Co-ordinated Observations of Forced and Resonant Field Line Oscillations at High Latitudes.

In this paper we examine the geomagnetic signatures of solar wind impulses using magnetometer array data, a bistatic HF radar, and a co-located imaging riometer. The impulses were associated with 1.6 mHz magnetic pulsations that were observed across a wide range of latitudes. These were most likely due to field line oscillations driven by incoming fast mode waves. At the same time, the radars recorded oscillatory velocity features in ground backscatter returns consistent with ionospheric motions driven by the downgoing waves. Accompanying decreases in cosmic noise absorption were evident in the imaging riometer data.

The purpose of this paper is to examine the magnetic and ionospheric signatures of solar wind pressure perturbations. Geomagnetic activity is driven by energy and mass transfer from the solar wind, and the outermost sunward field lines map most directly to the wave and particle entry regions. The injected particles may give rise to features of dayside auroras (Lui and Sibeck, 1991), while the waves may propagate large distances through the magnetosphere and couple to field line resonances and global cavity modes (Warnecke et al., 1990).

High latitude data sets are often characterised by the appearance of transient features, but the interpretation of these is controversial. Discussion in the literature has focused on radial displacements of the magnetopause due to perturbations in the solar wind dynamic pressure (Farrugia et al., 1989; Sibeck, 1990) and impulsive plasma transfer and magnetic field reconnection events across the magnetopause (Lemaire, 1977; Russell and Elphic, 1979; Saunders et al., 1984). Kelvin-Helmholtz surface waves driven by the velocity shear at the magnetopause may also produce similar signatures in ground magnetometer records (Pu and Kivelson, 1983). There is a need for detailed case studies in order to distinguish the causative mechanism. An important aspect of this is the intercalibration of signatures from different instruments.

In this paper we report the observation of high latitude transient events recorded on a selected day on the ground with the IMAGE and SAMNET magnetometer arrays and simultaneously in the ionosphere with the CUTLASS HF radar and an imaging riometer. The events were stimulated by solar wind pressure perturbations and we thus identify the magnetospheric response to this phenomenon.

Figure 1. Map in geographic coordinates showing IMAGE and SAMNET magnetometer stations and the CUTLASS beams of interest: Finland beam 6 and Iceland East beam 14 (broken lines). The IRIS imaging riometer was located at KIL. Shaded rectangular blocks denote regions in the ionosphere from which Doppler oscillations were recorded. (Larger version - 1MB)

Magnetic field observations were obtained from the IMAGE and SAMNET arrays spanning Scandinavia and Svalbard. Station locations are shown in Figure 1, where the triangles represent the IMAGE stations used here and the squares indicate the SAMNET locations. The IMAGE (International Monitor for Auroral Geomagnetic Effects) array comprises a coordinated multinational network of fluxgate magnetometers sampling the geographic X, Y and Z components of the geomagnetic field each 10 s with a resolution of 0.1-1 nT (Lühr, 1994; Lühr et al., 1998). For use here data were rotated into the geomagnetic H, D and Z reference frame; see Howard et al. (1999) for further details. Corrected geomagnetic coordinates (CGM; http://nssdc.gsfc.nasa.gov/space/cgm/cgm.html) are used throughout this paper.

SAMNET (UK Sub-Auroral Magnetometer NETwork) is operated by the University of York, U.K., and comprises fluxgate magnetometers that during the interval considered here sampled the geomagnetic H, D and Z components each 5 s with a resolution of 0.25 nT. Further details on SAMNET appear in Yeoman et al. (1990).

The IMAGE and SAMNET data used in this paper were obtained from the respective home pages, from which further details, including station coordinates, are available. These addresses are and http://samsun.york.ac.uk/samnet_home.html. Both magnetometer arrays are being continually upgraded, adding more stations and improving timing accuracy. For example, SAMNET now samples each 1 s and uses GPS timing.

