The Sun to the Earth – and Beyond: Panel Reports (2003)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

SSUMMARY 1 27 3.1 INTRODUCTION 132 3.2 SCIENCE THEMES AN D OPPORTU N ITIES FOR THE COMI NG DECADE 1 38 Earth as a Particle Accelerator 138 Earth's Electric Field 142 Volati le Weather in the Upper Atmosphere 1 46 Micro- and Mesoscale Control of Global Processes 1 53 Dynamics of Geomagnetic Storms, Substorms, and Other Space Weather Disturbances Sol ar Vari abi I ity and C I i mate 1 59 Magnetospheric, ionospheric, and Atmospheric Processes in Other Planetary Systems 161 3.3 SOCIETAL IMPACT OF SPACE WEATHER 164 Communications 1 65 Navigation 1 67 Electric Power Issues 168 Astronaut, Ai rl i ne, and Satel I ite Hazards 1 69 Satel I ite Drag and Col I ision Avoidance 1 69 3.4 EXISTI N G PROG RAMS AN D N EW I N ITIATIVES 3.5 TECH NOLOGIES FOR THE FUTU RE 1 71 Data Assimi ration 1 71 S pacec raft an d I n stru ment Tech n o l ogy 3.6 RECOMMEN DATIONS 173 Major NSF Initiative 1 74 NASA Orbital Programs 176 NASA Suborbital Program 178 Societal I mpact Program 1 78 Maxi mizi ng Scientific Return 1 80 BIBLIOGRAPHY 1 82 1 72 125

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS SUMMARY Earth is the single most interesting object in the uni- verse to its inhabitants, the only place where we can be certain that a suitable environment for life exists. Fur- thermore, its complex systems are close enough to study in the sort of detail we will never obtain elsewhere. Earth and its sister planets are embedded in the outer atmosphere of the Sun. This outer atmosphere is con- tinually being explosively reconfigured. During these explosive events, Earth is engulfed in intense high-fre- quency radiation, vast clouds of energetic particles, and fast plasma flows with entrained solar magnetic fields. Even though only a small fraction (generally <10 per- cent) of this energy penetrates into geospace, the effects are dramatic. Space science programs to date have given us a detai led understanding of the average behavior of the component parts of geospace, in effect providing us with climatologies upon which to base educated guesses about the dynamic behavior of the global system. To go beyond this and understand the coupling processes and feedback that define the instantaneous response of the global system is much more difficult. The atmosphere- ionosphere-magnetosphere (A-l-M) system occupies an immense volume of space. At the same time, processes on scales from micro to macro impact the global system response. GOALS AND OBJECTIVES The overarching goals are as follows: 1. To understand how Earth's atmosphere couples to its ionosphere and its magnetosphere and to the at- mosphere of the Sun and 2. To attain a predictive capability for those pro- cesses in the A-l-M system that affect human ability to live on the surface of Earth as well as in space. Researchers currently have a tantalizing glimpse of the physical processes controlling the behavior of some of the individual elements in geospace. Some of the crosscutti ng science issues are these: · The instantaneous global system response of the A-l-M system to the dynamic forcing of the solar atmo- sphere, · The role of micro- and mesoscale processes in control I ing the global-scale A-l-M system, 1 27 · The degree to which the dynamic coupling be- tween the geophysical regions controls and impacts the active state of the A-l-M system, · The physical processes that may be responsible for the solar forcing of climate change, · The origin of the multi-MeV electrons in the outer magnetosphere and the cause of the pronounced fluc- tuations in their intensity, and · The balance between internal and external forc- ing in the generation of plasma turbulence at low lati- tudes. These critical science issues thread the artificial boundaries between the disciplines. The maturity of the A-l-M disciplines leads to a close connection between A-l-M science and applications for the benefit of society. The application of space physics and aeronomy to soci- etal needs is now referred to as space weather. The space weather phenomena that most directly affect life and society include radiation exposure extending from space down to commercial airline altitudes, communi- cations and navigation errors and outages, changes in the upper atmosphere that affect satellite drag and or- bital decay, radiation effects on satellite electronics and solar panels, and power outages on the ground due to geomagnetical Iy induced currents (GlCs), to name a few. STRATEGY AND REQUIREMENTS The next decade may revolutionize our understand- ing of the dynamical behavior of the A-l-M system in response to driving from both the solar wind and the lower atmosphere. A carefully orchestrated collabora- tion between agencies with interest in space weather and space science research is required, since no one agency has the resources to provide the global view. Furthermore, new ground-based and space-based ob- serving programs are required that make use of innova- tive technologies to achieve a simultaneous global view, highly resolved in space and time. Clusters of satellites flying in close formation can resolve dynamical response and separate spatial from temporal variations. New data storage and handling technologies are necessary to man- age the shear volume of data generated, the multisatellite correlations, the mapping between in situ observations and images, searches across distributed databases, and other essenti al fu ncti ons that wi I I be necessary i n the next decade to achieve an understanding of the entire system. The systems view requires enhanced efforts to de- velop global theoretical models of the Sun-Earth system, including the simultaneous development of new soft-

1 28 ware technologies for efficient use of paral lel computing environments and adaptive grid technologies to address the large range in spatial and temporal scales character- istic of the global system structure and response. How- ever, the A-l-M system is not simply multiscale, but it also requires inclusion of additional physical processes of ionized and neutral gases made up of individual par- ticles. Data assimilation technologies are crucial for in- tegrating new observations into research and operational models of the space environment. The problems associ- ated with the transition of research models and data sets to operations must be specifically addressed in the plan- ning and implementation of research programs aimed at i mprovi ng space weather forecast) ng and specification. The National Science Foundation's (NSF's) highly successful Solar, Heliospheric, and Interplanetary Envi- ronment (SHINE) program, its Coupling, Energetics, and Dynamics of Atmosphere Regions (CEDAR) program, and its Geosphere Envi ran ment Model i ng (G EM) pro- gram, and the recent coordination of these groups into Sun-to-Earth analysis campaigns, highlight the need to focus this broad range of expertise on issues involved in coupling between the Sun, solar