Digisonde data from three different longitude sectors from Jicamarca (12°S, 76.8°W, -2.5° declination angle) from 2001–2016, Ascension Island (7.9°S, 14.4°W, -15.09° declination angle) from 2000–2014, Kwajalein (8.71°N, 167.7°E, 7.5° declination angle) from 2004–2012, has been processed and analyzed to determine statistical studies of equatorial spread F, a diagnostic of irregular plasma structure in the ionosphere. A new method of spread F detection for low latitude region is used to determine solar and seasonal variation over these three sites. An algorithm has been developed to detect the foF2 and hmF2 values from an ionogram and this has been validated using manually scaled ionograms, as well as comparisons to the SAMI2 and IRI models.
Research Article
Statistical analysis of Low latitude Spread F at the American, Atlantic and Pacific sectors using digisonde observations
https://doi.org/10.21203/rs.3.rs-3232628/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Digisonde data
three longitudinal sectors
low latitude spread F
spread F statistical study
Plasma density irregularities in the ionosphere are often observed as a spread pattern in data from radio sounding techniques such as digisondes and are commonly referred to as Spread F. Spread F can be caused by a number of different instability mechanisms, which are controlled in part by the angle of the magnetic field relative to the ionospheric plasma layer. In both cases, the height of the F layer increases before the spread F occurrence [Fejer et al., 1999 and references therein]. Potential mechanisms that influence the growth of spread F include gravity waves [Kil, 2022; Candido et al., 2011; Bowman and Mortimer, 2002, Bowman 2001, Fejer et al., 1999] as well as larger-scale waves [ eg.Takahashi et al., 2021; Aa et al., 2023] and disturbances from geomagnetic storms [e.g., Carter et al., 2014; Cherniak and Zakharenkova, 2022].. The various irregularities or spread patterns may be associated with multiple types of plasma structures, such as plumes, patches, bubbles, and blobs [Aarons, 2001]. This study focuses on the low-latitude region, where Equatorial Spread F is typically caused by equatorial plasma bubbles (EPBs) that are often observed as spread F plumes in radar echoes. Equatorial spread F has been observed for nearly eighty years [Booker & Wells, 1938], but we are still seeking to understand the physical mechanisms for why it forms on a given night [Carter et al., 2014, Klenzing et al., 2023].
Equatorial spread F has been measured using various instruments such as satellites, ionosondes, scintillation measurements, incoherent scatter radars, rockets and GPS. Afolayan et al., [2019] compared digisonde data for different longitudes to study spread F variation for different solar activities. Gonzalez [2022] used ionosonde and GPS data to study spread F characteristics over Tucuman. Saito and Maruyama [2007] conducted a study of equatorial spread F using ionosondes in south-east Asia. Aarons [1993] did a study of equatorial region sites for 2 years of data. The equatorial electric field pre-reversal enhancement is an essential condition to the development of large-scale irregularities at the equatorial region [e.g,. Candido et al., 2011, Abadi et al. 2022].
An ionosonde transmits radio waves in a series of varying frequency signals in the ionosphere and receives them back thereby providing us with the knowledge of density variations present in the ionosphere. The waves are reflected back on Earth when the wave frequency matches the plasma frequency it encounters. The information from the reflected waves, such as time and the frequency is used to obtain plots known as ionograms. A smooth, skinny trace on an ionogram will represent a smooth ionosphere while a thick, and a ragged trace will represent disturbances or density irregularities in the ionosphere. This is referred to as the spread condition and is known as spread F when present in the F layer of the ionosphere.
Section 2 discusses the ionosonde database and the algorithms used to distinguish between spread and non-spread events. Section 3 presents the statistical results. Section 4 discusses these results for both solar cycles 23 and 24 and seasonal variation of ESF events. Section 5 summarizes the conclusions and suggests possible future paths of inquiry.
The data used in this study has been obtained from three different low latitude regions; Jicamarca (12°S, 76.8°, -2.5° declination angle), Ascension Island (7.9°S, 14.4°W, -15.09° declination angle) from 2000–2014, Kwajalein (8.71°N, 167.7°E, 7.5° declination angle) from 2004–2012, This paper uses a previously established pattern recognition algorithm [Bhaneja et al., 2009, 2018] with slight modifications to automatically detect spread F on ionograms and another algorithm [Bhaneja & Bullett, 2023] with modifications to determine the critical frequency of the F2 layer (foF2).
