Skip to main content
NCBI home page
NASA Author Manuscripts logoLink to NASA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 16.
Published in final edited form as: Adv Space Res. 2019 Aug 23;64(10):1841–1853. doi: 10.1016/j.asr.2019.08.015

Variations in the ionosphere-thermosphere system from tides, ultra-fast Kelvin waves, and their interactions

Colin C Triplett 1,*, Thomas J Immel 1, Yen-Jung Wu 1, Chihoko Cullens 1
PMCID: PMC8050945  NIHMSID: NIHMS1686833  PMID: 33867620

Abstract

Large scale waves, such as the atmospheric tides and ultra-fast Kelvin waves (UFKW), have direct effects on the neutral wind and temperature fields of the ionosphere-thermosphere (I-T) system. In this study we examine the response of the I-T system to the atmospheric tides, one UFKW, and the secondary waves generated from their interactions using the Thermosphere-Ionosphere-Electrodynamics General Circulation Model (TIEGCM). We find that forcing an UFKW at the lower boundary of the TIEGCM is all that is required for it to setup in the model. We see variations around 10% in the zonal winds that lead to similar variations in the total electron content (TEC) depending on the phase of the UFKW. From these simulations, we expect the Ionospheric Connection Explorer (ICON) mission will be able to fully capture these wave interactions by observing winds and temperatures at the mesopause and above.

Keywords: Atmospheric tides, Ultra-fast Kelvin waves, Ionosphere, Thermosphere, TIEGCM, Ionosphere/atmosphere interactions

1. Introduction

Neutral wind control on the ionosphere has been a topic of study for many years (England, 2012; Immel et al., 2006; Takahashi et al., 2007; Vincent, 1993). Winds in the lower thermosphere drag plasma along and across magnetic field lines, the latter of which creates electric fields and subsequent ExB drifts. This global interaction affects the local plasma in the E-region as well as the F-region (e.g., Richmond, 1979). Waves from the lower atmosphere propagate upward into the lower thermosphere and affect local neutral winds. Through this process, waves extend into the I-T system with potential adverse effects on radio and other technologies (e.g., Hines, 1972; Immel et al., 2006). The spectrum of these waves contains a wide range of spatial and temporal scales with great variation throughout the year (e.g., England, 2012; Forbes et al., 2008; Liu, 2014; Snively, 2017; Zettergren et al., 2017). Because of their dominant influence, numerous studies have focused on atmospheric tides and their role in the dynamics of the ionosphere-thermosphere (I-T) system (e.g., Chang et al., 2008; England et al., 2010; Forbes et al., 2008; Hagan and Forbes, 2002; Haldoupis et al., 2004). Studies have also looked into other large scale waves that originate below the I-T system (e.g., Egito et al., 2018; Forbes, 2000; Forbes et al., 2018; Liu, 2014; Pogoreltsev et al., 2007; Nystrom et al., 2018). Nystrom et al. (2018) discussed the effect of one type of equatorially trapped large-scale wave, called an ultra-fast Kelvin wave (UFKW), on the mesosphere-lower thermosphere (MLT). They used reanalysis data to drive the Thermosphere-Ionosphere-Mesosphere Electrodynamics General Circulation Model (TIME-GCM) during a period when two UFKWs and secondary waves were identified. They conclude that secondary waves generated by nonlinear interactions between tides and UFKWs affect the zonal winds in the E-region which could substantially affect the F-region ionosphere. In this study we look into the effects of a single UFKW and secondary wave generation within the IT system itself without the additional information that comes from reanalysis data. We do this by using the Thermosphere-Ionosphere-Electrodynamics General Circulation Model (TIEGCM) forced by climatologically determined atmospheric tides and one UFKW whose properties are determined by observational data and linear wave theory. This isolates the I-T system from the lower atmospheric forcing other than the tides, the UFKW, and generated secondary waves to determine their direct effects. These simulations will be used to discuss the ability of NASA’s Ionospheric Connection Explorer (ICON) to see these effects on the wind field. The paper is organized as follows: Section 2 shows the setup of the TIEGCM experiment, Section 3 gives the results of the TIEGCM experiment, Section 4 discusses ICON’s ability to sample the results of the UFKW and secondary waves, and Section 5 gives conclusions.