wind, magnetosphere, and ionosphere/atmosphere regions. To this end, NSF recently funded the Science and Technology Center for I Integrated Space Weather Model i ng. NSF's i Information technology initiatives should be utilized as much as pos- sible to develop important collaboration technologies in support of such major community analysis efforts. The investigation of planetary A-l-M systems reveals details of value to understanding the terrestrial system. Future planetary missions should regularly be outfitted to carry out at least a baseline set of observations of the upper atmosphere, the ionosphere, and the magneto- sphere. In addition, theoretical studies linking our un- derstanding of the terrestrial environment with other planetary environments are an effective way of bringing extensive knowledge of plasma and atmospheric pro- cesses in the terrestrial environment to bear on the inter- pretation of planetary phenomena. While the National Oceanic and Atmospheric Ad- ministration (NOAA) and the Department of Defense (DOD) have pursued space environment forecasting for many years, their connection to the science community was facilitated by the inception of the National Space Weather Program (NSWP) in 1995 and NASA's new Liv- ing With a Star (LOOS) program. The NSWP is a multi- agency endeavor to understand the physical processes, from the Sun to Earth, that result in space weather and to transition scientific advances into operational applica- tions. NASA's new LWS program represents an impor- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS tent opportunity to provide measurements and develop models that will clarify the relationship between sources of space weather and their impact. Enhancements and innovations in infrastructure, data management and assi mi I ation, i nstru mentation, computational models, software technologies, and methods for transitioning research to operations are es- sential to support the future exploration of geospace. RECOMMENDATIONS In the next decade, NASA should give highest prior- ity to multispacecraft missions such as Magnetospheric Mu Itiscale (MMS), Geospace Electrodynamics Constel- lation (GEC), Magnetospheric Constellation (MagCon), and Living With a Star's geospace missions, which take advantage of adjustable orbit capability and the advanc- ing technology of smal I spacecraft. Missions that involve large numbers of simply instrumented spacecraft are needed to develop a global view of the system and should be encouraged. NSF, for its part, should support extensive ground-based arrays of instrumentation to give a global, time-dependent view of this system. Ground- and space-based programs should be coordinated as, for example, is being done in the Thermosphere-lono- sphere-Mesosphere Energetics and Dynamics (TIMED)/ CEDAR program to take advantage of the complemen- tary nature of the two distinct viewpoints. NASA, NSF, DOD, and other agencies should encourage the devel- opment of theories and models that support the goal of understanding the A-l-M system from a dynamic point of view. Furthermore, these agencies should work to- ward the development of data analysis techniques, us- ing modern information technology, that assimilate multipoint data into a three-dimensional, dynamic pic- ture of this complex system. Funding for the NASA Sup- porting Research and Technology (SR&T) program should be doubled to raise the proposal success rate from 20 percent to the level found in other agencies. SolarTerrestrial Probe (STP) flight programs should have their own targeted postlaunch theory, modeling, and data analysis support. Major NSF Initiative Simultaneous, multicomponent, ground-based ob- servations of the A-l-M system are needed in order to specify the many interconnecting dynamic and thermo- dynamic variables. As our understanding of the com- plexity of the A-l-M system grows, so does the require-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS ment to capture observations of its multiple facets. The proposed Advanced Modular Incoherent Scatter Radar (AMISR) will provide the opportunity for coordinated radar-optical studies of the aurora and coordinated in- vestigations of the lower thermosphere and mesosphere, a region not well accessed by spacecraft. Initial location at Poker Flat, Alaska, will allow coordination of radar with in situ rocket measurements of auroral processes. Subsequent transfer to the deep polar cap will enable studies of polar cap convection and mapping of pro- cesses deeper in the geomagnetic tail. 1. The National Science Foundation should extend its major observatory component by proceeding as quickly as possible with Advanced Modular Incoherent Scatter Radar (AMISR) and by developing one or more lidar- centered major facilities. Further, the NSF should begin an aggressive program to field hundreds of small auto- mated instrument clusters to allow mapping the state of the global system. Ground-based sensors have played a pivotal role in our understanding of A-l-M science and must continue to do so in the coming decade and beyond. Anchored by a state-of-the-art phased-array scientific radar, the $60 million AMISR is a crucial element for A-l-M. A distributed array of instrument clusters would provide the high temporal and spatial resolution observations needed to drive the assimilative models, which the panel hopes will parallel the weather forecasting models we now have for the lower atmosphere. Much of the neces- sary infrastructure for such a project has already been demonstrated in the prototype Suominet, a nationwide network of simple Global Positioning System (GPS)/me- teorology stations linked by the Internet. The proposed program would add miniaturized instruments, such as all-sky imagers, Fabry-Perot interferometers, very-low- frequency (VLF) receivers, passive radars, magnetom- eters, and ionosondes in addition to powerful GPS-based systems in a flexible and expandable network coupled to fast real-time processing, display, and data distribu- tion capabilities. Instrument clusters would be sited at universities and high schools, providing a rich hands-on environment for students and training with instruments and analysis for the next generation of space scientists. Data and reduced products from the distributed network would be distributed freely and openly over the Internet. An overall cost of $100 million over the 1 0-year plan- ning period is indicated. Estimated costs range from $50,000 to $1 50,000 per station depend) ng on i nstru- ments to be depl oyed. Adeq u ate fu nd i ng wou I d be i n- 1 29 eluded for the development and implementation of data transfer, analysis, and distribution tools and facilities. Such a system would push the state of the art in informa- tion technology as well as instrument development and . . . . m~n~atur~zat~on. Extendi ng the present radar-centered upper atmo- spheric observatories to include one or more lidar-cen- tered facilities is crucial if we are to understand the boundary between the lower and upper