------------------------------------------------ S2-01 GEOEFFECTIVENESS OF CMEs R. SCHWENN Max-Planck-Institut f\"ur Aeronomie, Lindau, Germany E-mail: schwenn@linmpi.mpg.de Among the several hundreds of CMEs observed by the LASCO instruments onboard the SOHO spaceprobe a significant percentage was of the ``halo" type, i.e., they were pointed towards or away from the earth. Only additional evidence from further instruments observing the disk appropriately, in particular EIT on SOHO, allows to predict which halo might become geoeffective. We measured the lateral expansion speeds of some tens of halo CMEs in order to infer their otherwise unaccessible radial speeds. We compared these numbers with average propagation speeds derived from the time difference between the CME and the occurrence of their associated shock waves at 1 AU. It turned out that the variance of expansion speeds close to the sun is significantly higher than that of the average propagation speeds. The geoeffectiveness of CMEs driving shocks towards the earth is extremely variable. It all depends crucially on the topology of the magnetic field topology within the shocked plasma and the succeeding driver plasma. The location and orientation of the magnetized source region of a CME might be used for more precise prediction of its probable geoeffectiveness. ------------------------------------------------ S2-02 3D MHD MODELING OF THE SOLAR DRIVERS OF SPACE WEATHER S. POEDTS (1), A. Csik (1,2), H. De Sterck (1,2), H. Deconinck (2), and D. Roose (3) (1) Centre for Plasma Astrophsyics, K.U.Leuven, Belgium (2) Von Karman Institute, St. Genesius-Rode, Belgium (3) Dept.~of Computer Science, K.U.Leuven, Belgium E-mail: Stefaan.Poedts@wis.kuleuven.ac.be Novel upwind residual distribution schemes have been developed for solving the time dependent three-dimensional magnetohydrodynamic (MHD) equations on unstructured grids. These monotone multidimensional schemes are especially developed for space weather applications, in particular for the simulation of the solar wind and the initiation and evolution of coronal mass ejections (CMEs) in the framework of a project investigating the physics of the solar drivers of space weather. The purpose of this project is to reveal the basic physics behind the recurrent structure, heating and acceleration of the solar wind and the formation and propagation of transients like CMEs and induced shocks from their birth in the solar corona up to their arrival at the Earth's magnetosphere. This should provide a basis for more reliable science-based space weather predictions. The novel numerical techniques are briefly discussed and their advantages and superiority are demonstrated by means of the first results of the above-mentioned project, viz. the simulation of the solar wind, complex MHD shock interactions associated with fast CMEs, and the solar wind interaction with the Earth's magnetosphere. ------------------------------------------------ S2-03 AN EMPIRICAL MODEL TO PREDICT THE ARRIVAL OF CORONAL MASS EJECTIONS AT 1 AU N. GOPALSWAMY (1), A. Lara (1), and M. L. Kaiser (2) (1) NASA GSFC and The Catholic University of America (2) NASA GSFC E-mail: gopals@fugee.gsfc.nasa.gov The interaction between coronal mass ejections (CMEs) and the solar wind can be quantified as an effective acceleration acting on the CMEs. Based on this acceleration, we obtain an empirical model to predict the arrival of CMEs at 1 AU. We test this model using past observations from Helios, P78-1, Pioneer Venus Orbiter and Solar Maximum Missions. We also discuss the geoeffectiveness of the CMEs that arrive at 1 AU. ------------------------------------------------ S2-04 PREDICTING BOW SHOCK AND MAGNETOPAUSE LOCATIONS DERIVED FROM EMPIRICAL MODELS AND REAL-TIME SOLAR WIND DATA S. M. PETRINEC Lockheed Martin Advanced Technology Center, Palo Alto, CA E-mail: petrinec@spasci.com Over the years many empirical models have been developed to describe the average locations and shapes of the Earth's bow shock and magnetopause. Several of these models are parameterized according to the external conditions of the solar wind. Because these models are comparatively simple, they can be calculated quickly and provide reasonable though crude estimates of the shapes of the geophysical boundaries. By describing the solar wind as a sequence of `fronts', one can estimate several shapes of the geophysical boundaries at a given moment of time. Assuming each boundary shape is most accurate at the location of each `front', one can piece together boundary shapes, thus providing shapes which coarsely represent the dynamics imposed by the solar wind. Real-time solar wind observations taken far upstream of the Earth are then used to predict boundary locations into the near future. We describe the strengths and weaknesses of using such empirical models in a dynamic manner to forecast the locations of the geophysical boundaries, and how they are being used for the purposes of space weather. ------------------------------------------------ S2-05 ON SPACE WEATHER ENERGY BUDGET H. E. J. KOSKINEN (1,2), E. I. Kallio (2), E. J. Kallio (2), and T. I. Pulkkinen (2) (1) University of Helsinki, Department of Physics (2) Finnish Meteorological Institute, Geophysical Research E-mail: Hannu.Koskinen@fmi.fi Determination of the magnetospheric energy budget is a difficult task because the energy input cannot be directly measured. The input-rate is often parameterized in terms of Akasofu's epsilon parameter, which is given as a function of upstream Poynting flux toward the magnetosphere. The scaling of the parameter as the input power was originally estimated by considering power of different energy sinks within the magnetosphere. While more recent studies have refined our knowledge of the sinks, epsilon has maintained its role as an easily determined input estimator. In this study we discuss certain issues about the epsilon parameter, which are not always appreciated when using the parameter. We analyze the Poynting flux in the upstream solar wind and discuss how it becomes incident on the magnetopause, considering the draped magnetic field configuration in the magnetosheath. While we cannot determine the exact amount of flux penetrating through the magnetopause, it is clear that the flux must be collected from a relatively large area if it really is to be considered as the source of magnetospheric energy. Furthermore, in order to be consistent with the interpretation of the epsilon parameter in terms of Poynting flux, it is necessary to exclude the Sun-Earth magnetic field component from the calculation. A further complication is that the effective area through which the Poynting flux can penetrate through the magnetopause cannot follow the changes in the upstream magnetic field direction with a constant time delay due to finite speed of the magnetosheath flow. Thus we argue that it is more physical to use the epsilon parameter as an integral over a given, e.g., storm or substorm time-period (energy) than as a measure of an instantaneous input rate (power). ------------------------------------------------ S2-06 MHD SIMULATIONS OF SUBSTORMS FOR SPACE WEATHER T. TANAKA Communications Research Laboratory E-mail: tanaka@crl.go.jp A goal of the Space Weather Program effort at Communications Research Laboratory (CRL) is to increase our physical understanding of the solar wind-magnetosphere-ionosphere (S-M-I) coupling system. In recent years, the global magnetohydarodynamic (MHD) simulation has become increasingly successful at reconstructing and predicting behaviors of the S-M-I system. It will implement a useful tool for practical space weather applications. To make these outcomes more fruitful, it is unavoidable to achieve a satisfactory success in reproducing the substorm in the numerical model. Starting from a stationary solution under a northward interplanetary magnetic field (IMF) condition with non-zero IMF B_y, our model reproduces the substorm after the southward turning of the IMF. At first, the plasma sheet thinning is promoted by the drain of closed flux from the plasma sheet under the enhanced convection. The onset occurs as an abrupt change of pressure distribution in the near-earth plasma sheet and an intrusion of convection flow into the inner magnetosphere. It is concluded that the dipolarization is not a mere pile up of the flux ejected from the NENL but the state (phase-space) transition of the convection system from a thinned state to a dipolarized state associated with a self-organizing criticality. ------------------------------------------------ S2-07 CAN WE TRACE THE SPACE-WEATHER CONDITIONS BY THE GROUND-BASED GEOPHYSICAL OBSERVATIONS? O. A. TROSHICHEV, A. V. Shirochkov, and L. N. Makarova Arctic and Antarctic Research Insitute, St.