Diurnal, seasonal and solar activity pattern of ionospheric variability from Irkutsk Digisonde data

Available online at www.sciencedirect.com

ScienceDirect Advances in Space Research 55 (2015) 2041–2047 www.elsevier.com/locate/asr

Diurnal, seasonal and solar activity pattern of ionospheric variability from Irkutsk Digisonde data K.G. Ratovsky ⇑, A.V. Medvedev, M.V. Tolstikov Institute of Solar-Terrestrial Physics of Siberian Branch of Russian Academy of Sciences, Irkutsk 664033, Russia Received 19 March 2014; received in revised form 16 July 2014; accepted 1 August 2014 Available online 9 August 2014

Abstract From 2003–2012 dataset of the Irkutsk DPS-4 Digisonde (52.3°N, 104.3°E), we obtained the diurnal, seasonal and solar activity pattern of the F2 peak density (NmF2) variability for both high- and low-frequency parts of the variability. The high-frequency part (periods from 0.5 to 6 h) is mainly caused by traveling ionospheric disturbances associated with internal gravity waves. The low-frequency part with periods more than 6 h is related to long-term variability. The year-to-year (solar cycle) variations of the total daytime variability were used to estimate the geomagnetic and meteorological activity contributions into the NmF2 variability. An explanation of the obtained variability pattern is proposed in terms of the geomagnetic, meteorological and wave activity contributions, and the sensitivity of the ionospheric electron density to the geomagnetic and meteorological impacts. Ó 2014 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Ionospheric variability; Diurnal–seasonal–solar activity behavior; Geomagnetic and meteorological activity

1. Introduction Ionospheric F2 layer variability has been studied in many papers (e.g. Forbes et al., 2000; Rishbeth and Mendillo, 2001; Araujo-Pradere et al., 2005; Altadill, 2007; Zhang and Holt, 2008; Deminov et al., 2013; and references therein). In the above-cited papers the variability is sample standard (or mean root square) deviation of an ionospheric characteristic from a reference level. Most often, the variability of the critical frequency (foF2) or the F2 peak density (NmF2) is studied (Forbes et al., 2000; Rishbeth and Mendillo, 2001; Araujo-Pradere et al., 2005; Deminov et al., 2013). Deminov et al. (2013) additionally considered the peak height (hmF2) variability. Zhang and Holt (2008) and Altadill (2007) investigated the variability of the electron density at different heights. ⇑ Corresponding author. Tel.: +7 3952 564539; fax: +7 3952 425557.

E-mail addresses: [email protected] (K.G. Ratovsky), medvedev@ iszf.irk.ru (A.V. Medvedev), [email protected] (M.V. Tolstikov). http://dx.doi.org/10.1016/j.asr.2014.08.001 0273-1177/Ó 2014 COSPAR. Published by Elsevier Ltd. All rights reserved.

The reference level was selected to be monthly mean values (Forbes et al., 2000; Rishbeth and Mendillo, 2001; AraujoPradere et al., 2005) or 15-day medians (Deminov et al., 2013) or the values given by the local empirical models (Altadill, 2007; Zhang and Holt, 2008). The monthly mean and median values were calculated using either the entire data set (Rishbeth and Mendillo, 2001), or only the data corresponding to geomagnetically quiet conditions (Forbes et al., 2000; Araujo-Pradere et al., 2005; Deminov et al., 2013). The main topics of this paper are the following: (1) to obtain the diurnal, seasonal and solar activity pattern of the NmF2 variability at Irkutsk with studying both high- and low-frequency parts of the variability; (2) to estimate solar/geomagnetic and meteorological activity contribution using the year-to-year (solar cycle) variations of the variability; (3) to compare the obtained variability statistic with the known results;