2.1 Spread F
Equatorial spread F is a nighttime phenomenon and thus the data for nighttime is processed for spread conditions. The measurements taken from the ionosonde include the reflected signal frequency and the time between the sending and receiving the signal. This time is used to calculate the virtual height by using the formula ct/2 where c is the speed of light as used for wave travelling in plasma. The plot showing these measurements is called an ionogram. Figure 1a shows an ionogram with a single trace indicating a quiet ionosphere. Figure 1b-d shows three different ionograms with multiple traces or thickness indicating a disturbed ionosphere. Note that there may be multiple traces visible at higher altitudes in the ionograms, which is due to double reflection of the pulse (from the ionosphere to the ground, back to the ionosphere and back to the receiver). These second hop traces contain no additional information, and these are ignored in the analysis conducted here.
Ionogram traces have a finite thickness due to the finite bandwidth of the ionosonde. This is an instrument effect. The thickness or spread indicating spread condition is an ionospheric effect. The equatorial spread F observed on the ionograms generated from Jicamarca data is of three different types: range, range and frequency spread F and spread F in the form of a big blob as observed in Fig. 1 (b), (c), (d) respectively. Equatorial plasma bubbles or EPBs cause spread F and the ionograms show an extended frequency and flat range spread echoes as shown in Fig. 1b [Candido et al., 2011, Sahai et al., 2000]. The spread signatures vary from early in the evening 1b with huge plumes and intense spread to 1d with spread reducing later in the evening to 1c where spread signatures are as observed for mid latitude regions later in the night. Range spread F refers to a condition in which there are multiple echoes at different ranges for each frequency. Frequency spreading is the case in which there are multiple echoes at different frequencies around the critical frequency for same height.
There are unusual F region traces as seen in the plots and can be classified into strong, weak and multi trace echo [Saito & Maruyama, 2007]. Plots 1 b and 1 d refer to the strong spread F where the F region trace is not visible. Afolayan et al., [2019] refers to spread in 1b as strong spread F. Weak spread F is where the F region peak is slightly visible as in plot c. These spread events are also defined as bottom-type spread F, bottomside spread F and topside spread F for the 3 different types of spread we see in ionograms [Hysell, 2000; Whalen, 2002; Chapagain et al., 2009]. Topside or radar plume goes all the way up and is elongated over F region like plot 1b. This is mostly observed for Jicamarca. High solar flux levels and geomagnetic activity are conducive for the formation of very high-altitude plumes. Plot 1 c and 1 d are also observed at Ascension Island.
Figure 2 shows Range-Time-Intensity (RTI) plot alongside ionograms for three equatorial spread F events. The RTI data has been obtained using JULIA data from Madrigal. All three days show radar plumes and corresponding intense spread signatures on the ionograms. The top plot is from October 29-30th night in reference to a rocket launch during midlatitude spread F event observed during a huge hurricane and a high kp observed few hours before midnight [Earle et al., 2010]. It is interesting that intense spread signatures are also observed at Jicamarca at the same time. The other two plots also show huge plumes and corresponding intense spread F signatures in the ionograms. Aol et al., [2020] drew similar comparisons observing plasma plumes coinciding with spread F signatures on ionograms between the JULIA radar data and the JRO ionosonde.
2.2 Data Analysis-methodology
This study includes data for Jicamarca (2001–2016), Ascension Island (2000–2014) and Kwajalein (2004–2012). Raw digisonde data are filtered and processed to generate ionograms for statistical analysis. First, the raw data are converted to a text format using the SAO explorer by doing Bulk Processing of ionograms for each day which gives the amplitude echoes of the ionograms. The amplitude echoes are then processed using a pattern recognition algorithm for the ordinary mode (O-mode) data because the O-mode does not require knowledge of the magnetic field, therefore it is trivial to incorporate this data into the pattern recognition algorithm.
Figure 1a shows a quiet ionogram where the foF2 can be determined easily. Figure 1b, c, d shows the spread F identified by two boxes. The heavy spread events like in plots1b & d, a clear trace is not present and so foF2 cannot be measured accurately for such ionograms. Edge detection is done on both the boxes (solid line boxes) to determine their location; this is done to find the bottom side of box 1 and right side of box 2.
