Thermospheric tidal effects on the ionospheric midlatitude summer nighttime anomaly using SAMI3 and TIEGCM

Transcription

1 JOURNAL OF GEOPHYSICAL RESEARCH: SPACE PHYSICS, VOL. 118, , doi: /jgra.50340, 2013 Thermospheric tidal effects on the ionospheric midlatitude summer nighttime anomaly using SAMI3 and TIEGCM C. H. Chen, 1 C. H. Lin, 1 L. C. Chang, 2 J. D. Huba, 3 J. T. Lin, 1 A. Saito, 4 and J. Y. Liu 2,5 Received 4 February 2013; revised 13 May 2013; accepted 15 May 2013; published 21 June [1] This paper is the first study to employ a three-dimensional physics-based ionosphere model, SAMI3, coupled with the National Center for Atmospheric Research Thermosphere Ionosphere Electrodynamics General Circulation Model (TIEGCM) and Global Scale Wave Model to simulate the mesospheric and lower thermospheric tidal effects on the development of midlatitude summer nighttime anomaly (MSNA). Using this coupled model, the diurnal variation of MSNA electron densities at 300 km altitude is simulated on both June solstice (day of year (DOY) 167) and December solstice (DOY 350) in Results show successful reproduction of the southern hemisphere MSNA structure including the eastward drift feature of the southern MSNA, which is not reproduced by the default SAMI3 runs using the neutral winds provided by the empirical Horizontal Wind Model 93 neutral wind model. A linear least squares algorithm for extracting tidal components is utilized to examine the major tidal component affecting the variation of southern MSNA. Results show that the standing diurnal oscillation component dominates the vertical neutral wind manifesting as a diurnal eastward wave-1 drift of the southern MSNA in the local time frame. We also find that the stationary planetary wave-1 component of vertical neutral wind can cause diurnal variation of the summer nighttime electron density enhancement around the midlatitude ionosphere. Citation: Chen, C. H., C. H. Lin, L. C. Chang, J. D. Huba, J. T. Lin, A. Saito, and J. Y. Liu (2013), Thermospheric tidal effects on the ionospheric midlatitude summer nighttime anomaly using SAMI3 and TIEGCM, J. Geophys. Res. Space Physics, 118, , doi: /jgra Introduction [2] The photoionization caused by solar extreme ultraviolet (EUV)/UV radiation is the main factor controlling the diurnal variation of electron density in the ionosphere and causes the maximum electron density during the daytime, and minimum electron density during the nighttime [e.g., Kelley, 1989; Burns et al., 2004; Forbes et al., 2006]. An anomaly showing maximum electron density during the nighttime is located near the Weddell Sea region during the southern hemisphere summer and was first observed by an ionosonde in the 1950s and followed by several later studies [Bellchambers and Piggott, 1958; Penndorf, 1965; Dudeney and Piggott, 1978]. This phenomenon is referred to as the 1 Department of Earth Science, National Cheng Kung University, Tainan, Taiwan. 2 Institute of Space Science, National Central University, Chung-Li, Taiwan. 3 Plasma Physics Division, Naval Research Laboratory, Washington, D. C., USA. 4 Department of Geophysics, Kyoto University, Kyoto, Japan. 5 Center for Space and Remote Sensing Research, National Central University, Chung-Li, Taiwan. Corresponding author: C. H. Lin, Department of Earth Science, National Cheng Kung University, No.1, University Road, Tainan 701, Taiwan. (charles@mail.ncku.edu.tw) American Geophysical Union. All Rights Reserved /13/ /jgra Weddell Sea anomaly (WSA). Studies of the WSA have been advanced in recent years through using satellite observations providing global observations of this phenomenon [e.g., Horvath and Essex, 2003; Horvath, 2006; Burns et al., 2008; Lin et al., 2009; Jee et al., 2009; He et al., 2009, 2010; Liu et al., 2010]. A summer evening increase in fof2 above Millstone Hill radar observatory (42.6 N, 71.5 W) in the northern hemisphere was first presented by Evans [1965], although the phenomenon was not associated with the WSA by the authors. While providing a clearer picture of the spatial extent of the WSA, scientists have discovered a similar phenomenon with WSA occurring around the Europe and Northeast Asia regions during the northern hemispheric summer using satellite and groundbased observations. These similar nighttime electron density enhancement phenomena in the northern and southern hemispheres have since been collectively referred to as the midlatitude summer nighttime anomaly (MSNA) or summer evening anomaly [Lin et al., 2009, 2010; Thampi et al., 2009, 2011; Liu et al., 2010; Burns et al., 2011; Hsu et al., 2011; de Larquier et al., 2011; Chen et al., 2012]. In this paper, the nighttime anomaly in the southern (northern) hemisphere is called as southern (northern) MSNA. [3] At magnetic midlatitude regions, the electron density enhancement could be produced by the longer duration of photoionization from solar radiation [e.g., Sojka et al., 1985; Horvath and Essex, 2003; Chen et al., 2011, 2012], equatorward neutral winds [e.g., Park, 1971; Dudeney and 3836