The ionosphere above northern Scandinavia and Svalbard can be studied with a number of instruments. One of these is the CUTLASS (Cooperative UK Twin Located Auroral Sounding System) bistatic HF coherent radar, with stations at Pykkvibær in Iceland and Hankasalmi in Finland (see Figure 1). The former is called the CUTLASS Iceland East radar, and since it looks across the magnetic meridian is used to identify features moving azimuthally. The Finland radar looks toward the magnetic pole and accordingly is better suited for examining equatorward or poleward moving features. Further details on CUTLASS are available from http://ion.le.ac.uk/cutlass/cutlass.html.

CUTLASS is a component of the international SuperDARN HF radar network (Greenwald et al., 1995). In standard operating mode these radars sweep over a 52^o azimuth sector using 16 evenly spaced beams. Each beam is gated into up to 75 range bins with spatial resolution of typically 45 km, giving a total field of view of order 3 x 10⁶ km². In standard mode, the integration time for each beam position is 7 s, and the cycle time 2.0 min.

The operation of HF radars was described by Greenwald et al. (1985). The radars detect echoes that are backscattered from decameter-scale field-aligned ionospheric plasma density irregularities. These are produced in the high latitude F-region by plasma drifts and density gradients (eg., Fejer and Kelley, 1980; Tsunoda, 1988) and drift with the ambient plasma motion (Villain et al., 1985). In each beam the HF radars measure the line-of-sight Doppler velocity, the backscattered power and the spectral width of the echo from these structures.

In addition to operation in standard modes, each SuperDARN radar also operates in discretionary modes for specific studies. The observations reported here were obtained when CUTLASS was operating in a non-standard scan mode optimised for high resolution studies of a specific target region. In this mode three adjacent beams are scanned with the central beam scanned again between each set. The Finland radar thus scanned beams 6, 7, 6, 5, 6, 7, 6, ... , and similarly the Iceland East radar scans were centered on beam 14. These beams were sampled each 7 sec and the first range gate was set at a distance of 180 km. The radio frequency was in the range 9.900 - 10.000 MHz. The central beams are represented in Figure 1 by broken lines, and their intersection region was in the F-region just east of Tromsø. The solid lines in Figure 1 represent the standard fan-shaped SuperDARN coverage.

An important feature of radars such as CUTLASS is measurement of the elevation angle of the backscatter returns. This is achieved using an interferometric technique and can provide considerable assistance in discriminating the direction of ground scatter and the altitude of irregularity structures (Milan et al., 1997a,b).

Information on the magnitude and structure of D-region absorption for the same location probed by CUTLASS was obtained from the IRIS (Imaging Riometer for Ionospheric Studies) experiment operated at Kilpisjärvi (KIL), Finland, by the University of Lancaster, U.K. IRIS examines ionospheric absorption of incoming galactic radio noise at a frequency of 38.2 MHz with 49 antenna beams in a 7 x 7 array and a sampling time of 1 s. The total field of view at 90 km altitude is of order 240 x 240 km. The experiment was described in detail by Browne et al. (1995), and data were obtained from http://www.dcs.lancs.ac.uk/iono/iris/. The location of KIL is also shown in Figure 1.

Solar wind data are obtainable from the WIND spacecraft. For the time of interest here this was located in the upstream solar wind near GSE(x, y, z) = (172.5, -5.6, 11.23) R_E. We used solar wind and magnetic field data from the SWE (Solar Wind Experiment) and MFI (Magnetic Fields Investigation) instruments respectively. Further information on WIND and its experiments is available from http://www-spof.gsfc.nasa.gov/istp/wind/, while the data used here may be accessed at http://cdaweb.gsfc.nasa.gov/.

General information on magnetospheric topology was obtained by reference to ion and electron flux data from the DMSP F12 spacecraft. This is one of a series of spacecraft in an approximately 830 km altitude sun-synchronous, 101 minute period polar orbit. Spectra of low energy ion and electron fluxes at high latitudes are produced by the SSJ/4 electrostatic analyzer instruments. Full details of the orbit and data, including spectrograms, are available at The use of DMSP particle spectra to determine the magnetospheric boundaries was discussed by Newell and Meng (1988, 1992) and Newell et al. (1989, 1991).