2042

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047

(4) to discuss the physical reasons of the obtained variability pattern. For our analysis we used 2003–2012 dataset of the DPS4 Digisonde (Reinisch et al., 1997) installed at Irkutsk, Russia (52.3°N, 104.3°E) in November, 2002. All ionogram data acquired since the beginning of Digisonde measurements have been manually scaled using an interactive ionogram scaling software, SAO Explorer (Reinisch et al., 2004; Khmyrov et al., 2008). As an ionospheric characteristic we selected the F2 peak density NmF2 calculated from the ionogram critical frequency. We used the indices of geomagnetic activity (Ap) and solar activity (F10.7) whose daily values are available from WDC-A in Boulder, Colorado (ftp://ftp.ngdc.noaa.gov/STP/GEOMAGNETIC_ DATA/INDICES/KP_AP). 2. Technique of variability calculation A disturbance of an ionospheric characteristic is considered as deviation of the observed value from a regular behavior. The regular behavior associated with climatological specifics of the diurnal, seasonal, and long-term solar activity variations can be represented by 27-day running medians. The 27-day running median of NmF2 for given local time (LT), day of year (D) and year (Y) is the median value for the set {NmF2(LT,D-13,Y), . . . ,NmF2(LT,D + 13,Y}. Thus, the NmF2 disturbance is the difference between observed NmF2OBS and the 27-day median NmF2MED value: DNmF2 ¼ NmF2OBS  NmF2MED ;

ð1Þ

DR NmF2ð%Þ ¼ DNmF2=NmF2MED  100%;

ð2Þ

where DNmF2 and DRNmF2 are the absolute and relative disturbances, respectively. The next step is separation of low- and high frequency components of the NmF2 disturbances. For the separation we used low-pass filtering with a cutoff period of 6 h. The high-frequency part with periods less than 6 h (denoted further as DNmF2TID, DRNmF2TID) is related to traveling ionospheric disturbances (TID) caused by internal gravity waves (IGW). The cutoff period was selected basing on Medvedev et al. (2013) statistics that showed existence of TIDs with periods up to 5 h. The low-frequency part with periods more than 6 h (denoted further as DNmF2LONG-TERM, DRNmF2LONG-TERM) is related to day-to-day variability and tidal variations with diurnal, semidiurnal, and terdiurnal components. Fig. 1 demonstrates the steps of the NmF2 disturbance calculation. The variability is considered as the root mean square of NmF2 disturbances: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rNmF2 ¼ hDNmF22 i; rR NmF2 ¼ hDR NmF22 i; ð3Þ where brackets denote some averaging. For different tasks we used different types of averaging. Annual averaging was used for studying year-to-year changes in the variability

(solar cycle variations). To study the difference between the day- and nighttime variability we made separate averaging for the day- and nighttime using ground terminator as a day–night boundary. To obtain the diurnal–seasonal pattern of the variability we performed averaging over years for each local time and day of year. 3. Diurnal, seasonal and solar activity pattern of the NmF2 variability 3.1. Year-to-year variations Fig. 2 shows the year-to-year variations of the absolute NmF2 variability (total, long-term and TID) both for the day- and nighttime. It is seen that the absolute variability depends strongly on the background (regular) NmF2. Main features of the absolute variability are similar to the regular NmF2 behavior: the daytime values exceed the nighttime ones, and all kinds of absolute variability show a clear increase with solar activity. Further we consider only the relative variability. The main reason is that the most well-known studies of the ionospheric variability (Forbes et al., 2000; Rishbeth and Mendillo, 2001; Araujo-Pradere et al., 2005; Altadill, 2007; Zhang and Holt, 2008; Deminov et al., 2013) were performed in terms of the percentage variability. In addition, Rishbeth and Mendillo (2001) and Forbes et al. (2000) estimated the contribution of different sources in terms of the percentage response of NmF2 to different kinds of activity. Fig. 3 shows the year-to-year variations of the relative NmF2 variability (total, long-term and TID) both for the day- and nighttime. Contrary to the absolute variability, the nighttime total and long-term relative variability is larger than the daytime one, and the TID relative variability shows close day- and nighttime values. The long-term relative variability is noticeably larger than the TID one and provides the main contribution to the total relative variability (year-to-year variations of the total and long-term relative variabilities are close together). From all kinds of relative variability, only the total and long-term daytime variability show a clear increase with the geomagnetic/solar activity. Thus, the geomagnetic/solar activity has no or weak influence on total/long-term nighttime relative variability and TID relative variability both for the day- and nighttime. This unexpected result will be discussed in Section 4. Using the year-to-year variations of the total daytime relative variability (Fig. 3) we may estimate solar/geomagnetic and meteorological activity contributions to the NmF2 variability. The geomagnetic contribution is the variability associated with effects of geomagnetic storms and geomagnetically disturbed conditions (e.g., Buonsanto, 1999; Prolss, 1993; Mikhailov, 2000; and references therein); whereas the meteorological contribution is the variability caused by processes in the lower atmosphere: planetary waves, tides, and internal gravity waves (Lastovicka, 2006).