atmosphere. Fortunately, a number of military and nonmilitary large- aperture telescopes may become avai fable for transition to lidar-based science in the next few years. Highest priority would be given to a facility at the same geo- graphic latitude as one of the existing radar sites. NASA Orbital Programs The Explorer Program has since the beginning of the space age provided opportunities for studying the geo- space environment just as the Discovery Program now provides opportunities in planetary science. The contin- ued opportu n ities for U n iversity-CI ass Explorer (U N EX), Smal I Explorer (SMEX), and Medium-Class Explorer (MIDEX) missions, practically defined in terms of their funding caps of $14 million, $90 million, and $180 million, respectively, allow the community the greatest creativity in developing new concepts and a faster re- sponse time to new developments in both science and technology. These missions also provide a crucial train- ing ground for graduate students, managers, and engi- neers. Imager for Magnetopause-to-Aurora Global Ex- ploration (IMAGE), launched in March 2000, is an example of a h igh Iy successfu I Ml DEX mission; it was preceded by the first two ongoing SMEX missions, Solar Anomalous and Magnetospheric Particle Explorer (SAMPEX) and Fast Auroral Snapshot Explorer (FAST), launched in 1992 and 1996, which have provided enor- mous scientific return for the investment. The Aeronomy of Ice in the Mesosphere (AIM) SMEX was recently se- lected for launch in 2006. The UNEX program, after the great success of the Student Nitric Oxide Explorer (SNOE), launched in February 1998, has effectively been cancelled. This least expensive component of the Ex- plorer program plays a role similar to that of the sound- ing rocket program, with higher risk accompanying lower cost and a great increase in the number of flight opportunities. An increase in funding to $20 million per mission with one launch per year would make this pro- gram viable with modest resources.

1 30 2. The SMEX and MIDEX programs should be vigor- ously maintained and the U N EX program should quickly be revitalized. The STP line of missions defined in the NASA Sun- Earth Connection (SEC) Roadmap (strategic planning for 2000 to 2025) has the potential to form the backbone of A-l-M research in the next decade. The missions that are part of the current program include TIMED, launched in February 2002, Solar-B, the Solar Terrestrial Relations Observatory (STEREO), MMS, G EC, and MagCon. After TIMED, launched in February 2002, the next A-l-M/STP mission, MMS, is in the process of instrument selection for a 2009 launch. The STP cadence, with one A-l-M mission per decade (TIMED was significantly delayed), has fallen behind the NASA SEC Roadmap projections. 3. The panel heartily endorses the STP line of missions and strongly encourages an increase in the launch ca- dence, with GEC and MagCon proceeding in parallel. The A-l-M research community has very success- fully utilized the infrastructure developed within the In- ternational Solar-Terrestrial Physics (ISTP) program. The integration of the data from spacecraft and ground-based programs beyond those funded by the ISTP itself such as those of NOAA, LANE, and the DOD have contrib- uted substantially to our understanding of the global system. Comparisons between the Sun-Earth system and other Sun-planet or stellar-planet systems provide im- portant insights into the underlying physical and chemi- cal processes that govern A-l-M interactions. Improved understanding of A-l-M coupling phenomena such as planetary and terrestrial auroras would benefit from such an approach. 4. The Sun-Earth Connection program partnership with the NASA Solar System Exploration program should be revitalized. A dedicated planetary aeronomy mission should be pursued vigorously, and the Discovery Pro- gram should remain open to A-l-M-related missions. NASA Suborbital Program The NASA Suborbital program has produced out- standing science throughout its lifetime. Many phenom- ena have been discovered using rockets, rockoons, and teal loons, and many outstand i ng problems brought to closure, particularly when space-based facilities are THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS teamed with ground-based facilities. These phenomena include the auroral acceleration mechanism, plasma bubbles at the magnetic equator, the charged nature of polar mesospheric clouds, and monoenergetic auroral beams. This program continues to generate cutting-edge science with new instruments and data rates that are more than an order of magnitude greater than typical satellite data rates. Both unique altitude ranges and very specific geophysical conditions are accessible only to sounding rockets and balloons, particularly in the cam- paign mode. Many current satellite experimenters were trained in the Suborbital program, and high-risk instru- ment development can occur only in such an environ- ment. To accomplish significant training, it is necessary that a graduate student remain in a project from start to finish and that some risk be acceptable; both are very difficult in satellite projects. The high scientific return, coupled with training of future generations of space- based experimenters, makes this program highly cost- effective. The sounding rocket budget has been level-funded for over a decade, and many principal investigators (Pls) are discouraged about the poor proposal success rate as well as the low number of launch opportunities. The sounding rocket program was commercialized in 2000; in this changeover, approximately 50 civil service posi- tions were lost and the cost of running the program increased. Approved campaigns were delayed by up to a year, and it is not yet clear whether the launch rate will ever return to precommercial ized levels. Effectively, commercialization has meant a significant decline in funding for the sounding rocket program. An additional concern is that, as currently structured i.e., with a fixed, 3-year cycle for all phases of a sounding rocket project funding is not easily extended to allow gradu- ate students to complete their thesis work, because it is generally thought that such work should fall under the SR&T program, already oversubscribed. The rocket pro- gram has a rich history of scientific and educational benefit and provides low-cost access to space for uni- versity and other researchers. Further erosion of this pro- gram will result in fewer and fewer young scientists with experience in building flight hardware and will ulti- mately adversely affect the much more expensive satel- lite programs. 5. The Suborbital program should be revitalized and its funding should be reinstated to an inflation-adjusted value matching the funding in the early 1980s. To fur- ther ensure the vibrancy of the Suborbital program, an independent scientific and technical panel should be