2. TIEGCM experiment setup

The TIEGCM is a self consistent numerical model of the thermosphere and ionosphere that solves the 3-D momentum, energy, and continuity equations for both neutral and ion species; a steady-state ionospheric electrodynamo is calculated from conductances and neutral dynamics within the model (Dickinson et al., 1984; Qian et al., 2014; Richmond et al., 1992; Roble et al., 1988). The model was modified for the ICON mission, called TIEGCM-ICON. TIEGCM-ICON updated TIEGCM V2.0 to allow for the Hough Mode Extension (HME) fitting of tides to be used as a lower boundary condition. For full details of TIEGCM-ICON see Maute (2017). For the remainder of this study TIEGCM-ICON will be referred to as TIEGCM. We used 2.5°-by-2.5° horizontal resolution with 57 log-pressure vertical levels. Because the vertical grid is in log-pressure, reported altitudes here are taken from TIEGCM’s geometric altitude output. The lower boundary is 97 km and the upper boundary varies with solar conditions. Two runs were performed for this study. Both were over ~ 90 days starting on day 194 of 2017, using real geophysical values for Kp and 10.7 cm radio flux (F10.7). The maximum Kp and F10.7 values for this time period are 8+ and 136.3 sfu, respectively. We used the Heelis Model for the ionospheric convection pattern which is build into TIEGCM (Heelis et al., 1982). A default lower boundary of zero wind and 181 K constant temperature is used to obtain a background atmosphere without external wave forcing. We subtract this background when comparing the two runs to aid in finding the UFKW and secondary waves in Section 3.1 We leave this background in when looking at the effects of the UFKW and secondary waves in Section 3.2 and examining ICON’s wind measurements in Section 4.

The first run has a lower boundary condition only forced by the Climatological Tidal Model of the Thermosphere (CTMT) tides, referred to hereafter as CTMT Only; The second run has a lower boundary forced by both the CTMT tides and a single UFKW with a flat-top Gaussian envelope, referred to hereafter as CTMT + KW. No other modifications to TIEGCM were done. CTMT consists of six diurnal and eight semi-diurnal tidal components (DW2, DW1, D0, DE1, DE2, DE3, SW4, SW3, SW2, SW1, S0, SE1, SE2, and SE3) (Oberheide et al., 2011). For clarification, D represents diurnal period tides, S represents semi-diurnal period tides, W are westward zonal wavenumbers, E are eastward zonal wavenumbers, and the digit represents the zonal wavenumber (e.g., DE3 is the diurnal eastward number 3 tide). A flat-top Gaussian is applied in time to the UFKW amplitude to better simulate natural wave propagation into the lower thermosphere. It has a full-width at half maximum of 12 days with a flatted top of value 1. The parameters of the UFKW were chosen to be consistent with observed values at 97 km (Chang et al., 2010; Egito et al., 2018; Timmermans et al., 2005). We force a 3.5-day period UFKW at 97 km with a zonal wind amplitude of 20 m/s, which corresponds to a temperature amplitude of 4.5 K. This UFKW has a wavenumber eastward 1. All other necessary parameters are calculated using linear wave theory (Andrews et al., 1987). All analysis presented in this study is done over seven days within the model runs, day 228 to day 234, when the UFKW is at its maximum amplitude and not suppressed by the flat-top Gaussian envelope. Over these seven days the maximum Kp is 3+ and F10.7 is 89.1 sfu. The total run length of 90 days was chosen so the CTMT Only run could be used in testing of the HME codes within the ICON data pipeline. TIEGCM data is outputted hourly.

3. TIEGCM results

3.1. Ultra-fast Kelvin wave forcing and secondary generation

To determine if TIEGCM supports the introduction of an UFKW through lower boundary forcing, we examine the temperature fields at altitudes above the boundary to see if the UFKW has propagated into the thermosphere. We compare the amplitude of 3.5-day period temperature perturbations in the two runs over the same seven day window from day 228 to day 234. A fast Fourier transform is done in time at each latitude, longitude, and altitude location. All frequencies are zeroed excluding the frequency of interest and then transformed back into the time domain. The amplitudes are found from these filtered data. This allows for the amplitudes to include all wavenumbers at the frequency of interest to see the effect of the UFKW on the superposition of all six diurnal tides within the CTMT. The results for 3.5-day period temperature perturbations are shown in Fig. 1. The right panel of Fig. 1 shows the amplitude of 3.5-day temperature perturbations for the CTMT + KW run between 40° north and 40° south from 97 km to 207 km. The left panel of Fig. 1 is the same as the right except for the CTMT Only run. The color1 scale is the same for both panels ranging from 0 K to 10 K. The structure in the right panel is the UFKW. Fig. 1 shows the UFKW propagating upwards to peak around 110 km which is consistent with other studies (Chang et al., 2010; Nystrom et al., 2018). The UFKW has the expected symmetry across the equator and decreases in amplitude towards the poles (Andrews et al., 1987). The left panel has little evidence of any structure. This shows that the UFKW forced at the lower boundary of TIEGCM propagated into the thermosphere. Fig. 1 also shows that the UFKW is still influencing the temperature field, and with it the zonal wind field, above its peak into the F-region with amplitudes of 3–4 K. For comparison, Fig. 2 shows the amplitude of 1.0-day period temperature perturbations in both model runs. These perturbations are the superposition of all 6 diurnal tides in the CTMT. Fig. 2 is over the same latitude band and altitude range as Fig. 1. The color scale for Fig. 2 is 4 times larger than Fig. 1. Fig. 2 shows that the overall strength and latitudinal structure of the diurnal tides change very little with the addition of the UFKW. Thus the UFKW is not greatly modulating the diurnal tides when they are introduced together at the lower boundary of TIEGCM. Observations of interactions in the mesosphere between the tides and UKFWs could mean that the effect of UFKWs could be greater than reported here (Egito et al., 2018; England et al., 2012). We show that TIEGCM can reproduce an UFKW by only modifying the lower boundary and is a useful tool for studying interactions between large-scale waves and tides within the I-T system.

