Pages 609-620

GROUND-BASED SUPPORTING PROGRAMMES FOR THE IASTP

Hermann J. Opgenoorth

Swedish Institute of Space Physics, Uppsala Division, S-75991 Uppsala, Sweden
(also visiting professor at: Finnish Meteorological Institute, Geophysical Research Division, Box 503, FIN-00101 Helsinki, Finland) , E-mail:opg@irfu.se

ABSTRACT

During the coming years there will be an unprecedented opportunity to study structure and dynamics of the magnetosphere. A fleet of specialised spacecraft will very much increase our possibilities to simultaneously monitor various magnetospheric regions. However, in spite of the local character of many magnetospheric processes, for example dayside reconnection and nightside substorms, they must still be considered as global features with respect to the effects that they have on magnetospheric energy storage, transport, and release. In addition most magnetospheric disturbances appear to occur in sporadic sequences at locations not a-priori predictable. Consequently even co-ordinated data from many and multiple satellites will depend on additional information from global networks of instrumentation on the ground. In this paper the preparations of the "ground-based" scientific community for the support of the IACG fleet are reviewed and perspectives are outlined for promising studies using multiple satellites and co-ordinated ground-based experiments.

INTRODUCTION

One of the main tasks of the InterAgency Solar Terrestrial Physics Program (IASTP) is the understanding of those physical processes which govern the flow - i.e. exchange and transport - of energy, mass and, momentum between the solar wind, the magnetosphere and the atmosphere, from its ionised topside down to the neutral atmosphere and biosphere. For this purpose an unprecedented fleet of national and international spacecraft - which are to large extent described in other articles in this issue - has been constructed and put into orbit. Even after the tragic loss of the Cluster satellites, during the coming years most of the key regions of the magnetosphere will be regularly monitored by many satellites in a co-ordinated fashion. However, any attempt to understand the global solar-wind / magnetosphere / ionosphere system would be incomplete without complementing measurements "onboard the very central satellite in this system", namely planet Earth itself. Today still only ground-based measurements can achieve an almost global coverage of measurements. They can provide continuous and simultaneous measurements from locations, which via magnetic field-lines map out to a vast region of space. A comparative coverage is intrinsically impossible with in-situ measurements from orbiting platforms.

Co-ordinated space and ground-based measurements have therefore been included in the concept of the IASTP right from the beginning. It is expected that ground-based measurements can complement the satellite observations by:

Understanding these important possibilities, ground-based investigators, representing many different types of instruments, have organised their methods of operation and the retrieval, presentation and dissemination of data in ways as to maximise the common return from the IASTP missions during the coming number of years. New key instruments have been constructed, respectively replaced or modernised. Relatively basic instruments such as e.g. magnetometers have been co-ordinated in international networks, and additional instruments were included as to widen the global coverage of such data. Common ways of data presentation (both in summary and final form) and their collection into central data bases has been organised by specially dedicated working groups and active individuals. Two working groups have been most prominent in this initiative, in the US a common Global Geospace Science (GGS) database was created at NSSDC (originally initiated by M. Teague; WWW-URL:http://iacg.org/iacg/ground_stations/ground_stations.html) and in Europe the working group for Cluster ground-based co-ordination ( Opgenoorth, 1993; Opgenoorth and Lockwood, 1996) created a database within the World Data Centre at the Rutherford Appleton Laboratories, UK (WWW-URL: http://www.gbdc. rl.ac.uk/gbdc/gbdc.html). It was recently decided that even after the loss of the four Cluster spacecraft this working group will continue its task for the benefit of IASTP and in preparation for an expected Cluster recovery mission. Both data centres in the US and in Europe are to large extent complementary, and will often make cross reference via WWW tools (or create links to the original location of the data at the various PI-groups) rather than actually duplicate the storage of summary data. It should be noted here that all original WWW home pages of the instruments and instrument arrays described below can be found and accessed via either of the two URLs above.   

