Characterization of the Thermospheric Mean Winds and Circulation During Solstice Using ICON/MIGHTI Observations

Abstract

Using the horizontal neutral wind observations from the Michelson Interferometer for Global High‐resolution Thermospheric Imaging (MIGHTI) instrument onboard NASA's Ionospheric Connection Explorer (ICON) spacecraft with continuous coverage, we determine the climatology of the mean zonal and meridional winds and the associated mean circulation at low‐to middle latitudes (10°S–40°N) for Northern Hemisphere summer solstice conditions between 90 and 200 km altitudes, specifically on 20 June 2020 solstice as well as for a one‐month period from 8 June–7 July 2020 and for Northern winter season from 16 December 2019–31 January 2020, which spans a 47‐day period, providing full local time coverage. The data are averaged within appropriate altitude, longitude, latitude, solar zenith angle, and local time bins to produce mean wind distributions. The geographical distributions and local time variations of the mean horizontal circulation are evaluated. The instantaneous horizontal winds exhibit a significant degree of spatiotemporal variability often exceeding ±150 m s⁻¹. The daily averaged zonal mean winds demonstrate day‐to‐day variability. Eastward zonal winds and northward (winter‐to‐summer) meridional winds are prevalent in the lower thermosphere, which provides indirect observational evidence of the eastward momentum deposition by small‐scale gravity waves. The mean neutral winds and circulation exhibit smaller scale structures in the lower thermosphere (90–120 km), while they are more homogeneous in the upper thermosphere, indicating the increasingly dissipative nature of the thermosphere. The mean wind and circulation patterns inferred from ICON/MIGHTI measurements can be used to constrain and validate general circulation models, as well as input for numerical wave models.

Full text

1. Introduction
Earth's thermosphere extending from ∼90km upwards is the outermost region of the atmosphere, where satellites
orbit the planet and a substantial portion of the solar ultraviolet (UV) radiation is absorbed by atmospheric gases.
This rarefied and highly dissipative region is influenced by a broad spectrum of internal atmospheric waves prop-
agating upward from the lower atmosphere (Dhadly, Emmert, Drob, McCormack, & Niciejewski,2018; Forbes
etal., 2021; Gavrilov etal.,2020; Hickey etal.,2011; Oberheide etal.,2015; Pancheva & Mukhtarov,2012;
Pancheva etal.,2020; Yiğit & Medvedev,2015) and by solar and geomagnetic processes (i.e., space weather)
Abstract Using the horizontal neutral wind observations from the Michelson Interferometer for Global
High-resolution Thermospheric Imaging (MIGHTI) instrument onboard NASA's Ionospheric Connection
Explorer (ICON) spacecraft with continuous coverage, we determine the climatology of the mean zonal and
meridional winds and the associated mean circulation at low-to middle latitudes (10°S–40°N) for Northern
Hemisphere summer solstice conditions between 90 and 200km altitudes, specifically on 20 June 2020 solstice
as well as for a one-month period from 8 June–7 July 2020 and for Northern winter season from 16 December
2019–31 January 2020, which spans a 47-day period, providing full local time coverage. The data are averaged
within appropriate altitude, longitude, latitude, solar zenith angle, and local time bins to produce mean wind
distributions. The geographical distributions and local time variations of the mean horizontal circulation are
evaluated. The instantaneous horizontal winds exhibit a significant degree of spatiotemporal variability often
exceeding ±150ms
−1. The daily averaged zonal mean winds demonstrate day-to-day variability. Eastward
zonal winds and northward (winter-to-summer) meridional winds are prevalent in the lower thermosphere,
which provides indirect observational evidence of the eastward momentum deposition by small-scale gravity
waves. The mean neutral winds and circulation exhibit smaller scale structures in the lower thermosphere
(90–120km), while they are more homogeneous in the upper thermosphere, indicating the increasingly
dissipative nature of the thermosphere. The mean wind and circulation patterns inferred from ICON/MIGHTI
measurements can be used to constrain and validate general circulation models, as well as input for numerical
wave models.
Plain Language Summary Atmospheric horizontal winds (i.e., motion of the neutral air),
composed of zonal (east-west) and meridional (north-south) components, play an important role for the
energy and momentum balance of the atmosphere and ionosphere. Due primarily to a lack of observations,
winds in the thermosphere are not well sampled. In this study we use the horizontal winds measured from
90 to 200km altitude by the Michelson Interferometer for Global High-resolution Thermospheric Imaging
instrument onboard NASA's Ionospheric Connection Explorer spacecraft to generate two-dimensional maps
of zonal and meridional winds, and of the resulting horizontal motion (or circulation) in the thermosphere for
Northern Hemisphere solstice conditions. Specifically, winds at solstice (20 June 2020) and a 1month Northern
summer solstitial period (8 June to 7 July 2020) and a 47-day winter solstitial period (16 December 2019 to 31
January 2020) have been analyzed. Mean winds show significant spatial variation as a function of time, often
demonstrating tidal variability.
YIĞIT ETAL.
© 2022 The Authors.
This is an open access article under
the terms of the Creative Commons
Attribution-NonCommercial License,
which permits use, distribution and
reproduction in any medium, provided the
original work is properly cited and is not
used for commercial purposes.
Characterization of the Thermospheric Mean Winds
and Circulation During Solstice Using ICON/MIGHTI
Observations
Erdal Yiğit1 , Manbharat Dhadly2, Alexander S. Medvedev3 , Brian J. Harding4 ,
Christoph R. Englert2 , Qian Wu5,6 , and Thomas J. Immel4
1Department of Physics and Astronomy, George Mason University, Space Weather Lab, Fairfax, VA, USA, 2U.S. Naval
Research Laboratory, Washington, DC, USA, 3Max Planck Institute for Solar System Research, Göttingen, Germany,
4UC Berkeley, Space Sciences Laboratory, Berkeley, CA, USA, 5High Altitude Observatory, NCAR, Boulder, CO, USA,
6COSMIC Program UCAR/UCP, Boulder, CO, USA
Key Points:
• Mean zonal and meridional winds
are derived for Northern summer
and winter solstice conditions from
Ionospheric Connection Explorer/
Michelson Interferometer for Global
High-resolution Thermospheric
Imaging observations
• Horizontal winds exhibit a significant
degree of spatiotemporal variability,
exceeding ±150ms
−1
• Zonal and meridional mean winds
are more homogeneous in the upper
thermosphere and exhibit reversal in
the lower thermosphere
Correspondence to:
E. Yiğit,
eyigit@gmu.edu
Citation:
Yiğit, E., Dhadly, M., Medvedev, A. S.,
Harding, B. J., Englert, C. R., Wu, Q.,
& Immel, T. J. (2022). Characterization
of the thermospheric mean winds and
circulation during solstice using ICON/
MIGHTI observations. Journal of
Geophysical Research: Space Physics,
127, e2022JA030851. https://doi.
org/10.1029/2022JA030851
Received 15 JUL 2022
Accepted 14 OCT 2022
Author Contributions:
Conceptualization: Erdal Yiğit,
Alexander S. Medvedev, Qian Wu
Formal analysis: Erdal Yiğit
Funding acquisition: Erdal Yiğit
Investigation: Erdal Yiğit, Manbharat
Dhadly, Alexander S. Medvedev,
Christoph R. Englert
Methodology: Erdal Yiğit, Manbharat
Dhadly, Alexander S. Medvedev, Brian
J. Harding
Project Administration: Erdal Yiğit
Resources: Erdal Yiğit
Supervision: Erdal Yiğit
Validation: Erdal Yiğit, Manbharat
Dhadly, Christoph R. Englert
10.1029/2022JA030851
RESEARCH ARTICLE
1 of 23

