Quasi-6-day wave effects in ionospheric E and F region during the recent solar maximum 2014–2015

We show the statistical characteristics of quasi-6-day wave (Q6DW) absolute amplitude in foE and foF2 during 2014–2015 by using six ionosondes at different latitudes. The results show that foE perturbations maximized at mid-latitudes during equinoxes, and the maximum amplitude of Q6DW in foF2 occurred near the northern crest of equatorial ionospheric anomaly (EIA). In addition, the absolute amplitude of Q6DW in foF2 increased with increasing solar activity. Our observations suggest that the dissipative Q6DW-like oscillations in the lower thermosphere may cause variations in the thermospheric neutral density via mixing effect and further result in foE disturbances in Q6DW events. Furthermore, the E region wind dynamo could also be modulated by the 6-day wave, thus leading to the disturbances in vertical plasma velocity via E × B drifts and F region electron density. Our observational investigation provides evidence of thermosphere–ionosphere coupling in the mid- and low-latitude region.


Introduction
As a significant oscillation in the middle atmosphere, planetary waves (PWs) from lower atmosphere have been investigated over the past several decades (Madden and Julian 1972;Riggin et al. 2006;Chang et al. 2010;Gu et al. 2013;Onohara et al. 2013). The westward propagating quasi-6-day wave (Q6DW) with period of 5-7 days, is one of the most prominent traveling PWs in the mesosphere-lower thermosphere (MLT) region. Previous studies have revealed the effect of Q6DW on the wind and temperature in the MLT region by both ground-based and space-based instruments (Rodgers 1976;Venne 1989;Talaat et al. 2001Talaat et al. , 2002Liu et al. 2004;Garcia et al. 2005;Sridharan et al. 2008;Gan et al. 2015;Gu et al. 2018). Study of Q6DW in winds by using Upper Atmosphere Research Satellite/High Resolution Doppler Imager (HRDI) data was shown in Wu et al. 1994. Their results illustrated that the maximum occurrence of Q6DW occurred during equinoxes in the equatorial and mid-latitude region in zonal and meridional winds, respectively. Gan et al. (2015) also presented the morphology of Q6DW in Thermosphere, Ionosphere, Mesosphere Energetics and Dynamics Doppler Interferometer (TIDI) horizontal wind and Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) temperature data during [2002][2003][2004][2005][2006][2007]. These global distributions of Q6DW occurrence presented in Gan et al. (2015) showed similar characteristics of Q6DW in the MLT region. Numerous research results have proposed that these seasonal variations of Q6DW response in the MLT region which reached a maximum during equinoxes were determined by background wave source, mean wind structure, baroclinic/barotropic instability, and critical layers of the wave (Meyer and Forbes 1997;Liu et al. 2004).
interaction between tide and Q6DW and the modulation of E region dynamo due to wind disturbances in the MLT region (Miyoshi 1999;Gu et al. 2014Gu et al. , 2018Gan et al. 2016Gan et al. , 2017Forbes and Zhang 2017;Yamazaki, 2018). Pedatella et al. (2012) and Forbes and Zhang (2017) have proposed that strong Q6DW-tide nonlinear interactions could produce secondary waves that vertically propagated into the E region. By using the Thermosphere-Ionosphere-Mesosphere Electrodynamics General Circulation Model (TIME-GCM), Gan et al. [2016] suggested that the zonal electric field in the equatorial E layer could be modulated by neutral wind perturbations due to Q6DW in the MLT region, which was confirmed in the work shown in Yamazaki et al. (2018) based on observations from Challenging Minisatellite Payload (CHAMP) and Swarm satellites. Yamazaki (2018) investigated the characteristics of Q6DW in total electron content (TEC) in the low-latitude and equatorial region over a solar cycle. Wave amplitude peaks appeared in equinoxes which agreed with those observed in neutral wind and temperature in the MLT region. Gu et al. (2018) further analyzed the characteristics of Q6DW in TEC and horizontal wind in the MLT region. They suggested that polarization electric fields modulated by zonal wind perturbations in the equatorial E region could affect daytime plasma velocity by E × B drift, which eventually resulted in Q6DW-like oscillations in TEC in the equatorial ionospheric anomaly (EIA) region via the equatorial fountain effect. Furthermore, the dissipation of PWs in the MLT region may also play a significant role in the vertical coupling between thermosphere and ionosphere. Based on multiple observations, Yamazaki et al. 2020 revealed 6-day variations simultaneously in the equatorial electrojet, F-region electron density, and TEC during the September 2019, which were attributed to the Q6DW simultaneously observed in the lower atmosphere. Yue and Wang (2014) first investigated the PWs-induced variations in the thermosphere/ionosphere using the TIME-GCM and suggested that mixing effects caused by PWs could lead to the changes of both O/N 2 ratio and TEC. The theoretical results were later confirmed by recent work based on satellite observations Gan et al. 2015. They showed the reduction of O/N 2 ratio and F 2 layer electron density during Q6DW events.
The 6-day variations in the critical frequency of F 2 layer (foF 2 ) have been studied in the past several decades (Laštovička 1996;Laštovička and Mlch 1996;Laštovička 2006a). Laštovička et al. (2003Laštovička et al. ( , 2006b investigated the planetary wave type oscillations in foF 2 by using ionosonde observations. These results presented that there was an obvious quasi-6-day oscillation amplitude in foF 2 . Altadill and Apostolov [2003] suggested the 6-day variations in foF 2 were attributed to the planetary wave activity in the MLT region. Laštovička and Mlch [1996] presented that the 6-day variations in foF 2 have also distinct seasonal and solar activity variations. These results showed that the quasi-6-day oscillation amplitude in foF 2 reached a minimum during the solar minimum in summer. Although there are many reports on 6-day variations of foF 2 in the context of vertical coupling by the Q6DW, a statistical analysis about Q6DW effects on the critical frequency of E layer (foE) has not been reported yet. By using ionosonde observations at different latitude region during 2014-2015, the statistical results of 6-day variations in foE and foF 2 are first presented in this study. The main objective of this work is to study the effect of Q6DW on ionospheric E and F layer by thermosphere-ionosphere coupling.