Fig. 1.

Fig. 1

Open in a new tab

Amplitude of the 3.5-day perturbations in temperature for CTMT Only (left) and CTMT + KW (right) runs for day 228 to day 234. Both panels share the same scale, between 0 K and 10 K, the same latitude band, between −40° and 40°, and the same altitude range, between 97 km and 207 km. The right panel shows a clear UFKW peaking near 110 km which is absent from the left panel.

Fig. 2.

Fig. 2

Open in a new tab

Same as Fig. 1 expect for the 1.0-day perturbations in temperature in CTMT Only (left) and CTMT + KW (right) runs. The scale goes from 0 K to 40 K. Both left and right panels show tidal structure peaking near 115 km. The difference between the two panels is very small showing the UFKW has little effect on the amplitude of the diurnal tides.

Atmospheric tides and UFKWs interact nonlinearly to produce secondary waves that have frequencies that are the sum and difference of the primary waves (Teitelbaum and Vial, 1991). To test whether these nonlinear interactions are occurring inside TIEGCM we look for secondary waves generated between the diurnal tides and the UFKW; we expect waves of 1.4-day and 0.8-day period to be generated through this interaction. The zonal wavenumbers of these secondary waves are also the sum and difference of the primary waves. With the CTMT diurnal wavenumbers, we expect 1.4-day period waves to have wavenumbers between westward 3 and eastward 2 and 0.8-day period waves to have wavenumbers between westward 1 and eastward 4. As an example, the strongest diurnal tide during this time period is the DE3. The interaction between the DE3 and the UFKW would produce a 1.4-day period wave with eastward 2 and a 0.8-day period wave with eastward 4. Secondary wave generation in the CTMT + KW run should decrease the energy of the primary waves compared to the CTMT Only run. We will focus on the decrease in energy of diurnal tides. Fig. 3 shows the percent change in the mean power of the diurnal temperature perturbations at 115 km, red, and 207 km, blue, relative to the CTMT Only run. Fig. 3 is over the same latitude band as Fig. 1. To produce Fig. 3, first the spectral power is found at each latitude, longitude, and altitude over the seven analysis days for 1.0-day perturbations. Then the mean is found over all longitudes for the given latitude. Finally, the percent change is calculated from these mean values. Fig. 3 shows that diurnal perturbations at their peak (115 km) have ~ 3% less power when the UFKW is present. The percent change in power in the F-region (207 km) have a median value of ~ 3% with a peak near the equator of almost 5%. This decrease in power at the diurnal period is a result of secondary wave generation.

Fig. 3.

Fig. 3

Open in a new tab

Mean percent change of the diurnal tides between the CTMT Only and CTMT + KW runs at 115 km (red) and 207 km (blue) over the latitudes −40° to 40°. Negative values indicate a loss in power of the diurnal tides. The red line shows that in the presents of the UFKW the tidal mean power decreases over the equatorial region ~ 3% at it’s peak altitude. The blue line shows that in the presence of the UFKW the tidal mean power decreases to a peak value of 4.5% at the equator.

We can determine the amplitude of these secondary waves using the method from Fig. 1. In Fig. 4 we plot the amplitude of 1.4-day period temperature perturbations. Fig. 4 shows a peak near the UFKW of ~ 3.5 K. In Fig. 5 we plot the amplitude of 0.8-day period temperature perturbations. Fig. 5 shows a peak near the UFKW of ~ 3.5 K. Figs. 4 and 5 are over the same altitudes and latitudes as Fig. 1. Peaks in the right panels of Figs. 4 and 5 show that the wave-wave interaction in TIEGCM is producing secondary waves. The peaks in both secondary waves are about one third the temperature amplitude of the UFKW. Like Figs. 1, 4 and 5 show changes in the temperature perturbations in the lower thermosphere, as well as small changes in the F-region. Similar figures to those above can be produced for semi-diurnal tides in the CTMT Only run and the CTMT + KW run that show secondary generation and decrease in mean power. These temperature perturbations have corresponding wind perturbations (not shown) which will affect the ionospheric wind electrodynamo. Changes in the electrodynamo will affect electron content in the model so we consider the total electron content (TEC) output to see what effects the simulation expects as a result of the UFKW.