The working groups and individuals behind these two data-centres are not only responsible for the collection, presentation and dissemination of the data, but also for the identification of favourable satellite conjunctions with key ground-based instrumentation, and the consequent planning and operation of suitable operational modes of the instruments. Both data centres therefore provide tools and dialogue forms, as to allow any interested scientist to participate in this extensive co-ordination. Any input is most welcome.

In the following we will not be able to give a complete overview of all ground-based data available and gathered in these two key data centres. The reader is instead advised to contact the given WWW-URL's for more detailed information, or to refer to the Cluster Ground-Based Co-ordination Source book, which will soon appear as an ESA SP issue. Instead, below we will give a concise overview of the scientific potential of some key instruments - probably with a slight bias for the European sector. We will also in short describe the format and content of the various datasets available. Most of the investigators of key instrumentation or networks of instruments provide summary data in an easily understandable and usable fashion, such as to enable even inexperienced users to take advantage of the benefits of co-ordinated multi-instrument data. For this purpose various types of quicklook data, but even more refined data presentations, have been designed, and will be available on the WWW. A new family of geomagnetic disturbance indices is provided in order to allow a quick IASTP event selection or alternatively, data categorisation for the purpose of statistical studies. Note that in the majority of cases these summaries are only meant for the purpose of scrutinising the availability, quality and information content of the data in question. For any more advanced use it is expected that a co-operation with the responsible investigator (or at least someone out of a group of Co-investigators) is initiated in order to obtain verified and more refined data for a complete study.

INSTRUMENTATION

Both data centres provide lists and maps of available ground-based instrumentation in an updatable manner, so instead of reproducing the momentary status here, the reader is advised to refer to any of the URLs above for the most recent status of ground-based instrumentation.

Magnetometers

A magnetometer, measuring the magnetic field-variations created by ionospheric and field- aligned current systems, is probably one of the cheapest, but throughout the last 200 years most reliable instrument for the study of ionosphere/magnetosphere coupling. Latest since the major advance of space physics during the International Magnetospheric Study (IMS) the power of magnetometer networks for the understanding of global and local three-dimensional current systems has been recognised. Nowadays the use of magnetometer data has been very much refined, and powerful analysis tools have been developed to derive the distribution of ionospheric and field-aligned currents, potential patters, Joule heating, etc. (Kamide, Richmond and Matsushita, 1981; Richmond and Kamide, 1988). Magnetometer networks are today operated mainly in four regions of the globe. Canada and northern US (CANOPUS and MACCS), Greenland (Danish Meteorological Institute network and MAGIC), Fenno-Scandinavia (IMAGE) and Antarctica (AGONET) are the key networks in which ground-based magnetometers are organised. Throughout Russian Siberia a ring of magnetometers, mainly at auroral latitudes, is operated in a number of co-operative international projects. Since the coverage of magnetometers in both hemispheres is constantly changing, but basically growing, the reader is referred to the up-dated maps at the two IASTP data-centres.

Individually each of the basic networks has the possibility to provide data in a number of different presentations: stacked magnetograms, vector plots, latitude profiles of magnetic disturbances etc. However as a common overview format most of the networks provide data summaries of their magnetic recordings in the form of latitude-versus-time plots of electrojet intensity (or location of maximum) for one or several key network meridians. In addition they also provide local magnetic AE-type indices, which are basically derived from the envelope of the maximum magnetic deflection observed at any of the stations in the network. The index is split up in an upper part of envelope (eastward electrojet, locally corresponding to global AU index) and a lower part of envelope (westward electrojet, locally corresponding to global AL index). As Kauristie et al. [1996 a, b] have shown, local magnetic indices are not only available much faster than global indices, but contain for localised individual studies of events occurring close to magnetic midnight, more exact information on the magnetospheric disturbance state than the global index, which is derived from quite coarsely spaced stations with virtually no latitude coverage.