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047

ΔNmF2, ΔNmF2LONG-TERM (105⋅cm-3) 4

2

2043

ΔNmF2TID (105⋅cm-3)

1 2 0

NmF2Obs , NmF2Med (105⋅cm-3)

0

9

-1

-2

6

0

6

12

18

24

-2

0

6

Universal Time ΔRNmF2, ΔRNmF2LONG-TERM (%)

3

0

0

6

12

18

24

Universal Time

90

45

60

30

30

15

0

0

-30

-15

-60

-30

-90

0

6

12

18

12

18

24

Universal Time

24

-45

ΔRNmF2TID (%)

0

6

Universal Time

12

18

24

Universal Time

Fig. 1. Steps of NmF2 disturbance calculation.

2.0

σNmF2 (105⋅cm-3) Day Night

F10.7

- - - - - - σNmF2LONG-TERM

σNmF2TID (105⋅cm-3) Day Night

150 0.6

1.6

130 0.5

1.2

110

F10.7

150 130

0.4

110

0.3 90

90

0.8

0.2

0.0

70

70

0.4

0.1 0.0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Years

Years

Fig. 2. Year-to-year variations of daytime (gray bars) and nighttime (black bars) absolute NmF2 variability with variations of annual running mean F10.7 index (black line with circles). Total (solid) and long-term (dashed) variability is on the left, and TID variability is on the right.

30

σRNmF2 (%) Day Night

Ap

σRNmF2TID (%) Day Night

25 10

26

20

22

15

Ap

25 20

8

15 6 10

10

18

4

10 6

5

5

14

- - - - - - σRNmF2LONG-TERM 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Years

2 0

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Years

Fig. 3. Year-to-year variations of daytime (gray bars) and nighttime (black bars) relative NmF2 variability with variations of annual running mean Ap index (black line with circles). Total (solid) and long-term (dashed) variability is on the left, and TID variability is on the right.

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047

3.2. Diurnal–seasonal pattern Fig. 4 shows the diurnal–seasonal pattern of the relative NmF2 variability (total and TID) resulting from averaging DRNmF2 and DRNmF2TID over 2003–2012 years for each local time and day of year. Despite a noticeable scatter of

Total var iability σRNmF2(% ) 21

Local time

Year of 2009 is characterized by extremely low level of the geomagnetic and solar activity (annual mean Ap = 4 nT, annual mean F10.7 = 71 s.f.u.). Neglecting the solar/geomagnetic contribution in this year we may consider that the variability is caused only by the meteorological impact. After that, we estimate that the meteorological contribution to the NmF2 daytime variability is about 14%. This estimation closely agrees with those obtained by different methods (15% from Rishbeth and Mendillo, 2001; and 13% from Deminov et al., 2013; and AraujoPradere et al., 2005). A detailed comparison with these methods is presented in Table 1. Table 1 shows that our estimation of the meteorological contribution exceeds by 1% the estimations based on using only geomagnetically quiet data. This exceeding may be associated with the socalled CIR-related minor-to-moderate magnetic storms observed under extremely low solar activity conditions (Buresova et al., 2014). Our estimation is noticeably less than 20% obtained by Forbes et al. (2000). The reason is that they did not separate the day and night variability and obtained a higher estimation due to the higher nighttime variability. If we average the day- and nighttime variabilities of 2009, we get a close estimation of 20%. The geomagnetic contribution may be estimated using the linear regression of the NmF2 daytime variability on the annual mean of Ap-index. As a result, we obtain that the geomagnetic contribution (percentage response of NmF2 to geomagnetic activity) is 0.8% per Ap unit, which is close but slightly less than the Rishbeth and Mendillo (2001) estimation (1% per Ap unit). Note, that the estimations were made by significantly different methods: we used the year-to-year variations of rRNmF2 at Irkutsk and Ap-index, whereas Rishbeth and Mendillo (2001) obtained their estimation by comparison of the semiannual amplitudes of rRNmF2 at Slough and Ap. Taking into account the different geomagnetic latitudes of the stations (Irkutsk, 42°N GLAT and Slough, 48°N GLAT), the agreement may be considered as satisfactory.

