errors decrease by 30%–50%. The strong neutral upwelling is caused by large Joule heating down to ∼120 km resulting from enhanced aurora and electric field. Data assimilation increases the height‐integrated Joule heating at Poker Flat to a level of 50–100 mW/m 2 while globally, its maximum value is comparable with the default run: the location of energy deposition becomes guided by data. Traveling atmospheric disturbances in the assimilation run show stronger magnitudes and larger extension leading to an increase of vertical wind variability by a factor of ∼1.5–3. Our work demonstrates that data assimilation of model drivers helps produce realistic storm‐time I‐T responses, which show richer dynamic range, scales, and variability than what has been simulated before. Plain Language Summary: Originated from activity of the Sun, space weather can be notoriously hazardous to space security and infrastructure on the ground. An accurate prediction of space environment during storm time is important to mitigate such impacts and enhance space situation awareness. A long‐lasting problem for space weather modeling using Ionosphere‐Thermosphere (I‐T) models is the lack of realistic forcing as a driver, which often causes an underestimation of storm impacts, especially locally. To conquer this difficulty, our work adopts and extends a data assimilation technique which has been used for lower‐atmosphere studies to assimilate aurora and electric fields. The I‐T model simulation driven by data assimilation captures realistic dynamics and electrodynamics with model results closer to observations than the one driven by empirical forcings (default run). The more realistic model simulation illustrates that space weather impacts are more dynamic, cause stronger disturbances, and show multi‐scale features, compared to previous understandings. Key Points: Thermosphere Ionosphere Electrodynamics General Circulation Model driven by data‐assimilated aurora and electric field better resolves the I‐T system responses to the 2015 St. Patrick's Day storm Strong neutral upward winds are induced by the enhancement of aurora and electric field, which increases local Joule heating in E‐region Realistic storm‐time I‐T responses should have larger dynamic range, richer scales, and stronger variability than default model simulations