Fig. 4.

Fig. 4

Open in a new tab

Same as Fig. 1 expect for the 1.4-day perturbations in temperature in the CTMT Only (left) and the CTMT + KW (right) runs. The scale goes from 0 K to 3.5 K. The right panel shows the presence of the secondary waves generated by the interaction between the UFKW and the diurnal tides. This secondary wave peaks near the UFKW seen in Fig. 1. The left panel shows no wave at the expected 1.4-day period of the secondary waves.

Fig. 5.

Fig. 5

Open in a new tab

Same as Fig. 1 expect for the 0.8-day perturbations in temperature in CTMT Only (left) and the CTMT + KW (right) runs. The scale goes from 0 K to 3.5 K. The right panel shows the presence of the secondary waves generated by the interaction between the UFKW and the diurnal tides. This secondary wave peaks near the UFKW seen in Fig. 1. The left panel shows no wave at the expected 0.8-day period of the secondary waves.

3.2. Response of total electron content to an ultra-fast Kelvin wave and secondary waves

We first look at zonal winds from the two model runs. Fig. 6 shows the CTMT Only zonal wind field at 153 km, the upper E-region, at 1700 local time (LT) for three consecutive days, day 231 to day 233. The range of all three days goes between −45 m/s and 45 m/s. A wave-4 structure is visible on all three days which varies very little between one day and the next. This is a signature of two tides within the CTMT (i.e., DE3 and SE2) though all CTMT tides are included. Fig. 7 is the residual zonal wind between the CTMT + KW and the CTMT Only runs for the same days, local time, and altitude as Fig. 6. The range goes between −5 m/s and 5 m/s. UFKW and all secondary waves generated that can be supported by TIEGCM are contained within this residual zonal wind field. It is hard to discern individual waves by eye in Fig. 7, but variations in the TEC should be related to all variations in the winds. We can see that the residual zonal winds are ~ 10% of the zonal winds seen in Fig. 6.

Fig. 6.

Fig. 6

Open in a new tab

Zonal winds at 153 km for the CTMT Only run for three consecutive days. The top panel is day 231. The middle panel is day 232. The bottom panel is day 233. All panels are showing 1700 LT for their respective days. All panels share the same scale between −45 m/s and 45 m/s. A wave-4 structure is clearly visible in all three panels. The wave-4 structure varies very little for one panel to the next.

Fig. 7.

Fig. 7

Open in a new tab

Residual of the zonal winds at 153 km between the CTMT + KW and CTMT Only runs for the same three consecutive days at 1700 LT as Fig. 6. All panels share the same scale between −5 m/s and 5 m/s. All panels are showing the zonal winds associated with the UFKW and secondary waves. The peak values of these zonal winds is ~ 10%.

We plot the percent change in TEC between the CTMT + KW and CTMT Only runs in Fig. 8. Fig. 8 shows the same days and local time as Figs. 6 and 7. The 0% contour is thickened. The magnetic equator is plotted as a dotted line. The peak percent changes are around ± 10%; a similar magnitude to the zonal wind variations in Fig. 7. Three features are visible in Fig. 8. First, there are large-scale positive and negative changes which are confined to longitude bands with apparent eastward propagation. Positive (negative) values indicate the TEC in the CTMT + KW run was greater (less) than the CTMT Only run. Second, within each of these large-scale regions we see peaks which fall on both sides of the magnetic equator around ± 20° magnetic latitude, away from the ionospheric anomaly which occurs at ± 10° magnetic latitude in these runs. Lastly, these large-scale regions show weakening or sign reversal around the magnetic equator. All of these features are from the electrodynamo reacting to the wind variations between the two model runs.

Fig. 8.

Fig. 8

Open in a new tab

Percent change in TEC for the same three consecutive days at 1700 LT as in Fig. 6. Positive (negative) values indicate higher (lower) TEC in the CTMT + KW compared to CTMT Only. All panels share the same scale between −10% and 10%. The 0% contour has been thickened. The magnetic equator is plotted as a dotted line. Three features are visible: large-scale structure in positive and negative changes, peaks poleward of TEC anomaly at ± 20° magnetic, and weakening/reversal of percent change near the magnetic equator.

The first feature can be understood by looking at the UFKW itself. Fig. 9 shows the UFKW at the lower boundary at 1700 LT from day 230 to day 232. These are the previous days to those shown in Figs. 6–8 because it takes about one day for the UFKW to reach 150 km. The structure seen in Fig. 9 is only the UFKW. Plotting in local time requires examination of 24 h of model runs from 0000 to 2300 universal time (UT), where at 0000 UT and 2300 UT the 1700 LT locale is in the American sector and so neighboring measurements there are obtained 24 h apart. Because of the 3.5-day period of the UFKW, it has moved a fraction of a wavelength in 24 h which appears as a discontinuity in Fig. 9. This effect is not obvious when viewing tides because they move one or more whole wavelengths in 24 h.