35 30

18 15

25

12 20

9 6

15

3 0 1

10

31

61

91 121 151 181 211 241 271 301 331 361

Days of year TI D var iability σRNmF2TID (% ) 21

Local time

2044

16 14

18 15

12

12

10

9

8

6

6

3 0 1

31

61

91 121 151 181 211 241 271 301 331 361

4

Days of year Fig. 4. Diurnal–seasonal pattern of relative NmF2 variability (total on the left and TID on the right) with superimposed sunrise–sunset lines (white dashed).

values due to averaging over a small number of years, some features are clearly seen. In the pattern of the total variability, the high and low values of DRNmF2 are separated approximately by sunrise–sunset lines superimposed on the pattern. This feature is well pronounced in winter and less pronounced in summer. On the whole, these diurnal– seasonal properties of the NmF2 variability agree with those reported by Araujo-Pradere et al. (2005). In the pattern of the TID variability, the highest values of DRNmF2TID are seen between sunset and 2 h after sunset, and between sunrise and 2 h before sunrise for a wintertime period (approximately November–February). This feature will be discussed in Section 4. For a detailed consideration of the seasonal behavior we performed 27-day smoothing of the data shown in Fig. 4 and averaging over day- and nighttime hours using ground terminator as a day–night boundary. The results are shown in Fig. 5. The nighttime total variability shows clear seasonal behavior with maximum in winter (31%), minimum in summer (19%), and intermediate values at equinoxes (24%). The winter–summer difference (r(Dec22)  r

Table 1 Estimations of meteorological contribution to NmF2 daytime variability. Reference

Estimation (%)

Data set

Method

This paper Rishbeth and Mendillo (2001) Deminov et al. (2013) Araujo-Pradere et al. (2005)

13.6 15.0

Irkutsk ionosonde, 2009 Slough ionosonde, 1973, 1979

12.7 13.3

Irkutsk ionosonde, 2007–2010 Ionosondes at 40–60° geomagnetic latitudes, 1981– 1988

Using Using Ap Using Using

only 2009 data semiannual amplitudes of rNmF2 and only geomagnetically quiet data only geomagnetically quiet data

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047

(Jun22)) is 10% and the equinox–solstice difference (r(Mar22) + r(Sep22)  r(Dec22)  r(Jun22)) is 1%. Such seasonal behavior of nighttime variability agrees qualitatively with the behaviors of mid-latitude stations reported by Rishbeth and Mendillo (2001), but there are some quantitative distinctions: the winter–summer difference is closer to Moscow and Slough stations that are closer to Irkutsk in geographic latitude, and the equinox– solstice difference is closer to Wakkanai and Tashkent stations that are closer to Irkutsk in geomagnetic latitude. The daytime total variability shows a multi-peak behavior with maxima in December, April–May and September– October (19%) and minima in July, February and November (16%). The winter–summer difference is 3% and the equinox–solstice difference is 2%. The daytime winter–summer difference is noticeably less than the nighttime one, and this looks like a common feature of mid-latitude stations reported by Rishbeth and Mendillo (2001). The positive daytime winter–summer difference observed at Irkutsk is not a common feature for mid-latitude stations, this difference may be both positive and negative (or zero) according to Rishbeth and Mendillo (2001). The daytime and nighttime TID variabilities show clear seasonal behavior with maximum in winter (11–12%), minimum in summer (6%), and intermediate values at equinoxes (7%). An explanation of such behavior will be discussed in Section 4. 4. Discussion 4.1. Total variability The nighttime total relative variability at Irkutsk is larger than the daytime one, both for annual mean and any season with the largest difference in winter. This feature agrees completely with the results reported by Rishbeth and Mendillo (2001), Deminov et al. (2013), AraujoPradere et al. (2005) and Altadill (2007). Such night–day difference may be explained by the photochemical control effect on the variability (Rishbeth and Mendillo, 2001; Araujo-Pradere et al., 2005). At the nighttime the photochemical control is weaker, and the ionospheric electron density is more sensitive to the geomagnetic and 35