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS formed to study how it might be changed to better serve the community and the country. Societal Impact Program The practical impact on society of variations in the A-l-M system falls into two broad categories: the well- established effects of space weather variations on tech- nology and the less clear yet tantalizing influence of solar variability on climate. The societal impacts of space weather are broad commu n Cations, navigation, human radiation hazards, power distribution, and sat- ellite operations are all affected. Space weather is of international concern, and other nations are pursuing parallel activities, which could be leveraged through collaboration. The role of solar variability in climate change remains an enigma, but it is now at least being recogn ized as i mportant to our understanding of the natural as opposed to anthropogenic sources of cli- mate variabi I ity. 6. The study of solar variability both of its short-term effects on the space radiation environment, communi- cations, navigation, and power distribution and of its effect on climate and the upper atmosphere should be intensified by both modeling and observation efforts. NASA's LivingWith a Star program should be imple- mented, with increased resources for the geospace com- ponent. Missions such as the National Polar-orbiting Operational Envi ran mental Satel I ite Systems (N POESS) and the Solar Radiation and Climate Experiment (SORCE) are needed to provide vital data to the science community for monitoring long-term solar irradiance. NPOESS should be developed to provide ionosphere and upper atmosphere observations to fill gaps in mea- surements needed to understand the A-l-M system. An L1 monitor should be a permanent facility that provides the solar wind measurements crucial to determining the response of the A-l-M system to its external driver, and the NSWP should be strengthened and used as a tem- plate for interagency cooperation. International partici- pation in such large-scope programs as LWS and NSWP is essential. 7. The NOAA, DOE/LAN L, and DOD operational spacecraft programs should be sustained, and DOD launch opportunities should be utilized for specialized missions such as geostationary airglow imagers, auroral oval imagers, and neutral/ionized medium sensors. 1 31 NASA's new Living With a Star program can, over the next decade, provide substantial new resources to address these goals. It is crucial that there be overlap between the geospace and solar mission components of LWS for the system to be studied synergistically, that resources be adequate for the geospace component, and that theory, modeling, and a comprehensive data sys- tem, which will replace the ISTP infrastructure, be de- fined at the outset, as called for in the Science Architec- tu re Team (SAT) report fi nd i ngs. NSWP, a mu Itiagency endeavor establ ished in 1 995, addresses the potential Iy great societal impact of physical processes from the Sun to Earth that affect the near-Earth environment in ways as diverse as terrestrial weather. The program specifi- cally addresses the need to transition scientific research into operations and to assist users affected by the space environment. Such multiagency cooperation is essential for progress in predicting the response of the near-Earth space environment to short-term solar variability. The interagency cooperation established in the NSWP is outstanding and is a model for extracting the maximum benefit from scientific and technical pro- grams. It has also been effective at bringing together different scientific disciplines and the scientific and op- erations communities. Interagency cooperation has worked well in the AFOSR/NSF Maui Mesosphere and Lower Thermosphere Program, and it has been key to the success of the NOAA GOES and POES programs of meteorological satellites with space environment moni- toring capabilities. International multiagency coopera- tion has been very successful for the ISTP program, which involves U.S., European, Japanese, and Russian space agencies. Global studies require such interna- tional cooperation. The panel recognizes that much more science can be extracted by careful coordination of ground- and space-based programs. Maximizing Scientific Return Funding for NASA's Supporting Research and Tech- nology program, including guest investigator studies and focused theory, modeling, and data assimilation efforts, is essential for maximizing the scientific return from large i nvestme nts i n s pacec raft h ardware. Supporting Research and Technology Wh i le spacecraft hardware projects are con- centrated at relatively few institutions, the NASA SR&T program is the primary vehicle by which independent investigations can be undertaken by the broader com-

1 32 munity. Likewise, NSF helps individual investigators to carry out targeted research through its Division of Atmo- spheric Sciences (ATM) base programs SHINE, CEDAR, and GEM. Such individual Pl-driven initiatives are the most inclusive, with data analysis as well as theoretical efforts and laboratory studies, and often lead to the high- est science return per dollar spent. The funding for such program elements falls far short of the scientific oppor- tunities, with the current success rate for submitted NASA SR&T proposals being 10 to 20 percent. Further- more, limited available SR&T funds have been used for guest investigator participation in underfunded STP-class flight programs. Without adequate MO&DA funding for NASA orbital and suborbital programs, the SR&T budget intended for targeted research on focused scientific ques- tions has been utilized to support broader data analysis objectives. 8. The funding for the SR&T program should be in- creased, and STP-class flight programs should have their own targeted postlaunch data analysis support. 9. A new small grants program should be established within NSF that is dedicated to comparative atmo- spheres, ionospheres, and magnetospheres (C-A-I-M). A new C-A-I-M grants program at NSF would allow the techniques (modeling, ground-, and space-based observations, and in situ measurements) that have so successfully been applied to A-l-M processes at Earth to be used to understand A-l-M processes at other planets. Such a comparative approach would improve our un- derstanding of these processes throughout the solar sys- tem, including at Earth. Currently, a modest $2 million planetary science program at NSF covers all of solar system science (except for sol ar and terrestrial stud ies), with only a small fraction going to planetary A-l-M re- search. Theory, Modeling, and Data Assimilation Theory and modeling provide the framework for in- terpreti ng, understand) ng, and visual izi ng diverse mea- surements at disparate locations in the A-l-M system. There is now a pressing need to develop and utilize data assimi ration techniques not only for operational use in specifying and forecasting the space environment but also to provide the tools to tackle key science questions. The modest level of support from the NSF base pro- grams (CEDAR, GEM, SHINE) and NASA SR&T has