Similarities and differences of sawtooth events and steady magnetospheric convection periods

Transcription

1 Similarities and differences of sawtooth events and steady magnetospheric convection periods N. Partamies, T. I. Pulkkinen, R. L. McPherron, K. McWilliams, C. Bryant, H. J. Singer, G. Reeves, and M. Thomsen Department of Physics and Astronomy, University of Calgary, Calgary, Alberta, Canada Finnish Meteorological Institute, Space Research Unit, Helsinki, Finland University of Los Angeles, California, USA University of Saskatchewan, Saskatoon, Canada NOAA, Space Environment Center, Boulder, Colorado, USA Los Alamos National Laboratory, NM, USA Manuscript submitted to Annales Geophysicae Manuscript-No.??? Offset requests to: N. Partamies Institute for Space Research University of Calgary Calgary Canada

2 Annales Geophysicae,,, SRef-ID: -/ag/-- European Geosciences Union LOGO Similarities and differences of sawtooth events and steady magnetospheric convection periods N. Partamies, T. I. Pulkkinen, R. L. McPherron, K. McWilliams, C. Bryant, H. J. Singer, G. Reeves, and M. Thomsen Department of Physics and Astronomy, University of Calgary, Calgary, Alberta, Canada Finnish Meteorological Institute, Space Research Unit, Helsinki, Finland University of Los Angeles, California, USA University of Saskatchewan, Saskatoon, Canada NOAA, Space Environment Center, Boulder, Colorado, USA Los Alamos National Laboratory, NM, USA Abstract. Both sawtooth events and steady magnetospheric convection (SMC) events are known to occur during strong to moderate solar wind driving. While sawtooth events are characterized by recurring substorm-like activity in the ionosphere, the SMC events are periods with elevated but rather steady auroral electrojet activity. This study presents a superposed epoch analysis of a set of sawtooth events and SMC periods from the years. We examine the similarities and differences of these two types of events by studying the solar wind driving, the ionospheric response of the auroral electrojets (AL), the ring current (ASY-H/SYM-H), and the polar cap potential and convection, as well as the magnetospheric response of the geostationary particle fluxes. This study demonstrates that the solar wind driving, as well as the ionospheric activity, is clearly stronger for the sawtooth events than for the SMC periods. The geostationary particle injections appear strong and widespread during the sawtooth events but very weak in association with the substorms initiating the SMCs. This suggests that substorms prior to the SMC periods are less intense and that magnetotail processes associated with SMCs rarely reach the geostationary orbit. The monthly occurrence of the SMCs is fairly even compared to that of the sawtooth events which is concentrated around the equinoxes. For the subset of events for which the solar wind motional electric field (or the Kp index) was chosen to be at the same moderate level, the superposed epoch analysis of the sawtooth and SMC events revealed strong similarities for all parameters but the solar wind speed and the electrojet index. Our results suggest that during otherwise similar driver conditions, low and steady wind speed results in steady convection while faster wind speed causes stronger and more periodic activity in the magnetosphere and ionosphere. Correspondence to: N. Partamies Introduction Steady magnetospheric convection (SMC) events, or convection bays, are periods during which the driving solar wind is particularly steady and the ionospheric convection is enhanced but substorm activity is not observed (McPherron et al., ; Sergeev et al., ). The solar wind is typically rather slow, and the interplanetary magnetic field (IMF) is moderate and stable. The minimum duration of an SMC event is typically required to be hours (Sergeev et al., ), which is clearly longer than the duration of an average substorm, but comparable to or slightly longer than the typical time ( hours) between recurring substorms (Borovsky et al., ). McPherron et al. () suggest that SMCs are periods when reconnection near the sub-solar region at the dayside magnetopause is balanced by tail reconnection in the nightside a scenario that was earlier speculated by Pytte et al. in. An SMC event often begins with a substorm (e.g. McPherron et al., ) that feature is thought to be related to the pre-conditioning of the magnetosphere for the steady convection (Sergeev et al., ). O Brien et al. () suggested that the solar wind and IMF also have an important effect in the pre-conditioning process, because prior to an SMC event the magnetosphere is usually moderately driven while the magnetosphere prior to an average substorm is often quiet. Most SMC periods also end with a substorm (e.g. McPherron et al., ). Many recent studies agree that SMCs are a specific group of events with a distinct response of the magnetosphere to the solar wind driving (e.g. Sergeev et al., ). Instead of using the steady solar wind as a selection criteria, O Brien et al. () identified their events from the magnetospheric data, in order to establish that the characteristic solar wind during the SMC events really is steady, even if it is not required by the selection criteria. A double auroral oval is typically observed during SMCs (Yahnin et al., ). The poleward boundary of the oval is often dominated by active and discrete aurora, some of which can be identified as poleward boundary intensifica-