In this paper we focus on observations from 23 February 1996, when Pc5 type magnetic pulsations were recorded across the IMAGE and SAMNET magnetometer arrays, and corresponding Doppler oscillations occurred in the CUTLASS records. These pulsations lasted for several hours, during which K_p was 3+.

Figure 2. Stacked H component time series for selected IMAGE and SAMNET stations, 0900-1300 UT, 23 February 1996. (Larger version - 133kB)

Figure 2 shows stacked H component time series plots of the magnetometer data from selected IMAGE and SAMNET stations for the interval 0900-1300 UT on this day, bandpass filtered between 1 and 40 mHz. The D component plots are similar and therefore not shown. Higher resolution plots have also been inspected but for brevity are not presented here. Several features are apparent in the plots.

• (i). A large, transient bipolar event commenced at 0932 UT at all stations.
• (ii). This was followed for some hours by wave-like activity that was remarkably similar at all stations <74^o latitude, although amplitudes increased with increasing latitude.
• (iii). Signals were somewhat different at latitudes $74^o.
• (iv). Shorter period pulsations were superimposed on the large, long period variations at all stations.

We now consider each of these features in turn.

WIND observations are presented in Figure 3, and show that the solar wind speed was in the range V = 400 - 430 km.s^-1 throughout this time. This includes a small, sharp increase from ~410 to ~430 km.s^-1 at 0850 UT. More importantly, the ion density decreased from n ~ 20 cm^-3 at 0841 UT to 13 cm^-3 at 0852 UT and 6 cm^-3 at 0855 UT. This represents a decrease in solar wind ram pressure, nV², by a factor of 2 - 3. Given the upstream location of WIND, this negative pressure pulse should reach the magnetopause around 0930 - 0935 UT. The corresponding WIND IMF data show B_z turning negative around 0838 UT, with B_y increasing from -6 to +5 nT over 0845 - 0852 UT.

Figure 3. Upstream WIND spacecraft data, 0730 – 1030 UT, 23 February 1996. Upper three panels show solar wind velocity, then ion density, ion pressure, and the IMF. (Larger version - 74kB)

There are further pressure perturbations at WIND after 0950 UT and in particular near 1130 UT (decrease by ~25%), 1140 (increase by ~50%), 1150 (decrease by ~7%), 1200 (decrease by ~17%). These perturbations exhibit 10 min periodicity (~1.7 mHz frequency). A large positive excursion in B_z and B_x between about 1140 and 1150 UT at WIND may be connected with the sudden decrease in pulsation activity at the ground between 1230 and 1244 UT.

Pressure pulses of the magnitude observed here are believed to be fairly common (Sibeck, 1990) and have been associated with ringing type magnetic pulsations on the ground with periods of a few minutes (Takahashi et al., 1988; Farrugia et al., 1989; Sibeck, 1990), including field line resonances (Potemra et al., 1989; Warnecke et al., 1990; Parkhomov et al., 1998; Prikryl et al., 1998). Accordingly, it seems likely the bipolar event seen in the magnetometer records at 0932 UT is due to the negative solar wind pressure impulse. This is followed by some hours of similar pulsation activity across a wide range of latitudes.

Stacked power spectra for the ground magnetometer H components are presented for a representative interval in Figure 4. The spectra have been normalized and weighted by f^1.0 to better illustrate higher frequency features. Pulsations at a frequency of 1.6 mHz dominate at all but the two highest latitude stations. A secondary peak around 8 mHz also occurs at many stations.