Data and methods
In this present study, the ionograms are obtained from January 2014 to December 2015, which is the most recent solar maximum period. The ionosondes are located at Guam ( Fig. 1. The ionograms are routinely recorded every 15 min at all stations, which are automatically scaled by the computer routine automatic real-time ionogram scaler with true height analysis (Reinisch et al. 2009). The ionospheric parameters, namely, foE and foF 2 are utilized for studying Q6DW effects on the ionosphere.
To obtain the information of wave in foE and foF 2 , the least square fitting model is adopted in the following formula within a 20-day running window at a 1-day time step: where A represents the background value; T represents the period of wave in days; t d represents the day of the year; t u represents the universal time in days; and B, C represent the coefficient of the cosine and sine term, respectively. The amplitude of wave y wave is written as: In this study, the parameters A, B, and C are calculated separately at different local times. The period T is considered with ranges of 4-8 days by increments of 0.125 day. (1) (2) y wave = B 2 (T ) + C 2 (T ). Figure Fig. 2. The maximum value of foE occurred at local noon in June solstice and minimum appeared in December solstice. foE value was greater at low-latitudes than that at middle-latitudes. Figure 3 shows the time series of foF 2 during 2014-2015. foF 2 peaks were observed during 12:00-18:00 LT in equinoxes at Guam and Sanya near the northern crest of EIA due to the equatorial fountain effect. In addition, the value of foF 2 increased with increasing solar activity (F 10.7 ≈ 146 in 2014 and F 10.7 ≈ 118 in 2015). It should be noted that there is a secondary maximum in foF 2 occurring around local midnight during the equinoxes occurring at all stations equatorward of Beijing. The reason may be attributed to the enhanced electric field (Liu et al. 2013;Zhao et al. 2008).

