Observations by the CUTLASS radar, HF Doppler, oblique ionospheric sounding,
and TEC from GPS during a magnetic storm
D. V. Blagoveshchensky1, V. A. Kornienko2, M. Lester3, I. I. Shagimuratov4, A. J. Stocker5 and
E. M. Warrington5
1
Department of Radioengineering, St. Petersburg University of Aerospace Instrumentation, Russia,
2
Department of Geophysics, Arctic and Antarctic Research Institute, St. Petersburg, Russia,
3
Department of Physics and Astronomy, University of Leicester, U.K.,
4
West Department of IZMIRAN, Kaliningrad, Russia,
5
Department of Engineering, University of Leicester, U.K.
Abstract. Multi-diagnostic observations, covering a significant area of north-west Europe, were made
during the magnetic storm interval (April 28-29 2001) that occurred during the High-rate SolarMax
IGS/GPS-campaign. HF radio observations were made with vertical sounders (St. Petersburg and
Sodankyla), oblique incidence sounders (Murmansk to St. Petersburg, 1050 km, and Inskip to
Leicester, 170 km), Doppler sounders (Cyprus to St. Petersburg, 2800 km, and Murmansk to St.
Petersburg), and using an HF radar (CUTLASS radar, Hankasalmi, Finland). These, together with
total electron content (TEC) measurements made at GPS stations from the Euref network in northwest
Europe, are presented in this paper.
1. Introduction
The High Rate GPS/GLONASS measuring campaign (HIRAC) was initiated by the International GPS
Service (IGS) and supported by COST 271 activities with the aim of analysing transionospheric
signals received from navigation satellites by the IGS ground station network in order to study the
behaviour of the ionosphere during the recent solar maximum. Measurements were undertaken during
the period 23-29 April 2001 with the period from 28 to 29 April 2001 being of particular interest since
a magnetic storm of moderate intensity took place at this time.
Whilst the purpose of the HIRAC campaign was to make measurements of total electron content (in
this paper from the Euref network of GPS stations in northwest Europe), it is interesting to compare
such measurements with those made contemporaneously with a range of other instrumentation. The
subject of this paper is HF radio measurements made with ionosondes at St. Petersburg and
Sodankyla, oblique incidence sounders on paths from Murmansk to St. Petersburg (1050 km) and
Inskip to Leicester (170 km), and Doppler measurements on paths between Cyprus and St. Petersburg
(2800 km) and between Murmansk and St. Petersburg. HF radar measurements were also made with
the CUTLASS radar located in Hankasalmi, Finland. The locations of these sites are shown in
Figure 1.
2. Character of the geomagnetic disturbance
Presented in Figure 2 are various geophysical parameters that were observed during the magnetic
storm on 28-29 April 2001. As indicated by the Dst-index (panel (a) of Figure 2) the storm
commenced at approximately 1200 UT on 28 April, peaked during the night and ended at about
1200 UT on 29 April. Since the Dst minimum was less than -50 nT, and Bz was less than -5 nT for at
least two hours, according to classifications given by Gonzalez et al.(1994), the storm is categorised as
moderate. The periods 0000–0600 UT on 28 April and 1200-2359 UT on 29 April may be considered
as either weakly disturbed or quiet.
The variations in the AE-index (panel c) show that the magnetic storm contained three pronounced
substorms, numbered 1, 2 and 3 in Figure 2:
1. A moderate substorm of 2.5 hours duration that has the onset of the expansion phase (To) at
0500 UT on 28 April with the end of this phase (Te) at around 0730 UT.
2. The strongest of the three substorms starts at 1300 UT and ends at 1700 UT.
3. The third substorm started at 0100 UT on 29 April and finished at 0500 UT.
It is evident from panel (d) of Figure 2 that the main maxima of Kp coincide with the three substorm
intervals with the largest values of Kp (~6) occurring during the second, most intense, substorm.