Fig. 9.

Fig. 9

Open in a new tab

The UFKW at the lower boundary of TIEGCM. Each panel is one day prior to the three days of Fig. 6 at 1700 LT: top is day 230, middle is day 231, and bottom is day 232. All panels share the same scale between −20 m/s and 20 m/s. The one day prior accounts for the time it takes the UFKW to propagate to 153 km. The westward (eastward) phase of the UFKW corresponds to the negative (positive) percent changes seen in Fig. 8.

Comparing Figs. 9 to 8 reveals that the westward (eastward) phase of the UFKW produces a negative (positive) percent change in TEC. The peaks in the negative (positive) percent changes show that less (more) plasma is making it to higher magnetic latitudes with the inclusion of the UFKW. The origin of this motion to higher magnetic latitudes is difficult to discern: either more or less plasma is moving in the electrodynamo, the UFKW is having an direct effect on the processes that move plasma away from the magnetic equator, or a secondary wave at higher latitudes (i.e., between the UFKW and the semi-diurnal tides) is moving the plasma. More work is needed to see which of these effects is driving this motion. All of the features discussed above indicate that the UFKW and/or secondary waves are having an effect on the structure of TEC more than the peak TEC value. The peak percent change of the TEC in Fig. 8 translates to ~ 1.5 TEC units (TECU). We also expect that these values are weaker than normal because of the exclusion of the mesosphere. For example, England et al. (2012) saw interaction between a UFKW and tides between 82–88 km and Egito et al. (2018) saw interaction at 91 km. Nevertheless TIEGCM still reproduces the nonlinear interaction between the UFKW, tides, and ionospheric plasma.

4. ICON wind results

As shown in Section 3, TIEGCM supports the addition of an UFKW and generated the expected secondary waves. NASA’s ICON mission will produce wind data between 90 km and 300 km and temperature data between 90 km and 105 km using the Michelson Interferometer for Global High-resolution Thermospheric Imaging (MIGHTI) instrument (Harding et al., 2017; Stevens et al., 2017). These wind and temperature data will be used in a HME fitting to force the lower boundary of TIEGCM (Forbes et al., 2017; Maute, 2017). ICON is expected to be launched into a circular orbit at 575 km with 27° inclination. In this orbit, ICON will sampling all local times at certain latitudes during each orbit and all local time at all locations in its 54-day precession cycle (Immel et al., 2017). MIGHTI will run through every orbit and only stop sampling during calibrations of other instruments every few day. MIGHTI wind data are retrieved using two different atomic oxygen lines; 557.7 nm (green) and 630.0 nm (red) (Harlander et al., 2017). For full details on MIGHTI and its retrieval methods see Englert et al. (2017), Harding et al. (2017), Harlander et al. (2017), Stevens et al. (2017). For full details on ICON see Immel et al. (2017) and references therein.

For this study we used the CTMT Only and CTMT + KW outputs to create artificial ICON wind data. These artificial data were created using realistic ephemeris for viewing geometries, but the values for the winds were taken from the two TIEGCM outputs at corresponding latitudes, longitudes, and altitudes. With these data we can determine if the winds changes between the two models would be visible to MIGHTI. We focus on green line zonal wind data which covers the altitude range from 90 km to 170 km during the day and from 90 km to 105 km at night. We produce the probability density of the difference between these zonal winds in the CTMT + KW run and the CTMT Only run in Fig. 10 for day 232. Fig. 10 has a bin size of 1 m/s and includes data from all green line altitudes. We see zonal wind differences are small and slightly skewed positive. This pattern is consistent when others days are used. Small values are expected as the tides dominate in the lower thermosphere and do not change significantly while interacting with the UFKW. However, Fig. 10 shows that larger differences do occur which could be detectable by ICON. Harding et al. (2017) discusses the expected precision of the MIGHTI data. In their study, Monte Carlo simulations were used to estimate MIGHTI’s precision during solar minimum conditions. They estimate a precision of ~ 5 m/s. This estimation is complex and the authors encourage those interested to read Harding et al. (2017) for full details. For the purposes of this study we will assume this to be the precision for all wind measurements. From Fig. 10 we find that around 45% of the differences are larger than ± 5 m/s which would be discernible to ICON.

Fig. 10.

Fig. 10

Open in a new tab

Probability density of artificial MIGHTI zonal wind differences for day 232 between the CTMT + KW and CTMT Only runs. Bins of 1 m/s are used. Probability sums to 1. The peak is at small values, but about 45% are greater than the ± 5 m/s precision of MIGHTI. These difference will be discernible to MIGHTI.