Figure 4. Stacked normalized H component power spectra, 1000-1100 UT, weighted by f^1.0. (Larger version - 60kB)

We now consider the activity at the highest latitude stations, $74^o CGM latitude (LYR and NAL). The DMSP F13 spacecraft passed over the arctic region north of Scandinavia around 0847 UT on this day. Particle energies and fluxes characteristic of the cusp were seen around 80^o geomagnetic latitude, and boundary layer particles just equatorward of this. Therefore, the magnetometer signals at the highest latitude IMAGE stations are probably associated with closed auroral oval field lines. This is confirmed by the radar observations presented in the next section. It is therefore not surprising that the magnetometer signals at these latitudes appear more noisy and less ordered than at lower latitudes.

The field line eigenfrequency was determined from the cross-phase between pairs of adjacent IMAGE stations, using the method described by Waters et al. (1995). The variation in eigenfrequency with latitude at 1030 UT is shown in Figure 5. It is seen that the resonant frequency decreases smoothly with increasing latitude, although power spectra show that power at the resonant frequency is low equatorward of ~70^o. Parameters of the resonance were estimated as described by Menk et al. (1999). The resonance width in the ionosphere is of order 90 - 150 km and the resonance Q is 1.5 to >3.

Figure 5. Variation in field line resonance frequency determined using cross-phase measurements between adjacent IMAGE stations, 1030 UT, 23 February 1996. (Larger version - 55kB)

Figure 6. Variation in pulsation amplitude and phase with latitude, at a frequency of 1.6 mHz (upper panels) and 8 mHz (lower panels), 0934 UT. Solid squares represent the H component and open triangles denote the D component. (Larger version - 166kB)

According to Figure 5, the 1.6 mHz signal that is prominent in the time series and power spectra is associated with a field line resonance around 71 - 72^o latitude. Further evidence of this is presented in Figure 6, which shows the variation in amplitude and phase with latitude at a frequency of 1.6 mHz (upper panels) and 8 mHz (lower panels). These parameters were determined over 12 min intervals using complex demodulation (Beamish et al., 1979). The 1.6 mHz signal H component peaks in amplitude around 72^o latitude, and the phase of both H and D components undergoes a 180^o reversal around 72 – 73^o latitude. This is characteristic of a field line resonance, but cannot explain the origin of the 1.6 mHz signals at the other stations. The most likely scenario is that 1.6 mHz waves are propagating into and throughout the magnetosphere, and couple to field line resonances where the eigenfrequency is the same, but elsewhere drive forced field line oscillations.

The smaller, 8 mHz pulsations were present before and after the solar wind impulse. Figure 6 shows that they have maximum amplitude at the highest latitude stations, but exhibit a secondary peak in amplitude and a reversal in phase around ~66^o latitude. We therefore presume the 8 mHz pulsations are also connected with inward propagating waves, that in this case stimulate field line oscillations over at least 20^o in latitude, and couple to field line resonances near 66^o latitude.

HF radar data are typically presented in the form of colour whole-day range-time parameter plots for a particular beam, where the parameters of interest are the power level of the echoes, the line-of-sight velocity, the elevation angle, and the spectral width (in m.s^-1) of the signal. Examples may be seen at the CUTLASS homepage address given earlier. The range-time-velocity images are usually plotted over a scale of order ±1000 m.s^-1. Backscatter regions of small velocity and spectral width are normally regarded as representing ground scatter and are therefore suppressed in the plots. In the following plots we have not suppressed the low velocity information, and in fact the plots cover a restricted velocity range of ±10 or 15 m.s^-1.

Figure 7(a) presents grey scale range-time-velocity plots for beams 13, 14 and 15 from the Iceland East radar, from 0900 to 1500 UT. Note that the non-standard high resolution mode did not commence until 1000 UT. Figure 7(b) shows the corresponding range-time-velocity plots for beams 5, 6 and 7 of the Finland radar, over the same time span. The Iceland East plot range axis is calibrated in 45 km range gates (commencing at 180 km), while for the Finland radar range is in magnetic latitude. The figures show many complicated features. Milan et al. (1999) have presented plots of the Finland beam 5 power, velocity and elevation angle for this day, and gave a detailed discussion of the features observed. Below we summarize the essential features.