3. Observations
3.1 Total electron content (TEC) measurements
Global Positioning System (GPS) observations provide an excellent method for studying the structure
and dynamics of the ionosphere. The total electron content (TEC) along the path between the satellite
and the ground receiver can be determined from observations of the time delay of the dual-frequency
(1.57 GHz and 1.23 GHz) GPS signals. A technique has been developed in the West Department of
IZMIRAN that allows TEC maps over Europe to be produced using the GPS measurements from
more than 60 stations of the Euref Network (Shagimuratov, 2002). The high density of stations in
Europe enables TEC maps to be derived with high temporal and spatial resolution between latitudes of
40 and 75o N and longitudes of 10° W and 40o E.
The TEC dynamics for a quiet day (27 April) and two disturbed days of the geomagnetic storm (28-
29 April) are presented in Figure 3. For the quiet day, as expected, the electron concentration
increases through the morning, there is a maximum close to noon, and a decrease in concentration in
the evening. At times (e.g. 0400-0800 UT), there is a strong east to west gradient. It is interesting to
compare TEC variations under quiet conditions with those under disturbed conditions. Prior to the
magnetic storm starting (0500 UT on 28 April), the TEC variations at 0000 UT and 0400 UT on 27
and 28 April are very similar. However, once the storm started, there are some differences between the
quiet and disturbed days for the same hours. For example, at 1600 UT and 2000 UT on 28 April, the
TEC is extremely low at high latitudes (60–75o) and somewhat decreased at mid latitudes (50-55o).
Similar behaviour persists until the end of the storm at 1200 UT on 29 April. After this time, the
geomagnetic conditions are quiet and the ionosphere recovers (e.g. the TEC at 2000 UT on 27 and 29
April are similar).
The TEC observed during the expansion phases of the three substorms is presented in Figure 4. For
substorm Number 1 (panel a) there is a smooth time variation in TEC with a growth from east to west.
Since similar behaviour is observed for quiet conditions then it appears, as mentioned above, that
substorm Number 1 did not make any significant changes to the ionosphere.
During the second, intense substorm (panel b), some large-scale irregularities and strong gradients in
the TEC have arisen. For example, there are bends in the contours at 1300 UT not present at 1200 UT
(Figure 4), the isolated structures at 1400 UT as well as the sharp gradients at 1500 UT.
There is a minimum in the TEC during the third substorm (panel c) at both mid (50–55o) and high
latitudes (60–75o), which is also found in the OIS observations (Figure 5 and Figure 6).
3.2 Oblique incidence soundings for the Murmansk – St. Petersburg path
Several key parameters indicative of the propagation conditions were determined from oblique
incidence sounding undertaken over the path from Murmansk to St. Petersburg (midpoint geographic
64.5o, geomagnetic 60°). These included F2MOF (maximum observed frequency for signals reflected
from the F2-layer) and F2LOF (lowest observed frequency for signals reflected from the F2-layer).
EMOF, ELOF, EsMOF, and EsLOF were similarly defined for signals reflecting from the E and
sporadic E (Es) layers respectively. An additional parameter, DeltaF2MOF, was defined as the
deviation of F2MOF from the 10-day quiet median. Note that the MOF depends on the ionospheric
electron density profile, while the LOF depends on the level of ionospheric absorption and on a range
of system parameters including the transmitter power and antenna gains.
The variations of the parameters defined above during the magnetic storm on 28-29 April are
presented in Figure 5. The first interval (0500–0730 UT) is characterized by an isolated substorm with
moderate intensity (Lyons, 1996) and increased absorption (see panels a, d and e). During this
substorm, vertical soundings made at St. Petersburg and Sodankyla indicate that the ionosphere is not
essentially altered. However, two or three hours before the onset of the substorm, DeltaF2MOF
increases reaching a maximum at To (0500 UT). During the substorm expansion phase DeltaF2MOF
decreases followed by a marked increase at and after Te (0730 UT). DeltaF2MOF then falls to zero
during the following few hours. Similar behaviour of DeltaF2MOF during an isolated substorm of
moderate intensity was reported by Blagoveshchensky and Borisova (2000), Blagoveshchensky et al.,
(1992, 1996 and 2003) and named the “main effect”.