2 Piggott, 1978; Su et al., 1994; Horvath and Essex, 2003; He et al., 2009; Thampi et al., 2009, 2011; Liu et al., 2010; Lin et al., 2010; Chen et al., 2011; 2012], and the downward diffusion of plasmaspheric plasma [Park, 1971; Bailey et al., 1991; Burns et al., 2008; Liu et al., 2010; Chen et al., 2011]. Many scientists have attempted to find the mechanisms of the MSNA electron density by carrying out theoretical model simulations. Thampi et al. [2011] simulated the northern MSNA along 135 E longitude by using the Sheffield University Plasmasphere Ionosphere Model with MU radar winds and obtained a better reproduced anomaly structure compared to that using winds from the Horizontal Wind Model 93 (HWM93). Chen et al. [2011] used various neutral wind and electric field conditions in the SAMI2 (Sami2 is Another Model of the Ionosphere) [Huba et al., 2000] model to simulate the electron density structure and further examine the mechanisms of the southern MSNA. Their simulation results showed that the equatorward neutral wind is the most important driver for the southern MSNA formation. Similarly, Ren et al. [2012] confirmed the effects of thermospheric meridional and zonal winds on the formation of the electron density enhancement using a three-dimensional model. [4] The MSNA is highly related to the thermospheric neutral winds that are produced by the in situ thermospheric circulations and upward propagating tides from the mesosphere and lower thermosphere (MLT) region. It is therefore desirable to perform a model simulation considering the relation between the thermospheric winds and the upward propagating tides. In this study, a three-dimensional physics-based model of the ionosphere (Sami3 is Also a Model of the Ionosphere, SAMI3) coupled with a global circulation model (National Center for Atmospheric Research Thermosphere Ionosphere Electrodynamics General Circulation Model, NCAR TIEGCM) and a tidal and planetary wave model (Global Scale Wave Model, GSWM) are used to simulate the mesospheric and lower thermospheric tidal effects on the formation of nighttime electron density enhancements and their longitudinal structure. The coupled model results are compared with a SAMI3 simulation using neutral winds from the empirical horizontal wind model (HWM93). 2. Model Description [5] The version of the NRL SAMI3 model used in this study is a global extension version of the two-dimensional ionosphere model SAMI2 that treats the dynamic plasma and chemical evolution of seven ion species (H +,He +,N +, O +,N + 2,NO +, and O + 2 ) in the low-latitude and midlatitude ionosphere over the altitude range 85 20,000 km [Huba et al., 2000]. The empirical thermospheric models, NRLMSISE00 [Picone et al., 2002] and HWM93 [Hedin et al., 1996] as well as the Scherliess-Fejer empirical E B drift model [Scherliess and Fejer, 1999] are used in the default SAMI3 model run. These models provide the time-dependent global specification of the atmosphere and solve the ionospheric equations along the magnetic field line considering the geometry of the eccentric dipole coordinates [cf. Huba et al., 2000]. An advantage of SAMI3 is that it includes the ion inertia term in the momentum equations [Huba et al., 2000]. Without including the ion inertia term, the plasma velocities at field lines with very high apex heights (field lines at poleward of around 50 magnetic latitude) may possibly become supersonic and result in unrealistic plasma distributions or may not reproduce the MSNA correctly. It is noted that there is also a version of SAMI3 that deals with the ionospheric electrodynamic self-consistently but with the central dipole magnetic field configuration. To avoid the complex results from changes of E B drift due to neutral wind driven by this additional imposed tidal effect and have a more realistic magnetic field configuration, the self-consistent electrodynamic version was not used for this study. According to Chen et al. [2011], the E B drift contributes little effect in the formation of WSA compared with the neutral wind effect through frictional force. Therefore, neglecting the imposed tidal effect to the E B drift is an appropriate assumption for this study. [6] Recently, several studies have indicated that ionosphere electron densities have a strong connection with the MLT nonmigrating tides, which can cause notable longitudinal diurnal tidal variation in the MLT region. To investigate the tidal effects on MSNA development, neutral winds from the NCAR TIEGCM with lower boundary tidal forcing produced by GSWM runs were used for specifying the neutral winds of the SAMI3 model runs. GSWM is a steady-state linear mechanistic model of planetary waves and solar tides in the Earth s atmosphere that has been used extensively to explore mechanisms of tidal and planetary wave structure and variability [Hagan et al., 1995, 1999; Hagan, 1996; Hagan and Forbes, 2002, 2003]. In order to capture the effects of thermospheric tides, the lower boundary of TIEGCM [Richmond et al., 1992] at around 97 km is specified by utilizing GSWM migrating and nonmigrating tidal amplitudes and phases, resulting in self-consistent thermospheric and ionospheric parameters in response to this lower boundary forcing. These TIEGCM simulations provide the neutral winds from 97 to 500 km and are incorporated into SAMI3. These results are then compared with SAMI3 runs using winds specified by HWM93. Note that the TIEGCM neutral winds above 500 km are assumed to be constant with altitude. According to the characteristics of the MSNA revealed by satellite observations [cf. Lin et al., 2009], we design two magnetic quiescent cases during June and December 2007 to simulate the MSNA electron density structure and examine its relationship with the thermospheric tides. The geophysical