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
2 of 23
from above (Deng etal.,2018; Dhadly, Emmert, Drob, Conde, etal.,2018; Emmert,2015; Schunk & Sojka,1996;
Shiokawa & Georgieva,2021; Ward etal.,2021; Yiğit, Knížová, etal.,2016). The forces acting on the neutral
flow are rarely in balance in the thermosphere, thus giving rise to an enhanced spatiotemporal variability with
turbulent to global scales.
The goal of this paper is to characterize the thermospheric mean zonal and meridional winds, and circulation
during solstice at low-to midlatitudes (10°S–40°N) using observations from the Michelson Interferometer for
Global High-resolution Thermospheric Imaging (MIGHTI) instrument (Englert et al.,2017) onboard NASA's
Ionospheric Connection Explorer (ICON) spacecraft during 2020 Northern Hemisphere solstice conditions. Due
primarily to poor observational coverage, neutral winds have been insufficiently characterized at thermospheric
altitudes. Horizontal winds and the associated circulation play an essential role in the energy and momentum
budget of the thermosphere and ionosphere. They modulate upwelling/downwelling of air, influence the critical
filtering and dissipation of internal atmospheric waves propagating upward, drive low-latitude ionospheric elec-
trodynamics, transport major chemical species (e.g., O, N2, NO), generate neutral drag on the ions, and redistrib-
ute thermospheric energy and momentum in general. Thermospheric winds are primarily horizontal, however, in
the regions of convergence or divergence, upwelling or downwelling (i.e., vertical motion) can occur as a conse-
quence of the principal of conservation of mass (Rishbeth etal.,1969; R. W. Smith,1998). Therefore, character-
ization of the mean winds is essential for our understanding of the thermosphere-ionosphere system as a whole.
Various methods were utilized to observe winds over a broad range of altitudes from the upper mesosphere to the
thermosphere. However, lower thermospheric winds are more routinely observed than in the upper thermosphere.
A summary of historical and current observations is presented by other researchers (e.g., Dhadly etal., 2019;
Drob etal., 2008,2015). Chemical release wind measurements carried out in different sites around the world
can provide profiles of wind velocity and high vertical resolution wind shear from ∼80–140km (Larsen,2002;
Lehmacher etal., 2022). Incoherent scatter radars around the world use measured ion drifts to derive neutral
winds from ∼90–130km (Hysell et al.,2014; Zhang et al.,2003). Meteor echoes and meteor radars are used
to retrieve wind profiles between ∼90–110km (Conte etal., 2022; Oppenheim et al., 2009). Various types
of ground-based Fabry-Perot Interferometers (FPI) have been employed since the 1980s across the globe to
understand neutral wind dynamics of the upper and lower thermospheric winds (Aruliah etal.,2010; Conde &
Smith,1995; Makela etal.,2012; Meriwether,2006).
Satellites can provide measurements of thermospheric winds at a broad range of thermospheric altitudes.
Dynamic Explorer 2 (DE2), was the first to monitor upper thermospheric neutral winds from space utilizing
an FPI (Killeen & Roble,1988). The wind imaging interferometer (WINDII) aboard the Upper Atmos-
phere Research Satellite (UARS) retrieved neutral winds based on the interfermetric limb measurements
of the visible airglow emissions of 557.7nm O
1S (green line) and 630.0nm O
1D (redline) between 90 and
300 km (Emmert etal.,2001; Shepherd etal., 2012). Thermosphere, Ionosphere, Mesosphere Energet-
ics and Dynamics/TIMED Doppler Interferometer (TIMED/TIDI) primarily focused on monitoring MLT
winds, launched in 2001 is still operational after 20years in orbit (Niciejewski etal., 2006). Cross-track
winds were derived from accelerometer measurements between 250 and 400km by the (Gravity Field and
Steady-State Ocean Circulation Explorer (GOCE) (Doornbos etal.,2014)) and CHAllenging Minisatellite
Payload (CHAMP) satellite (Lieberman etal.,2013; Liu etal.,2006).
Despite the extensive measurements by ground-based and space-borne instruments, thermospheric winds have
been insufficiently sampled so far. Much of the understanding of the thermospheric winds is based on dedicated
first-principle global scale modeling (e.g., Deng etal.,2018; Geisler,1966; Miyoshi etal., 2014; Richmond
etal.,1992; Vichare etal.,2012; Yiğit, Frey, etal.,2016) and empirical models are routinely used to study the
global behavior of winds (Dhadly etal.,2019; Drob etal.,2015). Depending on the type of model and observa-
tions, model-data agreement is often partially achieved (e.g., Tang etal.,2021). Thermospheric wind is an essen-
tial input parameter for ionospheric models and a better representation of neutral winds is needed for improved
space weather modeling (David etal.,2014). While models are powerful tools to study the different forces shap-
ing the winds, often the simulated winds are insufficiently constrained in models, therefore ground-based and
space-borne measurements of neutral winds are crucial for validating first principal models and to obtain a more
complete physical understanding of thermospheric dynamics.
Although ICON has started observing the thermosphere only recently, MIGHTI observations have already been
used to study thermospheric winds and to compare them to other ground-based and space-borne instruments.
Visualization: Erdal Yiğit
Writing – original draft: Erdal Yiğit
Writing – review & editing: Erdal
Yiğit, Manbharat Dhadly, Alexander S.
Medvedev, Brian J. Harding, Christoph R.
Englert, Qian Wu, Thomas J. Immel

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
3 of 23
Recent studies have validated some aspects of ICON horizontal winds with respect to Fabry-Perot interferometers
and meteor radars (Chen etal.,2022; Harding etal.,2021; Makela etal.,2021). Dhadly etal.(2021) compared
MIGHTI winds to the University of Michigan TIMED Doppler Interferometer (TIDI) level 3 data, contributing
to the validation of MIGHTI winds as well as providing guidance toward improving TIDI winds. In addition,
this study revealed the longitudinal variations in neutral winds associated with non-migrating tides, which are
currently missing from the existing wind climatologies. Forbes etal.(2021) analyzed coincident ICON measure-
ments of neutral horizontal winds, ion drifts, and densities and demonstrated a direct link between the day-to-day
variability of the wave-4 structure in the E-region and drifts and densities of ions in the F-region ionosphere.
More recently, England etal.(2022)'s analysis of MIGTHI winds have shown that strong wind shear is a common
feature of the lower thermosphere between 100 and 130km. Yamazaki etal. (2022) have analyzed concurrent
observations from COSMIC-2 and ICON/MIGHTI wind and found a correlation between the negative vertical
shear of the eastward wind and the occurrence rate of the sporadic E layer. Characterization of varying fields is
often represented in the form of appropriately defined mean, which requires a sufficient degree of observational
coverage in space and time. The averaging for a field variable ψ that is often conducted over time t and longitude x

(𝑧 )= 1
ΔΔ∫
+Δ

∫
+Δ

(𝑧 𝑧 𝑧 )