As an example, in Fig. 11 we plot the zonal winds at 100 km for the same single orbit for both model runs on the same three days as Fig. 6. The blue solid lines are zonal winds from the CTMT Only run and the red dashed lines are zonal winds from the CTMT + KW run. Orbits are approximately 24 h apart all starting near 12° south and 80° east. These data are plotted versus sample number. We add error bars of 5 m/s for the precision estimate. The overall structure for all six orbits is similar. We focus on three difference subsets in Fig. 11: sample numbers 0–60, 60–120, and 120–145. In sample range 0–60 we see CTMT + KW winds are more westward than CTMT Only winds on day 231. These stronger westward winds are greater than the 5 m/s precision. On day 232 we see that CTMT + KW winds are closer to CTMT Only winds with most points being within 5 m/s. On day 233 we see the opposite of day 231 with CTMT + KW winds being more eastward than CTMT Only winds. Again, most of these more eastward winds are greater than the 5 m/s. In sample range 60–120 most data are within the 5 m/s threshold on all three days. In sample range 120–145 we see that the CTMT + KW winds are within the 5 m/s precision on day 231, but day 232 and day 233 CTMT + KW winds are distinct from CTMT Only winds. This demonstrates that the ICON wind data will directly sample atmospheric changes due to UFKWs and other large scale waves that propagate into the lower thermosphere.

Fig. 11.

Fig. 11

Open in a new tab

Sample zonal wind data for the same three days as Fig. 6. Each panel is a single orbit at 100 km to ensure data over both day and night. All orbits start at 12° south 80° east. X-axis is the sample number in the orbit. Error bars are 5 m/s. The blue solid line is the CTMT Only run and the red dashed line is the CTMT + KW run. Samples where the effects of the UFKW are clearly visible. See text for full details.

5. Conclusion

In this study we present work done by introducing a single UFKW into the lower boundary of TIEGCM. It is shown that TIEGCM can support the propagation of an UFKW. This UFKW propagates into TIEGCM showing reasonable amplitude and peak location. We have also shown that TIEGCM produces secondary waves from nonlinear interactions of the UFKW and atmospheric tides imparted by CTMT. We show that UFKWs and secondary waves change the zonal wind fields by ~ 10% in the lower thermosphere. Larger variations may occur if the model included the mesosphere where the UKFW and tides also interact as they propagate. These small variations in the zonal winds become small variations in TEC. The phase of the UFKW is seen in the phase of the enhancement or depletion in TEC. With these data we show that NASA’s upcoming ICON mission will be able to sample the wind variation with enough precision to observe the UFKW and secondary waves. This will allow the study of large scale atmospheric waves, that are not tides, by using ICON data.

Acknowledgements

This study is supported by ICON through NASA’s Explorers Program contracts NNG12FA45C and NNG12FA421. The authors wish to thank the ICON science team for their helpful discussions, in particular B. Harding and A. Maute. TIEGCM is a community model with source code available through the Nation Center of Atmospheric Research: High Altitude Observatory or run requests through the NASA’s Community Coordinated Model Center. Magnetic data calculated using apexpy (https://pypi.org/project/apexpy).

Footnotes

1

For interpretation of color in Figs. 1 and 6, the reader is referred to the web version of this article.