3 Partamies et al.: Steady magnetospheric convection and sawtooth events tions (PBIs, Lyons et al., ), or auroral streamers, implying transient earthward plasma flows in the central plasma sheet (CPS) (e.g. Zesta et al., ). Unlike in the case of substorms, the aurora during a typical SMC does not include a surge. Qualitatively, the auroral features observed during SMCs are similar to those seen during a substorm recovery phase. Sawtooth events have also been reported as a separate class of magnetospheric activations. These activations are largeamplitude oscillations of energetic particle fluxes at geosynchronous orbit, recurring with a period of about hours (e.g. Henderson et al., ). The events typically occur during a geomagnetic storm when the solar wind driving is strong and the IMF is continuously southward for an extended period of time. A characteristic of these events is that the geosynchronous magnetic field can become highly stretched not only in the midnight sector but also in the evening sector reaching all the way to the dusk meridian (Pulkkinen et al., ). This can be observed as a reduction of the magnetic field inclination at geosynchronous orbit over a wide range of local time sectors, as well as in the strongly enhanced partial ring current as measured by the ASY-H index. A recent study by Pulkkinen et al. () presented a statistical comparison of the typical solar wind driver conditions and ionospheric activity of sawtooth events and substormlike auroral electrojet activations during geomagnetic storms. They concluded that sawtooth events are not a specific type of magnetic activity, and that the hour periodicity, strong stretching of the dusk sector field and strongly asymmetric ring current are also found in association with other types of storm-time activations. Furthermore, they demonstrate that the level of driving is very similar during the sawtooth events and other storm-time activations, while the auroral activity (AL index) is slightly lower in case of the sawtooth events. In this paper, we perform analysis similar to that of Pulkkinen et al. () to the SMC events to quantify the differences between sawtooth events and SMC intervals in both the driving conditions and the ionospheric and magnetospheric activity. Event and data descriptions. Event classification We used a data set of sawtooth events ( ) compiled by R. L. McPherron. Sawtooth events were visually identified as recurring, relatively dispersionless particle injections observed by multiple geosynchronous satellites in multiple magnetic local time sectors (Pulkkinen et al., ). In the present study we consider individual sawteeth as separate events. The solar wind and IMF conditions for a sequence of sawtooth events on October, are presented in Figure. Figure contains the ground indices as well as the energetic electron fluxes at geostationary orbit for the same day. These km/s cm /npa TW mv/m : : : : : : : : UT [hh:mm] Fig.. Solar wind measurements during the sawtooth events on October,. Panels from top to bottom are: IMF B X (black) and B Y (blue) components; IMF magnitude (black) and B Z component (blue); solar wind speed; solar wind density (black) and dynamic pressure (blue); ɛ parameter; dawn dusk electric field; and Alfvén Mach number. The vertical red lines mark the times of individual sawtooth onsets. sawtooth events occurred in the middle of a magnetic storm, which has been studied in detail by Pulkkinen et al. (). SMC events were identified in AE index data from the years. These SMC years were selected to correspond to the sawtooth event years ( ) to reduce bias due to variations in the solar wind driving conditions. An automated selection routine was developed, which required that during the SMC period the auroral electrojet activity is at the level of AE > and that the AL index is changing at a rate slower than per min (dal/dt > /min) (O Brien et al., ; McPherron et al., ). The threshold for the AE index has been chosen so that the auroral activity IMF Bx/By IMF B /Bz Speed Density/Pdyn Epsilon Ey Ma

4 Partamies et al.: Steady magnetospheric convection and sawtooth events kv Diff. flux Diff. flux : : : : : : : : UT [hh:mm] Fig.. Magnetospheric data for the sawtooth events on October,. Panels from top to bottom are: Auroral electrojet AU (upper) and AL (lower) indices; cross-polar cap potential; northern polar cap (PCN) index; asymmetric (ASY-H, upper) and symmetric (SYM-H, lower) ring current indices; energetic electron fluxes from the LANL instrument onboard geostationary spacecraft LANL-A in the evening sector and - in the morning sector. The vertical red lines mark the the times of individual sawtooth onsets. AU/AL PC potential PCN index SYM H/ASY H LANL A level is well above the quiet time values. The AL gradient restriction is applied to eliminate substorm occurrence (abrupt decreases of AL) during the steady convection events. In addition, we required these two criteria to hold for at least three hours. We found SMC events during the four-year period. Since most SMCs begin with a substorm, the substorm onset was chosen to be the reference time for our analysis. Each automatically found AL index curve was also visually inspected in order to find the onset of the substorm that initiates the SMC period. The onset was defined as an abrupt decrease of at least in the AL index within three hours from the beginning of the SMC period. As a result, events were discarded because a clear substorm could not be identified prior to the SMC. In a few cases, there was a clear substorm signature in the middle of the SMC, and these were also excluded from the analysis. In those cases where a substorm onset was found within an SMC, the AL index gradient was /min, which is below the threshold and therefore selected by the automated search routine. An example of observations made during an SMC period is shown in Figures and. Figure contains the solar wind and IMF conditions and the ground-based indices during an SMC event on May,, and geostationary orbit energetic electron measurements for the same event are shown in Figure.. Data sets The same data sources and analysis methods were used for both the sawtooth events and for the SMC events. For the SMC events and sawtooth events, the IMF and solar wind parameters were examined to determine the driving conditions. The ionospheric activity is characterized by the auroral electrojet index (AL), symmetric and asymmetric ring current indices (SYM-H and ASY-H), northern polar cap (PCN) index and the cross-polar cap (PC) potential. The magnetospheric behaviour is characterised by the geostationary orbit energetic electron fluxes. The solar wind parameters from Solar Wind Electron Proton Alpha Monitor (SWEPAM) instrument (McComas et al., ) and the IMF from the MAGnetic field experiment (MAG) (Smith et al., ) instrument, both on board the Advanced Composition Explorer (ACE) satellite were examined. ACE is located in the L point roughly R E upstream of the Earth. All the ACE data have been propagated to the magnetopause (to the distance of R E upstream of the Earth) using the upstream distance of the satellite from the magnetopause and the average solar wind speed during the interval of interest. We use the IMF X, Y and Z components and the total field magnitude as well as the solar wind number density, dynamic pressure and speed to calculate the epsilon parameter (Akasofu, ), the dawn-to-dusk electric field (E Y = V X B Z ), as well as the Alfvén Mach number.