The box 1 (dotted lines) boundary is between 200–500 km and 1.5-4 MHz. The bottom edge for the box is determined by counting the pixels for each altitude starting at 200 km and when the pixel count exceeds 6, the corresponding altitude is chosen as the bottom edge of box 1. The height of this box is fixed at 100 km. The limits on the box have been chosen to cover the spread F at night-time while being careful not to include sporadic E and second hop traces.
The height and width of box 2 (dotted lines) is changed for varying ionospheric conditions for the equatorial region as visible on the ionograms. Box 2 have been conditioned for different times of the day, seasons and years; dawn, day, dusk and night; winter and summer season; solar min and solar max years. This has been done to obtain more accuracy for the spread F, foF2 and height of the critical frequency (hmF2) values. The box 2 boundary varies between 400–800 km and 2–15 MHz for dawn, day and dusk times and 300–500 km and 1.2-8 MHz for night times. The right edge for box 2 is determined by counting the pixels for each frequency starting at 6 MHz, and when the pixel count exceeds 6, the corresponding frequency is chosen as the right edge of box 2. Bhaneja et al., [2018], gives the detailed methodology for box 2. Additional second box 2 was introduced in these plots to obtain a more accurate pixel count for frequency spread F and was taken from the left edge of the box 2 with 1.5 MHz width and same height as the boundary box. The average pixel count of both these boxes was taken as the total pixel count for determining frequency spread F. It was necessary as the spread types vary very differently for low latitude regions. The detailed methodology for foF2 will be explained in a future paper.
The total number of pixel count in the boxes is used to determine the spread condition by comparing it to a set threshold value. These pixel counts are shown in the plots for range spread (RS) and frequency spread (FS) for the two boxes respectively. The pixel counts of 6 for the edges and the thresholds were set by randomly choosing ionograms and checking their threshold counts. Spread events have been quantified, and type of spread is distinguished.
As spread F is a night-time phenomenon, only the data between 7 PM-7 AM local time is considered. The digisonde produces ionograms every 15 minutes, 24 hours a day. Thus, for our study, around 48 ionograms/night are analyzed. A spread event is counted as continuous spread detected across multiple ionograms without interruption of more than thirty minutes (or no more than two ionograms without spread). If more than one event is observed for a night, the longer event is saved and if two equal events are observed the first one is saved for further analysis. On determination of spread F, the onset time, duration and type of spread F event are recorded and archived for each night. This process is done for the entire set of data. Using this database, statistical analysis is performed to determine seasonal and solar cycle patterns for equatorial spread F. A prototype of a monthly summary plot for July 2011 is shown in Fig. 2.
Once the daily and monthly plots for spread F were made, the methodology was used towards the entire data set. Figure 4 validates the foF2 detection algorithm with SAMI2 and IRI models. Figures 5–6 shows the solar variation and Figs. 7–8 shows the seasonal variation of equatorial spread F. Figures 9 & 10 shows the pre-sunset hmF2 and foF2 vs. spread duration. Figure 11 shows the solar activity variation with spread duration.
3.1 foF2 accuracy and comparison with SAMI2 and IRI model
Figure 4 shows foF2 values for an entire day (0–23) UT at the three sites for various years including both solar minimum and solar maximum years. The plot shows the algorithm developed (P_code) values in comparison with foF2 values as calculated by the International Reference Ionosphere (IRI) model [Bilitza et al., 2014] and the SAMI2 model [Huba et al., 2000, Klenzing et al., 2022]. IRI provides an empirical model of the ionosphere, similar to a monthly average. SAMI2 provides a two-dimensional physical simulation of the ionosphere driven by empirical models of the neutral atmosphere (MSIS, Picone et al., 2002), neutral winds (HWM-14, Drob et al., 2015), and ExB electric fields (Scherliess and Fejer, 1999).
3.2 Solar cycle variation
Figure 5 shows the number of spread days for each year of available data between 2000–2016 for Jicamarca, Ascension Island and Kwajalein. The top panel shows the solar flux for all the years covering both solar maximum during 2000–2003 and 2011–2015 and solar minimum during 2004–2010 and 2016. The bottom three panels show the three sites from west to east in longitude. The left axis shows the number of spread days. The right axis shows the percentage of available data. The total number of longest spread F hours for each month is summed over a given year and then divided by the number of months to obtain average for each year. Black and white bars represent range and frequency spread F and the x symbol represents the percentage of data available.