References

  1. Andrews D, Holton J and Leovy C, Middle Atmosphere Dynamics. International Geophysics, 1987, Elsevier Science; San Diego, CA. [Google Scholar]
  2. Chang L, Palo S, Hagan M, Richter J, Garcia R, Riggin D and Fritts D, Structure of the migrating diurnal tide in the whole atmosphere community climate model (WACCM), Adv. Space Res 41, 2008, 1398–1407, 10.1016/j.asr.2007.03.035. [DOI] [Google Scholar]
  3. Chang LC, Palo SE, Liu HL, Fang TW and Lin CS, Response of the thermosphere and ionosphere to an ultra fast kelvin wave, J. Geophys. Res.: Space Phys 115, 2010, A00G04, 10.1029/2010JA015453. [DOI] [Google Scholar]
  4. Dickinson RE, Ridley EC and Roble RG, Thermospheric general circulation with coupled dynamics and composition, J. Atmos. Sci 41, 1984, 205–219, 10.1175/1520-0469(1984)041<0205:TGCWCD>2.0.CO;2. [DOI] [Google Scholar]
  5. Egito F, Buriti RA, Fragoso Medeiros A and Takahashi H, Ultrafast kelvin waves in the MLT airglow and wind, and their interaction with the atmospheric tides, Ann. Geophys 36, 2018, 231–241, 10.5194/angeo-36-231-2018. [DOI] [Google Scholar]
  6. England SL, A review of the effects of non-migrating atmospheric tides on the earth’s low-latitude ionosphere, Space Sci. Rev 168, 2012, 211–236, 10.1007/s11214-011-9842-4. [DOI] [Google Scholar]
  7. England SL, Immel TJ, Huba JD, Hagan ME, Maute A and DeMajistre R, Modeling of multiple effects of atmospheric tides on the ionosphere: an examination of possible coupling mechanisms responsible for the longitudinal structure of the equatorial ionosphere, J. Geophys. Res.: Space Phys 115, 2010, A05308, 10.1029/2009JA014894. [DOI] [Google Scholar]
  8. England SL, Liu G, Zhou Q, Immel TJ, Kumar KK and Ramkumar G, On the signature of the quasi-3-day wave in the thermosphere during the January 2010 URSI world day campaign, J. Geophys. Res.: Space Phys 117, 2012, A06304, 10.1029/2012JA017558. [DOI] [Google Scholar]
  9. Englert CR, Harlander JM, Brown CM, Marr KD, Miller IJ, Stump JE, Hancock J, Peterson JQ, Kumler J, Morrow WH, Mooney TA, Ellis S, Mende SB, Harris SE, Stevens MH, Makela JJ, Harding BJ and Immel TJ, Michelson interferometer for global high-resolution thermospheric imaging (MIGHTI): instrument design and calibration, Space Sci. Rev 212, 2017, 553–584, 10.1007/s11214-017-0358-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Forbes JM, Wave coupling between the lower and upper atmosphere: case study of an ultra-fast kelvin wave, J. Atmos. Sol. Terr. Phys 62, 2000, 1603–1621, 10.1016/S1364-6826(00)00115-2. [DOI] [Google Scholar]
  11. Forbes JM, Zhang X, Hagan ME, England SL, Liu G and Gasperini F, On the specification of upward-propagating tides for ICON science investigations, Space Sci. Rev 212, 2017, 697–713, 10.1007/s11214-017-0401-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Forbes JM, Zhang X, Maute A and Hagan ME, Zonally symmetric oscillations of the thermosphere at planetary wave periods, J. Geophys. Res.: Space Phys 123, 2018, 4110–4128, 10.1002/2018JA025258. [DOI] [Google Scholar]
  13. Forbes JM, Zhang X, Palo S, Russell J, Mertens CJ and Mlynczak M, Tidal variability in the ionospheric dynamo region, J. Geophys. Res.: Space Phys 113, 2008, A02310, 10.1029/2007JA012737. [DOI] [Google Scholar]
  14. Hagan ME and Forbes JM, Migrating and nonmigrating diurnal tides in the middle and upper atmosphere excited by tropospheric latent heat release, J. Geophys. Res.: Atmos 107, 2002, ACL 6–1–ACL 6–15, 10.1029/2001JD001236. [DOI] [Google Scholar]
  15. Haldoupis C, Pancheva D and Mitchell NJ, A study of tidal and planetary wave periodicities present in midlatitude sporadic E layers, J. Geophys. Res.: Space Phys 109, 2004, A2302, 10.1029/2003JA010253. [DOI] [Google Scholar]
  16. Harding BJ, Makela JJ, Englert CR, Marr KD, Harlander JM, England SL and Immel TJ, The MIGHTI wind retrieval algorithm: description and verification, Space Sci. Rev 212, 2017, 585–600, 10.1007/s11214-017-0359-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Harlander JM, Englert CR, Brown CM, Marr KD, Miller IJ, Zastera V, Bach BW and Mende SB, Michelson interferometer for global high-resolution thermospheric imaging (MIGHTI): monolithic interferometer design and test, Space Sci. Rev 212, 2017, 601–613, 10.1007/s11214-017-0374-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Heelis RA, Lowell JK and Spiro RW, A model of the high-latitude ionospheric convection pattern, J. Geophys. Res.: Space Phys 87, 1982, 6339–6345, 10.1029/JA087iA08p06339. [DOI] [Google Scholar]
  19. Hines CO, Gravity waves in the atmosphere, Nature 239, 1972, 73–78, 10.1038/239073a0. [DOI] [Google Scholar]