In addition to these local indices the Cluster Ground-Based Data Centre (GBDC) at RAL has developed a new family of geomagnetic disturbance indices (Opgenoorth and Lockwood, 1996; Opgenoorth et al., 1996; WWW-URL http://www. gbdc.rl.ac.uk/gbdc/ovals/plots). For quick analysis and event selection, or rapid statistical studies of IASTP events these new indices might be able to replace the standard AE geomagnetic index (Kamei and Maeda, 1981). The basic idea of this new family of indices is based on an attempt to improve both the timely availability and latitude coverage of the AE index. For many applications the exact energy state of the magnetosphere or the occurrence of disturbances during extremely quiet and very disturbed times is of interest. However, magnetic data from auroral latitudes only is bound to miss most of these types of magnetic disturbances. As the IASTP and the original Cluster mission objective aim at all kinds of magnetospheric energy states and disturbances, Opgenoorth et al. (1996) have designed a family of three geomagnetic AL-type of indices. These are derived from magnetometer data along three rings of approximately constant geomagnetic latitude in the northern hemisphere, corresponding to the locations of the Cusp or Contracted auroral Oval, the Standard auroral Oval, and the Expanded auroral Oval and consequently named CO, SO, and EO-index. Such index data from three different latitude regions will allow the identification of the magnetospheric energy state and the individual occurrence of very small or very large disturbances even during times of magnetospheric quiescence and storms. The timely availability of these indices for IASTP studies is improved by special agreements concerning rapid data delivery from the various involved principal investigators, and by an automated base-line subtraction and quality check. This method is admittedly less refined than the method used to derive the original AE index, however, it is sufficiently good to make the new indices useful for event selection, event timing and a general estimate of the magnetospheric energy state. In order to allow early studies even with an incomplete index data-base, we have also decided to make a preliminary (local) index available on the WWW as soon as data from at least two stations per ring has been received. Figure 1 gives an example of how such a preliminary index can look like, when data from only a few stations per ring have become available (the chosen example is from Christmas Eve, 1995, when Interball had a very interesting conjunction with Scandinavia). The longitudinal location of the contributing stations are indicated in the lower panel of each index plot. The thickness of the availability bar, indicates the station that contributes to the maximum envelope, i.e. provides the index for that particular time. In this example it can clearly be seen how a substorm, originating from the auroral oval between 1800 and 2000 UT expands to the high latitude region, and is reactivated again inside the auroral oval after 2100 UT. At 1000 UT an auroral oval disturbance has no effect on the high latitude stations, but is noticeable at subauroral latitudes, and at around 0500 UT there is a small disturbance, restricted to the high latitude region only. With the help of these new indices studies and event selections can be started more or less immediately, and the final index will improve together with the study, as more and more data comes into the data base (see Opgenoorth et al. (1996), Opgenoorth and Lockwood (1996), or the above URL for more details).

Concerning operational planning magnetometers need no further attention, since they are in permanent operation. For dedicated IASTP key periods, enhanced service and/or rapid recovery of failures is anticipated by most of the responsible investigators.

Optical and Other Imaging Instruments

An even more classical method of ground-based observations of magnetosphere ionosphere coupling is the visual imaging of auroral forms with different types of camera systems. Nowadays most of the earlier analogue cameras have been replaced by modern digital cameras, which makes data handling, analysis and reduction much easier. (During autumn 1996 also the Fenno-Scandinavian network of All Sky Cameras, operated by the Finnish Meteorological Institute in Helsinki is replaced with new digital cameras at locations in Finland, Sweden and Norway (Svalbard)). Optical auroral cameras are operated in all parts of the auroral zones and polar caps in both hemispheres, with less density, but roughly the same density distribution as magnetometers. Prime concentrations are again in Canada, Greenland, Scandinavia and Antarctica, with probably an extreme concentration on and around Svalbard, primarily because it is one of the few northern hemisphere locations to monitor dayside aurora in the vicinity of the cusp (again see the two basic URLs above for more details.)