5 Partamies et al.: Steady magnetospheric convection and sawtooth events km/s cm /npa TW mv/m... : : : : : : : : UT [hh:mm] Fig.. Solar wind data for the SMC event on May,. Panels from top to bottom are: IMF B X (black) and B Y (blue) components; IMF magnitude (black) and B Z component (blue); solar wind speed; solar wind density (black) and dynamic pressure (blue); ɛ parameter; dawn dusk electric field; and Alfvén Mach number. The vertical red line marks the substorm onset and the vertical green lines bracket the SMC period. IMF Bx/By IMF B /Bz Speed Density/Pdyn Epsilon Ey Ma kv Diff. flux Diff. flux : : : : : : : : UT [hh:mm] Fig.. Magnetospheric data during an SMC event on May,. Panels from top to bottom are: Auroral electrojet AU (upper) and AL (lower) indices; cross-polar cap potential; northern polar cap (PCN) index; asymmetric (ASY-H, upper) and symmetric (SYM- H, lower) ring current indices; energetic electron fluxes from the LANL instrument onboard geostationary spacecraft LANL-A and - (in the noon and afternoon sectors, respectively). The vertical red line marks the substorm onset and the vertical green lines bracket the SMC period. AU/AL PC potential PCN index SYM H/ASY H LANL A

6 Partamies et al.: Steady magnetospheric convection and sawtooth events The ɛ parameter is defined as ɛ = π µ V SW B l sin (Θ/), () where µ is the vacuum permeability, V SW is the solar wind speed, B is the magnitude of the IMF, l = R E is an empirical scaling parameter, and tan(θ) = B Y /B Z determines the IMF clock angle. All variables in the equation are given in GSM coordinates. The Alfvén Mach number M A is given as the ratio between the solar wind speed and the Alfvén speed: M A = V SW /V A = V SW B ρµ, () where the Alfvén speed is given by V A = B/ ρµ, where ρ is the solar wind mass density. The global AL index is used whenever available. For those events for which the global AL is not yet available (sawtooth events in ), a quasi-al index was calculated from the magnetic recordings of the ground-based networks IM- AGE (Viljanen et al., ) in Fennoscandia and CARISMA (old CANOPUS, Rostoker et al., ) in central and western Canada. Thus, the quasi-al only records substorm activity in and around these two sectors. The symmetric and asymmetric parts of the ring current are described by SYM- H and ASY-H indices, respectively. They are calculated as weighted averages (SYM-H) and maximum differences (ASY-H) of mid-latitude stations around the globe (Iyemori, ; Sugiura and Kamei, ). The PCN index is constructed from magnetic recordings at the Thule station located within the polar cap in Greenland (Troshichev et al.,, ). This index is generally used as a proxy for the ionospheric convection and thus, the reconnection rate. PC potential values were estimated from the Super Dual Auroral Radar Network (SuperDARN, Greenwald et al., ) measurements. Spherical harmonics are fitted to the recorded convection velocities to produce a smooth convection map over each hemisphere. The difference between the maximum and minimum voltages in the convection pattern is used as an estimate of the cross-polar cap potential (Ruohoniemi and Baker, ). The better the data coverage, the more reliable the fit. We examined separately the average behaviour of the PC potential for those SMC periods for which the average number of the data points in the convection pattern was greater than. Since the superposed epoch potential of this subset is similar to the one suggested by the entire data set we conclude that the polar cap potential values presented in this study are rather reliable. The geostationary orbit electron fluxes were obtained from the Los Alamos National Laboratory satellites. Synchronous Orbit Particle Analyzer (SOPA) data from the geostationary satellites -, -, -, LANL-A, LANL-A, and LANL-A (Belian et al., ) and used to monitor the magnetospheric activity. Electron flux enhancements and injections were examined in the energy range from to kev, corresponding to typical substorm injection energies. The temporal resolution of the ground-based indices is one minute, except the PC potential which is calculated once every two minutes. For all ACE parameters we use -second data. Superposed epoch analysis for SMC events The substorm onset preceding an SMC event was marked as the zero epoch time, and we examine a time period from two hours before to six hours after the onset. The superposed epoch values are median values of the single event curves at each epoch time. The IMF components and the total magnetic field measurements in Figure (black lines) do not contain significant variations and are moderate in magnitude. There is a slight decrease in the IMF B Z about one hour prior to the onset time that may be related to energy loading prior to the substorm. During the SMC period (starting hours after the zero epoch time) the IMF is typically steady and B Z is negative. The median values from the epoch analysis of the solar wind parameters in Figure (black lines) vary only slightly. The solar wind speed of km/s, pressure of. npa and density of cm agree with the typical average values of the solar wind. The average epsilon parameter varies between and GW, which is small but still around and above the substorm loading threshold of GW (Akasofu, ). The epsilon values do not exceed the storm threshold of GW for any individual event. Because the magnetic field components (Figure ) are small, the Alfvén Mach number is typically high for the SMCs, but still lower than an average solar wind Mach number of about (obtained assuming a typical Alfvén speed of km/s). In agreement with the decrease of the IMF B Z, the electric field increases somewhat within the hour before the substorm, but settles to a level of. mv/m for the steady convection period. The ground-based index epoch analysis results are shown in Figure (black lines). The AL index contains a clear substorm signature between the epoch hours and, which reflects the selection criterion of the substorm onset to define zero epoch time. After the subtorm recovery, the AL values remain steady between and. Similarly, ASY- H and PCN increase during the substorm and remain slightly elevated over the SMC period, while PC potential and SYM- H are fairly constant throughout the entire epoch. The higher values of the PCN hint for increased convection during the SMC events. The results of the superposed epoch analysis of the geosynchronous particle fluxes for three local time sectors, evening sector ( MLT), midnight sector ( MLT) and morning sector ( MLT) are shown in Figure. Although the zero epoch is chosen to be the substorm onset prior to the steady convection interval, there is only a slight change in the midnight sector fluxes, while a slightly larger flux increase is observed in the morning sector, which is strongest in the lowest-energy channel. This would seem