Figure 7. Range-time-velocity plots over 0900-1500 UT for (a) beams 13, 14 and 15 of the Iceland East radar over 0900 to 1500 UT, and (b) beams 5, 6 and 7 of the Finland radar. (Larger version - 902kB)

For the Iceland East radar:

• (i). Prominent bands of westward motion commenced around 0930 and extended over about 10 range gates (~450 km). These large scale flows showed a 10 min periodicity and lasted some hours, gradually moving westward, toward the radar. Line-of-sight velocities were around ±10 - 20 m.s^-1. Prikryl et al. (1998) reported similar flow features (but with much higher velocities) in HF backscatter data from the cusp region in connection with solar wind pressure pulses. They called these flow channel events (FCEs).
• (ii). A more detailed stacked velocity-time-range plot for beam 15 is presented in Figure 8(a). The velocity features are essentially simultaneous across the field of view.

  ---

  

  Figure 8a. Stacked velocity-time plots over 0900-1200 UT for several range gates from (a) beam 15 of the Iceland East radar, and (b) beam 5 of the Finland radar. (Larger version - 79kB)

  

  Figure 8b. (a) spectral width and (b) elevation angle for the Iceland East radar, 0900-1500 UT. (Larger version - 84kB)

  

  Figure 9a. Stacked spectra for selected range gates of (a) normalized power and (b) phase, for the Iceland East radar, 1020-1120 UT. Compare with magnetometer spectra in figure 4. (Larger version - 177kB)

  

  Figure 9b. Imaging riometer absorption maps for the interval 0920-1128 UT, with time moving from left to right then down the page. See text for details. (Larger version - 186kB)

  ---

• (iii). Range-time-power plots (not presented here) show that these bands were associated with enhanced power levels.
• (iv). Figure 9(a) shows the spectral width in the same format as Figure 7(a), while Figure 9(b) is a plot of elevation angle for beam 14. The spectral width from the region of interest is quite small, and elevation angle relatively high. These radar returns are therefore interpreted as ground backscatter (see Milan et al., 1997b).
• (v). Power and phase spectra of the velocity features (see Figure 9a) show a prominent peak at 1.6 mHz which has the same phase across at least 400 km in longitude.
• (vi). Plots for beams 13, 14 and 15 are similar although not identical.

For the Finland radar:

• (i). A prominent zone of high velocity poleward moving features around 76-80^o latitude at 0900 moving to 72-76^o latitude by 1300 UT (not shown in Figure 7b, but see Milan et al. 1999) was associated with high backscattered power and large spectral widths. This is the signature of the dayside auroral oval.
• (ii). Bands of equatorward moving features commenced ~0930 UT, extending from 66^o to >70^o in latitude. Line-of-sight velocities were of order ±20 m.s^-1.
• (iii). Figure 8(b) shows the velocity structures in more detail, for a number of range gates. The features obviously have very similar shape over 4 - 8 gates, ie. latitudinal extent of order 180 - 360 km.
• (iv). Similar results were obtained for the spectral width, elevation angles, power and phase spectra as for the Iceland East radar.

We emphasise that the features we are examining in the HF radar data are present in the ground backscatter, and are not returns from field-aligned irregularity structures in the usual sense.

Imaging riometer absorption maps for the interval 0920 - 1128 UT are presented in Figure 9b. The plots represent cosmic noise absorption in dB measured each 120 s, with time moving from left to right and down the page. Geographic north is to the top and geographic west to the left. The images clearly show two absorption features to the north and just east of centre from 0920 until 0930 UT. The general pattern changes after 0930, with lower levels of absorption across the field of view. Higher time resolution (15 s) images have also been examined.