Optical camera systems are naturally dependent on darkness, absence of cloud cover and the occurrence of aurora. Original image data is very memory demanding and for practical purposes a reduction of the raw data is necessary. One typical basic form of summary information is a record of availability of auroral data, which is provided by most of the investigators as a catalogue, often with some indication of the type or brightness of the aurora observed. Another, scientifically much more useful set of summary data is a so-called " keo-gram ", where the auroral recordings along the central (or another suitable) meridian through the camera field-of-view are plotted versus latitude and time. In this form the data resembles data from meridian scanning photometers, which is another often used technology to make quantitative auroral measurements, often using different filters for characteristic auroral emission lines. An example of a typical keogram, showing two consecutive events of equatorward expansion of the auroral oval and subsequent expansion of substorm aurora is presented in Figure 2. In particular in co-ordination with data from a meridional chain of magnetometers and/or other summary data representation from different types of radars (see below) auroral keograms have a very important information content for the understanding of magnetospheric activity, and are an excellent tool to understand the encounter of magnetospheric boundaries and/or processes of magnetosphere/ionosphere coupling by a satellite.

Fig. 1. Example plot of the Contracted, Standard and Expanded Oval indices. Below the index plots all stations contributing to the index are indicated. The station that contributes to the envelope function of the actual index is marked with a heavier bar. Plots based on roughly this number of stations are typically available one month after the date in question and will subsequently be upgraded as data from further stations are received.

Fig 2. Example of an auroral Keogram, representing auroral observations along a north south meridian versus latitude and time. This particular data example stems from a test operation of the new Finnish digital All - Sky Camera in Kiruna, Sweden. The scale is inverted, i.e. dark zones represent auroral emissions.

As in the case of magnetometer networks the original investigators can for a final study provide several different kinds of refined data sets, such as full movies of the auroral development and rectified maps of the location and altitude extent of auroral forms (see e.g. the Auroral Large Imaging System, ALIS, in Sweden). It is also possible to overlay optical data with data from other ground-based networks.

With the exception of ALIS, which is an instrument array operated in certain campaign modes and requires special scheduling, most of the auroral equipment is operated at all dark and cloud free hours, which makes special campaign planning unnecessary, except for special service alerts under IASTP prime campaigns.

In addition to auroral imagers, which monitor regions of particle precipitation with energies between a few hundred eV to a few tens of keV, a new type of imaging riometer (Relative-Ionospheric-Opacity-meter, which measures the ability of the ionosphere to absorb cosmic radio noise) can with an multitude of reception beams create an image of the ionospheric absorption over an area of several 100 by several 100 km (Stauning et al., 1995, and references therein). Since ionospheric absorption is created by either particle precipitation which ionises the ionospheric D-region (energies of several tens to hundreds of keV), or ionospheric instabilities created by very strong electric fields such instruments can add valuable information to other two dimensional ground-based datasets. The methods of data reductions and refinements are similar to the ones described for the auroral cameras. Imaging riometers are operated at locations in Canada, Greenland and Svalbard, and more instruments will be employed during the coming years. Check the ground-based URLs for future updates.

Both imaging and traditional integrating riometers are operated in a permanent mode of operation.

Coherent Radar Systems

It is well known that ionospheric plasma-irregularities of a multitude of amplitudes and wavelengths can either be caused by auroral particle precipitation or are driven by strong electric fields. When illuminated by HF or VHF radio waves, these irregularities can give rise to coherent wave reflection, leading to detectable radar echoes. Both the amplitude and the Doppler shift of the backscattered radio wave contains information on plasma parameters such as electron density and irregularity drift velocity in the region of backscatter. During the IMS a new type of coherent backscatter radar has been one of the most successful experiments. Utilising multiple radar beams from two positions the Scandinavian Twin Auroral Radar Experiment, STARE, (Greenwald et al., 1979) became one of the key ground-based experiments during the late 70's.

However, the use of radio waves in the VHF frequency range, which are sensitive to E-region irregularities, restricted the usefulness of such radars to the auroral zones up to latitudes, where the straight propagating radio waves could still be perpendicular to the magnetic field-lines. In order to cover even higher latitudes and a wider field-of-view a new system of bi-static HF radars has recently been developed. In the Dual Auroral Radar Network (DARN) pairs of HF radars utilise ionospheric refraction of the radar wave to achieve perpendicularity to magnetic field-aligned F-region irregularities in much wider regions and at much higher latitudes than the earlier VHF systems. As a result one pair of radars can monitor F-region plasma drifts over a region of 1000s of square kilometres.