been inadequate to build comprehensive, systems-level mod- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS els. Rather, individual pieces have been built, and first stages of model integration achieved with funding from such programs as NASA's ISTP program and its Sun- Earth Connections Theory Program (SECTP), the AFOSR MU RI program, NSF Science and Technology Center programs, and the multiagency support to such efforts as the Community Coordinated Model ing Center. Such pro- grams enable the development of theory and modeling infrastructure, including models to address the dynamic coupling between neighboring geophysical regions. Their value to the research community is clearly their provision of longer-term funding, which has been essen- tial to developing a comprehensive program outside the purview of SR&T. 10. The development and utilization of data assimila- tion techniques should be enhanced to optimize model and data resources. The panel endorses support for theory and model development at the level of the NASA Sun-Earth Connections Theory Program, the AFOSR/ ONR MURI program, NSF Science and Technology Center programs, and the multiagency support to such efforts as the Community Coordinated Modeling Cen- ter (CCMC). Support should be enhanced for large- scope, integrative modeling that applies to the cou- pling of neighboring geophysical regions and physical processes, which are explicit in one model and implicit on the larger scale. The preceding science recommendations can be grouped into three cost categories and prioritized (see Table 3.11. Equal weight is given to STP and LWS lines, as indicated by funding level. Smal I programs are ranked by resource allocation, while the Advanced Modular Incoherent Scatter Radar is the highest priority moderate initiative at lower cost than others. 3.1 INTRODUCTION Earth, unique in the universe as the only object known to support life, follows an orbit in the outer at- mosphere of the Sun an outer atmosphere that is con- tinually being explosively reconfigured. During these events, Earth is engulfed in intense high-frequency ra- diation, vast clouds of energetic particles, and fast plasma flows with entrained solar magnetic fields. Even though only a small fraction (generally <10 percent) of this energy penetrates into geospace, its effects are dra- matic.

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS TABLE 3.1 Panel's Recommended Priorities for New Initiatives Initiatives in Geospace Major SolarTerrestrial Probes (2) Living With a Star Discovery (1) Moderate Advanced Modular Incoherent Scatter Radar and Lidar Facilities Explorer Program (assume 3 missions in the 10 years will be devoted to AIM) L1 Monitor (excluding tracking) Small Instrument Distributed Ground- Based Network Recommended 10-Year Funding (million $) 800 500 350 Subtotal 1,650 92 300 50 100 Subtotal 542 Small Suborbital program NSF Supporting Research and Technology National Space Weather Program 50 NSF SHINE, CEDAR, GEM, C-A-I-M (new) Theory Living With a Star (geospace) Sun-Earth Connection Theory Program (geospace) DOD MURI (ionosphere) NSF STC (geospace) HPCC (geospace) NOAA, DOE/LANL, and DOD science for the A-l-M community 50 Subtotal 848 300 200 135 138 60 18 20 20 20 Total 3,065 To date, space science programs have provided detai led understanding of the average behavior of the component parts of geospace, in effect providing climat- ologies upon which to base educated guesses about the dynamic behavior of the global system. To go beyond this and understand the coupling processes and feed- back that define the instantaneous response of the glo- bal system is much more difficult. The A-l-M system occupies an immense volume of space. At the same time, processes on scale sizes from micro to macro im- pact the global system response. The ISTP program is the most ambitious program to date to explore the A-l-M system. ISTP samples the huge volume of the A-l-M system by simultaneous measure- ments from a handful of satellites. Despite the sparse a 1 33 coverage, analysis of data from ISTP satellites has al- lowed scientists to begin to glimpse the rich variety of coupling and feedback processes that define the global response of the geospace environment to solar wind disturbances. The first experiments with innovative im- aging technologies that view large regions of geospace in snapshots (e.g., from the IMAGE spacecraft) have al- ready provided insights into the instantaneous response, unattainable by past missions. The first attempts to achieve the high spatial and temporal resolution needed to survey the microscale controls of the global system (e.g., from the FAST spacecraft) have revealed new de- tails about acceleration processes and electrodynamic coupling. With these new missions, we are replacing our steady-state view of geospace regions with a dy- namical view. But we are far from understanding the complex coupling processes and interplay between components that dictate the integrated global system response. It is clear that the A-l-M system actively responds to the solar wind and that components of this system may be preconditioned or may interact in ways that redistrib- ute solar wind energy throughout the system, actively limiting the entry of solar wind energy into geospace during extreme events. A few examples are given in the next pages to illustrate the complexity of this interaction and the challenges that lie ahead. Life on this planet is protected from the high-energy radiation and dangerous particle clouds in interplan- etary space because Earth has its own magnetic field and is surrounded by an absorbing atmosphere. Earth's magnetic field presents a northward-directed magnetic field barrier to the oncoming solar wind in the ecliptic plane (Figure 3.1~. This barrier can be breached, how- ever, if southward-d i rected sol ar magnetic fields i mpact it and merge or reconnect with Earth's magnetic field. Fortunately, strongly southward-directed magnetic fields are not a persistent feature of the quiet interplanetary medium. They are mainly confined to structures gener- ated in explosive solar events and in high-speed plasma streams. The passage of southward interplanetary magnetic field (IMP) structures by Earth pumps energy into the near-space environment. The tightly coupled nature of the A-l-M system is clearly revealed by its response to interplanetary magnetic clouds (IMCs), which have strong and long-lived southward magnetic fields and drive the most intense magnetic storms. Intense convec- tion is produced, which brings particles from the mag- netotail storage region (called the plasma sheet) deep into the inner magnetosphere on open drift paths, ener-

1 34 THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS FIGURE 3.1 (a) A schematic of the magnetosphere showing major particle populations and current systems; (b) close-up of radiation belts (trapped particles), including inner and outer zone populations and trapped anomalous cosmic rays (interstellar matter).