7 Partamies et al.: Steady magnetospheric convection and sawtooth events Bx npa cm Density Pdyn By km/s Speed : : : : : : : : Epoch time [hh:mm] Btot Bz mv/m TW... : : : : : : : : Epoch time [hh:mm] Epsilon Ey Ma Fig.. Superposed epoch analysis results. Panels from top to bottom are: IMF B X; B Y ; B Z; and the IMF magnitude B T OT. The SMC events are shown in black and the sawtooth events in blue. The vertical red line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. Fig.. Superposed epoch analysis results. Panels from top to bottom are: Solar wind density; dynamic pressure; solar wind speed; ɛ-parameter; electric field; and Alfvén Mach number. The SMC events are shown in black and the sawtooth events in blue. The vertical red line marks the zero epoch, which is defined as the groundbased substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events.

8 Partamies et al.: Steady magnetospheric convection and sawtooth events SMC Sawtooth kv : : : : : : : : Epoch time [hh:mm] AL PC potential PCN ASY H SYM H e-flux e-flux e-flux - Epoch time [hours] Epoch time [hours] Fig.. Superposed epoch analysis results of the geosynchronous electron fluxes at MLT, MLT, and MLT for the SMC events (left panels) and for the sawtooth events (right panels). The electron fluxes are shown in the range kev in units of /cm /s/sr/kev. The vertical line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. - MLT - MLT - MLT Fig.. Superposed epoch comparison of SMC and sawtooth events. Panels from top to bottom are: AL index, PC potential; PCN index; ASY-H index; and SYM-H index. The SMC events are shown in black and the sawtooth events in blue. The vertical red line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. to indicate that the substorms initiating the SMC periods are not very strong, are not associated with significant electron energization, and rarely intrude into the geostationary orbit. No dipolarization signatures were evident in the superposed epoch results of the magnetic field inclination at the geostationary orbit (data not shown). Comparison to sawtooth events The zero epoch time in this study refers to the substorm onset for the SMC periods and to the individual sawtooth onset for the sawtooth events. We use an epoch interval of eight hours:

9 Partamies et al.: Steady magnetospheric convection and sawtooth events from two hours before the onset until six hours after the onset for each and every event (except for the geosynchronous data for the sawtooth events). The epoch curves are the median values of the single event data curves, which are of the same length even though the event itself may be longer or shorter. The blue curves in Figures,, and are the superposed epoch results for the sawtooth events. Both the solar wind plasma, the IMF, the ground-based indices, and the geostationary particle injections are clearly stronger during the sawtooth events. In addition to the sawtooth events being more strongly driven, the ground-based indices also show signs of a higher level of fluctuations. To quantify the fluctuations we applied the solar wind turbulence equations of Borovsky and Funsten (): δb = σb X + σb Y + σb Z () σb X + σb Y + σb Z δb B = () B where σ is the standard deviations for each IMF component, and B is the average magnitude of the IMF. Each standard deviation, δb and IMF magnitude was calculated for each event separately, and then averaged over the entire epoch period. For the sawtooth events the amount of fluctuations was characterized by δb =. and δb/b =. (Pulkkinen et al., ). The corresponding values for the SMC periods are δb =. and δb/b =.. The level of fluctuations is smaller for the SMCs than for the sawtooth events but the difference is not large. A larger difference between the two event categories is that while the magnitude of fluctuations varies from sawtooth to sawtooth (δb.. ), the range of fluctuations for the SMCs is smaller by a factor of two (δb.. ). Due to the smaller values of the IMF, the Alfvén Mach number is higher for the SMC events than for the sawtooth events. The solar wind driving is almost as steady during the sawtooth events as it is during the SMC events, but the speed, epsilon parameter, reconnection electric field and IMF B Z are clearly larger. It is also interesting to note that IMF B Y is typically negative for the SMCs and positive for the sawtooth events. This implies that also the IMF clock angle appears to be in the opposite phase for the two event groups. In the statistical convection pattern (Weimer, ) for the positive IMF B Y the evening cell extends over the midnight meridian to the dawn side. The coupling efficiency between the solar wind and the ionosphere parametrized by an AL-to-E Y ratio has also been calculated for the two event groups (for each event separately, and a superposed epoch analysis for the full data set). Although the sawtooth events are more strongly driven than the SMCs (E Y values of.. mv/m and.. mv/m, respectively), the coupling efficiencies for both event types gives very similar average values ( AL /EY ). In strong contrast to the SMC events, the sawtooth events are associated with (and defined by) very strong geostationary orbit particle injections. The electron fluxes for the sawtooth events in three different time sectors are given in the right panels of Figure. It can be seen that the fluxes prior to the sawtooth events are similar to those measured during the SMC events, but the strong geosynchronous particle injections at sawtooth onsets are seen at all local time sectors nearly simultaneously. This highlights the fact that the sawtooth events are associated with processes that occur over a wide local time sector in the near-geostationary region. As a measure of the global activity and magnetospheric convection (Thomsen, ), we present the distribution of the Kp indices during steady magnetospheric convection (light grey) and sawtooth events (dark grey) in Figure. For the sawtooth events, the Kp value is the three-hour value closest to the individual sawtooth onset. For the SMCs, the Kp indices are selected from the middle of the steady convection period. The peak Kp values for the SMCs are, while the ones for the sawtooth events range from. The tails of these distributions do overlap, but the average activity, and thus convection level, during the sawtooth events is roughly double compared to the activity during the SMCs. PCN increases by about one unit at the sawtooth onset. There is an increase of this index at the substorm onset prior to the SMCs but it is only about half-a-unit. In case of the sawtooth events, both PCN and AL indices show signs of -hour periodicity that has been discussed in more detail in a recent paper by Pulkkinen et al. (). The oscillating ionospheric activity is an interesting contrast to the steady solar wind driving. Another clear difference between these two event groups is the seasonal variation in their occurrence. This is demonstrated by the occurrence of the events per month in Figure. While the SMC events (light grey bars) are most frequent during the summer months (April to June), the sawtooth events (dark grey bars) are mainly observed in the autumn (August to November). The occurrence of the sawtooth events is in agreement with the seasonal behaviour of the geomagnetic activity and occurrence of the aurora (e.g Nevanlinna, ): minima around the solstices in July and December, maxima close to the equinoxes in March and September. This semi-annual variation has often been explained by the Russell-McPherron mechanism, where the activity occurrence maxima correspond to the periods of largest component of the IMF being antiparallel with the dayside geomagnetic field (Russell and McPherron, ). The summer maximum of the SMCs is most likely caused by the fixed threshold value of the AE index in selection criteria. During summer months the ionospheric conductivity is higher, the auroral electrojets are stronger, and the AE index threshold is more frequently exceeded. These data sets indicate that the sawtooth events are more concentrated around periods of highest geomagnetic activity during the equinoxes, while the SMC events have a more even distribution throughout the year (especially if the winter/summer asymmetry is corrected for).

10 Partamies et al.: Steady magnetospheric convection and sawtooth events SMC events SMC events Sawtooth events Sawtooth events Number of events Number of events Kp index Month Fig.. Distribution of the global activity in terms of Kp index. The light grey bars represent the Kp values recorded in the middle of the SMC periods, while the dark grey bars show the Kp values at the time of the sawtooth events. Fig.. Occurrence of the SMC and sawtooth events as a function of the month of the year. The light grey bars represent the monthly distribution of the SMC periods, while the dark grey bars show the monthly distribution of the sawtooth events. Effect of the driving solar wind electric field We analysed a subset of events, those in which SMCs and sawtooth events were most alike. The data sets were divided into subsets according to the average values of E Y and Kp. The average values for the SMCs include data from the entire steady convection periods, while the averages for the sawtooth events include data from the period from half an hour before to half an hour after the individual sawtooth onsets. We selected a subset of events where the solar wind motional electric field is roughly similar for the SMC and sawtooth events, as well as constant:. mv/m E Y. mv/m. This results in a set of SMCs and sawtooth events, whose epoch plots are shown in Figures, and. For this subset of sawtooth events both IMF, solar wind, and ground-based parameters are lower than what is typical for the entire set of sawtooth events. For this subset of SMCs, most of the parameters are higher than what is typical for the entire set of SMCs. This reflects the role of the solar wind electric field in setting the level of activity. The average Kp index values for the subsets of constant E Y are also very similar:. for the SMCs and. for the sawtooth events, compared with the corresponding mean values of. and., respectively, of the full data sets. The solar wind and IMF data (Figures and ) reveal interesting differences between these two subsets of sawtooth and SMC events. While the driving electric field is almost the same for both data sets (a selection criterion), the constituents of the electric field (V X and B Z ) are not. The solar wind speed is around km/s for the SMC events, while it is

11 Partamies et al.: Steady magnetospheric convection and sawtooth events Bx : : : : : : : : Epoch time [hh:mm] Fig.. Superposed epoch analysis results for a subset of SMC events (black) and sawtooth events (blue) for which. mv/m E Y. mv/m. Panels from top to bottom: IMF B X; B Y ; B Z; and the IMF magnitude B T OT. The vertical red line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. typically above km/s during the sawtooth events. Therefore, it seems that the magnetosphere is more stable when the solar wind is slow, and the higher speed drives a more dynamic magnetosphere. If we use Kp index as a proxy for the ionospheric convection (Thomsen, ) then setting Kp constant implies steady ionospheric convection. We therefor also examined subsets of events for which the average Kp index is at the same level: Kp between and. The superposed epoch curves of the resulting SMC periods and sawtooth events (not shown) appear very similar. In particular, this results in similar PC potential and the northern polar cap index values for the sub- By Bz Btot cm npa km/s TW mv/m... : : : : : : : : Epoch time [hh:mm] Fig.. Superposed epoch analysis results for a subset of SMC events (black) and sawtooth events (blue) for which. mv/m E Y. mv/m. Panels from top to bottom: Solar wind density; dynamic pressure; solar wind speed; ɛ-parameter; electric field; and Alfvén Mach number. The vertical red line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. Density Pdyn Speed Epsilon Ey Ma