conditions used in this study are as follows: case (1) Year = 2007, day of year (DOY) = 167 (16 June), Ap = 9, F10.7 = 70.2, and F10.7A = 74.9; and case (2) Year = 2007, DOY = 350 (16 December), Ap = 3, F10.7 = 79.1, and F10.7A = Simulation Results [7] Figure 1 shows the diurnal variations of electron densities at 300 km altitude in global constant local time simulated by SAMI3 with the default neutral wind model (HWM93) on (a) DOY 167 and (b) DOY 350 in It can be seen that the equatorial ionization anomaly (EIA) crests are well developed between 1000 and 1600 LT in both hemispheres. The asymmetric EIA crests show a typical ionospheric structure: winter EIA crest intensifies before the summer EIA crest. The hemispheric asymmetry of EIA crests is due to the seasonal effect [cf. Lin, 2005; Fang et al., 2006; Lin et al., 2007]. Figure 1a shows that a higher electron density region 3837

3 Figure 1. Diurnal variations of the SAMI3 with HWM93 simulated electron densities at 300 km altitude in global constant local time on (a) DOY 167, 2007 and (b) DOY 350, The black line denotes the geomagnetic equator. occurs between west Asia and Europe (around 60 E 100 E geographic longitude) starting at 2000 LT and persists till 0200 LT. The feature is similar to the northern MSNA, but the longitudinal location is different from the previous observations obtained from the satellite and ionosonde [Lin et al., 2009, 2010; Thampi et al., 2009, 2011; Liu et al., 2010; Burns et al., 2011; Hsu et al., 2011; de Larquier et al., 2011; Chen et al., 2012]. The simulation result shows only one MSNA enhancement region, while the observations show two MSNA enhancement regions around east Asia and central Europe in the northern hemisphere. Figure 1b reveals that the electron density enhancement of the southern MSNA occurs clearly between South America and Antarctica (around 150 E~ 50 E geographic longitude) after 2000 LT and lasts until 0600 LT, covering a narrower longitude range than that which occurs in the northern hemisphere. Furthermore, the simulated electron density enhancement of the southern MSNA shows a fixed location of the structure in the constant local time frame and is similar with the previous observations. However, the zonal eastward drift of the southern MSNA revealed in the previous studies using satellite observations [Burns, et al., 2008; He et al., 2009; Jee et al., 2009], although not being pointed out by these authors, is not reproduced here. [8] The neutral wind plays an important role in the variation of ionospheric F region parameters by sustaining (lowering) the F region ionization to higher (lower) altitudes where the ion-neutral recombination rate is significantly decreased (increased), leading to enhanced (reduced) electron densities [Park, 1971; Dudeney and Piggott, 1978; Kelley, 1989; Su et al., 1994; Horvath and Essex, 2003; He et al., 2009; Thampi et al., 2009, 2011; Liu et al., 2010; Lin et al., 2010; Chen et al., 2011, 2012]. The upward/downward fieldaligned transport of electron density by the neutral wind effect depends on the configuration of the magnetic field and the meridional and zonal wind conditions. It suggests that the different features of MSNAs obtained from simulations and observations (two peaks in the Northern Hemisphere and eastward drift in the Southern Hemisphere) might be related with the longitudinal variation of neutral wind and the geomagnetic configuration. The primary source generating the longitudinal variations of thermospheric neutral winds in the constant local time frame is thought to be the nonmigrating solar tides forced by tropospheric latent heating, as well as nonlinear tide-tide interactions [Angelats i Coll and Forbes, 2002; Hagan et al., 2009; Zhang et al., 2010a, 2010b]. The migrating/sun-synchronous tides, on the other hand, appear as part of the zonal mean in the constant local time. The HWM93 model is known to underestimate the migrating/sun-synchronous tides while lacking nonmigrating tides. The nonmigrating tides are known from observations [Forbes et al., 2008] and models [Hagan and Forbes, 2002] to be important factors in the dynamics of the MLT region and can sometimes have magnitudes comparable to those of the migrating tides. To examine the tidal effects on the thermospheric neutral winds and their potential effects in producing the MSNA, neutral winds generated by TIEGCM coupled with GSWM migrating and nonmigrating 3838

4 Figure 2. Same as Figure 1 but for the simulation using the TIEGCM neutral winds. tides at the lower boundary were used to specify the neutral winds of the SAMI3 model runs. [9] Figure 2 shows the associated diurnal variation of electron density in global constant local time at 300 km altitude simulated by SAMI3 with TIEGCM neutral winds. During the northern hemispheric summer (Figure 2a), it is clear that the northern MSNA appears around 2200 LT and continues to 0400 LT. The enhancement peak is located around 30 N 40 N magnetic latitude, showing lower latitude location but wider longitude range than the results of HWM93 (Figure 1a). The location of the MSNAs in Figure 2a is similar to previous observations [Lin et al., 2009, 2010; Liu et al., 2010; Hsu et al., 2011], but it only shows an MSNA enhancement region along the geomagnetic latitude. This might be due to the difference of the neutral winds between the TIEGCM and reality. Figure 2b shows that the nighttime electron