(1)
is generally referred to as “zonal mean,” if Δx spans all longitudes. In this paper, we perform averaging of ICON/
MIGHTI neutral winds over longitude, latitude and local time, and generally call the results “mean winds” as
well. Besides the physical importance of winds and their mean structure discussed above, they are routinely used
to validate theory and global scale models (or general circulation models) in the middle and upper atmosphere
(e.g., Dempsey etal.,2021; Garcia etal.,2007; Griffith etal.,2021; Koval etal.,2022; Lieberman etal.,2000;
Yiğit etal.,2021).
This paper analyzes the thermospheric horizontal winds between 90 and 200 km during December and June
solstice conditions as observed in 2019/2020 by ICON/MIGHTI. Next section describes the MIGHTI neutral wind
measurements used in this study; Section3 presents the results for the solstice wind distribution and circulation;
Section4 provides a discussion of the observed winds, and a summary and conclusions are given in Section5.
2. Materials and Methods
2.1. ICON Mission and Data
The ICON mission was launched in 2019 and has been surveying the low-latitude thermosphere-ionosphere
system above 90km in unprecedented detail. Its primary goal is to explore Earth's thermosphere-ionosphere
system and its connection to geospace as well as terrestrial drivers (Immel etal., 2018). The MIGHTI wind
observations are based on the Doppler shift measurements of the green line (λ = 557.7 nm) and red line
(λ=630nm) emissions of atomic oxygen. In this paper, we analyze the cardinal winds (i.e., zonal u and merid-
ional v components) from the MIGHTI instrument (Englert etal.,2017). The details of the wind retrieval algo-
rithm are described in the work by Harding etal.(2017). In the following, we outline the data selection, quality,
and spatiotemporal coverage.
In this study, we focus only on the MIGHTI green line neutral winds, which cover the altitude range from
∼90–200km during daytime and ∼90 to ∼115km both daytime and nighttime. Typically, MIGHTI daytime line
of sight (LOS) wind observations are available at a 30-s cadence, while nighttime LOS wind measurements are
available at a 60-s cadence. Northward and eastward components of the winds are obtained by pairing these LOS
wind measurements of MIGHTI A and MIGHTI B sensors, which are taken approximately 8min apart.
The quality of MIGHTI/ICON data was taken into account in the analyses. Each wind measurement was assigned
a data quality flag corresponding to “good,” “good, but use with caution,” and “bad.” We removed all data
with bad quality and also excluded outliers with wind magnitudes exceeding 300ms
−1. The result is shown in
FigureA1, where the zonal winds are plotted with different quality flags. The typical accuracy of winds derived
from the green line emission is 12ms
−1 or better, as compared with meteor radars (Harding etal.,2021). It is
seen that such procedure leaves a significant amount of data to maintain solid statistics.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
4 of 23
In general, strong mean winds (jets) develop in both hemispheres in the middle atmosphere during solstices. They
produce enhanced wave filtering and/or modify propagation conditions for internal atmospheric waves with great
dynamical implications for the upper atmosphere. Therefore, the solstice winds and circulation are highly varia-
ble and interesting to study. Gravity waves and tides significantly affect the mean flow in the whole atmospheric
system. During the Northern winter season additional effects of planetary waves and occurrence of sudden strat-
ospheric warming (SSW) events can complicate the interpretation of observations. Therefore, we have focused
in this study primarily on the Northern Hemisphere summer solstice (Sections 3.1–3.2). However, in order to
complement June/July solstice results, we have added a complementary analysis for December 2019–January
2020 Northern winter solstice (Section3.3.), which did not include sudden warming events.
2.2. ICON Coverage
Figure1 illustrates the spatiotemporal ICON data coverage for 20 June 2020, after having removed the bad quality
data. This subset for the analysis contains 2206 individual wind profiles. Panels a–c show the longitude, latitude,
and local time coverage as a function of UTC. It is seen that all longitudes and latitudes between −10° and +40°
are well covered. There are some nighttime data gaps, however, overall all local times were observed. Panels
(d and e) present the altitude coverage as a function of latitude and local time. It is seen that during daytime,
the altitude coverage extends into the upper thermosphere, while the nighttime observations are available only
between 90 and 115km. Also, latitudes between −10° and +10° are not observed above 115km on 20 June, since
they coincide with nighttime. The latitude-longitude distribution highlights the good spatial coverage between
−10° and +40° for all longitudes, with some data gaps around 300° longitude in the Southern Hemisphere
associated with South Atlantic Anomaly (SAA). The latitude-solar zenith angle distribution of measurements
(panel g) at ∼106km suggests that the Northern Hemisphere is covered primarily at daytime, while the Southern
Hemisphere latitudes are observed at nighttime. Longitude-solar zenith angle variations at ∼106km (panel h)
show that, for a given longitude, both nighttime and daytime data are available except in the region around 300°
longitude.
A single day observation is not sufficient to produce a wind climatology. Therefore, in order to obtain a more
consistent picture of mean winds and circulation, we have used one month of continuous ICON observations
Figure 1. Spatiotemporal coverage of Ionospheric Connection Explorer/Michelson Interferometer for Global High-resolution Thermospheric Imaging observations
during 20 June 2020. Only good quality data (2206 profiles) have been retained.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
5 of 23
from 8 June to 7 July 2020, representative of Northern Hemisphere summer solstice conditions. Neutral wind
measurements were binned with respect to altitude, latitude, longitude, local time and solar zenith angle with
bin sizes of 5km, 5°, 30°, 1–2hr, and 10° bins, respectively. Figure2 shows the latitude and solar zenith angle
coverage along with the corresponding distributions of the zonal and meridional winds at ∼106km during this
period. The daily mean F10.7cm solar radio flux and geomagnetic Kp-index variations shown in panels (e and f)
indicate quiet solar and geomagnetic conditions. The latitudes between 10°S and 40°N are continuously covered,
however, the latter period of June 2020, the southern latitudes are more sparsely sampled than northern latitudes.
Also, while all solar zenith angles between 10 and 140° are observed, the nighttime, in general, is observed more
sparsely than daytime. The zonal and meridional winds at ∼106km are generally up to ±150ms
−1 throughout
this period, exhibiting noticeable degree of day-to-day variability.
3. Results
3.1. Zonal and Meridional Winds on 20 June 2020
Figure3 shows variations of the zonal (upper panels) and meridional (lower panels) winds with altitude, longi-
tude, latitude, and local time on 20 June 2020. The profiles are plotted without any binning. The average of all
vertical profiles are shown with the red line in panels a and f. The zonal and meridional winds demonstrate a
high degree of spatiotemporal variability. They are generally faster at higher altitudes, occasionally exceeding
±150ms
−1. Also, the daytime winds observed in the Northern Hemisphere are faster than those during nighttime
at low-latitudes. Note also that the intermittent values of the winds are much larger than their average quantity.
Therefore, the latter should be treated with caution as not being representative of instantaneous numbers.
In order to study the observed wind variability, we have evaluated in Figure4 the occurrence rates of the wind
speeds binned in 5ms
−1 intervals on 20 June 2020 at three representative altitude layers 94–103km, 106–114km,
and 194–202km such that each layer included equal number of data points. The speeds shown as a function of
number of measurements exhibit a Gaussian distribution generally centered around slow speeds. The associated
standard deviations, σu and σv, which are a proxy for wind variability, are shown in the upper left corner. The
nighttime wind variabilities are greater than daytime ones and increase with height. It should be noted that the
reported standard deviations are the root mean-square of wind variability and observational errors, which vary
with height.
Figures5 and6 present the altitude distribution of the zonal and meridional winds, respectively, binned as func-
tions of latitude, longitude, local time, and solar zenith angle, where red and blue shadings represent positive
(eastward, northward) and negative (westward, southward) wind values, respectively. It is seen that the behavior
of the observed zonal and meridional winds in the lower thermosphere (90–110km) is remarkably different than
in the upper thermosphere. Zonal winds are predominantly eastward in the lower thermosphere, in particular,
in northern latitudes between 10 and 40°N, with magnitudes of up to 40ms
−1. Above 110km, they are clearly
westward, have speeds exceeding −70ms
−1 and essentially associated with daytime values. The eastward wind
regime in the lower thermosphere itself exhibits a substantial degree of variability, when viewed as a function of
longitude, local time, and solar zenith angle. For example, the daytime lower thermospheric winds are more east-
ward compared to those during nighttime. Meridional winds exhibit alternating patterns of flow direction with
altitude. They are predominantly southward during daytime in the lower thermosphere between 90 and 100km;
northward between 100 and 120km with speeds reaching 60ms
−1; southward between 120 and 190km, and
northward again around 190–200km, especially around noontime, where the winds above 110km correspond to
daytime measurements. The features near the terminator in Figures5c and5d are potentially an artifact associated
with low airglow signal, which are expected to be corrected in the next version of the wind data.
3.2. Thermospheric Mean Winds and Circulation During Northern Hemisphere Summer Solstice
A diurnal mean provides a short-term glimpse of the circulation, however it is not sufficient for deriving a more
accurate view of the climatology of mean winds due to limitations of the orbital coverage to 1day and intrinsic
variability of the wind field. Therefore, we analyzed one month of continuous ICON observations from 8 June to
7 July 2020, which are representative of the solstitial dynamics during the Northern Hemisphere summer.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
6 of 23
Figure 2. Spatiotemporal coverage of Ionospheric Connection Explorer from 8 June to 7 July 2020 (a–b) and winds (c–d) at 105.95km. Daily mean 10.7cm solar radio
flux in solar flux units (sfu) is shown in panel e and the geomagnetic activity is shown in terms of the daily mean Kp-index in panel f. The different colors are for the
different days.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
7 of 23
Figure7 shows the sequence of diurnal-mean altitude-latitude cross-sections of the zonal winds. It clearly shows
that the winds strongly vary from day-to-day. Eastward winds with 10–40ms
−1 speeds are a robust feature of the
midlatitude lower thermosphere between 90 and 110km. The westward winds during daytime dominate above
120km. Their speeds vary from −10 to −80ms
−1, depending on the latitude, and amplify toward the end of June
and beginning of July.
We next determine the Northern Hemisphere summer solstitial climatology of the horizontal winds by averaging
over all measurements between 8 June and 7 July shown in Figure7. The results are plotted in Figure8 in the form
of altitude-latitude and altitude-local time distributions for both the zonal (upper panels) and meridional (lower
panels) components. In the lower thermosphere, the eastward winds are up to 40ms
−1, and the daytime westward
mean zonal wind in the upper thermosphere can exceed −60ms
−1, especially between 15° and 40°N. The west-
ward mean zonal winds decrease around 160km, such that the jet exhibits a split in altitude, especially, between
10 and 40°N. In the lower thermosphere, meridional winds are weakly southward (up to −20ms
−1) between
∼90–105km and northward between ∼105–120km. Above 120km, the meridional winds are directed south-
ward with speeds occasionally exceeding −60ms