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS Sizing them to form the storm-time ring current, shown schematically in Figure 3.1. Under extreme conditions, the strong current produced by these particles cannot close upon itself in the equatorial plane (to form the ring of current that its name implies) but is forced to close through the subauroral and midlatitude ionosphere. This closure produces a strong electric field in the ionosphere called a polarization jet. The electric fields in the polar- ization jet map upward along the field lines to the mag- netosphere, producing a penetration electric field in the inner magnetosphere, further changing the drift paths of the ring-current ions. The plasmasphere (corotating plasma of ionospheric origin) responds strongly to the penetration electric fields and enhanced convection. Long plumes of plasmaspheric material snake out to the dayside magnetopause, draining thermal plasma into the dayside boundary layers. This plasma may later become a source for the plasma sheet (Figure 3.11. Along drift paths mapped into the ionosphere, storm-enhanced ion- ization moves toward the dayside polar cap, where geo- magnetic field lines connect to the interplanetary mag- netic field. Steep plasma density gradients at the edges of ionospheric density patches form in the changing con- vection pattern, playing havoc with technologies like the Global Positioning System, and ionospheric irregu- larities form, disrupting communication systems. There are indications that the ionosphere may, in turn, have a major impact on the dynamics of the mag- netosphere. Solar wind dynamic pressure variations trig- ger ionospheric outflows from the vicinity of the polar cap. These ionospheric "mass ejections" begin well be- fore the storm maximum, preconditioning the tail plas- mas with heavy ionospheric ions energized by the solar wind interaction. Enhanced convection during magnetic storms stresses the magnetotail, producing dramatic re- configurations of the basic structure of the magnetotail, called substorms. Auroral currents associated with sub- storms also produce an outflow of ionospheric ions di- rectly into the magnetotail. Substorms may actually sever the outer plasma sheet from the magnetotail, producing a major loss of plasma and energy. Dipolarization of the magnetic field during substorms, which reduces the stretching of magnetotail field lines, generates intense electric fields, which accelerate the storm-enhanced plasma sheet. Since this accelerated plasma drifts earth- ward under conditions of strong convection to form the ring current, there is a clear connection between mag- netotail dynamics and magnetic storm effects in the near- Earth region of the magnetosphere. Variations in plasma sheet density play an important role in modulating mag- netic storm intensity and substorm processes and repre- 1 35 sent another means by which the A-l-M system inter- nally modulates the geo-effectiveness of solar wind dis- turbances. The interplay between removal of plasma sheet material and refilling of the plasma sheet from the solar wind and the ionosphere to supply the ring current during storms is not understood. Even the basic mecha- nisms for refilling the plasma sheet (Figure 3.1 ) have not yet been determined. Earth's outer magnetosphere is often populated to a surprising degree by relativistic electrons, which pose a radiation hazard to space-based systems. The origin of the multi-MeV electrons in the outer zone is not known. They are generally correlated with increased substorm activity driven by high-speed solar wind and favorable coupl ing to southward interplanetary magnetic field, both of which have semiannual and solar-cycle varia- tions. Enhancements occur with relatively regular 27- day periodicity during the declining phase of the 11- year sunspot cycle and are wel I associated with high-speed solar wind stream structures. Flux variations at the solar activity maximum are dominated by coronal mass ejections (CM Es) (see the report of the Panel on Solar-Heliospheric Physics), which launch magnetic clouds toward Earth, producing geomagnetic storms. The mechan ism~s) by wh ich magnetospheric particles are accelerated to relativistic energies are unknown at present, although a number of interactions with plasma wave modes are promising candidates. The impact of solar wind shock structures on Earth's magnetosphere has been shown to generate induction-electric-field pulses that rapidly accelerate electrons and protons, gen- erating an entire new radiation belt in a matter of min- utes. An interesting coupling between the ring current and radiation belts results wherein magnetic fields gen- erated by the ring current cause scattering and loss of radiation belt particles. In addition, waves near the ion gyrofrequency generated by ring-current ions are be- lieved to contribute to electron precipitation losses in the dusk sector. Dramatic losses from the electron radia- tion belts (Figure 3.1b) also result from interaction of energetic electrons with lightning-generated waves, cal led whistlers. Lightning-induced electron precipita- tion events exemplify direct coupling of tropospheric weather systems with the radiation belts and the iono- spheric regions overlying thunderstorms. Much of the energy and momentum entering Earth's environment eventually finds its way into the upper at- mosphere. The in situ absorption of solar EUV radiation not on Iy protects I ife on Earth, but drives large day-night temperature and tidal wind variations in the upper at- mosphere, which vary dramatically with the solar cycle.