density enhancement appears around 2200 LT and continues to 0600 LT around the Weddell Sea region in the southern hemisphere, which is similar with the results of Figure 1b, except that the eastward movement of the southern MSNA is clearly seen in Figure 2b, consistent with observations. At 2200 LT, the density peak of southern MSNA is located around 100 E geographic longitude and extended to 30 E geographic longitude at 0600 LT with an eastward drift velocity of 251 m/s (= km/8 h; where, km is the length of 1 geographic longitude at 60 geographic latitude). This velocity is consistent with the FORMOSAT-3/COSMIC observation by Liu et al. [Liu et al., Three-dimensional observation on mid- and high-latitude electron density by using FORMOSAT-3/COSMIC, manuscript in preparation, hereinafter referred to as Liu et al., manuscript in preparation]. Their observations also indicate the eastward drift of MSNA in the northern hemisphere. However, the coupled simulation performed herein could not reproduce this feature in the northern hemisphere (Figures 1a and 2a). 4. Tidal Decomposition and Discussion [10] Since the feature of MSNA is mainly controlled by the neutral wind combined with the geomagnetic field configuration, the MSNAs occur around 45 N magnetic latitude, where neutral wind that produced upward effects are still comparable to the effect of gravitational force. To investigate the relationship between the MLT neutral wind tidal effect and the ionospheric MSNA, we perform tidal decomposition of the electron density at 300 km from Figure 2 and the neutral wind effect along the geomagnetic field line at 200 km and 45 N geomagnetic latitude. Using the linear least squares algorithm as described by Wu et al. [1995], the sampled electron density in each latitude bin in the local time frame is fitted to basic functions of the following form: Xðt LT ; lþ ¼ X þ X5 X 5 A n;s cos nωt LT ðn þ sþl þ y n;s n¼1 s¼ 5 þ X5 s¼1 A 0;s cos sl þ y 0;s Where X = electron density, t LT = local time, Ω = rotation rate of the earth (2p/24), l = geomagnetic longitude, X = zonal and time mean of electron density, n (= 1, 2,..., 5) denotes (1) 3839

5 Figure 3. The tidal spectrum of electron densities at 300 km altitude from Figure 2 at geomagnetic latitude (a) 45 N on DOY 167, 2007 and (b) 45 S on DOY 350, a subharmonic of 24 h, s (= 5, 4,..., 4, 5) is the zonal wave number, and the amplitude A n. s and phase y n,s are functions of altitude and latitude. The numbers of n = 1, 2, and 3 represent oscillation periods of 24 h (diurnal, denoted as D), 12 h (semidiurnal, denoted as S), and 8 h (terdiurnal, denoted as T) tides, respectively. The positive and negative numbers of s represent the eastward (denoted by E) and westward (denoted by W) propagation of global-scale wave propagating with respect to longitude. Using the naming convention described by Forbes and Wu [2006], the westward and eastward propagating diurnal tides are termed DWs and DEs, respectively, with zonal wave number s. The standing oscillations (s = 0) are termed D0, S0, and T0, and the stationary planetary waves (n = 0) with zonal wave number s are denoted as SPWs. Also, we can separate the decomposed tides into migrating (n + s = 0) and nonmigrating (n 6¼ s) tides. [11] The tidal decomposition results of electron density at 300 km in universal time, and geomagnetic coordinates are shown in Figure 3. Figure 3a shows a spectrum of the different tidal components of northern hemispheric electron density at fixed 45 N geomagnetic latitude on DOY 167, It clearly shows that the spectra have two peaks at zonal and time mean (n = 0 and s = 0, referred to hereinafter as the zonal mean) and the migrating diurnal tide (DW1), which are induced by the diurnal daily cycle of solar photoionization. Other stronger tidal components in electron density observed along 45 N geomagnetic latitude are the semidiurnal westward tide with zonal wave number s = 2 (SW2), standing diurnal oscillation (D0), diurnal eastward tide with zonal wave number s = 1 (DE1), and stationary planetary wave-1 oscillation (SPW1). According to equation (1), the migrating components (DW1 and SW2) are longitude independent (s 0 = n + s = 0, where s 0 is the wave number in the global constant local time frame) in the local time frame. The D0 and DE1 appear as diurnal eastward wave-1 (s 0 =1) and diurnal eastward wave-2 (s 0 = 2) structures of electron density in the local time frame, respectively. Figure 3b shows the tidal spectrum of electron density at 45 S geomagnetic latitude on DOY 350, The zonal mean and DW1 have stronger amplitude followed by D0 with an amplitude of ele/cm 3. A similar situation occurs with the spectra in the northern hemisphere (Figure 3a), though the SPW1 also has a relatively stronger amplitude. Differing with Figure 3a, the nonmigrating tidal components (DE1, DE2, and DE3) in Figure 3b also show notable amplitudes in the southern hemisphere. It is noted that the D0 component is an important contributor to the diurnal eastward wave-1 structure of southern hemisphere summertime electron density in the local time frame as seen in Figure 2b. [12] The vertical component of the neutral wind along the geomagnetic field line can be described by [13] Where V vert ¼ W geo m geo cosy þ W siny cosjjsin I jj I (2) V vert = vertical component of the neutral wind (positive for upward); = magnetic northern/southern hemisphere; W geo m = the geographically meridional neutral wind velocity (positive for northward); W geo z z = the geographically zonal neutral wind velocity (positive for eastward); y = declination angle of the geomagnetic field (positive for eastward); I = inclination angle of the geomagnetic field (positive for northern hemisphere; negative for southern hemisphere). [14] Figure 4 shows the results of tidal decomposition for the vertical component of neutral wind along the magnetic 3840