Figure 6 shows the number of spread days for each year of available data between 2000–2016 for Jicamarca, Ascension Island and Kwajalein. The top panel shows the solar flux for all the years covering both solar maximum during 2000–2003 and 2011–2015 and solar minimum during 2004–2010. The bottom three panels show the three sites from west to east in longitude. The left axis shows the average duration of spread days. The right axis shows the percentage of available data. The total number of longest spread F hours for each month is summed over a given year and then divided by the number of months to obtain average for each year. Black and white bars represent range and frequency spread F and the x symbol represents the percentage of data available.
3.3 Seasonal cycle variation
Figure 7 shows the average number of spread days for each month for all the years of available data between 2000–2016 for Jicamarca, Ascension Island and Kwajalein along with the angle between declination and terminator. The three panels show the three sites from west to east in longitude. The left axis shows the average number of spread days for each month for all the available years along with the years of data available for each month. The right axis shows the angle. The average is calculated by first obtaining the sum of spread F nights for each month for each available year. These sums for each year are divided by the number of available years of data to obtain the averages. Black and white bars represent range and frequency spread F while the x symbol represents the months of data available and the blue line represents the angle.
Figure 8 shows the average duration of spread days for each month for all the years of available data between 2000–2016 for Jicamarca, Ascension Island and Kwajalein. The three panels show the three sites from west to east in longitude. The left axis shows the average number of spread days for each month for all the available years. The right axis shows the number of years of data available for each month. The total number of longest spread F hours for each night for each month are counted and then divided by the total number of spread F nights. These monthly averages are then divided by the number of available years of data to obtain the averages. Black and white bars represent range and frequency spread F and the x symbol represents the years of months of data available.
3.4 hmF2 and foF2 variation for solar min and solar max
Figures 9 & 10 shows the pre-sunset hmF2 and foF2 vs. duration of spread days for all the available data between 2000–2016 for Jicamarca, Ascension Island and Kwajalein. It shows the duration of each spread day versus the presunset hmF2 of that day. The average hmF2 for three hours from 4–7 PM is obtained for each day and is plot versus the duration of both types of spread F for that night. It shows how hmF2 varies for spread F days for both solar maximum and solar minimum. The three panels show the three sites. The left-hand side shows solar maximum while right-hand side shows solar minimum. The red cross represents range spread F while blue circles represent frequency spread F.
3.5 Solar activity variation
Figure 11 shows pre- and post-midnight spread F duration for all the stations versus F10.7 for solar maximum (2000–2003, 2011–2015) and solar minimum (2004–2010, 2016) plotted versus F10.7. The red cross represents range spread F while blue circles represent frequency spread F.
Different types of spread F for low-latitude regions have been shown in Fig. 1. This has also been compared with the Range-Time-Intensity plots using JULIA coherent scatter radar data in Fig. 2. The life cycle of spread signatures observed start with huge plumes and intense spread to reduced spread later in the evening down to spread signatures similar to mid latitude spread F characteristics.
Comparison and Validation of foF2 data- The data from the detection algorithm has been compared with SAMI2 and IRI models and shown in Fig. 4. The detected values show a good match with the other values giving confidence in the algorithm. Both of the models provide an average behavior of the ionosphere and are not expected to capture small-scale day-to-day variation.
Solar Cycle Variation – The existence of Equatorial Spread F does not exhibit a strong variation with solar cycle variation for the three sites, however the values of foF2 and hmF2 tend to increase during solar maximum for all the sites. The high occurrence of spread F during solar minimum over Cachoeira Paulista in Brazil was observed by Candido et al., [2011]. Lower plasma densities during solar minimum is essential for development of gravity waves and this leads to the spread F development [Candido et al., 2011] which is similar to Fig. 1 (c) & (d). The spread F in Fig. 1(b) is due to equatorial plasma bubbles and is mainly during solar maximum years. More plasma bubbles occur during high solar activity during October and March [Sahai et al., 2000]. Spread F is mostly consistent throughout solar minimum and maximum and does not show much change except maybe radar plumes are more observed during solar minimum [Hysell and Burcham, 2002]. Figures 5–6 show another observation that all the stations tend to have more range spread F events during solar minimum, and more frequency spread F events during solar maximum. Abdu et al. [1983] and Candido et al. [2011] observe this similar pattern over Fortaleza, Brazil and Cachoeira Paulista, Brazil respectively.