σRNmF2 (%) Day Night

meteorological activity (Rishbeth and Mendillo, 2001). The photochemical control effect may also explain the winter–summer difference (see below), and an increase of the daytime electron density variability with height (Zhang and Holt, 2008; Altadill, 2007). The nighttime total relative variability at Irkutsk does not show a clear increase with the geomagnetic/solar activity. Below we give some examples of unexpected solar cycle variations of the nighttime variability. Deminov et al. (2013) reported that at Irkutsk the nighttime long-term NmF2 variability of the quiet ionosphere (2007–2010) is more than that under medium solar activity (1958–1992) and it is comparable with that at Slough under medium solar activity (1958–1992). Altadill (2007) concluded that the electron density variability at the base of the F-region increases with increasing sunspot activity from midnight to sunrise, and has no significant solar cycle dependence from sunset to midnight. Araujo-Pradere et al. (2005) revealed that the nighttime variability tends to increase with geomagnetic activity except for the high latitudes showing a clear reduction in variability. Araujo-Pradere et al. (2005) found this result surprising and assumed that this reduction is due to the increase of the chemical control (i.e. increase in the ion recombination rate caused by changes in chemical composition)), resulting in lower variability. Mikhailov et al. (2000) found that the post-midnight winter NmF2 enhancement amplitude decreased as solar activity increased and explained this as a result of an increase in the ion recombination rate (i.e. chemical control). Another manifestation of the chemical control effect is that the occurrence of the very strong foF2 nighttime enhancements decreases as solar activity grows (Pirog et al., 2011). Returning to an unclear geomagnetic activity pattern of the nighttime variability at Irkutsk, we may suppose that this pattern is explained by a competition between an enhancements caused by geomagnetic storm effects and a reduction associated with chemical control increase. The nighttime total relative variability at Irkutsk shows clear seasonal behavior with maximum in winter, minimum in summer, and intermediate values at equinoxes, which agrees qualitatively with the behavior of mid-latitude stations reported by Rishbeth and Mendillo (2001). The σ NmF2TID (%) Day Night 14 R

30

12

25

10

20

8

15

6

10

2045

4 1 31 61 91 121 151 181 211 241 271 301 331 361

1 31 61 91 121 151 181 211 241 271 301 331 361

Days of year

Days of year

Fig. 5. Seasonal behavior of daytime (gray) and nighttime (black) relative NmF2 variability (total on the left and TID on the right).

2046

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047

daytime variability at Irkutsk shows a multi-peak behavior with maxima in December, April–May and September– October and minima in July, February and November, and we cannot claim that it is a typical mid-latitude behavior, there are cases of both positive and negative difference both for winter–summer and equinox–solstice according to data presented by Rishbeth and Mendillo (2001) and Altadill (2007). We can consider the following factors affecting the seasonal behavior: (1) the photochemical control seasonal (Araujo-Pradere et al., 2005); (2) the geomagnetic activity seasonal (Rishbeth and Mendillo, 2001); and (3) the meteorological activity seasonal (Lastovicka, 2006; Altadill, 2007).