The negative solar wind pressure pulse near 0930 UT is associated with a decrease in cosmic noise absorption, indicated by the decrease in intensity of the two prominent absorption patches. These patches reappear around 1000 UT, when B_y (IMF) goes negative, and again 12 min later, when B_y becomes positive again.

The magnetometer observations show that the negative pressure impulse heralded the appearance of 1.6 mHz pulsations over a wide range of latitudes. Latitude independent 8 mHz were also recorded across the station array, before and after the impulse, but with smaller amplitude that the 1.6 mHz oscillations. The most likely explanation of these observations is that fast mode waves were propagating into and through the magnetosphere, driving forced field line oscillations. Hasegawa et al. (1983) showed that such fast mode waves may also couple to resonant transverse oscillations, so that the period of the resultant pulsations depends on the latitude of the observation site. Although the 1.6 mHz pulsations dominated the time series, the cross-phase analysis summarized in Figure 5 demonstrates that latitude dependent field line resonances were also observed.

It is not clear whether the pulsations were associated with global resonances of the magnetospheric cavity, or if the period of the incoming waves is simply the rate at which the magnetopause was driven by the solar wind pressure pulses. Note that on this day pressure pulses were sometimes seen 10 - 12 min apart, corresponding to the pulsation frequency. Our magnetometer observations are similar to those reported by Prikryl et al. (1998), also in connection with periodic solar wind pressure perturbations. They studied a day where compressional MHD waves in the IMF associated with solar wind pressure pulses exhibited spectral peaks around 0.8, 1.3, 1.7, 2.0, 2.4 and 2.9 mHz and stimulated multifrequency compressional oscillations in the magnetosphere. These coupled to discrete field line resonances.

The most likely explanation of the low velocity bands we observed from both radars is that the radar beams were refracted through the ionosphere to the ground, then some power was backscattered from the ground through the ionosphere to the receiver, and each time whilst traversing the ionosphere the beam experienced a Doppler shift, for example in response to motion of the ionospheric plasma (eg. Milan et al., 1997b). In this case the actual range is half that indicated in the range-time-velocity plots. The corresponding regions in the ionosphere where the velocity oscillations are occurring are indicated by the solid shading in Figure 1. This is the first report of such backscatter features, and suggests a new mode for examining small Doppler shifts.

The Doppler oscillations commenced at the same time as the negative solar wind pressure pulse and occurred over a wide spatial extent. These results suggest that inward propagating waves associated with solar wind pressure pulses can drive motions of the ionospheric plasma. A similar conclusion was reached by Yeoman et al. (1997), who examined transient pulsations in the ionosphere. The notable feature of their work was that those pulsations were only observable because the ionosphere had been stimulated by an RF heater.

The present results demonstrate that small amplitude ionospheric oscillations are likely to be associated with impulse events even when their signature is not apparent in conventional ionospheric HF backscatter plots. For example, while Prikryl et al. (1998) reported FCEs with similar appearance to our velocity features in the F-region cusp, we have observed these features specifically at lower latitudes.

Based on the work of others (eg. Sibeck, 1990; Sibeck and Samsonov, 1994) we may expect solar wind impulses to be associated with quasi-periodic enhancements in cosmic noise absorption measured by the IRIS experiment. Instead, we found the cosmic noise absorption was reduced by the initial impulse. This suggests that the negative pressure pulse resulted in a magnetospheric expansion, decreasing the energetic particle flux density. Subsequent enhancements and motions of absorption features are associated with variations in the B_y component of the IMF. The convection flow speeds we observed in the HF radar data, of order " 10 m.s^-1, would not result in appreciable motion of the absorption features.

IMAGE data were obtained through L Häkkinen at the Finnish Meteorological Institute (FMI), SAMNET data from D. Milling at the University of York, and IRIS data through S. Marple at Lancaster University. The IMAGE PI is Ari Viljanen at FMI. WIND data were obtained from CDAWeb and provided by K. Ogilvie (SWE) and R. Lepping (MFI) at NASA GSFC. This work was supported by the Australian Research Council and the Particle Physics and Astronomy Research Council.