Seasonal Variation - The three stations at different longitudes show varying seasonal patterns in Figs. 7–8. For Jicamarca most spread F is observed during vernal equinox and winter solstice; Ascension during autumn equinox and December; and Kwajalein during vernal equinox and summer. This may be due to the local angle of declination. Abdu at al., [1992] mentions how the declination angle affects the sporadic E and F layers. Abdu [2001] discusses similar results with angle of declination and its effects on ESF variation. When the terminator aligns with the magnetic field line, both the ends of the tube are in dark, it leads to maximum magnetic eastward wind and maximum longitudinal gradient in integrated conductivity. This causes maximum amplitude of pre-reversal enhancement in the vertical drift and therefore maximum irregularity or ESF occurrence [Abdu, 2001] and we see similar result for all the three sites. Tsunoda [1985] and Aarons [1993] show similar results with scintillation data. Afolayan et al., [2019] have similar seasonal variations for Jicamarca and Kwajalein.
An interesting observation from this study and Bhaneja et al., [2018] is that places located at negative declination tend to have most spread F conditions during northern winter seasons (December-February) while places at positive declination have most spread F during northern summer seasons (June-August). Jicamarca has similar seasonal pattern as Wallops Island [Bhaneja et al., 2009, 2018], because both Wallops 75.5°and Jicamarca 76.8° are on close longitudes. This suggests that same or close longitude locations have similar spread F occurrence statistics irrespective of their latitude locations. Equatorial Plasma Bubbles (EPBs) or Radar Plumes shown in Fig. 1(b) has the highest occurrence during equinoxes during the low angles for Jicamarca. Burke et. al., [2004] used DMSP data and found that the EPBs maximize at low angles and mostly occur during April and Aug-Sept periods.
ESF is almost always observed near dusk for all seasons and levels of solar activity and thus we see longer duration of spread F for Jicamarca data. For Jicamarca, less spread F during northern summer (May-August) is observed, probably due to downward drift velocity caused by westward electric field. Evening upward drift velocity and early reversal time causes increase in spread F during equinoxes and Sept-April time. More spread F over Jicamarca is observed between months of September and April and has also been observed by Fejer et al., [1999], Chen et al., [2006], Chapagain et al., [2009] and Aol et al., [2020]. During December strong post sunset spread F is observed for Jicamarca. The spread F during the equinox and December months occur mostly during quiet geomagnetic times, and similar for May-August months [Chapgain et al., 2009]. Aarons [1993] observes similar spread F patterns for Kwajalein and Ascension Island. Tsunoda et al., [2015] did seasonal study of ESF using data over Kwajalein and found more spread F during months of May-August for Kwajalein.
foF2 and hmF2 variation - Figs. 9 & 10 show variation of hmF2 and foF2 for the three stations and Saito & Maruyama [2007] observe variation of h’F with longitude. Jicamarca has the highest hmF2 for solar maximum for all the three stations. Ascension Island has the highest foF2 of all three stations and higher foF2 is observed during solar maximum for all the stations.
Solar activity variation – Fig. 11 shows the variation of spread F with F10.7. Jicamarca pre-midnight and high F10.7 values show more Range Spread F while post-midnight and high F10.7 show more Frequency Spread F. For Ascension Island, pre-midnight and high F10.7 values show more Range Spread F and post-midnight and high F10.7 show more Frequency Spread F. Kwajalein shows more frequency spread F for solar min. High solar flux levels and geomagnetic activity are conducive for the formation of very high-altitude plumes. Post midnight spread F is mostly plumes and during solar minimum and during solstices June and December months. Radar plumes are not observed during solar maximum summer June months.
A new method of determining spread F is used to find statistical pattern of equatorial spread F for three sites of Jicamarca, Ascension Island and Kwajalein. The spread F occurs throughout both solar minimum and maximum for all the stations but with varying seasonal patterns. The seasonal patterns vary due to different longitudes and declination angle for the three stations. Premidnight and high F10.7 values show higher range spread F for both Jicamarca and Ascension Island while more frequency spread F is observed for high F10.7 for Kwajalein. The foF2 values are validated with SAMI and IRI values and show good correlation.