~Petersburg, Russia E-mail: olegtro@aari.nw.ru The reliable and low-cost method of monitoring such important solar wind parameters as the interplanetary magnetic field (IMF) and its dynamic pressure can be ensured by the specially designed ground-based geophysical observations. The PC-index (derived from the ground-based polar cap geomagnetic observations) chracterizing magnetic activity in the polar cap is a dimensionless value parameterizing by the season, UT moment, and hemisphere. It has been calibrated for interplanetary electric field E = (V [B_z^2 + B_y^2]) value which merges into the magnetoshere. The PC-index calculated automatically with 1-minute resolution can serve as a proxy for such magnetospheric parameters as the cross-polar cap voltage, ionospheric electric field in the near-pole region and hemispheric Joule heat value. The daytime ionization level of the F2-region of the polar ionosphere is another easily measured parameter, which is proportional (with correlation degree more than 80) to the subsolar distance between the Earth and the magnetopause. There are the strong reasons to believe that this subsolar distance together with magnetopause shape determine amount of the solar wind energy transferred into the ``magnetosphere-ionosphere-atmosphere" system. Additionly, the riometer observations of the polar cap absorption events (PCA) are the indicators of the solar cosmic rays energy penetrated into the near-Earth space. Consequently, the regular geomagnetic, ionospheric and riometer observations in the near-pole region can provide an adequate information on the current state of the magnetosphere in the regime of quasi-real time. ------------------------------------------------ S2-08 REALTIME MONITOR FOR AURORAL KILOMETRIC RADIATION: RELATIONSHIP WITH SUBSTORMS, PROPAGATIONS IN THE VICINITY OF THE EARTH, AND REALTIME MONITOR SYSTEM T. MURATA (1), K. Tsutsumi (1), W. Kurth (2), K. Hasimoto (3), and H. Matsumoto (3) (1) Faculty of Engineering, Ehime University (2) Iowa University (3) Radio Science Center for Space \ Atmosphere E-mail: murata@cs.ehime-u.ac.jp Magnetospheric substorms often bring us a variety of troubles and disasters to human activities both on the Earth and in the space. One of the strongest impacts occurred during the last solar cycle in 1989, when the entire Province of Quebec went dark because a geomagnetic storm caused power lines to overload. In space, astronauts often face hazards when venturing outside the safety of the space shuttle and need to be alerted to the dangers of energetic particle emissions from the sun. Herein, we propose a system to monitor Auroral Kilometric Radiation (AKR) by spacecraft in the vicinity of the Earth. The AKR is one of the well-known plasma waves radiated form the Earth, and deeply related with substorm activities. With our system we can detect the onsets of the substorms within few minutes and this information is delivered through the Internet. In our talk, we first introduce event studies and statistic studies to demonstrate how the AKR is related to the substorm phenomena. Then, we show how the AKR looks like on the POLAR satellite orbits; there are regions where the AKR is unreachable in the vicinity of the earth. After comparison with AKR observations via GEOTAIL satellite, we can estimate the area where the AKR is occulted by the plasmapause both in the dayside and nightside. Clear relationships between the ``sizeh of the plasmapause and AKR occultation regions are found in both sides. Finally, taking into account the AKR properties, we propose the AKR onset monitor system. We hope this system will contribute to the space weather forecasting and substorm ``nowcastingh. ------------------------------------------------ S2-09 SCOSTEP S-RAMP SEPTEMBER 1999 SPACE WEATHER MONTH CAMPAIGN: OVERVIEW OF EVENTS, WORKSHOP INTERFACE, DATA SETS AND STUDIES UNDERWAY J. U. KOZYRA Space Physics Research Laboratory, University of Michigan, 2455 Hayward, Michigan 48109-2143 E-mail: jukozyra@engin.umich.edu September 1999 Space Weather Month was created by SCOSTEP/ S-RAMP to identify a one-month period during which international efforts could be focused on coordinated space weather observations and subsequent analysis. There were roughly three periods of shocks, ejecta and southward interplanetary magnetic fields that triggered magnetic storms during the core campaign interval in September 1999. This resulted in 4 moderate and 1 major magnetic storm. A fifth moderate storm period followed at the end of the month. A major magnetic storm in late October 1999 has also been folded into the campaign because of its impact on ground systems and the existence of data sets not available in September 1999. The campaign had a number of special features including: participation by a wide variety of ground-based observing facilities world-wide, optimized satellite coverage from the ISTP project and \Orsted, an embedded 3-day interval of coordinated incoherent scatter radar observations, and strong emphasis on forecasting and effects in addition to the core research efforts. An overview of the storm events, a description of the electronic workshop interface developed by the SPARC project for this campaign, the details of the available data sets and a summary of studies currently underway will be given. ------------------------------------------------ S2-10 SPACE WEATHER EVENTS DURING THE S-RAMP SPECIAL ANALYSIS INTERVAL: APRIL -- MAY 1998 D. N. BAKER Laboratory for Atmospheric and Space Physics, University of Colorado at Boulder E-mail: baker@lasp.colorado.edu Using a wide array of datasets from International Solar Terrestrial Physics (ISTP) spacecraft, from ground-based facilities, and from operational satellites, we have found evidence of highly disturbed solar, solar wind, and geomagnetic conditions in late April and early May 1998. The combination of coronal mass ejections, solar flares, and high speed solar wind streams during this interval led to a powerful sequence of solar wind drivers of magnetospheric processes at the Earth. The result of the compounding solar wind disturbances was to produce a deep, powerful, and long-lasting enhancement of the highly relativistic electron population throughout the outer terrestrial radiation zone. There was evidence that many space weather-related spacecraft anomalies occurred during this active interval. S-RAMP has collected and organized large sets of data and images for the April-May interval. These are widely available for analysis and interpretation. The kinds of disturbances witnessed during this remarkable interval are indicative of the types of events that commonly occur during the solar maximum period. It is most important to determine how well space systems can stand up to similar multifaceted effects of the space environment over the next several years as we pass through solar maximum. ------------------------------------------------ S2-11 ENERGETIC ELECTRON BEHAVIOR IN THE OUTER RADIATION ZONE DURING THE SPACE WEATHER MONTH T. OBARA and T. Nagatsuma Hiraiso Solar Terrestrial Research Center, CRL E-mail: T.Obara@crl.go.jp The Akebono (EXOS-D) satellite made good observations of the entire radiation belts in the evening sector in the space weather month (September 1999). During the campaign we had three magnetic storms; i.e. September 12-14, 22-23, and 27-28. On early September 23, the Dst index reached its minimum with -160 nT, showing a major magnetic storm, while both on September 13 and 27 the decrease of the Dst index was very gradual, having the minimum values of -70 nT and -65 nT, respectively. Even though the Dst variation was big during September 22-23, very little enhancement of MeV electrons was observed in the outer radiation belt during the storm recovery phase. On the contrary large enhancements in the intensity of MeV electrons took place for September 12-14 and 27-28, when the Dst variations were rather small. These large enhancements were made due to the prolonged magnetic activities during the storm recovery phase. The Russell-McPheron effect was evident; i.e. the IMF was actually away for the cases of September 12-14 and 27-28. The increase in the intensity of MeV electrons was first seen in the heart of outer radiation belt and in lower energy. As time progressed the intensity of MeV electrons in far regions (L>5) increased, suggesting that both an internal acceleration of the electrons and an extremely large outward diffusion are taking place in the outer radiation belt. ------------------------------------------------ S2-12 EFFECTS OF IONOSPHERIC SCINTILLATION: ITS SPECIFICATION AND FORECASTING S. BASU and K. M. Groves Air Force Research Laboratory, 29 Randolph Road, Hanscom AFB, MA 01731, U.S.A. E-mail: santimay@aol.com The ionized upper atmosphere often becomes turbulent and develops electron density irregularities. These irregularities scatter radio waves causing amplitude and phase scintillation of satellite signals. Amplitude scintillation, when intense, causes signals to fade below the noise level of the receiver. Under these circumstances, loss of signal occurs which causes communication links to encounter message errors and GPS navigation receivers to lose lock and spend additional time to reacquire the signal. Phase scintillations, imposing rapid changes in signal phase, introduce frequency fluctuations. When these fluctuations exceed the receiver bandwidth, it becomes difficult to maintain the phase lock in receivers. Scintillations are most intense in the equatorial region and occur between one hour after sunset to midnight. Scintillations in the polar region, though less intense than in the equatorial region, occur virtually at all times. At all locations, scintillations attain their maximum values during the solar maximum period. In the equatorial region, a recently developed scintillation specification system, known as the Scintillation Network Decision Aid (SCINDA), has been developed. SCINDA utilizes two geostationary satellites and performs two station measurements of scintillation and irregularity drifts. The dataset is brought to the user terminal by the internet and it drives a model to generate in near real-time three dimensional maps of scintillation structures. An equatorial satellite, Communication Navigation Outage Forecasting System (C/NOFS), has also been planned. This satellite, consisting of a multi-frequency beacon transmitter, GPS occultation sensor and a suite of in-situ sensors, that include the electric field, electron density and neutral wind sensors, is expected to specify and forecast equatorial scintillation. A possible specification and forecasting system for polar scintillation based on interplanetary magnetic field observations, and a suitably dispersed scintillation and macroscale plasma density sensors is also described. ------------------------------------------------ S2-13 MAGNETIC STORM INDUCED SCINTILLATIONS AT MID-LATITUDES DURING THE SPACE WEATHER MONTHS OF SEPT/OCT 1999 S. BASU (1), Sa. Basu (2), J. Foster (3), and A. Ridley (4) (1) National Science Foundation, Arlington, VA 22230, U.S.A. (2) Air Force Research Laboratory, VSBI, Hanscom AFB, MA 01731, U.S.A. (3) Atmospheric Sciences Group, MIT Haystack Observatory, Westford, MA, U.S.A. (4) University of Michigan, Ann Arbor, MI, U.S.A. E-mail: sbasu@nsf.gov Scintillation and total electron content (TEC) observations using geostationary and GPS satellites were made from stations at \sim 40 degrees N latitudes during the period Sept/Oct 1999 to study the effects of magnetic storms on communication and navigation systems. Coordinated DMSP satellite and incoherent scatter radar (ISR) data are utilized to study the background conditions associated with storm-induced TEC gradients and scintillations. Amplitude scintillations of 15 dB on 250 MHz signals from geostationary satellites and TEC fluctuations exceeding 5 TEC units (10^16 electrons m^-2) per minute were observed by GPS receivers in the International Geodynamic Service (IGS) network in conjunction with background TEC gradients of 5-10 TEC units per degree latitude. This impulsive onset of the stormtime component of mid-latitude scintillations is in all probability related to the direct penetration of the high latitude convection electric field into the plasmasphere. The ISR measurements from Millstone Hill have been used to determine the stormtime convection and plasma density gradients in the vicinity of the disturbed scintillation and TEC regions. Such comparisons are utilized to establish the relationship between the stormtime component of scintillations and plasma convection in the context of available models. The significance of the observed TEC fluctuations and scintillations for the Wide Area Augmentation System (WAAS), a satellite based navigation system for aircraft using GPS, is discussed. ------------------------------------------------ S2-14 RELATIONSHIP OF THE APR.