−1, for example, at low-latitudes of the Southern Hemisphere
during daytime after dawn and before dusk. The meridional winds around 20°−40°N are, generally, slower than
at low-latitudes. In the lower thermosphere, both zonal and meridional components exhibit a distinct local time
variability, when all observed latitudes are considered. Overall, the observed monthly mean daytime meridional
winds are directed southward and nighttime winds are northward.
Another view of the observed winds is presented in Figure9, where the latitude-local time cross-sections of the
mean zonal (left) and meridional (right) winds are shown at three representative thermospheric altitudes. Within
the 90–105km layer, the zonal winds are mainly eastward at midlatitudes around dawn and dusk and vary semi-
diurnally. At equatorial latitudes, the winds are eastward in the morning sector and westward in the afternoon,
suggesting a diurnal variation. Meridional winds are southward during day and northward at night, which is
indicative of a diurnal signal as well. Higher up in the lower thermosphere between 105 and 120km, mean zonal
winds exhibit a more complex latitude-local time variability, however, meridional winds overall maintain a diur-
Figure 3. Altitude, longitude, latitude, and local time variations of the zonal (upper panels) and meridional (lower panels) winds in ms
−1 during 20 June 2020
(i.e., Northern Hemisphere Summer solstice) as observed by Ionospheric Connection Explorer/Michelson Interferometer for Global High-resolution Thermospheric
Imaging. Longitude and latitude variations include data from all altitudes between 88 and 200km, while the local time variations of the winds are shown for the lower
thermosphere 88–114km and upper thermosphere separately 117–200km. Mean winds are shown with the red curve. See Figure1 for the details of the spatiotemporal
coverage.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
8 of 23
Figure 4. Distributions of the zonal and meridional wind velocities by day and night on 20 June 2020. Wind speeds are
binned in 5ms
−1 intervals at three representative altitudes in the thermosphere. Each altitude layer includes equal number of
data points. Standard deviations are given in each plot in the upper left corner.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
9 of 23
nal behavior at low-latitudes. In the upper thermosphere, the zonal winds are strongly westward at midlatitudes
around dawn exceeding 100ms
−1, moderately westward in general during day, and reverse their direction to
strong eastward flow before dusk. Again, the low airglow signal could have potentially affected the magnitude
of these winds at the terminator. Southward meridional winds dominate in the upper thermosphere during day,
similar to the lower thermosphere. They indicate the global north-to-south branch of the solstitial circulation cell
in the upper atmosphere.
Figure10 presents the latitude-longitude cross-sections of the zonal and meridional winds at three representative
altitudes in the thermosphere. It shows that the monthly mean morphology of the horizontal winds, that is, wind
magnitudes and directions, significantly changes in the lower thermosphere between 90 and 120km. Within the
90–105km layer, eastward winds of up to 40ms
−1 and southward winds of up to −30ms
−1 are prevalent in the
Northern Hemisphere. The Southern Hemisphere low-latitudes are characterized by relatively slow westward
winds. Around 105–120km altitude, zonal winds in the Northern Hemisphere reverse the direction to westward
and the northward flow becomes more prevalent compared to that at 90–105km. The upper thermosphere at
185–200km is dominated by westward and southward winds with speeds exceeding −60ms
−1 and −40ms
−1,
Figure 5. Altitude distributions of zonal wind speed (m s
−1) as a function of latitude, longitute, local time, and solar zenith
angle as observed by Ionospheric Connection Explorer/Michelson Interferometer for Global High-resolution Thermospheric
Imaging on 20 June 2020.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
10 of 23
respectively. The latter are a part of the pole-to-pole solstitial meridional circulation. These upper thermosphere
winds are, generally, much faster and more homogeneous than in the lower thermosphere.
Atmospheric circulation associated with neutral winds play an important dynamical role in redistribution of
momentum, energy, and mass in the thermosphere. The associated latitude-local time and latitude-longitude and
distributions of the mean horizontal circulation are seen in terms of velocity vectors in Figures11 and12 at three
representative altitudes in the thermosphere. This representation of the winds shows the direction as well as the
flow speeds. Upward and downward directed vectors represent northward (toward the North Pole) and southward
flow (toward the South Pole). While vectors directed to the right and left are for eastward and westward flow,
respectively, in latitude-longitude cross-sections, they facilitate an interpretation of the winds relatively to the
day-night sectors in latitude-local time plots. For example, winds flowing from the dayside to the nightside or
vice versa would be revealed. A large degree of spatiotemporal variability is seen especially within the lower
thermosphere. Eastward and southward winds prevail in the layer between 90 and 105km and westward and
northward as well as southward flows are found within 105–120km. Between 90 and 105km, dusk-to-dawn
and nighttime poleward flow is followed by a daytime equatorward flow. The mean circulation between 105 and
120km exhibits the greatest degree of complexity in terms of varying scale sizes and vortices, generally directed
westward during day and eastward during night diverging around the subsolar point. Generally flow speeds are
in the order of 50ms
−1. In the upper thermosphere, the observed daytime circulation is more easily discernible
Figure 6. Same as Figure5, but for the meridional winds.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
11 of 23
Figure 7. Day-to-day variations of the altitude-latitude distributions of the thermospheric zonal winds between 90 and 200km and 10°S–40°N in ms
−1 as observed
by Ionospheric Connection Explorer/Michelson Interferometer for Global High-resolution Thermospheric Imaging from 8 June–7 July 2020. For each day, all observed
longitudes and local times/solar zenith angles are included.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
12 of 23
with clear diverging patterns around the subsolar point. Geographically south-westward circulation prevails with
flow speeds exceeding 100ms
−1 at midlatitudes.
3.3. Thermospheric Mean Winds and Circulation During Northern Hemisphere Winter Solstice
So far we discussed the Northern Hemisphere summer season. Energy deposition from the Sun varies seasonally
changing the differential heating and the pressure gradient force, which alters the circulation patterns, hence
atmospheric wave propagation and its effects.
In order to complement the results presented above, we next analyze the winds and circulation patterns during
the Northern Hemisphere winter. Figure13 shows the altitude-latitude and altitude-local time distributions of
the mean zonal and meridional winds averaged from 16 December 2019–31 January 2020, which is represent-
ative of the December solstice (or Northern winter) conditions. They are presented in a manner similar to June
solstice mean winds (Figure8). Note that this period does not include any stratospheric sudden warming (SSW)
events (Roy & Kuttippurath,2022). It is seen that the zonal winds are westward (with magnitudes <40ms
−1)
Figure 8. Zonal and meridional wind climatology from 90 to 200km presented as altitude-latitude and altitude-local
time cross-sections based on Ionospheric Connection Explorer/Michelson Interferometer for Global High-resolution
Thermospheric Imaging data from 8 June–7 July 2020. Data include daytime and nighttime measurements below 110km and
only daytime observations above 110km.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
13 of 23
at low-latitudes and eastward at midlatitudes (with magnitudes <20ms
−1), relatively slower than the summer
winds. During winter, slow meridional winds are seen between 90 and 120km, and strong eastward daytime
winds prevail above 120 km. Local time variations in the lower thermosphere indicate a semidiurnal varia-
tion of the zonal winds and a diurnal variation of the meridional winds in the lower thermosphere, which are
qualitatively somewhat similar to summer conditions, however, the wintertime daytime northward and nighttime
southward phase of the meridional wind variations are different compared to the summer one. In the upper ther-
mosphere, daytime meridional winds are northward exceeding 60ms
−1 at midlatitudes and flow easterly at dawn,
and somewhat westerly and easterly at dusk.
Figure 9. Latitude-local time distribution of zonal and meridional winds at three representative thermospheric altitudes.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
14 of 23
Figure 10. Latitude-longitude distributions of the zonal and meridional winds at three representative thermospheric altitudes: (a and b) 90–105km, (c and d)
105–120km, (e and f) 185–200km. Note that winds at 185–200km altitude are only daytime measurements.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
15 of 23
Figure14 shows the associated horizontal circulation during Northern winter
season, as was done for the summer solstice (Figure12). In the lower thermo-
sphere, the circulation is overall easterly and southward (winter-to-summer),
with geographically varying intensity. The upper thermosphere during
daytime is dominated by a westward and northward (summer-to-winter) flow,
becoming dominated by meridional component at midlatitudes.
4. Discussion
We have presented the mean zonal and meridional winds on 20 June 2020
as well as averaged over a one-month summer solstice period (8 June–7 July
2020) and 47-day winter solstice period (16 December 2019–31 January
2020), using continuous measurements of ICON/MIGHTI. The climatology
of thermospheric horizontal circulation for the summer and winter solstice
periods have been constructed for the first time. We next discuss dynami-
cal forces influencing the winds and some of the noteworthy features of the
observations, comparing our analysis to previous studies.
4.1. Dynamical Forces That Control Upper Atmospheric Winds
Multiple observations demonstrate that winds are extremely variable, espe-
cially in the lower thermosphere (e.g., Larsen,2002; Larsen & Fesen,2009;
Lehmacher et al., 2022). In order to characterize their climatology, the
data have to be averaged using appropriate bins over multiple days. What
processes drive the mean and variable structure of the winds is of great inter-
est. For this, it is instructive to discuss first the dynamical forces that control
the motion of an air parcel. The complete momentum balance for neutral
winds is given by (Yiğit,2018,p.112)
(2)
where u=(u, v, w) is the neutral wind vector, P=ρRT/M is the thermody-
namic pressure, with temperature T, mass density ρ, universal gas constant
R, and molar mass M; g is gravitational acceleration, Ω is the rotation rate of
Earth, τ is the viscous shear stress, uk is the velocity of particles of species
k, which the neutrals with collision frequencies νnk collide with, and f′ is the
momentum deposition by eddies or small-scale waves. On the right hand
side of Equation2 from left to right the momentum balance terms are advec-
tive forcing, pressure gradient force, gravity, Coriolis force, viscous stress,
frictional drag due to the interactions of neutrals with charged particles, for
example, ion drag, and wave-induced momentum deposition. For an incom-
pressible atmosphere the viscous shear stress is proportional to the vector
laplacian of the wind velocity, that is, ρ
−1∇ ⋅τ=ν∇
2u, where ν=μ/ρ is the
kinematic viscosity and μ is the dynamic viscosity. In numerical models, the wave-induced momentum forcing
f′ is often not resolved and has to be parameterized (e.g., Medvedev & Yiğit,2019; Yiğit etal.,2008). With
improving resolution, global numerical models are increasingly able to capture a larger portion of subgrid-scale
wave effects (e.g., Miyoshi etal.,2014). Depending on the altitude, latitude, local time, and time scales of
motion, various combinations of these dynamical forces shape the neutral wind circulation. In general, atmos-
pheric dynamics is nonlinear, which can lead to winds and circulation over a broad spectrum of spatiotemporal
scales and complexity, as demonstrated by ICON/MIGHTI wind analysis presented earlier.