12 Partamies et al.: Steady magnetospheric convection and sawtooth events sets of sawtooth events and SMCs. The driving force of the convection, E Y, is also in the same range for the two event types when Kp is constant. Discussion kv : : : : : : : : Epoch time [hh:mm] Fig.. Superposed epoch comparison of a subset of SMC events (black) and sawtooth events (blue) for which. mv/m E Y. mv/m. Panels from top to bottom: AL index; PC potential; PCN index; ASY-H index; and SYM-H index. The vertical red line marks the zero epoch, which is defined as the ground-based substorm onset for the SMC periods and the geostationary orbit injection onset for the sawtooth events. AL PC potential PCN ASY H SYM H In this study we have analysed the solar wind, as well as indication of ionospheric and magnetosheric activity, for a set of steady magnetospheric convection periods. The results were compared with the level of activity during nearly the same number () of sawtooth events. We used superposed epoch analysis to describe the typical behaviour of the two different types of events. It is interesting to note that although the energy input from the solar wind to the magnetosphere during SMC periods exceeds the substorm threshold of GW (Akasofu, ), the magnetopheric convection remains steady and no substorms occur. As a comparison, the energy input during the sawtooth events is typically around the storm threshold of GW, and sawtooth events usually occur embedded in geomagnetic storms. The minimum of three-hour duration required for the SMC events is still comparable to the recurrence time of substorms, but it also seems to be a typical length of an SMC event in our data set. Over two thirds (%) of the events have lifetimes of three-to-four hours, and only about % of the events last longer than five hours. Since our epoch interval is hours, and no periodicity is observed in the superposed epoch curves, the steady convection events occur during periods that remain steady for longer than a typical substorm cadence. More SMC events are identified during the northern summer months (see Figure ), when a fixed AE value (AE < ) is used as a selection criterion. During summers the ionospheric conductivity is high, auroral electrojets are stronger, and thus the AE threshold is more likely to be exceeded. In that sense, the distribution presented here may not describe the true SMC occurrence throughout the year especially if one wants to compare the event numbers from the summer and winter months. To even out the event distribution, the AE limit value should be adjusted according to the month of the year by imposing a changing AE limit depending on the average level of activity during a given month. This problem has also been discussed by McPherron et al. () and should be examined in more detail in the future. The gradient threshold for AL of greater than /min seems to eliminate most of the substorms, but a gradient of /min is clearly large enough for a few substorm onsets to occur in the middle of steady convection periods. This is a difficult criterion to adjust, because similar auroral and magnetic features may appear during SMCs as well as stronger activity periods. The magnetic signature of a PBI can have a substorm onset-like transient structure and a very rapid decrease of AL. Furthermore, the AL index is the lower envelope of the magnetic signature of the auroral electrojet,