~29, 1998 HALO CME AND THE MAGNETIC CLOUD AND GEOACTIVITY ON MAY 2 -- 3 D. WEBB (1,2), R. Lepping (3), S. Plunkett (4), and S.-T. Wu (5) (1) Institute for Scientific Research, Boston College, Chestnut Hill, MA, USA (2) Also at Air Force Research Laboratory, Space Vehicles Directorate, Hanscom AFB, MA, USA (3) Laboratory for Extraterrestrial Physics, NASA Goddard Space Flight Center, Greenbelt, MD, USA (4) USRA, NASA Goddard Space Flight Center, Greenbelt, MD, USA (5) Center for Space Plasma and Aeronomic Research, The University of Alabama in Huntsville, Huntsville, AL, USA E-mail: david.webb@hanscom.af.mil We discuss the solar activity that was likely associated with the magnetic cloud, extended period of anomalous solar wind signatures and moderate magnetic storm observed at Earth on May 2-3, 1998. At the Wind spacecraft the cloud was preceded by an interplanetary shock on May 1, 2122 UT and had a duration of \sim 30 hr. It was well fit to a force-free flux rope model indicating a diameter of \sim 1/2 AU, Left handedness, and with its axis pointed toward the Sun and highly inclined by 55 deg. to the ecliptic. Within the cloud and possibly extending beyond it were one or more extended periods of flows of counterstreaming electrons and both unusually cold and hot material, including an extensive amount of He^+, considered a tracer of solar filament material in solar wind ejecta. Indeed, this ejecta was likely associated with a full halo CME observed by the SOHO LASCO coronagraphs on April 29 and an M7 long-duration flare and filament eruption arcade event on the surface 0.4 solar radii southeast of Sun center. We compare measurements of the orientation, helicity/chirality, magnetic flux and speed of the solar features and the magnetic cloud to test the utility of such events for forecasting space weather effects. We also use an MHD coronal streamer and flux rope model to predict the solar wind, cloud and storm parameters. ------------------------------------------------ S2-15 RAPID PROTOTYPING: APPLYING RESEARCH MODELS AND DATA TO OPERATIONAL SPACE WEATHER FORECASTING T. G. ONSAGER NOAA Space Environment Center, Boulder, Colorado E-mail: tonsager@sec.noaa.gov Advances in solar-terrestrial research are resulting in improved numerical models that can be applied to regions of space extending from the Sun to Earth. In addition to improved models, critical observations are also now available, some in near real-time, to drive and to validate the models. At the same time, available models must be carefully evaluated to determine their robustness, accuracy, and usefulness to space weather forecasting. This presentation will describe the models that are now being used operationally and those that are undergoing internal evaluation at the NOAA Space Environment Center. The various elements of the prototyping process used to evaluate model candidates and to transition selected models into operations will also be discussed. This process of model selection and transition includes: (1) model evaluation to assess the potential value to space weather operations and readiness to transition from research to operations; (2) implementation, modification, and evaluation in an operational environment; and (3) final transition to operational use. Specific operational applications of the existing models, and desired improvements to current capabilities will also be given. ------------------------------------------------ S2-16 TRANSITION OF RESEARCH RESULT TO OPERATIONAL ENVIRONMENT C.-I. MENG The Johns Hopkins University Applied Physics Laboratory E-mail: Ching.Meng@jhuapl.edu Prototyping space research results to real time practical space weather applications is an important aspect of the modern space science research. More importantly, it bears on its relevance to society benefits perceived by the public. I will discuss the synergy between scientific research and applications as well as present the importance of balancing basic research activity on space weather and dedicated effort in transitioning of existing space weather products. I will describe examples that have successfully transformed research products into operational uses. The first example is the visualization of boundaries of the instantaneous auroral oval. The second example is the production of a geomagnetic disturbance index on the fly for nowcasting of space weather activity. Potential utilization of these products will also be discussed. ------------------------------------------------ S2-17 THE COMMUNITY COORDINATED MODELING CENTER K. Baker (1), P. Bellaire (2), M. Bonadonna (3), J. Bredekamp (4), M. Heinemann (5), M. HESSE (6), T. Onsager (7), B. Robinson (1), J. Sharber (4), and K. Scro (8) (1) National Science Foundation, Arlington, VA (2) Air Force Office of Scientific Research, Arlington, VA (3) HQ USAF/XOWX, Arlington, VA (4) NASA Headquarters, Washington, DC (5) AFRL, Hanscom AFB, MA (6) NASA GSFC, Code 696, Greenbelt, MD (7) NOAA Space Environment Center, Boulder, CO (8) AF/SMC, Colorado Springs, CO E-mail: hesse@gsfc.nasa.gov The Community Coordinated Modeling Center (CCMC) is a multi-agency partnership aimed at the creation of next generation space weather models. The goal of the CCMC is to undertake the research and developmental work necessary to substantially increase the present-day modeling capability for space weather purposes, and to provide models for transition to the rapid prototyping centers at the space weather forecast centers. This goal requires substantial research community involvement. The physical regions to be addressed by CCMC-related activities range from the solar atmosphere to the Earth's upper atmosphere. The CCMC is an integral part of NASA's Living With a Star initiative, of the National Space Weather Program Implementation Plan, and of the National Security Space Architect's Transition Plan for space weather. CCMC includes a headquarters facility at NASA Goddard Space Flight Center, as well as distributed computing facilities provided by the Air Force. In this paper we will provide updates on CCMC status, on current plans, research and development accomplishments and goals, and on the level of community involvement already underway. In particular, we will discuss recent science results obtained by CCMC models. ------------------------------------------------ S2-18 THE USE OF DATA ASSIMILATION IN THE MAGNETOSPHERIC SPECIFICATION MODEL R. A. WOLF (1), R. W. Spiro (1), M. F. Thomsen (2), and H. Korth (2) (1) Rice Space Institue, Rice University (2) Los Alamos National Laboratory E-mail: garner@rice.edu This paper will present recent attempts to assimilate geosynchronous particle data into the Magnetospheric Specification Model (MSM), an operational model of the inner magnetosphere that is presently being used by the Space Environment Center to provide nowcasts. The MSM calculates particle fluxes by following drifts in data-driven, time-dependent models of the electric and magnetic fields. Data assimilation can be used to improve the performance of the operational model. We will discuss and test the effectiveness of two different data assimilation techniques, direct insertion and optimization of the input. The direct-insertion method overrides the model-calculated particle fluxes with measured values, in the vicinity of the measurement. The input-optimization method adjusts model input conditions (electric and magnetic field models, plasma boundary conditions) at each time step to optimize agreement with observations. ------------------------------------------------ S2-19 SUBSTORMS AND MAGNETIC STORMS FROM THE SATELLITE CHARGING PERSPECTIVE J. F. FENNELL, H. C. Koons, J. L. Roeder, and J. B. Blake The Aerospace Corporation E-mail: joseph.f.fennell@aero.orb Substorms and magnetic storms generate significant space weather effects in the inner magnetosphere. They change the dose rates experienced by satellites in many orbits and are directly linked to the occurrence of satellite charging. Substorms inject hot plasma into the nightside magnetosphere. The drifting electron component of this hot plasma can charge the surfaces of the satellites leading to electrostatic discharges and associated satellite anomalies and sometimes failures. These occur in regions that are consistent with the expected motions of the substorm injected particles. The high energy electron enhancements associated with many magnetic storms can be sufficient to cause charging of satellite elements even behind significant shielding. Not all magnetic storms result in flux enhancements sufficient to cause such ``internal" charging. Also, because the induced voltages from the ``internal" charging are usually not directly measured, the anomalies they cause are more difficult to link to the space environment and the magnetic storm related space weather. However, there are a few cases