1 36 During geomagnetic storms the additional energy and momentum input from the solar wind is initially focused at the poles but is so intense that ultimate effects are worldwide. The upper atmospheric regions underlying the polar cap and the adjacent auroral zone are not only heated but are also subject to momentum transfer from the solar wind. The ionosphere is the intermediary in this momentum transfer and acts like an ion engine driv- ing the wind in huge twin vortices in both hemispheres. Uninhibited by plasma boundaries, this wind permeates the globe, generating its own electric fields by dynamo action. This additional source of electrification can modify and even reverse the polarity of the ambient low-latitude electric field. It thus can turn on and off the ionospheric processes responsible for much of the space weather that adversely affects communication and navi- gational systems. Impulsive high-latitude energy inputs driven by substorm reconfigurations of the goemagnetic tail generate tsunami-like atmospheric waves, which propagate at near-supersonic speeds across the upper boundary of the atmosphere, disturbing the ionosphere along their path. The neutral atmosphere is an active player in the coupled system. It is heated at high latitudes by auroral currents and particle precipitation associated with the substorm current systems, expanding to higher altitudes and undergoing changes in composition. The tempera- ture and composition changes move in great sloshing waves from the conjugate auroral regions, meet at the equator, and pass through, modifying atmospheric con- ditions as they travel. The changing neutral atmosphere modifies the ionospheric plasma, creating traveling dis- turbances that have yet to be successfully modeled. These variations alter the conductivity of the ionospheric plasma and thus actively modulate the current flow be- tween the ionosphere and magnetosphere. Even without the energy and momentum input from the solar wind the ionosphere exhibits a rich variety of weather. I n fact, the h ighest d istu rbance level of trans- ionospheric or subionospheric communication channels and signal propagation occurs at low latitudes. Here the nighttime ionosphere is supported against gravity by electrical currents. But this equilibrium is unstable, and the slightest disturbance is enough to set off huge con- vective upwellings, much like thunderstorms, which rise to heights more than 1,000 km and last for hours. This phenomenon has been intensely studied by ground- based radars and more recently by rockets and satellites. Some of the earliest successes in computer simulations occurred when researchers applied codes developed for nuclear weapons effects to this explosive natural phe- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS nomenon. At midlatitudes we have been greatly sur- prised by the pictures of the ionosphere that are now available and that reveal much more structure than pre- viously thought. A highly remarkable coupling between atmospheric waves and ionospheric electric fields seems to be occurring, which we simply do not understand. Ground-based observations indicate that the upper/ middle ionosphere is driven from below by both me- chanical and electrodynamic inputs from the massive lower atmosphere. Gravity waves launched from thun- derstorms and orographic features on Earth's surface grow as they propagate upward and break in the meso- sphere, driving the large-scale flow and mixing of con- stituents. Lightning activity in tropospheric thunder- storms imposes huge transient electric fields on the middle atmosphere, leading to intense heating and ion- ization, manifested by transient luminous displays known as sprites and elves. These same electric fields can drive avalanche acceleration of relativistic electron beams, which produce gamma-ray emissions during their upward traverse and may escape into the radiation belts. Tropospheric thunderstorm activity may influence the electrical and chemical properties of the middle and upper atmosphere by means of these newly discovered phenomena and by upward conduction currents. The coldest temperatures on Earth, achieved at the summer menopause, are due to dynamical processes. There is much to learn about the dynamics and coupling between the upper/middle and lower atmospheric re- gions and even more about the possible impact on weather and climate. Noctilucent clouds (Figure 3.2), observed at 82 km altitude at 50 to 60 degrees latitude when temperatures drop below 140 K, are occurring more frequently, and sightings are moving to lower lati- tudes. These are part of a growing body of information that indicates the upper atmosphere has cooled over the past 20-50 years a cooling that is thought to be associ- ated with a warming trend at lower altitudes, possibly due to anthropogenic influences. Variations in the solar constant with solar activity seem too small to serve as the basis for such changes. Answers may lie in the solar cycle modulation of cloud nucleation through cosmic ray intensity, the effects of energetic particle precipita- tion on ozone chemistry, and solar cycle changes in the global electric circuit. A significantly improved under- standing of the links between the upper atmosphere and weather and climate is needed. NASA's TIMED space- craft mission, launched in February 2002, is providing an exploratory look at the response of the mesosphere and lower thermosphere to solar and magnetospheric inputs from above and atmospheric inputs from below.