Beamish, D., Hanson, H. W. and Webb, D. C. (1979). Complex demodulation applied to Pi2 geomagnetic pulsations. Geophysical Journal of the Royal Astronomical Society 58: 471-493.

Browne, S., Hargreaves, J. K. and Honary, B. (1995). An imaging riometer for ionospheric studies. Electronics and Communication Engineering Journal 7: 209-217.

Farrugia, C. J., Freeman, M. P., Cowley, S. W. H., Southwood, D. J., Lockwood, M. and Etemadi, A. (1989). Pressure-driven magnetopause motions and attendant response on the ground. Planetary and Space Science 37: 589-607.

Fejer, B. G., and Kelley, M. C. (1980). Ionospheric irregularities. Reviews of Geophysics 18: 401-454.

Greenwald, R. A., Baker, K. B., Hutchins, R. A. and Hanuise, C. (1985). An HF phased-array radar for studying small-scale structure in the high-latitude ionosphere. Radio Science 20: 63-79.

Greenwald, R. A., Baker, K. B., Dudeney, J. R., Pinnock, M., Jones, T. B., Thomas, E. C., Villain, J.-P., Cerisier, J.-C., Senior, C., Hanuise, C., Hunsucker, R. D., Sofko, G., Koehler, J., Nielsen, E., Pellinan, R., Walker, A. D. M., Sato, N. and Yamagishi, H. (1995). DARN/SuperDARN: a global view of the dynamics of high-altitude convection. Space Science Reviews 71: 761-796.

Hasegawa, A., Tsui, K. H. and Assis, A. S. (1983). A theory of long period magnetic pulsations, 3, Local field line oscillations. Geophysical Research Letters 10: 765-768.

Howard, T. A., Menk, F. W. and Morris, R. J. (1999). Hilat Pc3-4. ANARE Research Notes, Australian Antarctic Division (this issue).

Lemaire, J. (1977). Impulsive penetration of filamentary plasma elements into the magnetosphere of the Earth and Jupiter. Planetary and Space Science 25: 887-890.

Lühr, H. (1994). The IMAGE magnetometer network. STEP International Newsletter 4: 4-6.

Lühr, H., Aylward, A., Bucher, S. C., Pajunpää, A., Pajunpää, K., Holmbee, T. and Zalewski, S. M. (1998). Westward moving dynamic substorm features observed with the IMAGE magnetometer network and other ground-based instruments. Annales Geophysicae 16: 425-440.

Lui, A. T. Y. and Sibeck, D. G. (1991). Dayside auroral activities and their implications for impulsive entry processes in the dayside magnetosphere. Journal of Atmospheric and Terrestrial Physics 53: 219-229.

Menk, F. W., Waters, C. L. and Fraser, B. J. (1999). Field line resonances and waveguide modes at low latitudes; 1. Observations. Journal of Geophysical Research (in press).

Milan, S. E., Jones, T. B., Robinson, T. R., Thomas, E. C. and Yeoman, T. K. (1997a). Interferometric evidence for the observation of ground backscatter originating behind the CUTLASS coherent HF radars. Annales Geophysicae 15: 29-39.

Milan, S. E., Yeoman, T. K., Lester, M., Thomas, E. C. and Jones, T. B. (1997b). Initial backscatter occurrence statistics from the CUTLASS HF radars. Annales Geophysicae 15: 703-718.

Milan, S. E., Yeoman, T. K. and Lester, M. (1999). The cusp region as a hard target for coherent HF radars. Annales Geophysicae (in press).

Newell, P. T. and Meng, C.I. (1988). The cusp and cleft/boundary layer: Low-altitude identifications and statistical local time variation. Journal of Geophysical Research 93: 14549-14556.

Newell, P. T., Meng, C.-I., Sibeck, D. G. and Lepping, R. (1989). Some low altitude cusp dependencies on the interplanetary magnetic field. Journal of Geophysical Research 94: 8921-8927.