In a global co-operation project involving Canada, Finland, France, Japan, South Africa, Sweden, the UK, and the US, three such radar pairs now cover almost half of the polar regions of the northern hemisphere, and three overlapping radars cover almost the complete Antarctic continent (Greenwald et al., 1995). At least two additional radar pairs, one overlooking western Canada and one overlooking a southern hemispheric region conjugate to the Svalbard area are in the planning phase (see the URLs for maps of the viewing ranges of the existing and planned radar systems). This so-called SuperDARN network allows for the first time to monitor truly instantaneously large scale convection patterns and their dynamics in response to changes in the IMF and associated magnetospheric disturbances.

As the amount of data derived from these advanced radar systems is huge, again certain methods of data reduction are applied, and basic overview data is provided for IASTP users. The most concise summary data set, which is derived and provided for each individual radar via WWW, is illustrated in Figure 3. As in the earlier datasets of magnetometer and optical data the form of presentation is "plasma parameter versus latitude and time", and the chosen SuperDARN variables along the central meridian of each individual radar field of view are backscattered power, line-of-sight Doppler velocity and spectral width of the returned echo (top, central, and bottom panel of Figure 3, respectively).

This summary data set gives a first indication of the availability of data, and contains a coarse information about the observation of prevailing convection cells. It can also hint whether the auroral cusp was observable during that particular experiment, as cusp echoes are characterised by relatively wide spectral shapes. More refined datasets, which are available on request from SuperDARN Principal or Co-investigators, are latitude-versus-time plots of vectors of plasma velocity along the central meridians for each radar pair, or maps of convection vectors either for individual radar pairs or the merged total vector field for each hemisphere. The reader should refer to the basic URLs or the article by Greenwald et al. (1995) for examples of such refined data and the exact procedure to initiate a co-operative case study with members from the SuperDARN community.

As the radars are relatively versatile instruments they are not run in the same basic mode every day, even though the aim is to operate the radars in a permanent manner. About 50 percent of the time the radars are run in a well defined pre-determined common mode, typically well suited for the derivation of velocity vector maps. During the remaining 50 % of the time the radars can be operated in special experiment modes, either in a globally co-ordinated fashion or individually. These special modes are operated on request from SuperDARN investigators and can e.g. aim to derive data with higher time resolution, alternatively higher spatial resolution, at the expense of other parameters. Some experiments may be specially designed to serve only a very limited scientific interest, with no or little value for the purposes of the IASTP.

This means that the operation of the SuperDARN network needs to be scheduled for periods of prime IASTP satellite conjunctions, and other more general campaign periods. Such scheduling is usually taken care of by the SuperDARN community itself, however the IASTP working groups have the possibility of influencing the schedule. Again the interested reader is referred to the basic URLs for more information on the monthly SuperDARN schedule (both before and after the fact). At the same WWW location dialogue windows provide the IASTP participants with possibilities to take active part in the adjustment of the schedule to IASTP requirements that may have been overlooked by the co-ordinating working groups.

Fig. 3: Example of a daily overview plot from an individual SuperDARN radar station, giving a summary of range-versus-time observations of backscatter power (top panel), line-of-sight velocity (central panel) and spectral width (bottom panel). This data example comes from the western station (Hankasalmi, Finland) of the European radar pair CUTLASS, which is overlooking Svalbard.