6 Figure 4. Same as Figure 3 but for the vertical component of neutral wind at 200 km. field lines at geomagnetic 45 N latitudes and 200 km altitude in TIEGCM. Choosing 200 km altitude for tidal analysis of neutral wind is due to the following reasons. First, Chen et al. [2011] show that the upward plasma drift affected by neutral wind and the associated divergence of electron density flux are more prominent between 200 and 300 km altitude where the ion-neutral collision effect remains important. Second, the thermospheric neutral winds are almost constant above 200 km altitude. [15] Except for the component of the zonal mean, the DW1 component has the strongest amplitude in the northern hemisphere. It is noted that similar with the results of electron density shown in Figure 3a, the D0, DE1, and SW2 components also have notable amplitudes. It may also be mentioned here that compared with Figure 3a, the amplitude of SPW1 is lower than SPW2, and the amplitude of D0 is also lower than DE1 in Figure 4a. There are also stronger components (such as DW3 and SW4) seen in Figure 4a, but are not clear in Figure 5. Altitude-latitude variations of amplitude and phase for D0 component of vertical neutral wind effects on (a, b) DOY167 and (c, d) DOY350 in 2007, respectively. The horizontal axis is the magnetic latitude. 3841

7 Figure 6. Reconstruction of electron densities at 300 km altitude in the global constant local time frame by only the D0 component from Figure 3 on (a) DOY 167, 2007 and (b) DOY 350, Figure 3a, indicating that these components cannot propagate to upper atmosphere and cause the variations of MSNA. The decomposition results of southern vertical winds show significant amplitudes of DW1, D0, and SPW1, which correspond one-to-one with the components of electron density in Figure 3b. On the other hand, the strong D0 component can produce the feature of the diurnal eastward wave-1 structure of the MSNA in the local time frame. The direct relationship between vertical winds along geomagnetic field line in the thermosphere and electron density of MSNAs in the present study further confirms the relation suggested by previous studies [Horvath and Essex, 2003; He et al., 2009; Thampi et al., 2009, 2011; Liu et al., 2010; Lin et al., 2010; Chen et al., 2011; 2012]. [16] The nonmigrating tides can be excited by the asymmetries of Earth surface [Hagan and Forbes, 2003], nonlinear interactions between the migrating tides and the planetary waves [Hang and Roble, 2001], and the geomagnetic field configuration [Wu et al., 2012]. According to Hagan and Roble [2001], the nonlinear interactions between the migrating wave DW1 and the planetary wave SPW1 can generate the diurnal D0 and DW2 components (s = 1 1= 0or 2). However, compared to the D0 components, DW2 components shown in Figures 3 and 4 are very weak, which could be caused by different propagation conditions [Liu et al., 2007] and/or different geomagnetic field configurations. The D0 components are significantly stronger in the southern hemisphere than in the northern hemisphere and are likely to be caused by the different geomagnetic field configurations in both hemispheres. In general, the DW1 tidal component propagating upward from the lower atmosphere is forced primarily by diurnal insolation of the stratosphere and troposphere [Chang et al., 2013] and influenced by the background zonal mean winds. The DW1 tidal component in the upper thermosphere can also be attributable to the in situ response to the EUV solar radiation absorption [Hagan et al., 2009]. At the southern MSNA region (around 90 E geographic longitude), the geomagnetic equator has large offset from the geographic equator, which can be approximated as an SPW1 component. The neutral wind in the thermosphere can be modified by the effect of ion drag along the geomagnetic field line [Nozawa and Brekke, 1995]. Therefore, the neutral wind DW1 component could be modulated by the SPW1 geomagnetic filed configuration when translated from geographic to geomagnetic coordinates. This effectively acts as a nonlinear interaction to excite the D0 component around southern MSNA region. Since the range of the geomagnetic equator offset from the geographic equator around the northern MSNA region is not as large as the southern MSNA region, the D0 component is not significant in the northern MSNA region. [17] As mentioned above, we understand that the D0 component is established as a major contributor to the diurnal eastward wave-1 drift of electron density in the local time frame. In order to confirm whether the D0 component of vertical neutral wind can propagate up from lower atmosphere to affect the variations of MSNA or is generated in situ in the thermosphere, the variations of amplitude and phase for the D0 component of vertical neutral winds are calculated with altitude. The neutral winds are calculated 3842