13 Partamies et al.: Steady magnetospheric convection and sawtooth events and thus any single strong variation may have a significant contribution to the momentary index value. The characteristics of the SMC events presented here are in agreement with a previous SMC study by McPherron et al. (), both in average magnitude and in the steady behaviour, although they analysed more than events occurring over a ten-year period ( ). This favourable comparison suggests that our results are representative of a wide range of geomagnetic conditions, including solar cycle variations. In this analysis, we do not require the SMC events to end with a substorm, although a substorm at the beginning of the period is included in the search criteria. If the end of the SMC event was followed by substorm activity within two hours, the event was considered to end with a substorm. Visual examination of the individual AL curves was performed to determine whether or not the event ends with a substorm. As a result, events (%) were found to end with a substorm, while, for the rest of the SMC events, the activity just died out after the steady convection period. About the same number of discarded events, for which we did not find the substorm prior to the event, end with a substorm. Nearly half of the SMCs in the previous study by McPherron et al. () were also reported to end with a substorm. A clear difference in the substorms initiating steady convection intervals and the onsets of the sawtooth events is their location in the magnetotail. While the sawtooth oscillations occur nearly instantaneously in the magnetotail over a wide range of local times, as observed by geostationary satellites (Henderson et al., ), the superposed epoch curves for the SMC intervals indicate that the substorm activity prior to the SMC rarely reach geostationary orbit and that the injection energies are typically quite low. We also searched for solar wind and IMF triggers for the substorms that occurred prior to the SMC events. This was done by calculating -min averages of all parameters before and after the onset to avoid the uncertainty in timing. If the change of the average values before and after the onset was or more for IMF B Z, or larger than npa for the solar wind dynamic pressure the event was considered to be triggered. Only % of the substorms prior to the SMC periods were associated with a solar wind trigger, and this trigger was generally a small one. As a comparison, in % of the sawtooth events some kind of a trigger could be found. Thus, fewer SMC events han sawtooth events were associated with solar wind triggering. In addition to the entire data sets of SMCs and sawtooth events, we also analysed a number of subsets. It is interesting to note that one of the clearest IMF differences between SMCs and sawtooth events is the opposite sign of B Y (negative for SMCs and positive for sawtooth events, see Figure ). However, when subsets of average B Y > and average B Y < were analysed separately for both event groups, the results (not shown) are very similar to those of the full data sets (Figures ). This suggests that although the sign of IMF B Y rotates the ionospheric convection pattern, it does not have a significant effect on the ionospheric activity seen in the superposed epoch analysis. Other subsets that gave results similar to the entire event sets were: a subset of moderate solar wind speed ( km/s < V SW < km/s), a subset of high solar wind speed (V SW km/s), and a subset of moderately negative IMF B Z (-. < B Z < -. ). Subsets of SMC events and sawtooth oscillations consisting of moderate electric field (. mv/m E Y. mv/m) reveal interesting differences in the solar wind parameters: While the E Y values are similar, the solar wind speeds are different, with SMC events being associated with significantly lower speeds than the sawtooth events. In the ionosphere, the PC potentials are almost the same, while the AL activity is larger for the sawtooth events. This would indicate that low, steady solar wind speed drives steady convection while faster solar wind speed, associated with less negative IMF B Z leads to a sequence of loading unloading cycles in the magnetotail. These results are consistent with a simulation study by Pulkkinen et al. (b). They ran the Lyon Fedder Mobarry global magnetohydrodynamic simulation code for an SMC event and then repeated the runs by enhancing the driving E Y, first by increasing the IMF magnitude by % and then by increasing the speed by %. While increasing the IMF strength did not lead to qualitative changes in the magnetospheric behaviour, increasing the speed led to destruction of the steadiness of the convection pattern and periodic bursts of activity. Thus, it seems that the solar wind speed plays a major role not only in setting the level of convection as part of the electric field, but also in determining the type of activity that follows. Summary and conclusions We have examined the similarities and differences between the superposed epoch results of steady magnetospheric convection periods and sawtooth events. The superposed epoch analysis of all events show that the level of solar wind driving is higher and more turbulent for the sawtooth events than for the SMCs. For the SMC events, the energy input from the solar wind to the magnetosphere (the ɛ parameter by Akasofu ()) is at the substorm level ( GW) without substorms taking place, while in the sawtooth case the energy input is around the storm level ( GW). However, the average coupling efficiency defined by an AL-to-E Y ratio has a similar value of about for both event types. The ionospheric activity is also stronger for the sawtooth events than for the SMCs. The planetary Kp index, which can be used as a proxy for the ionospheric convection, is typically twice as high for the sawtooth events ( ) than for the convection periods ( ). A clear contrast between the two event groups is seen in the geostationary electron fluxes. A strong particle injection in multiple time sectors almost simultaneously is typically related to the sawtooth events, while only small changes in the low energy channels can be observed in the midnight and morning sector fluxes in case of the SMCs. Clear dipolar-