Incoherent Radar Systems

For ionospheric remote sensing from the ground the most refined and most modern equipment is probably the incoherent scatter radar. These radars utilise high power transmission and high sensitivity reception in the VHF and UHF frequency ranges. They make use of very faint scatter of radio waves from naturally occurring thermal plasma waves in the ionosphere. The returned echoes from various altitudes in the ionosphere contain information on several plasma parameters simultaneously, such as electron density, electron and ion temperature, and ion velocity. Using a few assumptions even ion composition, ionospheric conductivities and other plasma parameters can be derived as secondary results. Often the radars use complicated pulse-codes to gain spatial and temporal resolutions that are optimised for the altitude range investigated and fitted to the phenomenon studied. Antenna scans, multiple beams, beam-swinging and/or tri-static measurements from several sites are often utilised to derive two or three-dimensional vectors of the ionospheric plasma velocities. Incoherent scatter radars are so versatile, that only specialists can make full use of the instrument's capabilities. However, through years of experience a limited number of experiment modes (or combination of modes) have proven to be particularly well suited for multi-purpose experiments in co-ordination with magnetospheric satellites. For co-ordinated observations within the IASTP only a few well defined general purpose experiments will be carried out by the incoherent radars.

Fig. 4. Mode 1 (left panels): Topview (a) and sideview (b) of the EISCAT and ESR antenna pointing directions proposed for observations in association with satellite passages through the inner or outer cusp close to Svalbard. In the upper panel (topview) the magnetic meridians through Tromsö and Longyearbyen are indicated by dotted lines. The dotted circle marks the field-of-view of an All-Sky-Camera at Longyearbyen, for 630,0 nm emissions at 250 km altitude. The broken line indicates the viewing direction for the sideview in the bottom panel. We also indicate the combined field-of-view of the European pair of SuperDARN radars, CUTLASS. The filled square marks the altitude, resp. location, of the UHF tristatic electric field measurement. Mode 2 (right panels): As Mode 1, but for EISCAT / ESR observations within the nightside auroral oval and magnetotail regions.

The European Incoherent Scatter (EISCAT) facility in northern Scandinavia consists of three incoherent radar systems. On the Scandinavian mainland there is the tri-static UHF system (930 MHz) with a fully steerable transmitting and receiving antenna in Tromsö, Norway, and two remote receiving antennas in Kiruna, Sweden and Sodankylä, Finland. In Tromsö there is an additional VHF system (224 MHz) with a large antenna, which can be steered in elevation along a north-south meridian. The antenna can also be used to create two beams, separated either in elevation or in a limited azimuthal range. The new EISCAT Svalbard radar (ESR, 500 MHz) has a fully steerable mono-static receiver/transmitter antenna, and compared to the mainland UHF system it allows much faster scan motions. In the near future the EISCAT Svalbard radar will be upgraded with a doubled transmitter power (from 0.5 to 1 MW) and a second antenna (see the EISCAT WWW pages via the data centre URLs for more details).

To optimise the use of the combined EISCAT facilities the working group for Cluster Ground-Based Co-ordination has designed two basic combinations of experiment modes, which allow a good coverage of both polar cap and auroral latitude processes, for either day or nightside studies. The basic geometries for such combined EISCAT ESR experiments are sketched in Figure 4.

In mode 1 (left panels of Figure 4), which is dedicated to satellite conjunctions in the vicinity of the dayside cusp, the ESR and mainland UHF radars perform field-aligned experiments, which are best suited to derive complete altitude profiles of ionospheric plasma parameters, allowing a determination of the complete effect of precipitating particles in the ionosphere along one and the same field-line. This mode is also best suited to understand ionospheric responses such as particle heating and ionospheric ion outflow to the magnetosphere. The UHF system will use its tri-static capability to derive the three-dimensional ion drift velocity vector at one altitude along the same field line with high temporal resolution (indicated by the square on the field-aligned Tromsö beam u). Towards the south of the ESR a split-beam VHF experiment (denoted v1,v2) will provide two-dimensional ion drift vectors between the two field-aligned profiles of ionospheric parameters. In this experiment the motion of auroral features can be monitored by the detection of regions of high altitude electron heating, which is typically caused by auroral precipitation (see an example in Figure 5a below).