variations variations variations

The factor (1) enhances the winter–summer difference, whereas the factor (2) increases the amplitude of semiannual variations (the equinox–solstice difference). Only the factor (3) can explain the negative winter–summer difference. Most likely, the factor (1) dominates in the nighttime mid-latitude ionosphere, whereas the dominance of a factor at the daytime depends on the regional specifics. 4.2. TID variability The TID relative variability shows close day- and nighttime values, the highest values of DRNmF2TID are seen between sunset and 2 h after sunset and between sunrise and 2 h before sunrise for a wintertime period (approximately November–February). Contrary to the total (and long-term) variability, the night–day difference is small, and we may suppose, that the photochemical control has little effect on the TID variability. The higher values of DRNmF2TID in the evening and morning hours may be associated with appearance of terminator induced IGW Both the daytime and nighttime TID relative variabilities show clear seasonal behavior with maximum in winter, minimum in summer, and intermediate values at equinoxes. Such behavior agrees closely with the IGW activity at stratospheric and mesospheric heights (Alexander et al., 2010), the seasonal distribution of IGW events in the lower ionosphere (Oleynikov et al., 2007), and the seasonal distribution of wave-like disturbances at F2-layer heights (Medvedev et al., 2013). Thus, the most likely explanation of the TID variability seasonal behavior is the seasonal variations of the IGW activity. The TID relative variability does not show a clear increase with the geomagnetic/solar activity. The result is surprising because we expected an enhancement of the IGW activity associated with the IGW generation in the auroral ionosphere during the periods of geomagnetic storms (Hocke and Schlegel, 1996). Medvedev et al. (2013) showed that wave activity level (number of wavelike disturbances) tends to increase with both solar and geomagnetic activity for the considered 2004–2009 period.

Thus, the geomagnetic/solar activity behavior of the TID variability (root mean square of relative NmF2 disturbances in the IGW periods range) differs from that of the number of wave-like disturbances in the same periods range. The distinction is not clear and requires further research. 5. Conclusion From 2003–2012 dataset of the Irkutsk DPS-4 Digisonde (52.3°N, 104.3°E), we obtained the diurnal, seasonal and solar activity pattern of the NmF2 variability for both high- and low-frequency parts of the variability. The year-to-year (solar cycle) variations of the total daytime variability were used to estimate the geomagnetic and meteorological activity contributions into the NmF2 variability. We found our estimates to be close to the known results obtained by independent methods. The comparison with the NmF2 variability at other mid-latitude stations showed that the positive night–day difference and the positive nighttime winter–summer difference look like common features of the variability in the mid-latitude ionosphere, whereas the daytime seasonal behavior depends on the regional specifics. We proposed an explanation of the obtained diurnal, seasonal and solar activity pattern of the total variability in terms of the geomagnetic and meteorological effects, and the sensitivity of the ionospheric electron density to the geomagnetic and meteorological impacts. It should be noted that we proposed only a qualitative explanation that should be tested by model calculations. The seasonal behavior of the TID variability agrees closely with that of the IGW activity at different heights and thereby has a clear explanation. An unclear geomagnetic activity pattern of the TID variability requires further research. The obtained diurnal, seasonal and solar activity pattern of the NmF2 variability may be considered as a base for building a local empirical model of the electron density variability. Such model can serve as a complementation to the local model of regular variations (Ratovsky and Oinats, 2011) or to the global ionospheric model, e.g. the International Reference Ionosphere (Bilitza, 2001; Bilitza and Reinisch, 2008)..2 Acknowledgment The reported study was supported by RFBR, research project No. 14-05-00578. References Alexander, M.J., Geller, M., McLandress, C., Polavarapu, S., Preusse, P., Sassi, F., Sato, K., Eckermann, S., Ern, M., Hertzog, A., Kawatani, Y., Pulido, M., Shaw, T.A., Sigmond, M., Vincent, R., Watanabe, S., 2010. Recent developments in gravity-wave effects in climate models and the global distribution of gravity-wave momentum flux from