  20. Immel TJ, England SL, Mende SB, Heelis RA, Englert CR, Edelstein J, Frey HU, Korpela EJ, Taylor ER, Craig WW, Harris SE, Bester M, Bust GS, Crowley G, Forbes JM, Gérard JC, Harlander JM, Huba JD, Hubert B, Kamalabadi F, Makela JJ, Maute AI, Meier RR, Raftery C, Rochus P, Siegmund OHW, Stephan AW, Swenson GR, Frey S, Hysell DL, Saito A, Rider KA and Sirk MM, The ionospheric connection explorer mission: mission goals and design, Space Sci. Rev 214, 2017, 13, 10.1007/s11214-017-0449-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Immel TJ, Sagawa E, England SL, Henderson SB, Hagan ME, Mende SB, Frey HU, Swenson CM and Paxton LJ, Control of equatorial ionospheric morphology by atmospheric tides, Geophys. Res. Lett 33, 2006, L15108, 10.1029/2006GL026161. [DOI] [Google Scholar]
  22. Liu HL, 2014. WACCM-X simulation of tidal and planetary wave variability in the upper atmosphere. In: Huba J, Schunk R, Khazanov G (Eds.), Modeling the Ionosphere-Thermosphere System. American Geophysical Union (AGU), Washington, DC. chapter 16, pp. 181–199. 10.1002/9781118704417.ch16. [DOI] [Google Scholar]
  23. Maute A, Thermosphere-ionosphere-electrodynamics general circulation model for the ionospheric connection explorer: TIEGCM-ICON, Space Sci. Rev 212, 2017, 523–551, 10.1007/s11214-017-0330-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Nystrom V, Gasperini F, Forbes JM and Hagan ME, Exploring wave-wave interactions in a general circulation model, J. Geophys. Res.: Space Phys 123, 2018, 827–847, 10.1002/2017JA024984. [DOI] [Google Scholar]
  25. Oberheide J, Forbes JM, Zhang X and Bruinsma SL, Climatology of upward propagating diurnal and semidiurnal tides in the thermosphere, J. Geophys. Res.: Space Phys 116, 2011, A11306, 10.1029/2011JA016784. [DOI] [Google Scholar]
  26. Pogoreltsev A, Vlasov A, Fröhlich K and Jacobi C, Planetary waves in coupling the lower and upper atmosphere, J. Atmos. Sol. Terr. Phys 69, 2007, 2083–2101, 10.1016/j.jastp.2007.05.014. [DOI] [Google Scholar]
  27. Qian L, Burns AG, Emery BA, Foster B, Lu G, Maute A, Richmond AD, Roble RG, Solomon SC, Wang W, 2014. The NCAR TIE-GCM. In: Huba J, Schunk R, Khazanov G (Eds.), Modeling the Ionosphere-Thermosphere System. American Geophysical Union (AGU), Washington, DC, pp. 73–83. 10.1002/9781118704417.ch7 (chapter 7). [DOI] [Google Scholar]
  28. Richmond A, Ionospheric wind dynamo theory: a review, J. Geomagn. Geoelectr 31, 1979, 287–310, 10.5636/jgg.31.287. [DOI] [Google Scholar]
  29. Richmond AD, Ridley EC and Roble RG, A thermosphere/ionosphere general circulation model with coupled electrodynamics, Geophys. Res. Lett 19, 1992, 601–604, 10.1029/92GL00401. [DOI] [Google Scholar]
  30. Roble RG, Ridley EC, Richmond AD and Dickinson RE, A coupled thermosphere/ionosphere general circulation model, Geophys. Res. Lett 15, 1988, 1325–1328, 10.1029/GL015i012p01325. [DOI] [Google Scholar]
  31. Snively JB, Nonlinear gravity wave forcing as a source of acoustic waves in the mesosphere, thermosphere, and ionosphere, Geophys. Res. Lett 44, 2017, 12020–12027, 10.1002/2017GL075360. [DOI] [Google Scholar]
  32. Stevens MH, Englert CR, Harlander JM, England SL, Marr KD, Brown CM and Immel TJ, Retrieval of lower thermospheric temperatures from o2 a band emission: the MIGHTI experiment on ICON, Space Sci. Rev 214, 2017, 4, 10.1007/s11214-017-0434-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Takahashi H, Wrasse CM, Fechine J, Pancheva D, Abdu MA, Batista IS, Lima LM, Batista PP, Clemesha BR, Schuch NJ, Shiokawa K, Gobbi D, Mlynczak MG and Russell JM, Signatures of ultra fast kelvin waves in the equatorial middle atmosphere and ionosphere, Geophys. Res. Lett 34, 2007, L11108, 10.1029/2007GL029612. [DOI] [Google Scholar]
  34. Teitelbaum H and Vial F, On tidal variability induced by nonlinear interaction with planetary waves, J. Geophys. Res.: Space Phys 96, 1991, 14169–14178, 10.1029/91JA01019. [DOI] [Google Scholar]
  35. Timmermans RMA, van Oss RF and Kelder HM, Kelvin wave signatures in ECMWF meteo fields and global ozone monitoring experiment (GOME) ozone columns, J. Geophys. Res.: Atmos 110, 2005, D13104, 10.1029/2004JD005261. [DOI] [Google Scholar]
  36. Vincent R, Long-period motions in the equatorial mesosphere, J. Atmos. Terr. Phys 55, 1993, 1067–1080, 10.1016/0021-9169(93)90098-J, The First Pacific Regional Step Conference. [DOI] [Google Scholar]
  37. Zettergren MD, Snively JB, Komjathy A and Verkhoglyadova OP, Nonlinear ionospheric responses to large-amplitude infrasonic-acoustic waves generated by undersea earthquakes, J. Geophys. Res.: Space Phys 122, 2017, 2272–2291, 10.1002/2016JA023159. [DOI] [Google Scholar]

RESOURCES