In mode 2 (right panels of Figure 4), which is dedicated to nightside satellite conjunctions, the ESR radar is used to derive in principle the same parameters as the VHF experiment in mode 1, by utilising a southward directed beam-swing experiment (L1,2). The combined field-aligned UHF and vertical VHF experiment of the mainland system is optimised to monitor the dynamics and precipitation characteristics of auroral forms. The combined experiment in mode 2 will allow one to monitor substorm development in the auroral oval, and its consequent expansion towards the open/closed field-line boundary, which should be located somewhere inside the ESR split beam range.

For IASTP experiments EISCAT will provide overview data with one minute time resolution for 24 hour periods. In Figure 5 an example is given for data from a combined VHF and UHF experiment of the type as depicted in Mode 1 of figure 4. At this occasion (which was before the start of the ESR operation) the experiment was used to cover a nightside satellite conjunction. Figure 5a represents altitude-versus-time data for four plasma parameters measured with a low elevation VHF experiment (only data from one of the two beams shown). It can clearly be seen that at about 1500 UT a region of enhanced electron temperature (second panel) moves downward, that is in this very oblique geometry mainly equatorward. This is a signature of the expanding auroral oval during the substorm growth phase. At about 1600 UT the auroral oval precipitation is observed in the field-aligned UHF beam, as can be recognised in the start of enhanced electron density at auroral altitudes in Figure 5b. A substorm expansion takes place soon after 1600, indicated by the much wider range of precipitating electrons - the electron density increases at both higher and lower altitudes. At this time the VHF data in Figure 5a shows several poleward expansions of the auroral precipitation, again recognisable by the pattern in the electron temperature panel. We believe that this example illustrates how combined incoherent scatter experiments can provide simultaneous information on the plasma processes along one field-aligned altitude profile, and the spatial development in latitudinal direction. Combination with other ground-based data will furthermore improve the information content of this type of data. As usual, more refined datasets such as conductivities, electric field vectors, etc., will be available on the basis of co-operation with associate radar investigators. (Note that Figure 5 contains the remark "not for publication", in order to illustrate the exact format of the summary data available from the WWW.)

Other incoherent radars of interest for the IASTP are the Söndre Strömfjord radar in Greenland and the Millstone Hill radar in the US. Both radars will perform different kinds of scan experiments during interesting satellite conjunctions. In many cases it is planned that all incoherent radars shall operate simultaneously, thus delivering complementary data from different local time zones. It is obvious that instruments as versatile, complicated, power and manpower consuming as incoherent scatter radars can only be operated for a limited amount of time. For the EISCAT radars e.g. the total available operation time per year is limited to about 2000 hours.

For IASTP purposes the 7 associate countries in the EISCAT organisation (Finland, France, Germany, Japan, Norway, Sweden, and the UK) have created a common time pool for dedicated operation, and a relatively large amount of EISCAT common program operation is specially allocated for interesting satellite conjunctions during the coming years. The co-ordination of EISCAT operation with IASTP spacecraft is carried out by a special working group, chaired by the EISCAT Scientific Director, Dr. A. van Eyken, involving even members from the working group for Cluster ground-based co-ordination. Close co-operation with the other incoherent radars (which have, in comparison to EISCAT, considerably less complicated scheduling procedures ) is provided by the international Incoherent Scatter Working Group of URSI, and close and frequent personal contacts on director level. A common schedule for the high-latitude incoherent radars is now available on the WWW.

Fig. 5a) Daily overview plot of simultaneous EISCAT measurements of plasma parameters (electron density, electron and ion temperature and ion velocity) as a function of altitude (in this case rather indicating latitude variations) and time for an experiment with a north-pointing antenna. In these kinds of plots the expansion of precipitation regions can be identified as regions of enhanced electron temperature. Regions of enhanced ion temperature indicate strong electric fields.

Fig. 5b) Daily overview plot of EISCAT measurements of plasma parameters (electron density, electron and ion temperature and ion velocity) as a function of altitude and time for a field-aligned pointing experiment. Times and types of precipitation can easily be recognised in the electron density data.