K.G. Ratovsky et al. / Advances in Space Research 55 (2015) 2041–2047 observations and models. Q.J.R. Meteorol. Soc. 136 (650), 1103–1124. http://dx.doi.org/10.1002/qj.637. Altadill, D., 2007. Time/altitude electron density variability above Ebro, Spain. Adv. Space Res. 39 (5), 962–969. Araujo-Pradere, E.A., Fuller-Rowell, T.J., Codrescu, M.V., Bilitza, D., 2005. Characteristics of the ionospheric variability as a function of season latitude local time and geomagnetic activity. Radio Sci. 40, RS5009. http://dx.doi.org/10.1029/2004RS003179. Bilitza, D., 2001. International reference ionosphere 2000. Radio Sci. 36 (2), 261–275. Bilitza, D., Reinisch, B., 2008. International reference ionosphere 2007: improvements and new parameters. J. Adv. Space Res. 42 (4), 599–609. Buonsanto, M.J., 1999. Ionospheric storms: a review. Space Sci. Rev. 88 (3–4), 563–601. Buresova, D., Lastovicka, J., Hejda, P., Bochnicek, J., 2014. Ionospheric disturbances under low solar activity conditions. Adv. Space Res. 54 (2), 185–196. Deminov, M.G., Deminova, G.F., Zherebtsov, G.A., Polekh, N.M., 2013. Statistical properties of variability of the quiet ionosphere F2-layer maximum parameters over Irkutsk under low solar activity. Adv. Space Res. 51 (5), 702–711. Forbes, J.M., Palo, S.E., Zhang, X., 2000. Variability of the ionosphere. J. Atmos. Solar-Terr. Phys. 62, 685–693. Hocke, K., Schlegel, K., 1996. A review of atmospheric gravity waves and travelling ionospheric disturbances: 1982–1995. Ann. Geophys. 14 (9), 917–940. Khmyrov, G.M., Galkin, I.A., Kozlov, A.V., Reinisch, B.W., McElroy, J., Dozois, C., 2008. Exploring digisonde ionogram data with SAO-X and DIDBase. Radio sounding and plasma physics. AIP Conf. Proc. 974, 175–185. Lastovicka, J., 2006. Forcing of the ionosphere by waves from below. J. Atmos. Solar-Terr. Phys. 68 (3–5), 479–497.

2047

Medvedev, A.V., Ratovsky, K.G., Tolstikov, M.V., Alsatkin, S.S., Scherbakov, A.A., 2013. Studying of the spatial–temporal structure of wavelike ionospheric disturbances on the base of Irkutsk incoherent scatter radar and digisonde data. J. Atmos. Solar-Terr. Phys. 105, 350– 357. Mikhailov, A.V., 2000. Ionospheric F2-layer storms. Fis. Tierra. 12, 223– 262. Mikhailov, A.V., Fo¨rster, M., Leschinskaya, T.Y., 2000. On the mechanism of the post-midnight winter NmF2 enhancements: dependence on solar activity. Ann. Geophys. 18 (11), 1422–1434. Oleynikov, A.N., Sosnovchik, D.M., Kukush, V.D., Jacobi, Ch., Frohlich, K., 2007. Seasonal variation of space–time parameters of internal gravity waves at Kharkiv (49°300 N, 36°510 E). J. Atmos. Solar-Terr. Phys. 69 (17–18), 2257–2264. Pirog, O., Deminov, M., Deminova, G., Zherebtsov, G., Polekh, N., 2011. Peculiarities of the nighttime winter foF2 increase over Irkutsk. Adv. Space Res. 47 (6), 921–929. Prolss, G.W., 1993. On explaining the local time variation of ionospheric storm effects. Ann. Geophys. 11 (1), 1–9. Ratovsky, K.G., Oinats, A.V., 2011. Local empirical model of ionospheric plasma density derived from digisonde measurements at Irkutsk. Earth Planets Space 63 (4), 351–357. Reinisch, B.W., Haines, D.M., Bibl, K., Galkin, I., Huang, X., Kitrosser, D.F., Sales, G.S., Scali, J.L., 1997. Ionospheric sounding support of OTH radar. Radio Sci. 32 (4), 1681–1694. Reinisch, B.W., Galkin, I.A., Khmyrov, G., et al., 2004. Automated collection and dissemination of ionospheric data from the digisonde network. Adv. Radio Sci. 2, 241–247. Rishbeth, H., Mendillo, M., 2001. Patterns of F2-layer variability. J. Atmos. Solar-Terr. Phys. 63, 1661–1680. Zhang, S.-R., Holt, J.M., 2008. Ionospheric climatology and variability from long-term and multiple incoherent scatter radar observations: variability. Ann. Geophys. 26 (6), 1525–1537.