Range Spread (RS)
Frequency Spread (FS)
Range-Time Intensity (RTI)
International Reference Ionosphere (IRI)
Availability of Data
The digisonde data has been obtained using the National Geographic Data Center at NOAA (NGDC/NOAA) from ftp://ftp.ngdc.noaa.gov/ or https://data.ngdc.noaa.gov/instruments/remote-sensing/active/profilers-sounders/ionosonde/data/.
The IRI model was run online through https://ccmc.gsfc.nasa.gov/requests/requests.php. This work uses the SAMI2 ionosphere model written and developed by the Naval Research Laboratory. The SAMI2 modeled values were generated using sami2py 0.2.3 [Klenzing et al., 2020]. The sami2py software is freely available at www.github.com/sami2py/sami2py.
Julia data was obtained from Madrigal website.
Funding
Funding is provided by NASA Living With a Star NNH20ZDA001N-LWS.
Author’s Contributions
P.B did all the data analysis for the study and wrote most of the paper.
J.K helped with the writing of the introduction and discussion sections, editing and looking over the plots and results. Also provided the IRI and SAMI data for comparisons for Figure 4.
E.P provided Figure 2 and renamed the title along with various edits to the paper write up.
G.D looked over the results and helped in understanding the results and interpreting them along with ideas on what more to look with the analyzed data.
T.W helped in understanding the spread F in the ionograms and interpreting the types and life cycle of spread observed.
Acknowledgements Not applicable
Authors Information
The 1st author Preeti Bhaneja started working on this study a couple years ago when she was unemployed. She got the data available publicly on the noaa website and started to work with downloading the data and analyzing it. She asked J.K for some help for comparing the results and he provided the data for SAMI and IRI models. After being hired by NASA through USRA, P.B then started working on paper again with other co-authors. P.B has worked with digisonde and VIPIR data since 2005 from her PhD days. Her earlier papers were published on midlatitude study. This paper is on low latitudes and after this it will be high latitude study. So, it has a full database of digisonde study. The foF2 comparison in Figure 4 validates the results with other models. This will used for another paper for the full database comparison study.
- Aa, E., Zhang, S.-R., Liu, G., Eastes, R. W., Wang, W., Karan, D. K., et al. (2023). Statistical analysis of equatorial plasma bubbles climatology and multi-day periodicity using GOLD observations. Geo. Res. Lett., 50, e2023GL103510. https://doi.org/10.1029/2023GL103510
- Abadi, P., Ahmad, U.A., Otsuka, Y., Jamjareegulgarn, P., Martiningrum, D.R., Faturahman, A., Perwitasari, S., Saputra, R.E., Septiawan, R.R. Modeling Post-Sunset Equatorial Spread-F Occurrence as a Function of Evening Upward Plasma Drift Using Logistic Regression, Deduced from Ionosondes in Southeast Asia. Remote Sens. 2022, 14, 1896. https://doi.org/10.3390/rs14081896.
- Abdu, M. A., deMedeiros, R.T., and Nakamura, Y., Latitudinal and magnetic flux tube extension of the equatorial spread F irregularities, J. Geophys. Res., 88(A6), 4861–4868, doi:10.1029/JA088iA06p04861, 1983.
- Abdu, M. A., Batista, I.S., and Sobral, J.H.A., A new aspect of magnetic declination control of equatorial spread F and F region dynamo, J. Geophys. Res., 97(A10), 14897–14904, doi:10.1029/92JA00826, 1992.
- Abdu, M.A., Outstanding problems in the equatorial ionosphere-thermosphere electrodynamics relevant to spread F, J. Atmos. and Solar-Terres.Phys., 63, 869-884, 2001.
- Aarons, J., The longitudinal morphology of equatorial F-layer irregularities relevant to their occurrence, Space Science Reviews, 63, 209-243, 1993.
- Afolayan, A.O., Singh, M.J., Abdullah, M., Buhari, S.M., Yokoyama, T., and Supnithi, P., Observation of seasonal asymmetry in the range spread F occurrence at different longitudes during low and moderate solar activity, Ann. Geophys., 37, 733-745, 2019.
- Aol, S., Buchert, S., Jurua, E., and Milla, M., Simultaneous ground-based and in situ Swarm observations of equatorial F-region irregularities over Jicamarca, Ann. Geophys., 38, 1063-1080, 2020.