14 Partamies et al.: Steady magnetospheric convection and sawtooth events ization signatures usually occur together with the sawtooth particle injections, but they are not observed during the SMC intervals. These findings suggests that the substorms prior to the SMC periods are not intense and that the injections very seldom reach geostationary orbit. The monthly occurrence of the sawtooth and SMC events demonstrates a very different behaviour. While sawtooth events tend to take place around the equinoxes when the geomagnetic activity has its maximum, the SMC events are distributed more evenly throughout the year. The summer maximum in the SMC occurrence probably results from the constant AE threshold in the selection criteria. For moderate and roughly constant average solar wind motional electric field (. mv/m E Y. mv/m), we obtained subsets of SMC and sawtooth events. The results of the superposed epoch analysis for these subsets turned out very similar. However, the solar wind speed for the sawtooth events ( km/s) was higher than for the SMC events ( km/s). In the ionosphere, the AL index drops about further for the sawtooth events than for the SMCs. This suggests that the solar wind speed is the key parameter in determining the level of magnetospheric and ionospheric activity. Slow wind speed appears to support steady convection in the magnetosphere even when the energy input from the solar wind is large enough to produce substorm activity. Fast solar wind, on the other hand, drives a more disturbed magnetosphere. This finding is in agreement with a recent modelling results by Pulkkinen et al. (b). Acknowledgements. The work of N.P. was funded by the Academy of Finland, Alberta Ingenuity Fund, and Canadian Space Agency. We thank ACE PI teams for the MAG and SWEPAM data. ASY- H/SYM-H and AE indices where downloaded from Kyoto Geomagnetic Data Services, and the PCN data from the Danish Meteorological Institute web site. We thank Robin Barnes for providing the PC potential values based on SuperDARN radar measurements. Support for RLM was provided by NSF GEM grant ATM--, NASA grant NNG-GAG, and the Center for Integrated Space Models(CISM)through a subcontract to the University of Colorado at Boulder ATM-. References Akasofu, S.-I., Energy coupling between the solar wind and the magnetosphere, Space Sci. Rev.,,,. Belian, R. D., G. R. Gisler, T. Cayton, and R. Christensen, High- Z Energetic Particles at Geostationary Orbit During the Great Solar Proton Event Series of October, J. Geophys. Res.,,,,. Borovsky, J.E., Nemzek, R.J. and Belian, R.D., The occurrence rate of magnetospheric substorm onsets: random and periodic substorms, J. Geophys. Res.,,,. Borovsky, J.E., and Funsten, H.O., Role of solar wind turbulence in the coupling of the solar wind to the Earth s magnetosphere, J. Geophys. Res.,, doi:./ja,. Greenwald et al., DARN/SuperDARN, A global view of the dynamics of high-latitude convection, Space Sci. Rev.,,,. Henderson, M.G., Reeves, G.D., Skoug, R., Thomsen, M.F., Denton, M.H., Mende, S.B., Immel, T.J., Brandt, P.C., and Singer, H.J., Magnetospheric and auroral activity during the April sawtooth event, J. Geophys. Res.,, AS, doi:./ja,. Iyemori, T., Storm-time magnetospheric currents inferred from mid-latitude geomagnetic field variations, J. Geomag. Geoelectr.,,,. Lyons, L. R., Nagai, T., Blanchard, G. T., Samson, J. C., Yamamoto, T., Mukai, T., Nishida, A., and Kokubun, S., Association between Geotail plasma flows and auroral poleward boundary intensifications observed by CANOPUS photometers, J. Geophys. Res.,,,. McComas, D.J., Bame, S.J., Parker, P., Feldman, W.C., Phillips, J.L., Riley, P., and Griffee, J.W., Solar Wind Electron Proton Alpha Monitor (SWEPAM) for the Advanced Composition Explorer, Space Sci. Rev.,,,. McPherron, R.L., O Brien, T.P. and Thomson, S.M., Solar wind drivers for steady magnetospheric convection, in Multiscale Coupling of Sun-Earth Processes, edited by A.T.Y. Liu, Y. Kamide, and G. Consolini,,. Nevanlinna, H., Results of the Helsinki magnetic observatory, Ann. Geophys.,,,. O Brien, T.P., Thompson, S.M., and McPherron R.L., Steady magnetospheric convection: Statistical signatures in the solar wind and AE, Geophys. Res. Lett.,,./GL,. Pulkkinen, T.I., Ganushkina, N. Yu., Tanskanen, E.I., Kubyshkina, M., Reeves, G.D., Russell, C.T., Singer, H.J., and Slavin, J.A., Magnetospheric current systems during stormtime sawtooth events, J. Geophys. Res.,, AS, doi:./ja,. Pulkkinen, T.I., Partamies, N., McPherron, R.L., Henderson, M., Reeves, G.D., Thomsen, M.F, and Singer, H., Statistical analysis of stormtime activations and sawtooth events, J. Geophys. Res.,, A, doi:./ja,. Pulkkinen, T. I., C. C. Goodrich, and J. G. Lyon, Solar wind electric field driving of magnetospheric activity: Is it velocity or magnetic field?, submitted to Geophys. Res. Lett.,. Pytte, T., McPherron, R.L., Hones, E.W., Jr.,and West, H.I., Jr., Multiple satellite studies of magnetospheric substorms: distinction between polar magnetic substorms and convection-driven negative bays, J. Geophys. Res.,,,. Rostoker, G., Samson, J. C., Creutzberg, F., Hughes, T. J., Mc- Diarmid, D. R., McNamara, A. G., Vallace Jones, A., Wallis, D. D., and Cogger, L. L, CANOPUS - A ground based instrument array for remote sensing in the high latitude ionosphere during ISTP/GGS program, Space Sci. Rev.,,,. Ruohoniemi, J. M. and Baker, K. B., Large-scale imaging of highlatitude convection with Super Dual Auroral Radar Network HF radar observations, J. Geophys. Res.,,,. Russel, C.T. and McPherron, R.L., Semiannual variation of geomagnetic activity, J. Geophys. Res.,,,. Sergeev, V.A., Pellinen, R.J. and Pulkkinen, T.I., Steady magnetospheric convection: a review of recent results, Space Sci. Rev.,,,. Smith, C. W., Heureux, J. L., Ness, N. F., Acuna, M. H., Burlaga, L. F., and Scheifele, J., The ACE magnetic fields experiment, Space Sci. Rev.,,,. Sugiura, M., and Kamei, T. Equatorial Dst index, in: IAGA bulletin, edited by A. Berthelier and M. Menville, ISGI Publ. Off., Saint Maur,. Thomsen, M.F., Why Kp is such a good measure of the

15 Partamies et al.: Steady magnetospheric convection and sawtooth events magnetospheric convection, Space Weather,, S, doi:./sw,. Thomsen, M.F., McComas, D.J., Reeves, G.D., and Weiss, L.A., An observational test of the Tsyganenko (Ta) model of the magnetospheric field, J. Geophys. Res.,,,. Troshichev, O.A., Kuznetsov, B.M. and Dmitrieva, N.P., Polar cap magnetic activity as a signature of substorm development, Planet. Space Sci.,,,. Troshichev, O. A., Lukianova, R., Yu.,Papitashvili, V. O., Rich, F. J., and Rasmussen, O., Polar cap index (PC) as a proxy for ionospheric electric field in the near-pole region, Geophys. Res. Lett.,,,. Viljanen, A. and Häkkinen, L., IMAGE magnetometer network. In: Satellite-Ground Based Coordination Sourcebook, edited by M. Lockwood, M.N. Wild and H.J. Opgenoorth, ESA publications SP-,,. Weimer, D., Models of high-latitude electric potentials derived with a least error fit of spherical harmonic coefficient, J. Geophys. Res.,,,. Yahnin, A.G., Malkov, M.V., Sergeev, V.A., Pellinen, R.J., Aulamo, O., Vennerström, S., Friis-Christensen, E., Lassen, K., Danielsen, C., Craven, J.D., Deehr, C., and Frank, L.A., Features of steady magnetospheric convection, J. Geophys. Res.,,,. Zesta, E., Loyns, L.R., Donovan, E., The auroral signature of earthward flow bursts observed in the magnetotail, Geophys. Res. Lett.,,,.