in which either the anomaly statistics were sufficient to show linkage or the anomaly was reproduced by laboratory testing using energetic electron beams. Data from different satellites will be used to show the measurement of surface charging from different regions of space and link the charging to electrostatic discharges and anomalies. Similarly, we will show the magnetic storm related variability of the high energy electron fluxes and provide a look at some of the evidence that these penetrating fluxes can lead to spacecraft anomalies. ------------------------------------------------ S2-20 A STATISTICAL LINK BETWEEN MAGNETIC STORMS AND SPACECRAFT ANOMALIES M. P. FREEMAN (1), M. Daws (1,2), and R. B. Horne (1) (1) British Antarctic Survey, Cambridge, U.K. (2) Cambridge University, U.K. E-mail: M.P.Freeman@bas.ac.uk We present a statistical method to test whether there is an association between magnetic storms and spacecraft anomalies and, if so, to find what proportion of spacecraft anomalies are associated with magnetic storms. The method is a superposed epoch analysis combined with information on the known observed statistical distribution of magnetic storms and presumed statistical distribution of spacecraft anomalies. For a general database of 5033 spacecraft anomalies collected over 30 years and compiled by NOAA's National Geophysical Data Center, the results give the percentage of all spacecraft anomalies and of electrostatic discharge anomalies that are statistically associated with magnetic storms. ------------------------------------------------ S2-21 MONITORING EQUIVALENT DOSES RECEIVED BY AIR CREW P. LANTOS Observatoire de Paris-Meudon, France E-mail: Pierre.Lantos@obspm.fr By application of an EU Directive, European air transport companies have to monitor annual equivalent doses received by air crew, starting in 2000. Calculations based on cosmic ray observation are the solution accepted by French Authorities. We will present the project, which is a typical Space Weather application. ------------------------------------------------ S2-22 GROUND EFFECTS OF SPACE WEATHER R. PIRJOLA Finnish Meteorological Institute, Helsinki, Finland E-mail: risto.pirjola@fmi.fi At the Earth's surface a space weather event manifests itself as a rapid variation of the geomagnetic field. The variation induces a geoelectric field that drives ``geomagnetically induced currents" (GIC) in technological systems, like electric power transmission grids, pipelines, phone cables and railway systems. In power grids, saturation of transformers may occur, which can have several harmful consequences and even result in a black-out of the whole system and in permanent damage of transformers. In pipelines, GIC are accompanied by fluctuations of the pipe-to-soil potential. This interferes with corrosion control surveys and also changes the electrochemical conditions at the pipe-soil interface enhancing possibilities of corrosion of the pipe steel. Telecommunication systems have also experienced trouble due to GIC, like overvoltages and even fires. On railways, misoperations of traffic signals have occurred. So far, Finland has not suffered from GIC problems but, due to the high-latitude location, GIC constitute a potential risk in the country. Therefore, the phenomenon is studied as a collaboration between the Finnish Meteorological Institute and the Fingrid power and Gasum pipeline companies. An active GIC research was started in Finland already more than twenty years ago. The research has contained GIC recordings, theoretical model calculations and statistical analyses of the occurrence of GIC in the Finnish systems. In a recent study, a special attention was paid to the possibility of large GIC at several 400 kV or 220 kV transformers simultaneously since such a situation is critical regarding reactive power consumption in the system. A power transmission system is a discretely-earthed network while a buried pipeline is earthed continuously through its coating, so modelling methods are quite different in these two cases. ------------------------------------------------ S2-23 HIGH VOLTAGE POWER TRANSMISSION LINE DISTURBANCES DURING LARGE GEOMAGNETIC STORMS P. STAUNING Solar-Terrestrial Physics Division, Danish Meteorological Institute E-mail: pst@dmi.dk oindent On March 13, 1989 a large geomagnetic storm caused extensive disruptions of high voltage power transmission circuits especially in the Quebec province of Canada but also to a lesser degree in Scandinavia. Similar events have occurred earlier a.o. during the great storms of February 8 - 9, 1986 and July 13 - 14, 1982. Some of these cases are connected to the chock-like disturbances accompanying the compression of the front of the magnetosphere by a sudden enhancement in the solar wind plasma. Other cases are related to extraordinarily intense substorm events. Using ground geomagnetic recordings it is attempted to trace the varying ionospheric current systems causing the power line failures in Scandinavia. The effects of ionospheric currents on extended conducting structures at the ground such as power transmission lines, signal lines and oil or gas pipe lines have previously been described by various models of geomagnetically induced currents (GIC). Some of these models have been applied to the storm events considered. The correspondance between model predictions and actual disturbances will be discussed as part of the presentation. Corresponding observations from the recent large magnetic storm on April 6-7, 2000, will also be presented. ------------------------------------------------ S2-24 STUDY OF GEOELECTRIC FIELD AND GEOMAGNETICALLY INDUCED CURRENTS DURING RECENT SPACE WEATHER EVENTS L. TRICHTCHENKO and D. H. Boteler Geomagnetic Laboratory, Geological Survey of Canada, Ottawa E-mail: larisa@geolab.nrcan.gc.ca Disturbances of the Earth's magnetic field induce electric fields and currents in the ground and in the network of power lines, pipelines and phone cables spread on the surface of the Earth. Geomagnetically induced current (GIC) data from various systems across Canada together with the geoelectric and geomagnetic field measurements were collected and analyzed for different space weather events. Geomagnetic field observations and layered earth conductivity structure were used to model the geoelectric field. The modelling results showed a good correlation with the GIC data. We studied how the earth conductivity structure and the spatial characteristics of the magnetic field disturbance affects the accuracy of the modelling. This shows that better knowledge of the ionospheric electrojet position and therefore the spatial characteristics of geomagnetic field disturbance can improve our modelling of the geoelectric field and GIC in the ground systems. ------------------------------------------------ S2-25 NOWCASTING SPACE WEATHER EFFECTS IN THE HIGH-LATITUDE IONOSPHERE WITH THE SUPERDARN HF RADARS J. M. RUOHONIEMI, R. A. Greenwald, and R. J. Barnes The johns Hopkins University applied Physics Laboratory E-mail: mike\_ruohoniemi@jhuapl.edu The SuperDARN network of HF radars in the northern hemisphere has recently been expanded with the addition of three radars to include coverage of western Canada, Alaska, and eastern Siberia. At times, effects in the ionosphere can be observed over 2/3 of the entire high-latitude region. The convection velocity data can be synthesized into global maps of the convection pattern. These patterns portray the Space Weather within the ionospheric plasma windsand provide a useful diagnostic of the overall state of the magnetosphere-ionosphere system. Internet links to the radars make it possible to perform the analysis on the fly. In this presentation we review the activity during selected periods of intense Space Weather in SuperDARN nowcastmode. The onset and progress of disturbances are easily monitored. In addition to showing regions of intensified electric fields and currents, the various maps available at the site show the movement of auroral boundaries, the distribution of HF clutter, and the progress of HF absorption. The web page provides real-time estimates of the total cross polar cap potential variation and the effective delay time from the arrival of effects at an upstream solar wind monitor (ACE) to impact on the ionosphere. We discuss the operation of the SuperDARN nowcast facility and the application of its products to the validation and correction of predictive Space Weather services. ------------------------------------------------ S2-26 A WEB-BASED EMPIRICAL MODEL OF THE EARTH'S IONOSPHERE USING INCOHERENT SCATTER RADAR DATA J. HOLT and R. Sitar MIT Haystack Observatory E-mail: jmh@haystack.mit.edu A Web-based regional empirical model of the ionosphere using incoherent scatter radar data from 1970 to the present is being developed. This model includes two parts: first, a model of basic and derived scalar parameters as measured at Millstone Hill, including electron density, electron and ion temperature, neutral meridional wind and exospheric temperature; and second, a model of the E ~ \mbox\boldmath B plasma drifts and corresponding electrostatic potential patterns as obtained from measurements at Millstone Hill and the Sondrestrom facility in Greenland. The scalar model is keyed to solar and geomagnetic indices chosen by multiple regression, so that deviations of actual data from the model represent the remaining day-to-day variability due to such causes as tidal forcings, gravity waves and uncertainties in the solar EUV flux. The electric field model is keyed to the interplanetary magnetic field. It represents the average response to solar wind induced changes in the magnetospheric convection, thus providing a baseline from which more variable effects of substorms, storms and disturbed neutral winds (the disturbance dynamo) can be isolated. The combined model will be valuable in several ways to space weather studies and systems. Incoherent scatter radar data are available only a few days a month due to budgetary restrictions, but the model will always be available to provide a climatological representation of the state of the ionosphere and its variability