Substorm Number 2, which was very strong, is accompanied by high absorption and changes in the
ionosphere measured by vertical incidence sounding at St. Petersburg and Sodankyla. A sharp
decrease of DeltaF2MOF takes place and the main phase of the magnetic storm begins when Dst and
Bz values are negative (Figure 2). Here the “main effect” is absent since the substorm Number 2 is
intense and not isolated (Lyons, 1996). Similar behaviour is observed during substorm Number 3.
After the third substorm, the ionosphere gradually recovers over about 10 hours and by 1200 UT on
29 April, a quiet period begins with insignificant variations of DeltaF2MOF.
Panels (b) and (c) on Figure 5 illustrate the range of operational frequencies suitable for radio
communications for signals reflected from the F2-layer and by sporadic-E respectively. During the
main phase of magnetic storm (from 1200 UT on 28 April to 1200 UT on 29 April) the frequency
range F2MOF–F2LOF quickly becomes narrow and there are two intervals when the signal is absent.
At the same time, sporadic-E is the principal cause of reflections from the E-region, appearing on the
oblique ionograms at the same time as on the vertical ionograms at Sodankyla and St. Petersburg. It is
interested to compare the undisturbed interval 0000–0600 UT on 28 April with the disturbed interval
0000–0600 UT on 29 April and the undisturbed interval 1200–2359UT on 29 April with the disturbed
interval 1200–2359 UT on 28 April.
3.2.1 0000–0600 UT
For the quiet period (0000 – 0600 UT on 28 April), signal reflections from the normal E-layer take
place and the range EMOF–ELOF is rather wide (Figure 5, panel c). However, for the same period on
April 29 there are only Es-reflections and frequency range is narrow because EsLOF increases with
the increase of absorption.
During the disturbance (29 April, substorm Number 3) there is a sharp narrowing of the frequency
range F2MOF–F2LOF and the signal is absent from 0100 to 0500 UT. The absence of the signal may
result from either very high absorption (Figure 5, panel c) or because EsMOF is higher than F2MOF
since the signals might then only reflect from the Es-layers.
3.2.2 1200–2359 UT
Like the 0000–0600 UT interval, a sharp narrowing of the range F2MOF–F2LOF and the loss of OIS
signal also occurs during substorm Number 2 (from 1900 to 2200 UT). The causes of signal
disappearance are the same, i.e. high absorption and existence of sporadic Es-layers.
The change of signal reflection from E- to Es-reflections and the decrease of the frequency range
EsMOF–EsLOF also occurs during substorm Number 2 from 1300 to 1700 UT on April 28.
3.3 Comparison with the oblique incidence soundings for the Inskip – Leicester path
The midpoint of the Inskip-Leicester path (170 km) is located at mid-latitudes (geographic latitude
53°, geomagnetic latitude 51°), although propagation can sometimes be affected by auroral features to
the north (Siddle et al., 2004a; b), as opposed to the subauroral path from Murmansk to St. Petersburg
(geomagnetic latitude 60°). The main results are presented in Figure 6, in a similar way to those for
Murmansk–St. Petersburg (Figure 5). The intervals when the three substorms (1, 2 and 3) identified
earlier occurred are also shown.
3.3.1 Figure 5a and Figure 6a.
At night during the magnetic storm on 28-29 April deltaF2MOF < 0 due to atmospheric circulation at
both high and mid latitudes. However, as expected, the disturbance is less intense and of shorter
duration at mid than at high latitudes. The “main effect” occurs during all three substorms. Namely, a
few hours before To, the ionization in the F2-layer increases. During the substorm expansion phase
(To to Te), the ionization decreases followed by an increase at the end of the expansion phase Te. As
noted by Blagoveshchensky and Borisova (2000), Blagoveshchensky et al. (1992, 1996 and 2003), the
“main effect” commonly occurs during substorms at both middle and high latitudes.