- Bhaneja, P., and Bullett, T.W., Detection of foF2 using digisonde and VIPIR instruments, In Preparation for submission, 2023.
- Bhaneja, P., Earle, G.D., and Bullett, T. W., Statistical Analysis of Midlatitude Spread F using Multi-Station Digisonde Observations, J. Atm. Solar-Terres. Phys., 167, 146-155, doi.org/10.1016/j.jastp.2017.11.016, 2018.
- Bhaneja, P., Earle, G.D., Bishop, R.L., Bullett, T.W., Mabie, J., Redmon, R., A statistical study of midlatitude spread F at Wallops Island, Virginia, J. Geophys. Res., 114, A04301, doi:10.1029/2008JA013212, 2009.
- Bilitza, D., Altadill, D., Zhang, Y., Mertens, C., Truhlik, V., Richards, P., McKinnell, L-A., and Reinisch,B., The International Reference Ionosphere 2012 – a model of international collaboration, J. Space Weather Space Clim., 4 (2014) A07 DOI: 10.1051/swsc/2014004, 2014.
- Booker, H. G., & Wells, H. W. (1938). Scattering of radio waves by the F-region of the ionosphere. Terrestrial Magnetism and Atmospheric Electricity, 43(3), 249–256.
- Bowman, G.G., A comparison of nighttime TID characteristics between equatorial-ionospheric-anomaly crest and midlatitude regions, related to spread F occurrence, J.Geophys. Res., 106, A2, 1761-1769, 2001.
- Bowman, G.G and Mortimer I.K., Ionospheric coupling, especially between ionogram-recorded spread-F and sporadic-E enhancements at an equatorial-anomaly crest station, Chung-Li, J. Geophys. Res., 107, A10, 1292, doi:10.1029/2001JA007549, 2002.
- Burke, W.J., Huang, C.Y., Gentile, L. C., and Bauer, L., Seasonal-longitudinal variability of equatorial plasma bubbles, Ann. Geophysicae, 22, 3089-3098, 2004.
- Candido, C.M.N., Batista, I.S., Becker-Guedes, F., Abdu, M.A., Sobral, J.H.A., and Takahashi, H., Spread F occurrence over a southern anomaly crest location in Brazil during June solstice of solar minimum activity, J. Geophys. Res., 116, A06316, doi:10.1029/2010JA016374, 2011.
- Carter, B. A., Yizengaw, E., Retterer, J. M., Francis, M., Terkildsen, M., Marshall, R., Norman, R., & Zhang, K. (2014). An analysis of the quiet time day-to-day variability in the formation of postsunset equatorial plasma bubbles in the Southeast Asian region. Journal of Geophysical Research: Space Physics, 119(4), 3206–3223. https://doi.org/10.1002/2013JA019570
- Chen, W. S., C. C. Lee, J. Y. Liu, F. D. Chu, and B. W. Reinisch, Digisonde spread F and GPS phase fluctuations in the equatorial ionosphere during solar maximum, J. Geophys. Res., 111, A12305, doi:10.1029/2006JA011688, 2006.
- Chapagain, N.P., Fejer, B.G., and Chau, J.L., Climatology of postsunset equatorial spread F over Jicamarca, J. Geophys. Res., 114, A07307, doi:10.1029/2008JA013911, 2009.
- Cherniak, I., & Zakharenkova, I. (2022). Development of the storm-induced ionospheric irregularities at equatorial and middle latitudes during the 25–26 August 2018 geomagnetic storm. Space Weather, 20, e2021SW002891. https://doi.org/10.1029/2021SW002891
- Drob, D., J. Emmert, J. Meriwether, J. Makela, E. Doornbos, et al., An update to the Horizontal Wind Model (HWM): The quiet time thermosphere. Earth and Space Science, 2(7). 10.1002/2014EA000089, 2015.
- Earle, G. D., Bhaneja, P., Roddy, P.A., C. M. Swenson, C.M., Barjatya, A., Bishop, R.L., Bullett, T.W., Crowley, G., Redmon, R., Groves, K., Cosgrove, R., Sharon L. Vadas, A comprehensive rocket and radar study of midlatitude spread F, J. Geophys. Res., 115, A12339, doi:10.1029/2010JA015503, 2010.