 = −(⋅∇) −
1
∇+−2×+
1
∇⋅−∑

 (−)+′
,
Figure 11. Latitude-local time distribution of the mean horizontal circulation
at three representative thermospheric altitudes averaged over longitudes for 8
June–7July 2020. Note that the magnitude of the vector is 50ms
−1 in panels (a
and b), while it is 100ms
−1 in panel (c).

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
16 of 23
4.2. Thermospheric Winds and Circulation During Solstice
In order to illustrate one possible way of studying wind variability seen in Figure3, we have plotted the zonal
and meridional wind speeds as a function of their occurrence rates on 20 June 2020 (Figure4) They generally
exhibit a Gaussian distribution centered around slow speeds. Larger standard deviation found at night suggests
that winds are more variable during night, which, in turn, indicates an elevated atmospheric wave activity. This is
also suggested by more variable wind vectors at night (Figure11). The atmosphere is in general less dissipative
at night due, for example, to smaller molecular viscosity, providing more favorable propagation conditions for
small-scale waves. The wind variability is found to increase with height, which could be linked to larger wind
speeds at higher altitudes and growing with height wave amplitudes.
If dissipative effects such as wave breaking/saturation and viscosity are ignored in Equation2, then the large-scale
behavior of the thermospheric mean winds is controlled primarily by pressure gradient force generated by the
differential heating by the Sun and, to a secondary degree, by inertia (advection), ion drag, and Coriolis force,
with the latter being negligible at equatorial latitudes. This simplified force balance should yield westward (east-
ward) mean zonal winds in the summer (winter) hemisphere and summer to winter mean meridional flow, with
associated upwelling in the summer hemisphere and downwelling in the winter hemisphere as a consequence
of mass continuity (Forbes,2007). However, as observed by ICON/MIGHTI, solstitial zonal mean winds are
consistently eastward in the upper mesosphere and lower thermosphere (MLT) between 90 and 110km (Figures5
and8) with increasing magnitude from low-to middle-latitudes. This feature of the zonal winds is associated
with the eastward gravity wave momentum deposition in the MLT, as has been demonstrated in a number of
general circulation modeling studies (Garcia etal.,2007; Griffith etal.,2021; Lilienthal etal.,2020; Miyoshi &
Yiğit,2019; Yiğit etal.,2009). Eastward mean winds of up to 10–40ms
−1 were also seen around 10–40°N in
the monthly mean wind climatologies compiled as part of the Upper Atmosphere Research Satellite Reference
Atmosphere Project UARS (Swinbank & Ortland,2003) and at northern midlatitudes in meteor and MF radar
measurements (Conte etal.,2017; Portnyagin & Solovjeva,2000; Tang etal.,2021). They are a general feature of
Figure 12. Same as Figure11 but for the latitude-longitude distributions of the mean horizontal circulation at three
representative thermospheric altitudes. Note that winds at 185–200km altitude are only daytime measurements.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
17 of 23
the summer midlatitudes in the MLT (Drob etal.,2008; A. K. Smith,2012). The observed dawn-dusk asymmetry
in the lower thermospheric winds is noticeable as well (Figure5c).
Lower thermospheric winds have been routinely observed by incoherent scatter radars (ISR) and meteor radars.
Portnyagin etal.(1999)'s analysis of seasonal variations of the mean zonal wind observed by ground-based radars
and WINDII at 95km shows eastward winds of 40–50ms
−1 at midlatitudes for Northern Hemisphere summer
conditions, which is similar to ICON/MIGHTI measurements. Zhang et al. (2003) used the Millstone Hill
incoherent scatter radar (42.6°N) to study the seasonal climatology of zonal and meridional winds in the iono-
spheric E-region (94–130km) and compared with WINDII observations. These results shown in their Figures2
and3 demonstrate semidiurnal variations during all seasons. Our monthly mean zonal and meridional winds
during the solstice viewed as altitude-local time cross-sections indicate a semidiurnal variation as well (Figure8).
Although the Millstone ISR provides data at a fixed latitude and here we have included all latitudes between 10°S
and 40°N, the phases of the semidiurnal variations in mean winds are quite similar. Overall, differences in the
magnitudes are probably due to differences in latitude and seasonal binning.
ICON/MIGHTI provides evidence that the mean meridional winds in the MLT reverse their direction. The
northward mean meridional flow (i.e., from winter to summer) seen in the MLT is opposite to the radiatively
Figure 13. Same as Figure8 but for the Northern Hemisphere winter season based on Ionospheric Connection Explorer/
Michelson Interferometer for Global High-resolution Thermospheric Imaging observations from 16 December 2019–31
January 2020.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
18 of 23
driven mean meridional flow, which is due to the direct response of the circulation to the eastward gravity wave
momentum deposition in the MLT (Holton,1983). A simple force balance based on the Transformed Eulerian
Mean (TEM) analysis illustrates the role of the zonal wave forcing in driving the mean residual circulation at
midlatitudes:
 𝜕

−0𝜕 ∗=𝜕 
,
(3)
where
 𝐴 
is the total zonal force due to small-scale and non-zonal eddies/waves, f0 is the Coriolis parameter at
midlatitudes, and
 𝐴 ∗
is the meridional component of the residual circulation
( ∗,
∗)
, where

∗
is the vertical
component, and the overbar indicates an appropriate averaging as seen in Equation1. For steady-state conditions,
(Equation3) implies that the mean meridional circulation (or transport) is primarily driven by wave dissipation,
that is,
 𝐴 ∗≈−𝐴 ∕0
. Around the midlatitude MLT, small-scale gravity waves are the primary contributor to the
zonal body force. Diurnal migrating tide-gravity wave interactions are an important mechanism of wind vari-
ability, especially in the low-latitude MLT (Miyahara & Forbes,1991; Watanabe & Miyahara,2009; Yiğit &
Medvedev,2017).
Above 120km, radiative processes and ion-neutral coupling (i.e., ion drag), which is proportional to the relative
velocity of neutrals and ions moving within the magnetic field, become increasingly important in driving the
horizontal circulation. ICON/MIGHTI observations show that the zonal winds are predominantly westward and
southward (i.e., directed from summer to winter hemisphere) in the upper thermosphere (Figures5–12). This
large-scale flow is maintained by the pressure gradient force, modulated by ion drag and Coriolis effect. Molec-
ular viscosity, which increases exponentially with height, smooths out smaller-scale motions, as can be seen in
daytime circulation and winds between 185 and 200km (Figures11 and12).
Figure 14. Same as Figure12 but for the Northern Hemisphere winter season based on Ionospheric Connection Explorer/
Michelson Interferometer for Global High-resolution Thermospheric Imaging observations from 16 December 2019–31
January 2020.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
19 of 23
Recently, Drob etal.(2015) have updated the Horizontal Wind Model (HWM14) and provided a comparison
against WINDII climatologies of Emmert etal. (2002) and an older version of HWM. Our altitude-local time
analysis of mean winds at 200km (Figures 9e and9f) can qualitatively be compared with their June solstice
climatology (day 180) at 250km. There is a general agreement concerning the morphology of the mean winds.
Westward zonal winds prevail during day and eastward winds around dusk at low-to midlatitudes in the Northern
Hemisphere. Meridional winds are generally northward during morning and southward during afternoon.
In order to complement the northern summer solstice results, we have also compiled northern winter winds and
circulation, including data from 16 December 2019–31 January 2020, during which no SSWs occurred (Roy &
Kuttippurath,2022). Although the data sampling is not the same between these seasons, the results can be at
least qualitatively compared. There is certainly a certain degree of asymmetry between the summer and winter
seasons in the Northern Hemisphere in terms of the magnitude and spatial distribution of winds and circulation.
The mean zonal winds are much weaker during winter than summer. The lower thermospheric westward wind
reversal (Figure13a) is also weaker in magnitude than the eastward reversal during summer (Figure8), which
qualitatively agrees with previous wind compilations in this region.
The summer-winter differences in circulation can be caused by a combination of seasonal changes in pressure
forces, upward propagation conditions for gravity waves, tides, and planetary waves. Since dynamical conditions
(westerly winds) are favorable for the upward propagation of planetary waves in the winter hemisphere and not
in the summer hemisphere with easterlies (Yiğit & Medvedev, 2016), presumably planetary waves contribute
significantly to the differences between the two seasons.
5. Summary and Conclusions
We have presented the mean behavior of the thermospheric zonal and meridional winds at 90–200km as observed
by the MIGHTI instrument onboard NASA's ICON spacecraft between 10°S and 40°N. A comprehensive picture
of the mean zonal and meridional winds and horizontal circulation has been shown for a single solstice day, 20
June 2020, and using a month of continuous observations from 8 June–7 July 2020, representative of Northern
Hemisphere solstice conditions, and a 47-day Northern Hemisphere winter solstice period from 16 December
2019–31 January 2020, primarily focusing on the northern summer solstice season.
The main inferences of our analysis of ICON/MIGHTI northern summer solstice winds are as follows:
1. Altitude, longitude, latitude, and local time profiles of winds show that the typical instantaneous zonal and
meridional winds during solstice can exceed ±150ms
−1, with magnitudes increasing with altitude.
2. We have evaluated the occurrence rates of the wind speeds observed by ICON/MIGHTI on 20 June 2020
at three representative altitude bins 94–103km, 106–114km, and 194–202km, such that each altitude bin
included equal number of data. The speeds as a function of number of measurements exhibit a Gaussian
distribution centered around small values. The nighttime magnitudes of the wind are greater than during day
and increase as a function of height. Larger standard deviation at night suggests more variability during night,
which indicate more atmospheric wave activity.
3. Thermospheric mean winds are up to ±80ms
−1 and depend strongly on altitude, latitude, longitude, and local
time. Local time variations of the mean winds exhibit diurnal and semidiurnal variations.
4. Mean winds and circulation change substantially within the lower thermosphere (90–120km). Eastward and
southward flow between 90 and 105km change to a northward and westward flow within 105–120km.
Upper thermospheric winds are generally characterized by a westward and southward (i.e., directed from the
summer-to-winter) flow in the Northern Hemisphere with diverging flow from the post-noon sector at midlat-
itudes. The upper thermospheric wind system is more homogeneous compared to the lower thermospheric
one, which exhibits spatial variations at smaller scales and vortex patterns, especially around 105–120km.
5. The observed eastward mean flow and the northward (winter-to-summer) meridional flows in the lower ther-
mosphere are a consequence of eastward gravity wave momentum forcing there. ICON/MIGHTI observations
are capable of demonstrating vertical coupling induced by waves. These features are in a good agreement with
previous observations and modeling.
6. Despite the seasonal differences in the data sampling size due to data gap issues, we have qualitatively
compared the northern summer and winter winds. Zonal winds during winter are clearly slower than summer
zonal winds. Overall, an asymmetry between the summer and winter seasons in terms of the magnitude and