3.3.2 Figure 5b and Figure 6b.
While the variation of the F2MOF – F2LOF range during a magnetic storm is approximately the same
there are some differences. Firstly, at mid latitudes, the any increased absorption does not lead to the
complete disappearance of the signal. Secondly, the range of operational frequencies F2MOF –
F2LOF at middle latitudes is wider than at high ones.
3.3.3 Figure 5c and Figure 6c.
The behaviour of the ionization in the E-layer during a magnetic storm is different at middle and high
latitudes. The Es-layers observed continuously at high latitudes are not present at all times at middle
latitudes (for example, at 1500 UT and 1800 UT on 29 April). Rises of ELOF values at midday are
caused by absorption in the ionospheric D-layer.
3.4 Doppler measurements for the Murmansk–St. Petersburg path
On 28-29 April 2001, 60 records of Doppler measurements of hourly duration were made over the
Murmansk–St. Petersburg (5930 kHz) path. For one of these, the absorption was high and the signal
was not detected. Measurements were taken during part of the expansion phase and the recovery phase
of substorm Number 2 (1600–2100 UT on 28 April; 15 sonograms) and during the full expansion
phase and recovery phase of substorm Number 3 (0100–1000 UT on 29 April; 26 sonograms).
Measurements were also obtained during a quiet interval (1800–2359 UT on 29 April; 18 sonograms).
Three important characteristics can be identified from the sonograms: the width of the Doppler
spectrum ΔfD (Hz), wave disturbances with an average quasi-period TW (min) and amplitude fDmax
(Hz), and additional traces characterized by the amplitude of their frequency deviation (Hz). Since not
all of the measurements can be shown here, a few example sonograms will be presented, and the
following general conclusions can be drawn.
1. ΔfD is in the range 1–3 Hz during quiet conditions, increasing to a maximum of 10–30 Hz during
the substorm expansion and recovery phases. The precise value of ΔfD depends to some extent on
the propagation mechanism. If the operational frequencies exceed the MOF, there is a transition
from signal reflection to scattering from irregularities and the signal spectrum widens (Table 1).
2. During the substorm expansion and recovery phases Tw lies between about 2 and 30 minutes. As a
rule, under quiet conditions wave disturbances are not observed. Here the mode mechanism of
propagation is also significant which is F2- or Es-reflection of signal (see Table 1).
3. Additional traces on the sonograms are only observed during the expansion phase of a substorm or
immediately after it (see Table 1).
4. The Doppler characteristics are presented together with parameters derived from the oblique
soundings (Figure 5) in Table 1. These results are discussed in more detail below.
3.4.1 28 April, substorm Number 2
1600–1700 UT The expansion phase. A sonogram is presented in Figure 7. The signal is strong and
reflected from the F2-layer. Wave processes were present only on additional tracks
and were absent on the direct signal.
1700–1800 UT The recovery phase. Signal is strong and reflected from the F2-layer or the Es-layer.
1800–1900 UT The recovery phase. Signal is weak and spread if its reflection takes place near
F2MOF or above it. Signal is narrow if reflected from the Es-layer.
1900–2000 UT The recovery phase. Signal is reflected only from the Es-layer and is spread in
Doppler.
2000–2100 UT The recovery phase. Signal is reflected from the Es-layer and is narrow.
3.4.2 29 April, substorm Number 3
0100–0300 UT The expansion phase. The signal is diffuse, and is reflected near EsLOF.
0300–0400 UT The expansion phase. The signal is absent since its frequency (5930 kHz) lies out of
the range EsMOF – EsLOF.