- Fejer, B.G., Scherliess, L., de Paula, E.R., Effects of the vertical plasma drift velocity on the generation and evolution of equatorial spread F, J. Geophys. Res., 104, A9, 19,859-19,869, 1999.
- Gonzalez, G., Spread-F characteristics over Tucuma´n near the southern anomaly crest in South America during the descending phase of solar cycle 24, Adv. Space Res., 69, 3, 1281-1300, 2022.
- Huba, J. D., G. Joyce, and J. A. Fedder, Sami2 is Another Model of the Ionosphere (SAMI2): a new low-latitude ionosphere model. Journal of Geophysical Research: Space Physics, 105(A10), 23,035–23,053. 10.1029/2000ja000035, 2000.
- Hysell, D.L., and Burcham, J.D., Long term studies of equatorial spread F using the JULIA radar at Jicamarca, J. Atmosphere. Sol-Terr. Phys., 64, 1531-143, 2002.
- Hysell, D.L., An overview and synthesis of plasma irregularities in equatorial spread F, J. Atmosphere. Phys., 62, 1037-1056, 2000.
- Kil, H., The occurrence climatology of equatorial plasma bubbles: A review, J. Astron. Sci., 39(2), 23-33, 2022.
- Klenzing, J., Smith, J., Hirsch, M., and Burrell. A.G., sami2py/sami2py: Version 0.2.2 (v0.2.2). Zenodo. https://doi.org/10.5281/zenodo.3950564, 2020.
- Klenzing, J., Smith, J. M., Halford, A. J., Huba, J. D., & Burrell, A. G. (2022). sami2py — Overview and applications. Frontiers in Astronomy and Space Science, 1–11. https://doi.org/10.3389/fspas.2022.1066480.
- Klenzing, J., Halford, A. J., Liu, G., Smith, J. M., Zhang, Y., Zawdie, K., Maruyama, N., Pfaff, R., & Bishop, R. L. (2023). A system science perspective of the drivers of equatorial plasma bubbles. Frontiers in Astronomy and Space Sciences, 9, 1–6. https://doi.org/10.3389/fspas.2022.1064150.
- Picone, J. M., NRL MSISE-00 empirical model of the atmosphere: Statistical comparisons and scientific issues. Journal of Geophysical Research, 107(A12), 1468–SIA 15–16. 10.1029/2002JA009430, 2002.
- Sahai, Y., Fagundes, P.R., and Bittencourt, J.A., Transequatorial F-region ionospheric plasma bubbles: solar cycle effects, J. Atmos. Solar-Terres. Phys., 62, 1377-1383, 2000.
- Saito, S., Maruyama, T., Large-scale longitudinal variation in ionospheric height and equatorial spread F occurrences observed by ionosondes, Heophys. Res. Lett., 34, L16109, doi:10.1029/2007GL030618, 2007.
- Scherliess, L., and B. G. Fejer, Radar and satellite global equatorial F region vertical drift model. Journal of Geophysical Research: Space Physics, 104(A4), 6829–6842. 10.1029/1999JA900025, 1999.
- Takahashi, H., Essien, P., Figueiredo, C. A. O. B., Wrasse, C. M., Barros, D., Abdu, M. A., Otsuka, Y., Shiokawa, K., and Li, G. Z. (2021). Multi-instrument study of longitudinal wave structures for plasma bubble seeding in the equatorial ionosphere. Earth Planet. Phys., 5(5), 368–377, eepp2021047. http://doi.org/10.26464/epp2021047
- Tsunoda, R. T., Control of the seasonal and longitudinal occurrence of equatorial scintillations by the longitudinal gradient in integrated E region Pederson conductivity, J. Geophys. Res., 90, A1, 447-456, 1985.
- Tsunoda, R. T., T. T. Nguyen, and M. H. Le, Effects of tidal forcing, conductivity gradient, and active seeding on the climatology of equatorial spread F over Kwajalein, J. Geophys. Res.Space Physics, 120, 632–653, doi:10.1002/2014JA020762, 2015.
- Whalen, J.A., Dependence of equatorial bubbles and bottomside spread F on season, magnetic activity, and E x B drift velocity during solar maximum. J. Geophys. Res., 107, A2, 1024, 10.1029/2001JA000039, 2002.