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
20 of 23
spatial distribution of winds and circulation is evident, which is presumably due to the differences in planetary
wave propagation and interaction with the mean flow as well as to differences in the generation, propagation,
and filtering in gravity waves and tides between winter and summer seasons.
7. In the upper thermosphere, the morphology of the ICON/MIGHTI mean winds is, in general, in a good agree-
ment with previous wind climatologies based on WINDII and HWM14.
The mean wind and circulation patterns inferred in this study using ICON/MIGHTI measurements can be used
to constrain and validate general circulation models or as an input for numerical wave models. They also serve
for an indirect validation of parameterized subgrid-scale processes, which control the large-scale winds and
circulation. This work is expected to contribute toward filling in the observational gap with horizontal winds in
the thermosphere.
Appendix A: Data Quality
We have removed data with quality less than 0.5 and filtered outliers by removing wind magnitudes exceeding
300ms
−1. We illustrated the effect of this filtering for zonal winds in FigureA1.
Figure A1. The effect of filtering the data according to quality is shown, where wq=1 stands for “good,” wq=0.5 is for
“good, but use with caution” and wq=0 is for “bad” zonal wind measurements.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
21 of 23
Data Availability Statement
The horizontal wind data (version 4) used in this study are available at the ICON data center (https://icon.ssl.
berkeley.edu/Data).
References
Aruliah, A. L., Griffin, E. M., Yiu, H.-C. C. I., McWhirter, I., & Charalambous, A. (2010). SCANDI—An all-sky Doppler imager for studies of
thermospheric spatial structure. Annales Geophysicae, 28(2), 549–567. https://doi.org/10.5194/angeo-28-549-2010
Chen, Z., Liu, Y., Du, Z., Fan, Z., Sun, H., & Zhou, C. (2022). Validation of MIGHTI/ICON atmospheric wind observations over China region
based on meteor radar and horizontal wind model (HWM14). Atmosphere, 13(7), 1078. https://doi.org/10.3390/atmos13071078
Conde, M., & Smith, R. W. (1995). Mapping thermospheric winds in the auroral zone. Geophysical Research Letters, 22(22), 3019–3022.
https://doi.org/10.1029/95GL02437
Conte, J. F., Chau, J. L., Liu, A., Qiao, Z., Fritts, D. C., Hormaechea, J. L., etal. (2022). Comparison of MLT momentum fluxes over the
Andes at four different latitudinal sectors using multistatic radar configurations. Journal of Geophysical Research: Atmospheres, 127(4),
e2021JD035982. https://doi.org/10.1029/2021JD035982
Conte, J. F., Chau, J. L., Stober, G., Pedatella, N., Maute, A., Hoffmann, P., etal. (2017). Climatology of semidiurnal lunar and solar tides at
middle and high latitudes: Interhemispheric comparison. Journal of Geophysical Research: Space Physics, 122(7), 7750–7760. https://doi.
org/10.1002/2017JA024396
David, M., Sojka, J. J., & Schunk, R. W. (2014). Sources of uncertainty in ionospheric modeling: The neutral wind. Journal of Geophysical
Research: Space Physics, 119(8), 6792–6805. https://doi.org/10.1002/2014JA020117
Dempsey, S. M., Hindley, N. P., Moffat-Griffin, T., Wright, C. J., Smith, A. K., Du, J., & Mitchell, N. J. (2021). Winds and tides of the Antarctic
mesosphere and lower thermosphere: One year of meteor-radar observations over Rothera (68°S, 68°W) and comparisons with WACCM and
eCMAM. Journal of Atmospheric and Solar-Terrestrial Physics, 212, 105510. https://doi.org/10.1016/j.jastp.2020.105510
Deng, Y., Sheng, C., Tsurutani, B. T., & Mannucci, A. J. (2018). Possible influence of extreme magnetic storms on the thermosphere in the high
latitudes. Space Weather, 16(7), 802–813. https://doi.org/10.1029/2018SW001847
Dhadly, M. S., Emmert, J. T., Drob, D. P., Conde, M., Doornbos, E., Shepherd, G., et al. (2018). Seasonal dependence of geomagnetic
active-time northern high-latitude upper thermospheric winds. Journal of Geophysical Research: Space Physics, 123(1), 739–754. https://
doi.org/10.1002/2017JA024715
Dhadly, M. S., Emmert, J. T., Drob, D. P., Conde, M. G., Aruliah, A., Doornbos, E., etal. (2019). HL-TWiM empirical model of high-latitude
upper thermospheric winds. Journal of Geophysical Research: Space Physics, 124(12), 2019JA027188. https://doi.org/10.1029/2019JA027188
Dhadly, M. S., Emmert, J. T., Drob, D. P., McCormack, J. P., & Niciejewski, R. J. (2018). Short-term and interannual variations of migrating diur-
nal and semidiurnal tides in the mesosphere and lower thermosphere. Journal of Geophysical Research: Space Physics, 123(8), 7106–7123.
https://doi.org/10.1029/2018JA025748
Dhadly, M. S., Englert, C. R., Drob, D. P., Emmert, J. T., Niciejewski, R., & Zawdie, K. A. (2021). Comparison of ICON/MIGHTI and TIMED/
TIDI neutral wind measurements in the lower thermosphere. Journal of Geophysical Research: Space Physics, 126(12), e2021JA029904.
https://doi.org/10.1029/2021JA029904
Doornbos, E., Bruinsma, S. L., Fritsche, B., Koppenwallner, G., Visser, P., Van Den Ijssel, J., & Teixeira da Encarnação, J. (2014). GOCE+
Theme 3: Air density and wind retrieval using GOCE data final report. Technical Report. 4000102847/NL/EL, TU Delft.
Drob, D. P., Emmert, J. T., Crowley, G., Picone, J. M., Shepherd, G. G., Skinner, W., etal. (2008). An empirical model of the earth’s horizontal
wind fields: Hwm07. Journal of Geophysical Research, 113(A12). https://doi.org/10.1029/2008JA013668
Drob, D. P., Emmert, J. T., Meriwether, J. W., Makela, J. J., Doornbos, E., Conde, M., etal. (2015). An update to the Horizontal Wind Model
(HWM): The quiet time thermosphere. Earth and Space Science, 2(7), 301–319. https://doi.org/10.1002/2014ea000089
Emmert, J. T. (2015). Thermospheric mass density: A review. Advances in Space Research, 56(5), 773–824. https://doi.org/10.1016/j.
asr.2015.05.038
Emmert, J. T., Fejer, B. G., Fesen, C. G., Shepherd, G. G., & Solheim, B. H. (2001). Climatology of middle- and low-latitude daytime F region
disturbance neutral winds measured by Wind Imaging Interferometer (WINDII). Journal of Geophysical Research, 106(A11), 24701–24712.
https://doi.org/10.1029/2000ja000372
Emmert, J. T., Fejer, B. G., Shepherd, G. G., & Solheim, B. H. (2002). Altitude dependence of middle and low-latitude daytime thermospheric
disturbance winds measured by WINDII: WINDII daytime disturbance winds. Journal of Geophysical Research, 107(A12), SIA19-1–SIA19-
15. https://doi.org/10.1029/2002JA009646
England, S. L., Englert, C. R., Harding, B. J., Triplett, C. C., Marr, K., Harlander, J. M., etal. (2022). Vertical shears of horizontal winds in
the lower thermosphere observed by ICON. Geophysical Research Letters, 49(11), e2022GL098337. https://doi.org/10.1029/2022GL098337
Englert, C. R., Harlander, J. M., Brown, C. M., Marr, K. D., Miller, I. J., Stump, J. E., etal. (2017). Michelson interferometer for global
high-resolution thermospheric imaging (MIGHTI): Instrument design and calibration. Space Science Reviews, 212(1), 553–584. https://doi.
org/10.1007/s11214-017-0358-4
Forbes, J. M. (2007). Dynamics of the upper mesosphere and thermosphere. Journal of the Meteorological Society of Japan, 85B, 193–213.
https://doi.org/10.2151/jmsj.85b.193
Forbes, J. M., Zhang, X., Heelis, R., Stoneback, R., Englert, C. R., Harlander, J. M., etal. (2021). Atmosphere-ionosphere (A-I) coupling as
viewed by ICON: Day-to-Day variability due to planetary wave (PW)-Tide interactions. Journal of Geophysical Research: Space Physics,
126(6), e2020JA028927. https://doi.org/10.1029/2020JA028927
Garcia, R. R., Marsh, D. R., Kinnison, D. E., Boville, B. A., & Sassi, F. (2007). Simulations of secular trends in the middle atmosphere. Journal
of Geophysical Research, 112(D9), D09301. https://doi.org/10.1029/2006JD007485
Gavrilov, N. M., Kshevetskii, S. P., & Koval, A. V. (2020). Thermal effects of nonlinear acoustic-gravity waves propagating at thermo-
spheric temperatures matching high and low solar activity. Journal of Atmospheric and Solar-Terrestrial Physics, 208, 105381. https://doi.
org/10.1016/j.jastp.2020.105381
Geisler, J. E. (1966). Atmospheric winds in the middle latitude F-region. Journal of Atmospheric and Terrestrial Physics, 28(8), 703–720. https://
doi.org/10.1016/0021-9169(66)90020-1
Griffith, M. J., Dempsey, S. M., Jackson, D. R., Moffat-Griffin, T., & Mitchell, N. J. (2021). Winds and tides of the Extended Unified Model in the
mesosphere and lower thermosphere validated with meteor radar observations. Annales Geophysicae, 39(3), 487–514. https://doi.org/10.5194/
angeo-39-487-2021
Acknowledgments
This work was supported by NASA
(Grant 80NSSC22K0016). ICON is
supported by NASA's Explorers Program
through contracts NNG12FA45C and
NNG12FA42I.