0400–0500 UT The expansion phase. Signal is weak, with reflection occurring near to EsLOF and
F2MOF. There are short-term variations in Doppler frequency shift with a period of
about 30 s, which are associated with magnetic pulsations. Magnetic pulsations are
known to be a special feature of magnetic storms and substorms.
0500–0600 UT The recovery phase. A sonogram is presented in Figure 8. The signal is strongly
reflected from the Es- and F2-layers. Some short-term variations in Doppler shift
are observed.
3.5 Doppler measurements for the Cyprus–St. Petersburg path
For the Cyprus–St. Petersburg path (9410 kHz and 12095 kHz), the variations of Doppler spectra at
9410 kHz are well correlated with those on 12095 kHz. Furthermore, ray tracing calculations showed
that the point of reflection of the wave with a frequency 9410 kHz in the F2-layer of ionosphere was
about 22 km lower than the point of reflection of the wave with a frequency 12095 kHz. Therefore,
only sonograms for 9410 kHz signals will be presented here.
A sonogram taken during the expansion phase of substorm Number 3 (0200 to 0300 UT on
29 April 2001) for this path is presented in Figure 9. Here the Doppler width Δ(fD) is 4 Hz and there
are wave disturbances with a quasiperiod (Tw) of 30 minutes and amplitude (fDmax) 2 Hz but, as with
the Murmansk – St. Petersburg path at this time, there are no additional traces. The wave processes
are likely to be medium-scale travelling ionospheric disturbances (TIDs). From the Doppler
measurements in Figure 9, it is possible to estimate some values of TID parameters. According to
Afraimovich (1982), tentative estimations of wave disturbance amplitude M, low border of horizontal
length of wave Λ, velocity of wave screen movement V and intensity of disturbance of electron
concentration δ can be obtained from the following
M = Тw∙ (fDmax) ∙λ∙(1/2π),
Λ = 2π∙Zo∙(M/2Zo)1/2,
V = Λ/ Тw,
δ = M/2Zo,
where Тw is a period of wave disturbance, fDmax is a wave amplitude by data of Doppler
measurements, λ is a wave-length of the HF radio wave, and Zo is height of reflection.
From Figure 9 the values of parameters are: Тw = 30 min, fDmax = 2 Hz, Zo = 250 km, and λ = 32 m.
These give the following results: M = 18 km, Λ = 300 km, V = 167 m/s, δ = 3,6 10-2. These values
are similar to those obtained by other observation methods ( Stocker et al., 2000).
For Figure 7, the ionospheric velocity near the signal reflection position is determined by
V = (λ∙ fDmax)/ 2∙Cosφ ,
where φ is a wave incidence angle on the ionosphere. Calculations give V = 175 m/s.
3.5 Observations by the CUTLASS radar
CUTLASS is an HF radar constructed and operated by the Radio and Space Plasma Physics Group at
the University of Leicester to study the high latitude ionosphere (Jones and Thomas, 1997; Yeoman et
al., 2001) and which forms part of the SuperDARN network. There are two CUTLASS radars, one
located in Iceland and one in Finland with both radars observing a volume over and to the north of
Scandinavia. Combination of the measurements from two radars enables the ionospheric convection
velocity vector perpendicular to the Earth’s magnetic field to be determined.
Observations made with the Hankasalmi, Finland radar during 28-29 April 2001 from 1200 UT on
28 April to 1100 UT on 29 April are presented in Figure 10. This interval corresponds to the most
disturbed part of magnetic storm described above (Figures 2 and 5). The geographic latitudes
indicated in Figure 10 range from 60°, which corresponds to that of St. Petersburg to 69°,
corresponding to Murmansk. Although the CUTLASS radar data can be compared with Doppler and
OIS measurements on the Murmansk–St. Petersburg path, the midpoint of Cyprus–St. Petersburg path
is located well to the south of the radar field-of-view. Beam 15 of the radar was chosen since it is
directed northward and is approximately parallel to the Murmansk–St. Petersburg path. In Figure 10,
the expansion phases of substorms Number 2 and Number 3 are indicated by the vertical lines, and the
times at which the Doppler records presented in Figures 7-9 are shown.