for any set of conditions. The model covers a wide range of latitude and thus is useful for validation of global theoretical and empirical models. It will also be used in investigations of specific scientific problems, such as storm-time density gradients and enhancements. The model is available over the World Wide Web for user-specified input parameters, and a real-time version is planned. ------------------------------------------------ S2-27 THE REAL TIME AMIE TECHNIQUE: HOW IT WORKS AND HOW WE CAN MAKE IT BETTER A. RIDLEY Space Physics Research Laboratory, University of Michigan E-mail: ridley@umich.edu The real-time version of the assimilative mapping of ionospheric electrodynamics (AMIE) is a model which ingests ground-based magnetometer data which is available in near-real-time to provide a now-cast of the ionospheric electric potential, currents, and particle precipitation pattern. The model then makes a 1 hour prediction of the ionospheric state. We discribe the data gathering and processing techniques involved in this effort. In addition, we discuss the validation of the now-cast and forecast. Included in this discussion are suggestions for realistic locations for new (real-time available) ground based magnetometers which would significantly improve the specifications and forecasts provided by this model. ------------------------------------------------ S2-28 THE SPACE WEATHER REQUIREMENTS FOR INDIA--A PERSPECTIVE B. M. REDDY (1) and D. R. Lakshmi (2) (1) National Geophysical Research Institute (2) National Physical Laboratory E-mail: postmast@csngri.ren.nic.in India, the most populous tropical country is presently a combination of old traditions and state of the art in science and technology. While launching sophisticated satellites for communications and remote sensing, HF communications remain as the workhorse for a number of organisations in the country.In view of this, the space weather requirements listed below are at once traditional and very modern: (1) Long term solar and Ionospheric predictions for High Frequency (HF) communications (2) Solar cycle predictions to plan low orbiting satellite launch planning and for deciding orbital parameters (3) Short term solar and magnetic event predictions for communications and satellite tracking (4) Equatorial scintillation predictions to warn on satellite link degradation (5) Intense Magnetic storm and particle event prediction to issue alerts on single event upsets at the geostationary orbit (6) Real time indexing of geomagnetic activity to aid in Earth resources survey and in other remote sensing applications. The present status and future plans in all the above areas will be presented. ------------------------------------------------ S2-P01 ON THE RELATIONSHIP BETWEEN CORONAL MASS EJECTIONS (CMEs), INTENSIVE SOLAR FLARES, SOME MAGNETOSPHERIC PARAMETERS AND DIFFERENT TYPE AURORAS DURING GREAT MAGNETIC STORMS Y. P. MALTSEV and L. S. Yevlashin Polar Geophysical Institute E-mail: maltsev@pgi.kolasc.net.ru Six giant magnetic storms, with the minimum value of Dst variation less than -250 nT were analyzed. They are on Febr. 8-9, 1986; March 13-14, 1989; Oct. 21-22, 1989; March 24-25, 1991; Oct., 28-29, 1991, and Nov. 8-9, 1991. All of these superstorms were preceded by solar flares and, in some events, by Coronal Mass Ejections (CMEs) as well. During these superstorms, over large areas of the Earth surface, there were registered bright auroras, the luminosity spectrum of which depended on the type of heliospheric source. In some events bright green auroras occurred, whereas the A-type red ones were observed in others. Some suggestions have been made as to the nature and mechanisms of origin of similar types of luminosity. Unfortunately, data on the parameters of the solar wind generating those storms were either not available or extremely scarce. So, on the basis of the available data on temporal variations of the Dst index, we restored the IMP B_z component and estimated the polar cap electric voltage. Quite large values of the latter were obtained, which agrees fairly well with some direct measurements. ------------------------------------------------ S2-P02 THE RELIABILITY OF PREDICTIONS OF LARGE SOLAR WIND DISTURBANCES BY AN UPSTREAM MONITOR P. A. DALIN (1), A. J. Lazarus (2), G. N. Zastenker (1), K. I. Paularena (2), and J. D. Richardson (2) (1) Space Research Institute, Moscow, Russia (2) Center for Space Research, MIT, Cambridge, MA, USA E-mail: pdalin@iki.rssi.ru Comparison of simultaneous solar wind plasma observations by several widely-separated spacecraft (INTERBALL-1, IMP 8, and WIND) has shown that the average value of their cross-correlations was about 0.7 and 15-20 % of cases had the very poor correlations, less than 0.5. This result means that predictions of solar wind disturbances coming to the Earth based on upstream measurements (at L1, for example) may be incorrect. We studied the dependence of the reliability of these predictions on the amplitude of solar wind disturbances and found that if we calculate the correlations for large and abrupt disturbances (for example, the increasing of the solar wind density or ion flux in 1.5-2 times or more during several minutes), we obtain average correlation values as large as 0.9-1.0. ------------------------------------------------ S2-P03 GEOEFFICIENCY OF CORONAL MASS EJECTIONS DURING A RISING SOLAR CYCLE K. E. J. Huttunen (1), H. E. J. KOSKINEN (1,2), R. Schwenn (3), and O. C. St. Cyr (4) (1) University of Helsinki, Department of Physics (2) Finnish Meteorological Institute, Geophysical Research (3) Max-Planck-Insitut f\"ur Aeronomie (4) Naval Research Laboratory, Solar Physics Branch E-mail: Hannu.Koskinen@fmi.fi We have analyzed the geoefficiency of halo and partial halo coronal mass ejections (CME) observed by the SOHO/ LASCO instrument during 1996-1999 with the exception of the second half of 1998 when the spacecraft was inoperative. This period covers most of the rising phase of cycle 23. Near the minimum most storms with Dst < -50 nT were driven by shocks and/or CMEs. Toward the maximum the number of disturbances due to high-speed streams increases. Furthermore, the distribution of CMEs becomes spread over a larger range of latitudes, which reduces the relative amount of earthward directed CMEs. If the observed halo or partial halo CMEs would have been used to forecast magnetospheric storms, the predictions had been much better in 1996-1997 but no more in 1998. The reason for this appears to be that near minimum the axes of the CMEs tend to be closer to the ecliptic plane whereas during increasing activity they can be oriented in any direction. Thus good space weather forecasting needs detailed observations of the alignment of the CME axis as well as a good prediction how far the CME axis will pass the Earth. ------------------------------------------------ S2-P04 ENERGETIC ELECTRON VARIATION IN THE OUTER RADIATION ZONE DURING EARLY MAY 1998 MAGNETIC STORM T. OBARA (1), Y. Miyoshi (2), and A. Morioka (2) (1) Hiraiso Solar Terrestrial Research Center, CRL (2) Planetary Plasma and Atmospheric Research Center, Tohoku University E-mail: T.Obara@crl.go.jp Using NOAA and Akebono observations we examined variations of the energetic electron flux in the outer radiation zone during May 2 and 4, 1998 magnetic storm, which was two-step storm. Both a flux dropout and an inward shift of the outer belt MeV electrons were recorded during the main phase of May 2 magnetic storm. A very big injection of the intermediate energy (30 keV - 100 keV) electrons to the heart of the outer radiation zone took place during the main phase of the storm. During the recovery phase of the storm an increase in the MeV electron flux was seen, which surpassed the pre-storm level in one day. Comparison of NOAA and Akebono observations yields that the injected electrons with energy of 100 keV seeded a subsequent enhancement of the MeV electrons in the outer radiation zone. A more inward shift of the peak position as well as a flux dropout occurred during the main phase of the May 4 magnetic storm. No significant injection of the intermediate energy (30 keV - 100 keV) electrons was, however, seen during the main phase of the May 4 magnetic storm. A remarkable increase of the MeV electron flux was seen during the recovery phase of the storm. The pre-existing intermediate electrons seeded the increase of the MeV electrons near the new peak portion. The increase propagated to higher L region with a significant time delay, suggesting an enhanced radial diffusion. ------------------------------------------------ S2-P05 ENHANCEMENTS OF ENERGETIC ELECTRON FLUX AT GEOSYNCHRONOUS ORBIT DURING THE RECOVERY PHASE OF GEOMAGNETIC STORM: IMPORTANCE OF THE SUBSTORM ACTIVITY HISTORY M. Fukata (1), S. Taguchi (1), T. OKUZAWA (1), and T. Obara (2) (1) Department of Information \ Communication Engineering, University of Electro-communications (2) Hiraiso Solar Terrestrial Research Center, Communications Research Laboratory E-mail: okuzawa@ice.uec.ac.jp Relativistic (>2 MeV) electron flux at geosynchronous orbit(GSO) enhances during the recovery phase of geomagnetic storm after its rapid diminishment in the main phase, and this enhancement often exceeds the pre-storm level. In this study we demonstrate that this electron flux enhancement during the recovery phase is strongly associated with the history of substorm activity represented by \Sigma AL, where \Sigma denotes the summation from the time of a Dst minimum peak. We used electron flux data that were obtained with Space Environment Monitor onboard Geostationary Meteorological Satellite. Our approach is to model the temporal variations of the electron flux with the technique of Elman artificial neural network (ANN). The recent history of AL, \Sigma AL, Dst, \Sigma Dst, and time from the Dst minimum were used as initial input parameters, and the ANN was trained using the data with 1043 hours of 9 typical storms. The performance of the model was evaluated by comparing output electron fluxes with one-hour ahead observations for 20 storm events that were selected from the period of 1978-1994. The best prediction is identified at Oct. 