SUMMARY AND CONCLUSIONS

A unique combination of ground-based instruments is prepared to meet the opportunities for science progress, which are provided by the fleet of IASTP spacecraft. Well functioning databases supply a wealth of overview data, and guide interested scientists to more refined datasets, on the basis of international collaboration.

The combined ground-based data gives its users a possibility to simultaneously monitor magnetospheric processes in vast regions of space. From the magnetopause to the plasmapause, from the ring current via the inner edge of the plasma sheet to the distant neutral line, and from the cusps throughout the magnetospheric lobes - ground-based instruments can detect signatures of magnetospheric processes, and give information to satellite experimenters, where and when with respect to certain phenomena their detailed in situ measurements were carried out. Temporal sequences in ground-based observations must reflect temporal sequences in magnetospheric processes, and the combined satellite and ground-based observations should be able to address many cause-and-reaction questions, still open in our field. The full dynamical range of cusp processes, the alternative developments of magnetospheric substorms (occurring in sequence or in parallel), dayside trigger events of nightside disturbances - all these features can be better understood if global measurements from the ground are used to complement the interpretation of satellite results.

ACKNOWLEDGEMENTS

Nothing described in this paper would have been possible without the work of the numerous individual scientists and staff teams responsible for the various ground-based experiments. The co-ordination of observations is strongly supported and stimulated by the active members of the working groups for ground-based co-ordination and satellite teams. The data centres would not have been possible to create and maintain without a dedication of the staff, which goes far beyond the line of duty. Last not least only many different national funding agencies have made the individual experiments possible.

We are indebted to the Director and staff of EISCAT for operating the facility and supplying the data. EISCAT is an international association supported by Finland (SA), France (CNRS), the Federal Republic of Germany (MPG), Japan (NIPR), Norway (Nfr), Sweden (NFR), and the United Kingdom (PPARC).

REFERENCES

Greenwald, R.A., W.Weiss, E.Nielsen, and N.R.Thomson, STARE: A new radar auroral backscatter experiment in northern Scandinavia. Radio Sci., 13, 1021, 1978

Greenwald, R.A., K.B.Baker, J.R.Dudeney, M.Pinnock, T.B.Jones, et al., DARN/SuperDARN, A global view of the dynamics of high latitude convection, Space Sci. Rev., 71, 761-796, 1995

Kamei, T., and H. Maeda, Auroral electrojet indices (AE) for January-June 1978, Data Book 3, Fac. Sci., Kyoto Univ., Kyoto, Japan, 1981

Kamide, Y., A.D.Richmond, and S.Matsushita, Estimation of ionospheric electric fields ionospheric currents and field-aligned currents from ground magnetic records, J. Geophys. Res., 86, 801, 1981

Kauristie, K., T.I.Pulkkinen, R.J.Pellinen, and H.J.Opgenoorth, What can we tell about AE from a single magnetometer chain, Ann. Geophys., 14, 1177-1185, 1996a

Kauristie, K., T.I.Pulkkinen A.Viljanen, R.J.Pellinen, D.Wallis, and S.Vennerström, Preliminary Studies on a new oval index: Comparisons of AE and meridional magnetometer chains, in Cluster Ground-Based Co-ordination Source Book, European Space Agency, ESA-SP, in press, 1996b

Opgenoorth, H.J., Co-ordination of ground-based observations with Cluster, in Cluster: Mission, Payload and Supporting Activities, ESA-SP-1159, pp.301-305, 1993.

Opgenoorth, H.J, and M.Lockwood, Opportunities for magnetospheric research with co-ordinated Cluster and Ground- Based observations, Space Science Rev., in press, 1996

Richmond, A.D., and Y.Kamide, Mapping electrodynamic features of the high-latitude ionosphere from localised observations: Technique, J. Geophys. Res., 93, 5741, 1988

Stauning, P., H.Yamagishi, M.Nishino, and T.J.Rosenberg, Dynamics of cusp latitude absorption events observed by imaging riometers. J. Geomag. Geoelectr., 47, 823, 1995