The top panel in Figure 10 illustrates the ionospheric backscattered power. Prior to 1400 UT on
28 April, there was no ionospheric scatter observed. However, during substorm Number 2, there is an
area where intense irregularities exist, located at the same latitude as the edge of diffuse precipitation
(Blagoveshchensky et al., 1992). These irregularities disappear after the end of the expansion phase of
a substorm. The Doppler observations in Figure 7 can be explained by existence of pronounced
irregularities at the end of the expansion phase of the substorm Number 2. These irregularities are
concentrated near the midpoint of path Murmansk–St. Petersburg on φ = 64.5° (see Figure 10). For
substorm Number 3, there is a stretched area of irregularities but these are less intense than those
formed during substorm Number 2. Simultaneously, southward moving TIDs were observed in the
Cyprus–St. Petersburg Doppler data (Figure 9). During the recovery phase of substorm Number 3, the
ionospheric irregularities remain intense and there are short periodic variations in the Doppler
frequency measured on the Murmansk–St. Petersburg path (Figure 8). These variations are known to
arise sometime in the beginning or at the end of magnetic storms or strong substorms (Lyons, 1996).
From 0800 UT, 29 April, until the end of the magnetic storm, the irregularities become weaker and the
ionosphere returns to an undisturbed state.
The irregularity velocity (Figure 10, middle panel) at 1630 UT, 28 April is less than 200 m/s. This is
consistent with the value of 175 m/s derived from the Doppler data in Figure 7.
The bottom panel illustrates the variation of the spectral width during the disturbance. This parameter,
which can be looked at in a variety of ways, depends to some extent on the nature of the irregularities.
One view is that the spectral width represents the decorrelation time of the irregularities, such that
large values indicate a rapid decorrelation while small values represent a much longer decorrelation
time. This has been investigated using Heater generated irregularities. Another view is that the
spectral width represents the variability of the Doppler velocity within the range beam cell with large
values indicating a significant variability of the Doppler velocity. There is also some evidence that the
spectral width can be broadened in association with enhanced electron temperatures probably due to
particle precipitation. It is evident from Figure 10 that intense irregularities are formed during the
expansion phases of substorms Number 2 and Number 3 that can produce spectral broadening in both
the radar observations and the Doppler measurements (Table 1).
4. Concluding remarks
During the High Rate GPS/GLONASS measuring campaign (HIRAC), simultaneous experimental
measurements were made using GPS, oblique and vertical incidence sounding, HF Doppler, and
CUTLASS radar. The most interesting results were obtained during a magnetic storm on 28-
29 April 2001 and the following conclusions drawn.
1. The behaviour of TEC derived from measurements of GPS signals during a magnetic storm and
substorms can be studied over a wide geographical area. The observed variations in TEC are
consistent with changes in the Delta F2MOF determined from OIS measurements at both mid and
high latitudes.
2. The behaviour of the oblique sounding characteristics on the subauroral path Murmansk–
St. Petersburg were primarily determined by geophysical factors with ionospheric variations
leading to changes in the radio propagation mechanisms and by the level of absorption in the low
ionosphere. The propagation behaviour at high and mid latitudes was similar when the signal was
reflected from the F2-layer, but different when the signal was reflected from the E-layer.
3. The Doppler spectral width is at a maximum (10–30 Hz) during the substorm expansion and
recovery phases, while under the quiet conditions the width is 1 – 3 Hz. The spectral broadening
is associated with the presence of ionospheric irregularities during disturbed conditions also
depends to a degree on the propagation mode.
The wave disturbances observed during the substorm expansion and recovery phases were found
to have periods between 2 and 30 min. Under quiet conditions, wave disturbances were not
usually observed. The parameters of the medium-scale travelling ionospheric disturbances (TID)
derived from the measurements are similar to those previously reported.