Newell, P. T., Wing, S., Meng, C.-I. and Sigilleto, V. (1991). The auroral oval position, structure, and intensity of precipitation from 1984 onwards: An automated online data base. Journal of Geophysical Research 96: 5877-5882.

Newell, P. T. and Meng, C.I. (1992). Mapping the dayside ionosphere to the magnetosphere according to particle precipitation characteristics. Geophysical Research Letters 19: 609-612.

Parkhomov, V. A., Mishin, V. V. and Borovik, L. V. (1998). Long-period geomagnetic pulsations caused by the solar wind negative pressure impulse on 22 March 1979 (CDAW-6). Annales Geophysicae 16: 134-139.

Potemra, T. A., Lühr, H., Zanetti, L. J., Takahashi, K., Erlandson, R. E., Marklund, G. T., Block, L. P., Blomberg, L. G. and Lepping, R. P. (1989). Multisatellite and ground-based observations of transient ULF waves. Journal of Geophysical Research 94: 2543-2554.

Prikryl, P., Greenwald, R. A., Sofko, G. J., Villain, J. P., Ziesolleck, C. W. S. and Friis-Christensen, E. (1998). Solar-wind driven pulsed magnetic reconnection at the dayside magnetopause, Pc5 compressional oscillations, and field line resonances. Journal of Geophysical Research 103: 17307-17322.

Pu, Z.-Y. and Kivelson, M. G. (1983). Kelvin-Helmholtz instability at the magnetopause: solutions for compressible plasmas. Journal of Geophysical Research 88: 841-852.

Russell, C. T. and Elphic, R. C. (1979). ISEE observations of flux transfer events at the dayside magnetopause. Geophysical Research Letters 6: 33-36.

Saunders, M. A., Russell, C. T. and Sckopke, N. (1984). Flux transfer events: scale size and interior structure. Geophysical Research Letters 11: 131-134.

Sibeck, D. G. (1990). A model for the transient magnetospheric response to sudden solar wind dynamic pressure variations. Journal of Geophysical Research 95: 3755-3771.

Sibeck, D. G. and Samsonov, S. N. (1994). Transient CNA events and magnetospheric compressions. In: Baker, D. N., Papitashvili, V. O. and Teague, M. J. (Eds). COSPAR Colloquia 5, COSPAR. Pp 323-326.

Takahashi, K., Kistler, L. M., Potemra, T. A., McEntire, R. W. and Zanetti, L. J. (1988). Magnetospheric ULF waves observed during the major magnetospheric compression of November 1, 1984. Journal of Geophysical Research 93: 14369-14382.

Tsunoda, R. T. (1988). High-latitude F region irregularities: a review and synthesis. Reviews of Geophysics 26: 719-760.

Villain, J. P., Caudal, G. and Hanuise, C. (1985). A SAFARI-EISCAT comparison between the velocity of F region small-scale irregularities and the ion drift. Journal of Geophysical Research 90: 8433-8443.

Warnecke, J., Lühr, H. and Takahashi, K. (1990). Observational features of field line resonances excited by solar wind pressure variations on 4 September 1984. Planetary and Space Science 38: 1517-1531.

Waters, C. L., Samson, J. C. and Donovan, E. F. (1995). The temporal variation of the frequency of high latitude field line resonances. Journal of Geophysical Research 100: 7987-.

Yeoman, T. K., Milling, D. K. and Orr, D. (1990). Pi2 polarization patterns on the U.K. Sub-Auroral Magnetometer Network (SAMNET). Planetary and Space Science 38: 589-602.

Yeoman, T. K., Wright, D. M., Robinson, T. R., Davies, J. A. and Rietveld, M. (1997). High spatial and temporal resolution observations of an impulse-driven field line resonance in radar backscatter data artificially generated with the Tromsø heater. Annales Geophysicae 16: 634-644.