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
22 of 23
Harding, B. J., Chau, J. L., He, M., Englert, C. R., Harlander, J. M., Marr, K. D., etal. (2021). Validation of ICON-MIGHTI thermospheric wind
observations: 2. Green-line comparisons to specular meteor radars. Journal of Geophysical Research: Space Physics, 126(3), e2020JA028947.
https://doi.org/10.1029/2020JA028947
Harding, B. J., Makela, J. J., Englert, C. R., Marr, K. D., Harlander, J. M., England, S. L., & Immel, T. J. (2017). The MIGHTI wind retrieval
algorithm: Description and verification. Space Science Reviews, 212(1), 585–600. https://doi.org/10.1007/s11214-017-0359-3
Hickey, M. P., Walterscheid, R. L., & Schubert, G. (2011). Gravity wave heating and cooling of the thermosphere: Sensible heat flux and viscous
flux of kinetic energy. Journal of Geophysical Research, 116(A12), A12326. https://doi.org/10.1029/2011JA016792
Holton, J. R. (1983). The influence of gravity wave breaking on the general circulation of the middle atmosphere. Journal of the Atmospheric
Sciences, 40(10), 2497–2507. https://doi.org/10.1175/1520-0469(1983)040<2497:tiogwb>2.0.co;2
Hysell, D. L., Larsen, M. F., & Sulzer, M. P. (2014). High time and height resolution neutral wind profile measurements across the mesosphere/
lower thermosphere region using the Arecibo incoherent scatter radar. Journal of Geophysical Research: Space Physics, 119(3), 2345–2358.
https://doi.org/10.1002/2013JA019621
Immel, T. J., England, S. L., Mende, S. B., Heelis, R. A., Englert, C. R., Edelstein, J., etal. (2018). The ionospheric connection explorer mission:
Mission goals and design. Space Science Reviews, 214(1), 553–584. https://doi.org/10.1007/s11214-017-0449-2
Killeen, T. L., & Roble, R. G. (1988). Thermosphere dynamics: Contributions from the first 5 years of the dynamics explorer program. Reviews
of Geophysics, 26(2), 329–367. https://doi.org/10.1029/RG026i002p00329
Koval, A. V., Gavrilov, N. M., Pogoreltsev, A. I., & Kandieva, K. K. (2022). Dynamical impacts of stratospheric QBO on the global circulation up
to the lower thermosphere. Journal of Geophysical Research: Atmospheres, 127(4), e2021JD036095. https://doi.org/10.1029/2021JD036095
Larsen, M. F. (2002). Winds and shears in the mesosphere and lower thermosphere: Results from four decades of chemical release wind measure-
ments. Journal of Geophysical Research, 107(A8), SIA28-1–SIA28-14. https://doi.org/10.1029/2001JA000218
Larsen, M. F., & Fesen, C. G. (2009). Accuracy issues of the existing thermospheric wind models: Can we rely on them in seeking solutions to
wind-driven problems? Annales Geophysicae, 27(6), 2277–2284. https://doi.org/10.5194/angeo-27-2277-2009
Lehmacher, G. A., Larsen, M. F., & Zanetti, J. (2022). Mesoscale spatial variability of lower thermospheric winds during the anomalous transport
rocket experiment. Journal of Geophysical Research: Space Physics, 127(5), e2022JA030378. https://doi.org/10.1029/2022JA030378
Lieberman, R. S., Akmaev, R. A., Fuller–Rowell, T. J., & Doornbos, E. (2013). Thermospheric zonal mean winds and tides revealed by CHAMP.
Geophysical Research Letters, 40(10), 2439–2443. https://doi.org/10.1002/grl.50481
Lieberman, R. S., Smith, A. K., Franke, S. J., Vincent, R. A., Isler, J. R., Manson, A. H., etal. (2000). Comparison of mesospheric and lower ther-
mospheric residual wind with High Resolution Doppler Imager, medium frequency, and meteor radar winds. Journal of Geophysical Research,
105(D22), 27023–27035. https://doi.org/10.1029/2000JD900363
Lilienthal, F., Yiğit, E., Samtleben, N., & Jacobi, C. (2020). Variability of gravity wave effects on the zonal mean circulation and migrating
terdiurnal tide as studied with the middle and upper atmosphere model (MUAM2019) using a nonlinear gravity wave scheme. Frontiers in
Astronomy and Space Sciences, 7, 588956. https://doi.org/10.3389/fspas.2020.588956
Liu, H., Lühr, H., Watanabe, S., Köhler, W., Henize, V., & Visser, P. (2006). Zonal winds in the equatorial upper thermosphere: Decom-
posing the solar flux, geomagnetic activity, and seasonal dependencies. Journal of Geophysical Research, 111(A7), A07307. https://doi.
org/10.1029/2005JA011415
Makela, J. J., Baughman, M., Navarro, L. A., Harding, B. J., Englert, C. R., Harlander, J. M., etal. (2021). Validation of ICON-MIGHTI thermo-
spheric wind observations: 1. Nighttime red-line ground-based Fabry-Perot interferometers. Journal of Geophysical Research: Space Physics,
126(2), e2020JA028726. https://doi.org/10.1029/2020JA028726
Makela, J. J., Meriwether, J. W., Ridley, A. J., Ciocca, M., & Castellez, M. W. (2012). Large-scale measurements of thermospheric dynamics
with a multisite Fabry-Perot interferometer network: Overview of plans and results from midlatitude measurements. International Journal of
Geophysics, 2012(3), 872140. https://doi.org/10.1155/2012/872140
Medvedev, A. S., & Yiğit, E. (2019). Gravity waves in planetary atmospheres: Their effects and parameterization in global circulation models.
Atmosphere, 10(531), 531. https://doi.org/10.3390/atmos10090531
Meriwether, J. (2006). Studies of thermospheric dynamics with a Fabry–Perot interferometer network: A review. Journal of Atmospheric and
Solar-Terrestrial Physics, 68(13), 1576–1589. https://doi.org/10.1016/j.jastp.2005.11.014
Miyahara, S., & Forbes, J. M. (1991). Interactions between gravity waves and the diurnal tide in the mesosphere and lower thermosphere. Journal
of the Meteorological Society of Japan, 69(5), 523–531. https://doi.org/10.2151/jmsj1965.69.5_523
Miyoshi, Y., Fujiwara, H., Jin, H., & Shinagawa, H. (2014). A global view of gravity waves in the thermosphere simulated by a general circulation
model. Journal of Geophysical Research: Space Physics, 119(7), 5807–5820. https://doi.org/10.1002/2014JA019848
Miyoshi, Y., & Yiğit, E. (2019). Impact of gravity wave drag on the thermospheric circulation: Implementation of a nonlinear gravity wave param-
eterization in a whole atmosphere model. Annales Geophysicae, 37(5), 955–969. https://doi.org/10.5194/angeo-37-955-2019
Niciejewski, R., Wu, Q., Skinner, W., Gell, D., Cooper, M., Marshall, A., etal. (2006). TIMED Doppler interferometer on the thermosphere iono-
sphere mesosphere Energetics and dynamics satellite: Data product overview. Jour nal of Geophysical Research, 111(A11), A11S90. https://
doi.org/10.1029/2005JA011513
Oberheide, J., Shiokawa, K., Gurubaran, S., Ward, W. E., Fujiwara, H., Kosch, M. J., etal. (2015). The geospace response to variable inputs from
the lower atmosphere: A review of the progress made by task group 4 of CAWSES-II. Progress in Earth and Planetary Science, 2(1), 1–31.
https://doi.org/10.1186/s40645-014-0031-4
Oppenheim, M. M., Sugar, G., Slowey, N. O., Bass, E., Chau, J. L., & Close, S. (2009). Remote sensing lower thermosphere wind profiles using
non-specular meteor echoes. Geophysical Research Letters, 36(9), L09817. https://doi.org/10.1029/2009GL037353
Pancheva, D., & Mukhtarov, P. (2012). Global response of the ionosphere to atmospheric tides forced from below: Recent progress based on
satellite measurements. Space Science Reviews, 168(1–4), 175–209. https://doi.org/10.1007/s11214-011-9837-1
Pancheva, D., Mukhtarov, P., Hall, C., Meek, C., Tsutsumi, M., Pedatella, N., etal. (2020). Climatology of the main (24-h and 12-h) tides
observed by meteor radars at Svalbard and Tromsø: Comparison with the models CMAM-DAS and WACCM-X. Journal of Atmospheric and
Solar-Terrestrial Physics, 207, 105339. https://doi.org/10.1016/j.jastp.2020.105339
Portnyagin, Y. I., & Solovjeva, T. V. (2000). Global empirical wind model for the upper mesosphere/lower thermosphere. I. prevailing wind.
Annales Geophysicae, 18(3), 300–315. https://doi.org/10.1007/s00585-000-0300-y
Portnyagin, Y. I., Solovjova, T. V., & Wang, D. Y. (1999). Some results of comparison between the lower thermosphere zonal winds as seen by the
ground-based radars and WINDII on UARS. Earth Planets and Space, 51(7), 701–709. https://doi.org/10.1186/BF03353228
Richmond, A. D., Ridley, E. C., & Roble, R. G. (1992). A thermosphere/ionosphere general circulation model with coupled electrodynamics.
Geophysical Research Letters, 19(6), 601–604. https://doi.org/10.1029/92gl00401
Rishbeth, H., Moffett, R., & Bailey, G. (1969). Continuity of air motion in the mid-latitude thermosphere. Journal of Atmospheric and Terrestrial
Physics, 31(8), 1035–1047. https://doi.org/10.1016/0021-9169(69)90103-2