Additional tracks on sonograms are observed during the expansion phase of a substorm or
immediately after it, while at other times the tracks are absent.
4. The spectral width and velocity of the irregularities detected by the CUTLASS radar are
consistent with those observed by the Doppler method.
The results from these experiments and the various derived characteristics could be useful in Space
Weather, for organization of radio communications at middle and high latitudes as well as for
ionospheric forecasting during moderate magnetospheric disturbances.
Acknowledgements
The authors would like to express their gratitude to colleagues of Sodankyla observatory and the
compilers of the NASA catalogue edited by J. H. King for a possibility of using their data accessed
over the INTERNET, as well as to M. A. Sergeeva and A. V. Piatkova, students at St. Petersburg
University of Aerospace Instrumentation, for their help in data processing.
The CUTLASS Finland radar is supported by PPARC, the Swedish Institute for Space Physics,
Uppsala, and the Finnish Meteorological Institute.
This work was facilitated through financial support on NATO grant PST/CLG No980327.
References
Afraimovich E. L. Interference methods in ionospheric radio sounding. – M.: Nauka, 1982, 197 p. (in
Russian).
Blagoveshchensky D. V., Borisova T. D. Substorm effects of ionosphere and HF propagation. Radio
Science, 35, (5), 1165-1171, 2000.
Blagoveshchensky D.V., Egorova L.V. and Lukashkin V.M. High-latitude ionospheric phenomena
diagnostics by high-frequency radio wave propagation observations. Radio Science, 27, (2), 267-274,
1992.
Blagoveshchensky D.V., Borisova T. D., and Egorova L.V. Pre- and after substorm situations in the
ionosphere and decameter radio wave propagation . Geomagn. Aeron., 36, No4, 125 - 134, 1996 (in
Russian).
Blagoveshchensky D.V., Maltseva O.A. and Rodger A.S. Ionosphere dynamics over Europe and
western Asia during magnetospheric substorms 1998-99, Ann. Geophys.. 21, 1141 – 1151, 2003.
Gonzalez W. D., J. A. Joselyn, Y. Kamide, et al., What is a geomagnetic storm?, Journal of Geophys.
Res., 99, No. A4, 5771-5792, 1994.
Jones, T.B. and Thomas, E.C. CUTLASS: A new HF auroral radar. In: Proceedings of the IEE
Seventh International Conference on HF Radio Systems and Techniques, July 1997, pp 325-329.
Lyons, L.R. Substorms: Fundamental observational features, distinction from other disturbances, and
external triggering, Journal of geophysical research, vol. 101, No. A6, pp. 13011-13025, 1996.
Shagimuratov I. I. Monitoring of the ionosphere with using GPS. In: Book of abstracts of the
International conference on problems of geocosmos, June 3-8, 2002, St. Petersburg, Russia, pp. 102 -
103
Siddle D.R., A.J. Stocker and E.M. Warrington. The time-of-flight and direction of arrival of HF
radio signals received over a path along the mid-latitude trough: observations. Radio Science
(submitted), 2004a.
Siddle D.R., N.Y. Zaalov, A.J. Stocker and E.M. Warrington. The time-of-flight and direction of
arrival of HF radio signals received over a path along the mid-latitude trough: theoretical
considerations. Radio Science (submitted), 2004b.
Stocker A.J., N.F. Arnold, and T.B. Jones, The synthesis of travelling ionospheric disturbance (TID)
signatures in HF radar observations using ray tracing, Ann. Geophys., 18 (1), 56-64, 2000.
Warrington E.M. and A.J. Stocker. Measurements of the Doppler and multipath spread of HF signals
received over a path oriented along the mid-latitude trough. Radio Science, 38, (5), 1080,
doi:10.1029/2002RS002815, 2003.