8 Figure 7. Same as Figure 6 but for the D0 and SPW1 components from Figure 3. by the upward propagating migrating and nonmigrating tidal components from the troposphere in the GSWM [Hagan and Forbes, 2002] and used as the lower boundary of TIEGCM at around 97 km in this study. Figure 5 shows the altitude variations of amplitude and phase on DOY167 and DOY350 in 2007, respectively. It is clearly shown in amplitude results (Figures 5a and 5c) that there are relatively larger amplitudes of the D0 component at altitudes higher than 200 km, around the midlatitude regions of the two hemispheres. According to the phase variation results of D0 component (Figures 5b and 5d), the D0 component can only propagate up to around 150 km altitude from the lower atmosphere. This result suggests that the D0 component of vertical neutral winds higher than 150 km are generated in situ in the thermosphere from a nonlinear interaction between DW1 component and SPW1 component. [18] To examine the role of the D0 component in producing the eastward drift of the southern MSNA, we reconstruct the electron density distributions by extracting the D0 component amplitudes (shown in Figure 3) and phases (not shown). Figure 6 shows the reconstruction of diurnal electron density at 300 km altitude by the D0 component from Figure 3 in the global constant local time frame. Figure 6a is the electron density distribution on DOY 167, 2007 in the northern hemisphere. It clearly reveals that the electron density enhancement at MSNA latitudes has eastward drift and reaches the middle Asia region around 0000 LT. However, this kind of eastward drift of electron density is not significant in Figure 2a, which shows the superposition of all tidal and SPW components. Figure 6b shows the reconstruction result on DOY 350, 2007 in the southern hemisphere. The electron density enhancement also shows a clear eastward drift and reaches the southern MSNA region around 0000 LT. The location and the local time of eastward drift of electron density are consistent with previous model simulations and satellite observations. Furthermore, the feature of eastward drift of electron density is restricted to the magnetic midlatitude region, consistent with the location of the southern MSNA. This indicates that the D0 component is the major contributor to the diurnal eastward wave-1 drift of southern MSNA in the local time frame. [19] Although Figure 6b does successfully reconstruct the eastward drift of the southern MSNA by utilizing only the D0 component, the electron density of the southern MSNA is still close to an order of magnitude smaller than that shown from the combined results of Figure 2b. Since the SPW1 component can also produce zonal electron density variation in the local time frame, we reconstruct the structure of electron density at 300 km altitude using both the D0 and SPW1 components. Figure 7 shows the reconstruction results by the D0 and SPW1 components from Figure 3 on (a) DOY 167, 2007 and (b) DOY 350, As shown in Figure 7, the combination of D0 and SPW1 components causes the summer nighttime electron density enhancement feature of MSNA around the magnetic midlatitude regions in both hemispheres. In the northern hemisphere, the maximum value occurs around the middle Asia region at 2000 LT with a density around ele/cm 3, which is around 21% ( / ) of the maximum density in Figure 2a. However, the zonal mean in Figure 2 can serve to globally change electron densities everywhere; it is difficult to compare the electron densities reconstructed by D0 3843

9 and SPW1 components and those in Figure 2. After removing the zonal mean in Figure 2a, the reconstructed rate of Figure 7a is up to 48% ( / ). On the other hand, Figure 7b reveals that the electron density in the southern hemisphere reaches its maximum value around LT with a density around ele/cm 3, which is around 20% ( / ) of the maximum density in Figure 2b. The reconstructed rate is up to 35% ( / ) after removing the zonal mean. It is also noted that the occurrence time of maximum density in the northern hemisphere is earlier than that in the southern hemisphere. The result of asymmetric occurrence time in both hemispheres is consistent with satellite and groundbased observations indicating that the feature of MSNAs in the ionosphere is strongly controlled by the combined effect of both D0 and SPW1 components. 5. Conclusion [20] The present paper is the first study of the SAMI3 run with TIEGCM neutral winds including nonmigrating tidal components on the MSNA/WSA in the midlatitude ionosphere. The TIEGCM can provide the SAMI3 model with a more complete ensemble of MLT migrating and nonmigrating tides based on GSWM forcing at the model lower boundary. The simulation results successfully reproduce the diurnal variation of the southern MSNA structure and its feature of an eastward moving electron density enhancement. However, in the northern hemisphere, the SAMI3 simulation could only reproduce one MSNA structure compared with satellite observations. This difference might be caused by the reality of model neutral winds. An observationally based lower boundary condition (such as TIMED satellite) for the TIEGCM neutral wind can provide more accurate and realistic neutral wind variations in the lower atmosphere [Wu et al., 2012] and trigger the corresponding variation of ionospheric electron density. In future work, instead of the GSWM model, neutral winds from the observationally based lower boundary condition of TIEGCM will