Results
The spectrum of foE at 12:00 LT at all stations in 2014 is presented in Fig. 4, except for Wuhan and Beijing due to data shortage. foE disturbance amplitude with period of ~ 6.5 days reached a maximum with about 0.2 MHz at Mohe in April, which indicates that there was a prominent Q6DW event in MLT region at Mohe during Spring. Furthermore, somewhat weaker disturbance in foE with period of 4.5-7.5 days was also observed at GU station during October, at SA station during March and October, at MH station during July and November and at YA station during November, respectively. Figure 5 illustrates the foF 2 spectrum during 2014 at a fixed local time of 12:00 LT, which was clearly dominated by a PW with period of ~ 6 days. The maximum amplitude of Q6DW in foF 2 occurred during April and November at all stations. The strongest absolute response in foF 2 appeared near the northern crest of EIA (GU and SA stations) with a maximum amplitude of ~ 1.4 MHz which accounted for ~ 8% of background foF 2 . Our results suggest that the latitude-season variations of Q6DW amplitude in foF 2 agreed well with the morphology of background foF 2 . Furthermore, compared to the large variation in Q6DW periods seen in foE in Fig. 4, the Q6DW periods in foF 2 are all quite similar for all events observed at all the stations. It should be noted that somewhat weaker disturbance in foF 2 with period of 5 -7 days was also observed at all stations during summer (June, July and August). The reason may be that foF 2 reached a minimum at all stations during summer, resulting in the weaker absolute response in foF 2 at all stations during summer.
The morphology of spectrum in foE at 12:00 LT at all stations in 2015 is shown in Fig. 6, expect for Wuhan due to data shortage. Similar characteristics of Q6DW in foE are also found in Fig. 6. The maximum amplitude of Q6DW in foF 2 with period of ~ 7.0 days occurred during May at Mohe. Furthermore, somewhat weaker disturbance in foE with period of 4.5-7.5 days was also observed at GU station during November, at SA station during March and September, at BP station during September, at MH station during February and at YA station during April and October, respectively. Similar temporal and spatial distribution of Q6DW in 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11  foF 2 in 2015 is also found in Fig. 7. Q6DW amplitude in foF 2 reached a maximum at GU and SA stations during equinoxes. Compared to the morphology of Q6DW in foF 2 during 2014 in Fig. 5, the weaker amplitude of Q6DW in foF 2 during 2015 may reveal the influence of solar activity, which was consistent with previous studies [e.g., Yamazaki 2018]. Figure 8 shows the local time versus month structures of Q6DW in foF 2 during 2014. In this study, the maximum amplitude of PWs with period of 5-7 days was adopted as the amplitude of Q6DW at different local time. Amplitude peaks were observed around 12:00-18:00 LT in the northern EIA crest region (at GU and SA stations) during equinoxes, which agreed well with the ionospheric fountain effect. The maximum amplitude of Q6DW was ~ 1.5 MHz at Sanya at 16:00 LT, while amplitude reached minima during 06:00-08:00 LT. At the mid-latitude region, the absolute amplitude decreased with increasing latitude. Similar variations of Q6DW amplitude in 2015 are also shown in Fig. 9.