Yeoman T.K., D.M. Wright, A.J. Stocker, and T.B. Jones. An evaluation of range accuracy in the
SuperDARN over-the-horizon HF radar systems, Radio Science, 36, (4), 801-813, 2001.
Figure captions
Figure 1. Map showing the locations and paths of the various HF instrumentation employed in these
experiments. The GPS measurements fall in the range between latitudes of 40 o and 70oN and
longitudes of 10oW and 40oE.
Figure 2. Variations of geophysical parameters during a magnetic storm on April 28-29, 2001, (a)
Dst-index, (b) Bz-component of the interplanetary magnetic field, (c) AE-index, and (d) Kp. The three
substorms that occurred during this period are marked by vertical lines and indicated as 1, 2, and 3.
Figure 3. Observed TEC (bold UT values indicate a storm interval), (a) quiet day of 27 April, (b)
disturbed day of 28 April, and (c) disturbed day of 29 April. Numbers on the scale are TEC in units of
1016 m-2.
Figure 4. Observed TEC during the expansion phases of three substorms: (a) Number 1 of 28 April,
(b) Number 2 of 28 April, and (c) Number 3 of 29 April. Numbers on the scale are TEC in units of
1016 m-2.
Figure 5. Oblique sounding measurements made on the path Murmansk–St. Petersburg during a
magnetic storm of (a) DeltaF2MOF, (b) F-region MOF and LOF, and (c) E-region MOF and LOF.
Measurements made at Sodankyla, Finland (near to path mid-point) of (d) X-component of
geomagnetic field, and (e) riometer absorption A (dB) at 32 MHz.
Figure 6. Oblique ionospheric measurements made on the path Inskip–Leicester, 28-29 April, 2001,
of (a) DeltaF2MOF, (b) F-region MOF and LOF, and (c) E-region MOF and LOF.
Figure 7. Sonogram for Murmansk–St. Petersburg between 16:26 and 16:45 UT on 28 April, 2001.
Numbers on the colour scale are power spectrum values, arbitrary units.
Figure 8. Sonogram for Murmansk–St. Petersburg between 05 :00 and 05:46 UT on 29 April, 2001.
Numbers on the colour scale are power spectrum values, arbitrary units.
Figure 9. Sonogram for Cyprus–St. Petersburg between 02:00 and 02:46 UT on 29 April, 2001.
Numbers on the colour scale are power spectrum values, arbitrary units.
Figure 10. Ionospheric scatter observed by CUTLASS Finland radar during a magnetic storm 28-29
April 2001, (a) power, (b) velocity, and (c) spectral width.
Table 1. Doppler characteristics and OIS data on the path Murmansk – St. Petersburg. fop is the
frequency of the Doppler signal (5930 kHz).
Doppler characteristics
Date Time, Amplitude Oblique
UT ΔfD, Hz Тw, min fDmax, Hz of ionospheric
additional sounding
traces, Hz
1600 - 1700 6 6 5-7 -10 ÷ +5 fop ≤ F2MOF
1700 - 1800 10 3 2 -4 ÷ +8 fop ≤ F2MOF
fop < EsMOF
28 April 2001 1800 - 1900 30 2 1,5 -5 ÷ +8 fop < EsMOF
fop ≥ F2MOF
1900 - 2000 30 - - - fop < EsMOF
2000 - 2100 16 - - - fop < EsMOF
0100 - 0200 25 - - -6 ÷ +18 fop<
fop ≤ EsLOF
0200 - 0300 4 2 1 - fop << EsMOF
fop < EsLOF
0300 - 0400 Signal is not detected fop << EsLOF
29 April 2001 10 5 3 -4 ÷ 0 fop ≤ F2MOF
0400 - 0500 magnetic magnetic fop ≤ EsLOF
pulsations: pulsations:
30 s to 8Hz
9 5 3 - fop << EsMOF
0500 - 0600 magnetic magnetic fop < F2MOF
pulsations: pulsations:
40 s to 8Hz