Journal of Geophysical Research: Space Physics
YIĞIT ETAL.
10.1029/2022JA030851
23 of 23
Roy, R., & Kuttippurath, J. (2022). The dynamical evolution of Sudden Stratospheric Warmings of the Arctic winters in the past decade
2011–2021. SN Applied Sciences, 4(4), 105. https://doi.org/10.1007/s42452-022-04983-4
Schunk, R. W., & Sojka, J. J. (1996). Ionosphere-thermosphere space weather. Journal of Atmospheric and Terrestrial Physics, 58(14), 1527–
1574. https://doi.org/10.1016/0021-9169(96)00029-3
Shepherd, G. G., Thuillier, G., Cho, Y. M., Duboin, M. L., Evans, W. F., & Gault, W. A. (2012). The wind imaging interferometer (WINDII) on
the upper atmosphere research satellite: A 20 year perspective. Review of Geophysics, 50(2). https://doi.org/10.1029/2012RG000390
Shiokawa, K., & Georgieva, K. (2021). A review of the SCOSTEP’s 5-year scientific program VarSITI—Variability of the Sun and its terrestrial
impact. Progress in Earth and Planetary Science, 8(1), 21. https://doi.org/10.1186/s40645-021-00410-1
Smith, A. K. (2012). Global dynamics of the MLT. Surveys in Geophysics, 33(6), 1177–1230. https://doi.org/10.1007/s10712-012-9196-9
Smith, R. W. (1998). Ver tical winds: A tutorial. Journal of Atmospheric and Solar-Terrestrial Physics, 60(14), 1425–1434. https://doi.org/10.1016/
s1364-6826(98)00058-3
Swinbank, R., & Ortland, D. A. (2003). Compilation of wind data for the upper atmosphere research satellite (UARS) reference atmosphere
project. Journal of Geophysical Research, 108(D19), 4615. https://doi.org/10.1029/2002JD003135
Tang, Q., Zhou, Y., Du, Z., Zhou, C., Qiao, J., Liu, Y., & Chen, G. (2021). A comparison of meteor radar observation over China region with
horizontal wind model (HWM14). Atmosphere, 12(1), 98. https://doi.org/10.3390/atmos12010098
Vichare, G., Ridley, A., & Yiğit, E. (2012). Quiet-time low latitude ionospheric electrodynamics in the non-hydrostatic global ionosphere–
thermosphere model. Journal of Atmospheric and Solar-Terrestrial Physics, 80, 161–172. https://doi.org/10.1016/j.jastp.2012.01.009
Ward, W., Seppälä, A., Yiğit, E., Nakamura, T., Stolle, C., Laštovička, J., etal. (2021). Role of the Sun and the middle atmosphere/thermo-
sphere/ionosphere in climate (ROSMIC): A retrospective and prospective view. Progress in Earth and Planetary Science, 8(1), 47. https://doi.
org/10.1186/s40645-021-00433-8
Watanabe, S., & Miyahara, S. (2009). Quantification of the gravity wave forcing of the migrating diurnal tide in a gravity wave–resolving general
circulation model. Journal of Geophysical Research, 114(D7), D07110. https://doi.org/10.1029/2008JD011218
Yamazaki, Y., Arras, C., Andoh, S., Miyoshi, Y., Shinagawa, H., Harding, B. J., etal. (2022). Examining the wind shear theory of sporadic
E with ICON/MIGHTI winds and COSMIC-2 radio occultation data. Geophysical Research Letters, 49(1), e2021GL096202. https://doi.
org/10.1029/2021GL096202
Yiğit, E. (2018). Atmospheric and space sciences: Ionospheres and plasma environments (Vol. 2). Springer. https://doi.org/10.1007/978-
3-319-62006-0
Yiğit, E., Aylward, A. D., & Medvedev, A. S. (2008). Parameterization of the effects of vertically propagating gravity waves for thermosphere
general circulation models: Sensitivity study. Journal of Geophysical Research, 113(D19), D19106. https://doi.org/10.1029/2008JD010135
Yiğit, E., Frey, H. U., Moldwin, M. B., Immel, T. J., & Ridley, A. J. (2016). Hemispheric differences in the response of the upper atmosphere
to the august 2011 geomagnetic storm: A simulation study. Journal of Atmospheric and Solar-Terrestrial Physics, 141, 13–26. https://doi.
org/10.1016/j.jastp.2015.10.002
Yiğit, E., Knížová, P.K., Georgieva, K., & Ward, W. (2016). A review of vertical coupling in the atmosphere-ionosphere system: Effects of waves,
sudden stratospheric warmings, space weather, and of solar activity. Journal of Atmospheric and Solar-Terrestrial Physics, 141, 1–12. https://
doi.org/10.1016/j.jastp.2016.02.011
Yiğit, E., & Medvedev, A. S. (2015). Internal wave coupling processes in Earth’s atmosphere. Advances in Space Research, 55(5), 983–1003.
https://doi.org/10.1016/j.asr.2014.11.020
Yiğit, E., & Medvedev, A. S. (2016). Role of gravity waves in vertical coupling during sudden stratospheric warmings. Geoscience Letters, 3(1),
1–13. https://doi.org/10.1186/s40562-016-0056-1
Yiğit, E., & Medvedev, A. S. (2017). Influence of parameterized small-scale gravity waves on the migrating diurnal tide in earth’s thermosphere.
Journal of Geophysical Research: Space Physics, 122(4), 4846–4864. https://doi.org/10.1002/2017JA024089
Yiğit, E., Medvedev, A. S., Aylward, A. D., Hartogh, P., & Harris, M. J. (2009). Modeling the effects of gravity wave momentum deposition on
the general circulation above the turbopause. Journal of Geophysical Research, 114(D7), D07101. https://doi.org/10.1029/2008JD011132
Yiğit, E., Medvedev, A. S., & Ern, M. (2021). Effects of latitude-dependent gravity wave source variations on the middle and upper atmosphere.
Frontiers in Astronomy and Space Sciences, 7, 614018. https://doi.org/10.3389/fspas.2020.614018
Zhang, S. P., Goncharenko, L. P., Salah, J. E., Roble, E. G., & Shepherd, G. G. (2003). Climatology of neutral winds in the lower thermo-
sphere over Millstone Hill (42.6°N) observed from ground and from space. Journal of Geophysical Research, 108(A1), 1051. https://doi.
org/10.1029/2002JA009512