be used to reproduce and analyze the MSNA features in the northern hemisphere. [21] Tidal decomposition via a linear least squares algorithm is utilized to understand the influence of various tidal components on the variation of MSNAs. Results show that the D0 component of vertical neutral wind is produced in situ in the thermosphere and causes the diurnal eastward wave-1 drift of the southern MSNA in the local time frame. Furthermore, the SPW1 component can cause the diurnal variation of the nighttime electron density enhancement around the midlatitude ionosphere. As a result, the combined effect of D0 and SPW1 components can reconstruct around 20% of the overall amplitude of the southern MSNA and 35% amplitude after removal of the zonal mean. [22] Acknowledgments. This research is partially supported by the grant NSC M to the National Cheng Kung University from the National Science Council of Taiwan. L.C.C. was supported by Taiwan NSC grant NSC M MY2. [23] Robert Lysak thanks the reviewers for their assistance in evaluating this paper. References Angelats i Coll, M., and J. M. Forbes (2002), Nonlinear interactions in the upper atmosphere: The s = 1 and s = 3 nonmigrating semidiurnal tides, J. Geophys. Res., 107(A8), 1157, doi: /2001ja Bailey, G. J., R. Sellek, and N. Balan (1991), The effects of interhemispheric coupling on night-time enhancements in ionospheric total electron content during winter at solar minimum, Ann. Geophys., 9, Bellchambers, W. H., and W. R. Piggott (1958), Ionospheric measurements made at Halley Bay, Nature, 182, , doi: / a0. Burns, A. G., T. L. Killeen, W. Wang, and R. G. Roble (2004), The solarcycle-dependent response of the thermosphere to geomagnetic storms, J. Atmos. Sol. Terr. Phys., 66(1), 1 14, doi: /j.jastp Burns, A. G., Z. Zeng, W. Wang, J. Lei, S. C. Solomon, A. D. Richmond, T. L. Killen, and Y.-. H. Kuo (2008), Behavior of the F2 peak ionosphere over the South Pacific at dusk during quiet summer condition from COSMIC data, J. Geophys. Res., 113, A12305, doi: / 2008JA Burns, A. G., S. C. Solomon, W. Wang, A. D. Richmond, G. Jee, C. H. Lin, C. Rocken, and Y. H. Kuo (2011), The summer evening anomaly and conjugate effects, J. Geophys. Res., 116, A01311, doi: / 2010JA Chang, L. C., C. H. Lin, J. Y. Liu, N. Balan, J. Yue, and J. T. Lin (2013), Seasonal and local time variation of ionospheric migrating tides in FORMOSAT-3/COSMIC and TIE-GCM total electron content, J. Geophys. Res., doi: /jgra.50268, in press. Chen, C. H., J. D. Huba, A. Saito, C. H. Lin, and J. Y. Liu (2011), Theoretical study of the ionospheric Weddell Sea Anomaly using SAMI2, J. Geophys. Res., 116, A04305, doi: /2010ja Chen, C. H., A. Saito, C. H. Liu, and J. Y. Liu (2012), Long-term variations of the nighttime electron density enhancement during the ionospheric midlatitude summer, J. Geophys. Res., 117, A07313, doi: /2011ja de Larquier, S., J. M. Ruohoniemi, J. B. H. Baker, N. Ravindran Varrier, and M. Lester (2011), First observations of the midlatitude evening anomaly using Super Dual Auroral Radar Network (SuperDARN) radars, J. Geophys. Res., 116, A10321, doi: /2011ja Dudeney, J. R., and W. R. Piggott (1978), Antarctic ionospheric research, in Upper Atmosphere Research in Antarctica, Antarct., Res. Ser., vol. 29, edited by L. J. Lanzerotti and C. G. Park, pp , AGU, Washington, D. C. Evans, J. V. (1965), Cause of the midlatitude evening increase in fof2, J. Geophys. Res., 70(5), , doi: /jz070i005p Fang, T. W., J. Y. Liu, A. D. Richmond, P. Y. Chang, and C. H. Lin (2006), Seasonal variation of the global ionosphere, Eos Trans. AGU, 87(52), Fall Meet. Suppl., Abstract SA33B Forbes, J. M., and D. Wu (2006), Solar tides as revealed by measurements of mesosphere temperature by the MLS experiment on UARS, J. Atmos. Sci., 63(7), Forbes, J. M., S. Bruinsma, and F. G. Lemoine (2006), Solar rotation effects on the thermospheres of Mars and Earth, Science, 312(5778), , doi: /science Forbes, J. M., X. Zhang, S. Palo, J. Russell, C. J. Mertens, and M. Mlynczak (2008), Tidal variability in the ionospheric dynamo region, J. Geophys. Res., 113, A02310, doi: /2007ja Hagan, M. E. (1996), Comparative effects of migrating solar sources on tidal signatures in the middle and upper atmosphere, J. Geophys. Res., 101, 21,213 21,222. Hagan, M. E., and J. M. Forbes (2002), Migrating and nonmigrating diurnal tides in the middle and upper atmosphere excited by tropospheric latent heat release, J. Geophys. Res., 107(D24), 4754, doi: / 2001JD Hagan, M. E., and J. M. Forbes (2003), Migrating and nonmigrating semidiurnal tides in the upper atmosphere excited by tropospheric latent heat release, J. Geophys. Res., 108(A2), 1062, doi: /2002ja Hagan, M. E., and R. G. Roble (2001), Modeling the diurnal tidal variability with the National Center for Atmospheric Research thermosphereionosphere-mesosphere-electrodynamics general circulation model, J. Geophys. Res., 106, 24,869 24,882. Hagan, M. E., J. M. Forbes, and F. Vial (1995), On modeling migrating solar tides, Geophys. Res. Lett., 22, Hagan, M. E., M. D. Burrage, J. M. Forbes, J. Hackney, W. J. Randel, and X. Zhang (1999), GSWM-98: Results for migrating solar tides, J. Geophys. Res., 104, Hagan, M. E., A. Maute, and R. G. Roble (2009), Tropospheric tidal effects on the middle and upper atmosphere, J. Geophys. Res., 114, A01302, doi: /2008ja He, M., L. Liu, W. Wan, B. Ning, B. Zhao, J. Wen, X. Yue, and H. Le (2009), A study of the Weddell Sea Anomaly observed by FORMOSAT-3/ COSMIC, J. Geophys. Res., 114, A12309, doi: /2009ja He, M., L. Liu, W. Wan, J. Lei, and B. Zhao (2010), Longitudinal modulation of the O/N2 column density retrieved from TIMED/GUVI measurement, Geophys. Res. Lett., 37, L20108, doi: /2010gl Hedin, A. E. et al. (1996), Empirical wind model for the upper, middle and lower atmosphere, J. Atmos. Terr. Phys., 58, , doi: / (95)