5-9, 1985 storm for which the correlation coefficient is 0.95, and the prediction efficiency reaches 80. The average of the correlation coefficient for the 20 cases is 0.84, the prediction efficiency 71, and the rms error of the network 0.47 It is found that these high evaluation marks are obtained only when \Sigma AL is included in the input parameters. ------------------------------------------------ S2-P06 FORMATION OF NEW PROTON RADIATION BELT ASSOCIATED WITH SOLAR PROTON EVENTS AND INTERPLANETARY SHOCKS M. DEN (1), T. Obara (1), and T. Onsager (2) (1) Hiraiso Solar Terrestrial Research Center, CRL (2) Space Environment Center, NOAA E-mail: den@crl.go.jp It was revealed that a formation of a new proton radiation belt is closely associated with solar energetic protons and an interplanetary shock passage. A solar proton event was detected by GOES satellite at 14:20 UT on May 2, 1998 and the flux (6.3 MeV-15 MeV) enhancement appeared in region around L=4.5 in the data of Akebono satellite at 17:25 UT on May 2. The Akebono also observed the increase in the higher energy (15 MeV-29 MeV, 29 MeV-62 MeV) proton flux there at 20:27 UT, which indicates the penetration of the energetic protons to near Earth region. The ACE satellite detected multiple interplanetary shock passages from Apr 30 to May 4, and two major magnetic storms occurred during that period. Solar proton event was on going during the first magnetic storm and a strong interplanetary shock hit the geomagnetic field. A new proton radiation belt was formed in the region around L=3.0 due to this shock. This phenomenon indicates that SEPs are source population of the new proton radiation belt, and the interplanetary shock and a strong compression of the Earth's geomagnetic field play an important role in the formation of the new proton radiation belt. This radiation belt persisted for several months, causing significant bad effects on spacecraft and astronauts. ------------------------------------------------ S2-P07 RADIAL DEPENDENCE OF RELATIVISTIC ELECTRON FLUXES DURING THE STORM MAIN PHASE H.-J. KIM, G. Rostoker, and Y. Kamide Solar-Terrestrial Environment Laboratory, Nagoya University E-mail: hjk@stelab.nagoya-u.ac.jp Satellite observations have revealed that the flux variations of outer belt relativistic electrons during the main phase of magnetic storms exhibit a strong radial dependence. This dependency on L can be characterized as small increases or decreases at the inner edge of the belt (L\sim 2.5-3.5) and a large decrease in its outer region (L \sim 5-6). In this work, we extend the study by Kim and Chan [1997] of relativistic electron flux decreases at geostationary orbit and investigate the characteristic radial dependence in terms of the fully adiabatic response of relativistic electrons to magnetic field perturbations. We calculate storm time fluxes of equatorially mirroring electrons by adiabatically evolving the prestorm values using Liouville's theorem and the conservation of the first and third adiabatic invariants. Calculations of fully adiabatic fluxes successfully reproduce the radial variation of relativistic electron fluxes during the storm main phase. It is the radial structure of magnetic field perturbations and the spatial and energy dependence of the quiet time electron distribution that affect the main phase fluxes in an adiabatic process. In response to the field perturbations, adiabatic flux changes become larger at larger L shells where electrons experience strong deceleration and large radial displacement. The slight increase of the inner edge electron flux can be attributed to the non-monotonic energy spectrum of the quiet time electron distribution. The increasing electron flux with energy can yield a flux increase even during adiabatic deceleration. We conclude that a fully adiabatic treatment can explain the observed variation of relativistic electron fluxes across the outer radiation belt during the storm main phase. ------------------------------------------------ S2-P08 MAGNETIC FIELD VARIATIONS AT GEOSYNCHRONOUS ORBIT AND ITS RELATIONS TO RELATIVISTIC ELECTRON FLUX T. NAGATSUMA and T. Obara Hiraiso Solar Terrestrial Research Center, CRL E-mail: tnagatsu@crl.go.jp Magnetic field variations at geosynchronous orbit can give us information of magnetospheric current systems, Since magnetic field variations at geosynchronous orbit are caused by magnetospheric current systems including magnetopause currents, ring currents, cross tail currents, and field-aligned currents. We have statistically examined dynamic pressure, pressure-corrected Dst, and dipole tilt angle dependence of magnetic field variations at geosynchronous orbit with using 1-hour averaged magnetic field data obtained from GOES-8, 9, and 10. The results of our data analysis suggest that the contribution from tail currents is significant in the midnight sector and the tail current region seems to approach geosynchronous orbit during the large storm while the contribution from ring currents and magnetopause currents are significant in the noon sector. On the contrary, current relativistic electron flux is one of the important information for satellite operation. However, observations of relativistic electron flux are not fully covered at whole longitude of geosynchronous orbit. Therefore some assimilative technique is needed for estimating flux at whole orbit. For this point of view, we start studying the relationships between magnetic field and relativistic electron flux using the results of our statistical data analysis. Because L-shell value is changing depending on local time and magnetic activity due to magnetic field variations. The result of our study and possibility for estimating relativistic electron flux at geosynchronous orbit will be presented. ------------------------------------------------ S2-P09 MULTI-SATELLITE OBSERVATIONS OF GEOSYNCHRONOUS MAGNETOPAUSE CROSSINGS D. YOSHIDA and T. Araki Department of Geophysics, Graduate School of Science, Kyoto University E-mail: daiki@kugi.kyoto-u.ac.jp The magnetopause sometimes moves across the geosynchronous orbit. These events are called Geosynchronous Magnetopause Crossings (GMCs). Here we show 3 case studies of GMC events. The first event occurred on February 8, 1986. The magnetopause moving earthward was observed first by AMPTE/CCE satellite at 1433 UT around 8.2 R_E and 10 LT and 5 min later by GOES-5 around 6.6 R_E and 9.5 LT. The magnetic field increased both at the geosynchronous orbit and on the ground suggesting effect of the solar wind dynamic pressure. In the second event which occurred 2013--2326 UT also on February 8, 1986, the bow shock was observed at 7.5 R_E by AMPTE/CCE, and a magnetopause crossing was observed at 5.2 R_E by AMPTE/CCE. The average velocity of the inward motion was about 30-40 km/sec. The third GMC event occurred on March 10, 1998. GOES-9 located in the morning side (9h LT) went out into the magnetosheath from 1754 to 1807 UT but GOES-8 around noon (13h LT) stayed in the magnetosphere. The IMF-B_z took a large negative value (-15 nT) and the SYM index decreased to -110 nT. Since the solar wind dynamic pressure was not so high (about 5 nPa), magnetic erosion may be a main cause of this GMC. ------------------------------------------------ S2-P10 CHANNELS OF INFLUENCE OF THE SHORT-TERM CHANGES IN SOLAR ACTIVITY ON STATE OF THE LOWER ATMOSPHERE O. TROSHICHEV, A. Shirochkov, A. Frank-Kamenetsky, I. Gabis, L. Egorova, L. Makarova, and V. Vovk Arctic and Antarctic Research Institute, St.~Petersburg, Russia E-mail: olegtro@aari,nw.ru Analysis of long standing measurements of atmospheric perturbations in the Arctic and Antarctic regions has revealed that the polar lower atmosphere is affected by fluctuations of solar and galactic cosmic rays, changes in the solar wind pressure and variations of the interplanetary magnetic field (IMF). The cosmic rays variations crucially influence temperature and pressure regimes above the Antarctic plateau and can dramatically change the wind system above Antarctica. Changes in the UV irradiation have been found in connection with such manifestations of solar activity as the active regions, solar proton events, and passage of regions responsible for the Forbush decrease. It is significant that UV-irradiance starts to increase 5-7 days before the key day of solar proton events (SPE) or Forbush decrease, that can explain ahead reaction of the lower atmosphere to the Forbush decrease and SPE. Finding of the quasi-biennial periodicity in level of the solar UV irradiance implies that just change in solar UV radiation specifies the west or east direction of zonal winds in the Earth's equatorial stratosphere i.e., well-known quasi-biennial oscillations-QBO). Temperature and circulation regimes in the lower stratosphere are also crucially affected by the location of the magnetopause, i.e., by the solar wind pressure. The IMF parameters determine the electric potential patterns in the polar region and,influence, correspondingly, the global electric circuit generated by tropical thunderstorms. These four channels act simultaneously and inconsistently; mechanisms of their acion remain unclear. ------------------------------------------------ S2-P11 ON NONLINEARITY FEATURES OF THE CLIMATE SYSTEM Z. VOROS, A. Prigancova, and D. Jankovicova Geophysical Institute SAS, Slovak Republik E-mail: geomag@geomag.sk The issues of the Sun-climate connections attract the interest within the global change studies. To solve global problems an application of global approaches implying the consideration of as complete set of forcing factors as possible is required. To infer the patterns of the climate system response in a more reliable manner the plausible space weather signatures in the meteorological parameter changes are analysed. For this purpose the available time series of air temperature and indices of solar and/or geomagnetic activity on a daily value basis are used as input data for nonlinear analysis to reveal specific features of the climate system dynamical behaviour. The results obtained are discussed from the viewpoint of the nonlinear characteristics derived. ------------------------------------------------ S2-P12 NEW PROSPECTS IN PATTERN RECOGNITION OF SPACE WEATHER CONDITIONS Z. VOROS, D. Jankovicova, P. Dolinsky, and F. Valach GEOPHYSICAL INSTITUTE SAS E-mail: geomag@geomag.sk The Earth's magnetosphere exhibits considerable spatial changes and large temporal fluctuations within a wide range of scales which are directly related to the conditions present in the solar -- terrestrial system. In order to contribute to the understanding of solar wind -- magnetosphere interactions linear and nonlinear techniques will be used in the paper. The scaling and singularity properties of the field fluctuations will be modeled by multifractals. Simultaneously, multivariate statistics and neural networks representing powerful pattern -- recognition techniques for an improved classification of space weather conditions will be also applied. ------------------------------------------------ S2-P13 DIURNAL VARIATION OF GEOMAGNETIC ACTIVITY AND ITS ROLE IN SPACE WEATHER FORECAST W. LYATSKY (1,2) and A. M. Hamza (1) (1) Physics Department, University of New Brunswick, Fredericton, N.B., E3B 5A3 Canada (2) Polar Geophysical Institute, Apatity, 184200 Russia E-mail: lyatsky@unb.ca We calculated the diurnal variations both in the occurrence of large AE indices and in the AO index. Both methods show similar diurnal variations in geomagnetic activity with a deep minimum around (3-7) UT in winter and equinoctial months. In winter months, the probability of high geomagnetic activity versus UT varies by as much as a factor of 3, and in equinoctial months it can vary by a factor of 1.5. The observed UT variation is consistent with earlier results of other scientists, but it is very different from that expected from mechanisms proposed earlier to explain the seasonal variation. The existence of the diurnal variation in geomagnetic activity shows that some UT sectors are more or less favorable to the development of geomagnetic disturbances than other UT sectors. We calculated correlation patterns for the AE index versus solar wind parameters inside and outside the (02-07) UT sector related to the minimum in geomagnetic activity. The correlation patterns appear indeed different in these two sectors. It shows that it is possible to improve significantly the reliability of Space Weather forecast by taking into account the dependence of geomagnetic activity not only on solar wind parameters but also on universal time and season. Our test shows that a simple account for the dependence of geomagnetic activity on universal time can improve the reliability of Space Weather forecasting by at least two times in the (02-07) UT sector. ------------------------------------------------ S2-P14 NEAR-REAL TIME Kp ESTIMATES K. TAKAHASHI (1), B. A. Toth (1), J. V. Olson (2), and B. J. Anderson (1) (1) The Johns Hopkins University Applied Physics Laboratory (2) Geophysical Institute, University of Alaska E-mail: kazue.takahashi@jhuapl.edu The existing Kp index is the result of an effort to quantify geomagnetic disturbances in a simple manner. The index is derived from magnetic field data acquired at 13 ground stations distributed worldwide at moderately high latitudes (49-62 degrees). Kp is widely used to empirically specify the location of the cusp, plasmapause, and other plasma regions/boundaries, and also as input to various models of the magnetosphere and ionosphere. The official Kp index is delivered with a delay of many days so it is not useful for near-real time monitoring of the state of the magnetosphere. For operational purposes, one could obtain Kp estimates with much smaller delay using data available in near-real time from some stations. However, there are problems with the estimates: (1) the magnetometer data contain spikes and other anomalies and these need to be removed; (2) there is no clear definition of the quiet day curve that gives the baseline for the magnetic field perturbation at each station; (3) longitudinal distribution of the stations is not the same as those for the official Kp. In this paper we present a practical procedure to handle these problems. Specifically, we describe a data-adaptive data cleaning procedure, an automated procedure to define the quiet day curve, and a method to relate the magnetic field perturbation to the K index for individual stations. Test runs of our algorithm using data from 9 INTERMAGNET stations indicate that the correlation coefficient between the estimated and official Kp is higher than 0.9 with a standard error of approximately 0.5 (in Kp unit). ------------------------------------------------ S2-P15 SPACE ENVIRONMENT SIMULATOR FOR THE RESEARCH OF THE SPACECRAFT-PLASMA INTERACTIONS H. USUI, H. Matsumoto, and Y. Omura Radio Science Center for Space and Atmosphere, Kyoto University E-mail: usui@kurasc.kyoto-u.ac.jp One of the important issues to be studied in the space weather research is the influence of the solar activity variation to the spacecraft environment. In order to investigate the spacecraft-plasma interactions, we would like to propose the space environment simulator which enables us to perform large-scaled numerical simulations of the PIC (Particle-In-Cell) model. Since the simulator can solve the spatial and temporal evolution of the electromagnetic fields and plasma dynamics in the self-consistent manner, we can examine the physical process of the spacecraft-environment interactions including kinetic effects such as charging/discharging at the surface, plasma and field responses not only in the steady state but also in the transient state. A prototype of this simulator is planned to be installed as one of the components of the space simulation net laboratory. This net laboratory is now being developed and will enable us to perform numerical simulations through the network by providing the plasma environment of interest as the initial and boundary conditions. In the presentation, in addition to the concept of the simulator and the net simulation laboratory, we will present some results of the simulations associated with the spacecraft-plasma interactions such as observed in the electrodynamic tethered satellite experiments. ------------------------------------------------ S2-P16 THE STUDY OF IONOSPHERIC RESPONSE TO SOLAR FLARE OCCURRED ON NOV. 22, 1998 WITH GPS METHOD D. ZHANG, Z. Xiao, and Q. Chang Department of Geophysics, Peking University E-mail: zjeffery@263.net Ionospheric disturbances have been widely studied using dual-frequency GPS receivers. With the method a lot of results have been obtained in recent years. It is known that large solar flare can cause an ionospheric TEC enhancement, but the relation between them has not been fully established. Besides, since compared with the background TEC, the increase due to flare effect in general takes only a small percentage, there need a proper method to distinguish the effect. In this paper, a method is put forward to minimize the effect of noises so that small TEC changes caused by flares can be recognized( to a certain degree). Using data from 4 dual-frequency GPS receivers distributed over China, the temporal TEC variations of the ionosphere during the solar flare occurred on Nov 22, 1998 are calculated. The X-ray level of the flare is X3.7. The results indicate that an obvious TEC enhancements in the ionosphere does occur during the flare, the largest value of the TEC enhancement caused by this flare is about 1.25 TECU. Discussions on the features and validity of this method in studying such ionospheric disturbances are made in some detail in the end of this paper. It is concluded that this method has advantages over some traditional methods in the study of the ionospheric disturbances caused by solar flares and it can be used in analyzing the global ionospheric disturbance evolution using the data from GPS receivers scattered all over the world. ------------------------------------------------ S2-P17 RECENT OBSERVATIONS AND MODELING OF THE FORMATION OF POLAR CAP PATCHES C. E. VALLADARES (1) and T. Pedersen (2) (1) Institute for Scientific Research, Boston College (2) VSBI, Air Force Research Laboratory E-mail: cesar@dL5000.bc.edu Two campaigns aimed to measure plasma densities and electric fields in the regions where polar cap patches are formed were conducted at very high latitudes. The first campaign was performed on February 1996 and included simultaneous operations of the Sondrestrom and the EISCAT incoherent scatter radars. The second campaign was carried out on January 1999 and comprised the Sondrestrom and Svalbard radars. On both occasions the SuperDARN network of coherent radars observed the convection pattern in a large part of the auroral oval polar cap region. During the first campaign two different types of events were seen. In both events the Sondrestrom radar registered the formation and evolution of large-scale density structures. The first event consisted of the passage of traveling convection vortices, and the other event occurred in association with the development of large plasma jets (LPJ) embedded in the sunward convection part of the dusk cell. It was observed that on both types of events a section of the plasma density was eroded by a factor of 2. The measurements suggest that the number density reduction is caused by an enhancement in the O^+ recombination due to en elevated Ti which was produced by the much higher frictional heating inside the vortices or within the plasma jets. The data also provide evidence for abrupt changes in the location of the LPJ that creates regions containing dayside plasma almost detached from the rest of the oval density. Numerical simulations of the role of LPJ on producing density depletions indicate that transport of low-density plasma from earlier (or much later) local times can contribute to `60 % of the depletion. ------------------------------------------------ S2-P18 AURORAL ZONE GPS TEC MEASUREMENTS AND IONOSPHERIC BACKSCATTER FROM SUPERDARN P. PRIKRYL (1), H. G. James (1), S. Skone (2), and D. Andre (3) (1) Communications Research Centre, Ottawa, Canada (2) Dept.