Discussion
In the above analysis, we mainly investigated the statistical features of Q6DW in foE and foF 2 based on 2014-2015 data set obtained by six ionosondes in different latitude region. The general characteristics of Q6DW were basically similar to previous studies (Gu et al. 2018, Yamazaki 2018). Our results have suggested that the thermosphere-ionosphere coupling may be considered as the main reason for the effect of Q6DW on ionosphere.
In our reported results, the maximum amplitude of Q6DW in foE occurred at Mohe in equinoxes during 2014-2015. The generation mechanism of E region may help understand foE perturbations due to Q6DW in the MLT region. Daytime E layer at mid-latitudes is mainly produced in the ionization process of neutral O 2 by solar EUV radiation (Kelley 2009). Following Ivanov-Kholodny and Nusinov 1979, the expression of foE is given as follows: where q m and α ′ m are the maximum ionization rate and effective dissociative recombination coefficient in the E layer, respectively.I ∞ is the incident ionizing flux. σ i and σ a are the ionization and absorption cross sections. χ is the solar zenith angle. H is the O 2 scale height. e is elementary charge. According to Eq. (3), the variations of q m , I ∞ , σ i , σ a , and χ depend on local time and solar (3) activity. Thus, we propose that the morphology of Q6DW in foE may be explained by the variations of α ′ m or H during Q6DW events.
Ion compositions in the daytime mid-latitude E region are dominated by NO + and O 2 + (Kelley 2009). Danilov (1994) and Danilov and Smirnova (1997) have suggested that α ′ m depended on the NO + /O 2 + ratio, which was mainly controlled by [NO] concentration in the E region. In our cases, the maximum amplitude of Q6DW in foE was ~ 0.2 MHz which accounted for ~ 6% of background value. According to incoherent scatter radar observations, Mikhailov et al. (2007) proposed that 5% variations in foE may need the changes with a factor of 8-10 in [NO], which were impossible for the quiet daytime midlatitude E region. Therefore, we suggest that the effect of α ′ m variations during Q6DW events on foE disturbances could be neglected.
The effect of scale height H variations on foE changes has been studied by previous researches (Ivanov-Kholodny and Nusinov 1979;Mikhailov 1983;Nusinov 1988;Rishbeth 1990;Mikhailov et al. 2017). Mikhailov (1983) proposed that foE variations due to H variations were more likely to be induced by neutral particles vertical motion rather than neutral temperature changes.   It should be noted that the period of foE disturbance is not strictly within 6 days, which varies from 4.5 days to 7.5 days. Liu et al. (2004) suggested that this phenomenon may be caused by the effect of Doppler shift. In addition, the large variation in Q6DW periods seen in foE could also be caused by ionospheric parameter foE measurement errors, which may be induced due to the presence of sporadic E layer (E S ). Kelley [2009] proposed that an enhanced metallic-ion layer (a factor of 2-5 in foE) is frequently observed in ionospheric E region in mid-and low-latitude regions. The high-density Es could result in the measurement errors in foE, which could lead to a deviation in the period of foE disturbance during the Q6DW events.
Our statistical results also presented that the latitude structures of 6-day oscillation in foF 2 reached a maximum near the northern crest of EIA, which indicated the effect of Q6DW on the equatorial fountain effect via modulating E region dynamo in the MLT region (Gan et al. 2016Gu et al. 2018;Yamazaki 2018;Yamazaki et al. 2018). Liu et al. (2010) showed that zonal gradients of zonal and meridional wind perturbations from PWs could generate convergence/divergence of Pedersen and Hall currents, resulting in eastward polarized electric fields and equatorial vertical ion drifts. According to the dynamo theory, the wind-driven eastward currents are given as follows: where J P x and J H x are eastward Pedersen and Hall currents, respectively. σ P and σ H are Pedersen and Hall conductivities. u and v are zonal and meridional winds in the magnetic coordinates. B represents the geomagnetic field. I is the magnetic dip angle. From Eq. (4), it is found x is larger at magnetic low-latitude and equatorial region compared to J P x due to the finite dip angle. In addition, based on TIDI observations, zonal wind perturbations in the MLT region during Q6DW events were much stronger than those of meridional wind at low-latitude and equatorial region Gan et al. 2015. Therefore, we propose that the zonal wind component of Q6DW in the MLT region contributes more to foF 2 variations at lowlatitude and equatorial region by thermosphere-ionosphere coupling.
Recent studies have presented that the nonlinear interaction between tide and Q6DW in the coupling process also resulted in the variations in the F region ionosphere, especially the diurnal and semidiurnal variability. In the recent modeling work, Gan et al. (2016) illustrated migrating diurnal and semidiurnal tidal disturbances in F region electron densities during Q6DW events. Their results suggested that short-term variability of F region electron densities may result from strong Q6DW-tide nonlinear interaction via thermosphere-ionosphere coupling. Similar conclusions were also shown in Liu et al. (2010) and Yue et al. (2016). Furthermore, mixing effect induced by the dissipation of Q6DW in the MLT region could cause not only foE variations, but also foF 2 changes. With the decreased O and increased O 2 and N 2 propagating into the F region by molecular diffusion, O/N 2 ratio presented in Yue and Wang (2014) was decreased by about 16% and 20% near the ionospheric F 2 peak at low-latitudes and mid-latitudes. According to previous study, F 2 region electron density was proportional to O/N 2 ratio (Rishbeth 1998). Therefore, the reduced O/N 2 ratio could also produce the variations of foF 2 during Q6DW events.
Yamazaki 2018 presented the effect of solar activity on the absolute amplitude of Q6DW in TEC based on a long-term data set during 2004-2017. They showed that absolute amplitude at low-latitudes increased with increasing solar activity. In this work, we presented similar features of Q6DW absolute amplitude in foF 2 during 2014-2015. The weaker amplitude occurred in 2015 due to lower solar activity. Our study provides observational evidences of the solar activity dependence of absolute amplitude in foF 2 response to Q6DW from lower atmosphere.

Summary
In this work, we reported observations of six ionosondes during the most recent solar maximum 2014-2015. Morphology of Q6DW absolute amplitude in foE and foF 2 at different latitudes was investigated during this period. We also discussed the effects of Q6DW in the MLT region on ionosphere via thermosphere-ionosphere coupling. The conclusions are summarized as follows: 1. In our statistical results, the maximum amplitude of Q6DW in foF 2 occurred around 12:00-18:00 LT in equinoxes near the northern EIA crest region. The absolute amplitude in foE reached the maximum in equinoxes at mid-latitudes. In addition, the solar activity dependence of absolute amplitude in foF 2 due to Q6DW was also shown in our study. These results were in agreement with previous studies.
2. It is suggested that the mixing effect induced by the dissipation of Q6DW in the MLT region could change E region neutral composition and reduce effective scale height H, which eventually caused foE disturbances during Q6DW events. Furthermore, equatorial zonal wind disturbances due to Q6DW in the MLT region could modulate E region dynamo and vertical plasma drifts, resulting in F region electron density changes in the northern crest of EIA via equatorial fountain effect. Our study proposes that the thermosphere-ionosphere coupling plays a key role in ionospheric variations due to Q6DW from lower atmosphere.

Key points
• Q6DW signatures in foE and foF 2 are simultaneously examined using an array of ionosondes. • The latitudinal changes of Q6DW in foE and foF 2 may be caused by mixing effects and modulation of E region dynamo due to PWs, respectively. • Thermosphere-ionosphere coupling may play a significant role in the effect of Q6DW on ionosphere in the mid-and low-latitude region.