10 Horvath, I. (2006), A total electron content space weather study of the nighttime Weddell Sea Anomaly of 1996/1997 southern summer with TOPEX/ Poseidon radar altimetry, J. Geophys. Res., 111, A12317, doi: / 2006JA Horvath, I., and E. A. Essex (2003), The Weddell Sea Anomaly observed with the Topex satellite data, J. Atmos. Sol. Terr. Phys., 65, , doi: /s (03)00083-x. Hsu, M. L., C. H. Lin, R. R. Hsu, J. Y. Liu, L. J. Paxton, H. T. Su, H. F. Tsai, P. K. Rajesh, and C. H. Chen (2011), The OI nm airglow observations of the midlatitude summer nighttime anomaly by TIMED/GUVI, J. Geophys. Res., 116, A07313, doi: /2010ja Huba, J. D., G. Joyce, and J. A. Fedder (2000), Sami2 is Another Model of the Ionosphere (SAMI2): A new low-latitude ionosphere model, J. Geophys. Res., 105, 23,035 23,053, doi: /2000ja Jee, G., A. G. Burns, Y.-H. Kim, and W. Wang (2009), Seasonal and solar activity variations of the Weddell Sea Anomaly observed in the TOPEX total electron content measurements, J. Geophys. Res., 114, A04307, doi: /2008ja Kelley, M. C. (1989), The Earth s Ionosphere, Academic, New York. Lin, C. H. (2005), Low-latitude ionosphere variations during magnetic disturbances, PhD dissertation, Natl. Cent. Univ., Chung-Li, Taiwan. Lin, C. H., J. Y. Liu, T. W. Fang, P. Y. Chang, H. F. Tsai, C. H. Chen, and C. C. Hsiao (2007), Motions of the equatorial ionization anomaly crests imaged by FORMOSAT-3/COSMIC, Geophys. Res. Lett., 34, L19101, doi: /2007gl Lin, C. H., J. Y. Liu, C. Z. Cheng, C. H. Chen, C. H. Liu, W. Wang, A. G. Burns, and J. Lei (2009), Three-dimensional ionospheric electron density structure of the Weddell Sea Anomaly, J. Geophys. Res., 114, A02312, doi: /2008ja Lin, C. H., C. H. Liu, J. Y. Liu, C. H. Chen, A. G. Burns, and W. Wang (2010), Midlatitude summer nighttime anomaly of the ionospheric electron density observed by FORMOSAT-3/COSMIC, J. Geophys. Res., 115, A03308, doi: /2009ja Liu, H.-L., T. Li, C.-Y. She, J. Oberheide, Q. Wu, M. E. Hagan, J. Xu, R. G. Roble, M. G. Mlynczak, and J. M. Russell III (2007), Comparative study of short-term diurnal tidal variability, J. Geophys. Res., 112, D18108, doi: /2007jd Liu, H., S. V. Thampi, and M. Yamamoto (2010), Phase reversal of the diurnal cycle in the midlatitude ionosphere, J. Geophys. Res., 115, A01305, doi: /2009ja Nozawa, S., and A. Brekke (1995), Studies of the E region neutral wind in the disturbed auroral ionosphere, J. Geophys. Res., 100, , doi: /95ja Park, C. G. (1971), Westward electric fields as the cause of nighttime enhancements in electron concentrations in midlatitude F region, J. Geophys. Res., 76, , doi: /ja076i019p Penndorf, R. (1965), The average ionospheric conditions over the Antarctic, in Geomagnetism and Aeronomy, Antarct. Res. Ser.,, vol. 4, edited by A. H. Waynick, pp. 1 45, AGU, Washington, D. C. Picone, J. M., A. E. Hedin, D. P. Drob, and A. C. Aikin (2002), NRLMSISE- 00 empirical model of the atmosphere: Statistical comparisons and scientific issues, J. Geophys. Res., 107(A12), 1468, doi: /2002ja Ren, Z., W. Wan, L. Liu, H. Le, and M. He (2012), Simulated midlatitude summer nighttime anomaly in realistic geomagnetic fields, J. Geophys. Res., 117, A03323, doi: /2011ja Richmond, A. D., E. C. Ridley, and R. G. Roble (1992), A thermosphere/ ionosphere general circulation model with coupled electrodynamics, Geophys. Res. Lett., 6, Scherliess, L., and B. G. Fejer (1999), Radar and satellite global equatorial F region vertical drift model, J. Geophys. Res., 104, , doi: /1999ja Sojka, J. J., W. J. Raitt, R. W. Schunk, and J. L. Parish (1985), Diurnal variations of the dayside, ionospheric, mid-latitude trough in the southern hemisphere at 800km: Model and measurement comparison, Planet. Space Sci., 33, , doi: / (85) Su, Y. Z., G. J. Bailey, and N. Balan (1994), Night-time enhancements in TEC at equatorial anomaly latitudes, J. Atmos. Terr. Phys., 56, , doi: / (94) Thampi, S. V., C. Lin, H. Liu, and M. Yamamoto (2009), First tomographic observations of the midlatitude summer nighttime anomaly over Japan, J. Geophys. Res., 114, A10318, doi: /2009ja Thampi, S. V., N. Balan, C. Lin, H. Liu, and M. Yamamoto (2011), Mid-latitude summer nighttime anomaly (MSNA) Observations and model simulations, Ann. Geophys., 29, , doi: /angeo Wu, D. L., P. B. Hays, and W. R. Skinner (1995), A least squares method for spectral analysis of space-time series, J. Atmos. Sci., 52, Wu, Q., D. A. Ortland, B. Foster, and R. G. Roble (2012), Simulation of nonmigrating tide influences on the thermosphere and ionosphere with a TIMED data driven TIEGCM, J. Atmos. Sol. Terr. Phys., 90-91, 61 67, doi: /j.jastp Zhang, X., J. M. Forbes, and M. E. Hagan (2010a), Longitudinal variation of tides in the MLT region: 1. Tides driven by tropospheric net radiative heating, J. Geophys. Res., 115, A06316, doi: /2009ja Zhang, X., J. M. Forbes, and M. E. Hagan (2010b), Longitudinal variation of tides in the MLT region: 2. Relative effects of solar radiative and latent heating, J. Geophys. Res., 115, A06317, doi: /2009ja