~of Geomatics Engineering, University of Calgary (3) ISAS, University of Saskatchewan E-mail: prikryl@cancrc.dgrc.crc.ca The dispersive and highly structured ionosphere can limit the accuracy of positions determined by Global Positioning System (GPS) satellite navigation systems and even result in a loss of carrier lock and intermittent GPS receiver operation. The auroral ionosphere is subject to space weather which can cause significant spatial and temporal variations of electron density and density gradients resulting in highly variable total electron content (TEC) and radio scintillation. Remote sensing of the ionospheric irregularities and their drift motions by SuperDARN provides maps of ionospheric HF backcatter. The ionospheric structure is thus characterized near the GPS line-of-sight ionospheric pierce point for GPS ground receivers measuring TEC. Some preliminary results from analysis of the two kind of data collected during a storm event will be presented. ------------------------------------------------ S2-P19 SPACE WEATHER PRODUCTS FROM SUPERDARN R. A. GREENWALD, R. Barnes, J. M. Ruohoniemi, and S. Shepherd Johns Hopkins University, Applied Physics Laboratory E-mail: ray.greenwald@jhuapl.edu SuperDARN provides near-continuous radar coverage of much of the high-latitude ionosphere and offers an excellent means of monitoring the auroral zone and polar cap. For the past two years, data from the northern-hemisphere radars of this network have been used to provide global views of high-latitude plasma convection as well as real-time determinations of the polar-cap potential drop. The quantities have appeared on the JHU/APL SuperDARN web site in near real-time and with two-minute temporal resolution. In this presentation, we consider other space-weather parameters that might be defined from the SuperDARN observations. In particular, recent studies have indicated that the equatorward boundary of backscatter observed with the radars is closely associated with the equatorward boundary of auroral precipitation. This boundary can also be associated with the equatorward boundary of high-latitude radar clutter and the equatorward boundary of the high-latitude scintillation zone. The flow reversal boundary in the convection maps is closely related to the poleward boundary of auroral precipitation (effectively the polar cap boundary). Thus, the SuperDARN radars provide a continuous identification of both the auroral zone and polar cap, in addition to a determination of the convection pattern. Using the SuperDARN receivers as riometers and combining data from them with data from high-latitude riometer networks, we have found that we can provide a global specification of the structure and dynamics of high-latitude absorption. Finally, intercomparison of the operating frequencies and backscatter characteristics of the various radars provides a real-time diagnostic of high-latitude propagation conditions. In this presentation, we will present examples of these various capabilities. ------------------------------------------------ S2-P20 STATISTICAL INVESTIGATION OF THE SATURATION EFFECT IN THE IONOSPHERIC FOF_2 VERSUS SUNSPOTS, SOLAR RADIO NOISE, AND SOLAR EUV J. Y. LIU (1) and Y. I. Chen (2) (1) Institute of Space Science, National Central University (2) Institute of Statistics, National Central University E-mail: jyliu@jupiter.ss.ncu.edu.tw This study explores the possible saturation effects in the ionospheric foF_2 due to smoothed sunspot number R12, smoothed solar radio noise (10.7 cm) flux F(10.7), and smoothed solar EUV. To locate the R12, F(10.7) or EUV value at which the foF_2 values are saturated, a two-segmented regression model is built based on the data of the strictly rise period of the 21st solar cycle recorded by eight ionosonde stations scattering roughly between 40^\circN and 40^\circS geomagnetic latitude. The regression model is then fitted into the foF_2 data observed at Chung-Li station to investigate the hourly variation of the saturation effect. To check the solar cycle variation of the saturation effect, the same model is further built based on the foF_2 data observed at Chung-Li station in the rise period of the 22nd solar cycle. Results show that clear saturation features appear around the equatorial anomaly crest region. ------------------------------------------------ S2-P21 CONTINUOUS MONITORING AND FORECASTING OF SPACE WEATHER BY USING ON-LINE COSMIC RAY DATA FROM THE WORLD NETWORK OF STATIONS L. I. DORMAN (1,2), N. Iucci (3), and G. Villoresi (4) (1) Israel Cosmic Ray Center and Emilio Segre Observatory, affiliated to Tel Aviv University, Technion and Israel Space Agency (2) Cosmic Ray Department of IZMIRAN, Troitsk, Moscow reg. (3) Terza Universita di Roma, Dipartimento di Fisica ``E. Amaldi" (4) IFSI/CNR c/o Terza Universita di Roma, Dipartimento di Fisica ``E. Amaldi" E-mail: lid@physics.technion.ac.il, lid1@ccsg.tau.ac.il, icrc@ccsg.tau.ac.il The main idea of International Cosmic Ray Service (ICRS) is to use on-line cosmic ray data from many stations of the existing world-wide network. In many investigations of historical events it was shown that by cosmic rays it is possible to obtain very important information on the space weather and dynamic processes in the Heliosphere. Moreover, we show that ground-based cosmic ray data (exchanged in real time in the frame of ICRS), can be effectively used for obtaining continuous information on the electromagnetic and radiation situation in the interplanetary space and in the Earth's magnetosphere, for forecasting of great geomagnetic storms associated with Forbush-decreases, for prediction of big increases of radiation hazard and other dangerous phenomena. We show that ICRS can predict also extremely big increases of radiation hazards very dangerous for the Earth's civilization due to extremely powerful solar flares and local supernova explosions. In the frame of ICRS, after some additional investigations of high energy cosmic-ray distribution function outside the Heliosphere, it could be possible in future to solve more complicated problems: to determine in combination with astrophysical methods the location and velocity of nearest dust-molecular galactic clouds with frozen-in magnetic fields and predict the expected time of the Sun capturing by some clouds with possible great global changes of Earth's climate. ------------------------------------------------ S2-P22 SPACE WEATHER IMPACTS ON THE EARTH: INCREASING OF THE FREQUENCY OF MYOCARDIAL INFARCTIONS, BRAIN STROKES AND TRAFFIC ACCIDENTS IN MOSCOW AND IN ST.~PETERSBURG IN PERIODS OF SPACE MAGNETIC STORMS ASSOCIATED WITH COSMIC RAY FORBUSH-DECREASES L. I. DORMAN (1,2), N. Iucci (3), N. G. Ptitsyna (4), and G. Villoresi(5) (1) Israel Cosmic Ray Center and Emilio Segre Observatory, affiliated to Tel Aviv University, Technion and Isael Space (2) Cosmic Ray Department of IZMIRAN, Russian Academy of Sciences (3) Terza Universita di Roma, Dipartimento di Fisica ``E. Amaldi" (4) SpbFilial of IZMIRAN, Russian Academy of Sciences (5) IFSI/CNR c/o Terza Universita di Roma, Dipartimento di Fisica ``E. Amaldi" E-mail: lid@physics.technion.ac.il, lid1@ccsg.tau.ac.il, icrc@ccsg.tau.ac.il On the basis of a great amount of medical information obtained in Moscow and St. Petersburg we show that cosmic ray Forbush-decreases can be considered as important indicators of space weather dangerous phenomena regarding their impact on the Earth. We analyzed 6304000 ambulance cases in Moscow for 1979-1981 (among them 85819 myocardial infarctions and 98625 brain strokes), 1314200 ambulance cases in St. Petersburg for 1981 (among them 14248 myocardial infarctions), 17005 heavy traffic accident cases in St. Petersburg for 1987-1989, and 15543 myocardial infarctions in hospitals of St. Petersburg in 1989-1990. We found that, regarding terrestrial impacts, it is important only the decreasing phase of cosmic ray Forbush-decreases (one-two days after the sudden commencement of geomagnetic storm). In these periods the frequency of myocardial infarctions increased in average by 12.5 \pm 1.5 % in Moscow and by 14.0 \pm 5.5 % in St.~Petersburg, the frequency of brain strokes increased by 7.0 \pm 1.7 % in Moscow and frequency of traffic accidents (with emergency service) increased by 17.4 \pm 3.1 % in St.~Petersburg. The forecasting of space phenomena causing terrestrial impacts and cosmic ray Forbush-decreases can be done by on-line data, precursory effects, changes in the 3-D cosmic ray anisotropy and changes in the cosmic ray spectrum of scintillations before the sudden commencement of geomagnetic storm. ------------------------------------------------ S2-P23 LIGHTNING -- AN INDEX OF SPACE WEATHER DIAGNOSTICS R. N. SINGH Emeritus Fellow, C.S.J.M. University, Kanpur-208016 India E-mail: vinodks@iitk.ac.in Solar radiations play an important role in controlling the space weather. The cloud formation, their movements and charge separation are known to result into lighting phenomena. The lighting phenomena has played a dominant role in the diagnostics of earth's lower and upper atmosphere. Initially the phenomena of lighting was thought to be arising from lower lying clouds which is known to move upwards as well as downwards. However, the diagnostic measurements were primarily confined to lighting radiations from cloud-to-ground (CG) discharges. In recent years, extensive study of lighting generated electromagnetic waves and their propagation have been carried out. It is now established that CG and cloud-to-cloud (CC) lighting discharges depict almost equal probability despite their dominantly different radiated electromagnetic wave power and associated polar distribution in space. The EM-waves propagate upwards undergoing multiple reflections between earth-ionosphere wave guide. The higher frequency components propagate outward and are monitored by space-borne devices. The variational features of CG and CC discharges are known to provide important details of ionized regions. Some of these details are being worked. However, it seems that we have spent more time in projecting nomenclature of some of these phenomena and the in-depth study of electromagnetic radiations and their propagation features are not yet fully developed. Some of these features are developed and discussed in detail.