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS Lo an June 22/23 1999 t41 6° Ni g , ~ ~ 1 37 FIGURE 3.2 Image of noctilucent clouds (NLC) observed at midlatitudes from Logan, Utah (41.6 N. 1 11.8 W), recorded at 10:30 p.m. on June 22 and 23,1 999.The NLC consisted of two bright patches containing diffuse and billow-type wave structures that appeared bluish white in the twilit sky (the dark silhouettes are due to tropospheric clouds). Faint NLC are also evident toward the image center. NLC have been observed at high latitudes (>55 degrees) for more than 100 years, with recent evidence for migration to midlatitudes suggesting climate change. Courtesy of M.J.Taylor, Utah State University. The AIM Explorer, scheduled for launch in 2006, will investigate the causes of high-altitude noctilucent clouds and serve as a baseline for further study of long-term changes in the upper atmosphere. Finally, the solar system is a natural laboratory for comparing planetary A-l-M systems. Comparison of the Sun-Earth system to other Sun-planet systems provides insights into the broad atmospheric, ionospheric, and magnetospheric responses under a variety of conditions. The physical and chemical processes that control the atmosphere and plasma environments of the planets and their satellites are essentially the same throughout the solar system, but they are manifested in very different ways due to differing heliocentric distances, planetary sizes, atmospheric composition, and intrinsic magnetic fields. Studies of the A-l-M systems on other planets are significant in a number of ways. First, the interpretation of plasma processes on other planets is often aided by knowledge of similar processes on Earth. Second, pro- cesses that have counterparts on Earth are often modi- fied in interesting ways in the extreme environments of other planets (such as possible substorms on Mercury in the absence of an ionospheric source of plasma and substorms on Jupiter to replace magnetic flux drifting out under the influence of outflowing torus plasma). Third, by studying a wide variety of solar wind-mag- netosphere environments, a range of scenarios is built up for use in interpreting conditions on extrasolar plan- ets. Within this category falls the study of plasma pro- cesses with no direct analogue at Earth. Lastly, the study of basic plasma physics processes within these plan- etary environments supplies important information on processes operating within astrophysical plasmas throughout the universe. A-l-M science themes for the coming decade are organized not by isolated regions in geospace but by crosscutting themes that describe global system behav- ior. These include the Earth as a particle accelerator; the global electrical environment; a new view of the neutral atmosphere; micro- and mesoscale control of the be- havior of the global system; the response of the global system to forcing, exemplified by magnetic storms and substorms; the possible impact of the A-l-M system on global climate; and comparative aspects of planetary A-l-M systems. The maturity of the A-l-M discipline leads to a close connection between A-l-M science and applications for the benefit of society. The application of space physics and aeronomy to societal needs is now referred to as space weather. Space weather phenomena that most

1 38 directly affect life and society include radiation expo- sure extending from space down to commercial airline altitudes; communications and navigation errors and outages; changes in the upper atmosphere that affect satellite drag and orbital decay; radiation effects on sat- ellite electronics and solar panels; and power outages on the ground due to geomagnetically induced currents (G ICs), to name a few. H igh-frequency satel I ite-based navigation systems, of which the Global Positioning Sys- tem is an example, rely on propagating signals through the ionosphere from ground to satellite. Any outage or degradation due to space weather is a threat to life as we l I as to m i I itary advantage. Ionospheric i rregu I ariti es, therefore, are just as important for navigation applica- tions as they are for satellite communication. Precise positioning can also be compromised by the total elec- tron content (TEC) of the ionosphere. The highly dy- namic nature of the system introduces an uncertainty for single-frequency GPS users (for instance, commercial airlines). HF is the communication channel most sus- ceptible by far to space weather, with absorption refrac- tion, multipath, and scintillation all adding to the prob- lems. Many mi I itary systems depend on H F and U H F communications systems, as do commercial airlines. Systems that use HF waves that refract from the bottom of the ionosphere are vulnerable to space weather ef- fects. During large geomagnetic storms, rapidly time-vary- ing electric fields and currents, often co-located with the aurora, intensify and expand to lower latitudes, some- times producing power outages that affect transmission I i nes th rough G ICs. With i mproved forecast) ng, power companies can take protective measures to reduce or eliminate effects on their systems. Astronauts are subject to considerable radiation even in low Earth orbit. When the International Space Station orbit was changed to accommodate Russian launches, it entered a more dan- gerous portion of the radiation belt. Even polar airplane flights have high radiation levels. Likewise, instruments and entire satellites are placed at risk because of space- craft charging by energetic electrons, loss of computer- ized control systems, and severe magnetic torques. In addition, satellites in low Earth orbit are significantly affected by atmospheric drag. During a magnetic storm the atmosphere is heated and expands. After one such event, NORAD lost track of many space objects and had to reacquire them. Our knowledge of the timing of such events and their prediction could greatly assist in keep- ing track of all the space debris in orbit. THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS 3.2 SCIENCE THEMES AND OPPORTUNITIES FOR THE COMING DECADE EARTH AS A PARTICLE ACCELERATOR High-energy particles, photons, and radio waves permeate the cosmos and provide glimpses of particle acceleration processes throughout the universe. Remark- ably, Earth is also capable of generating powerful beams of this sort. A classic example arose when researchers using Compton Gamma Ray Explorer data found that thunderstorms near the surface of Earth were the source of the intensely energetic photons. The advantage of near-space regions of Earth as a cosmic particle accel- erator and the source of photons is that we are free to delve deeply into the origins and effects of these pro- cesses. Detailed study is possible and we are not forced simply to blend theory with the small amounts of infor- mation we can obtain from distant sources. In the next sections the panel discusses a few of the more spectacu- lar acceleration mechanisms that have been found in the near-space regions of Earth and that we hope to understand by the end of this decade. Auroral Physics Earth's aurora was one of the earliest motivations for the study of space plasmas, and it remains one of the most fu ndamental areas of space research (Figu re 3 .31. With in situ measurements provided by rockets and sat- ellites, great progress has been made in understanding these stunningly beautiful light displays. Nonetheless, the physics of auroral processes is among the most com- plex in all of space physics, spanning the entire range from macroscale to microscale. The broad outlines of auroral physics are generally understood. Auroral electrons are accelerated out of Earth's plasma sheet to energies of between 1 and 10 keV, and when they strike the atmosphere, they excite atmospheric atoms, which then emit the characteristic greenish blue and reddish purple colors of the aurora. The source of energy for this process appears to come from the magnetotail and is closely related to disturbed geomagnetic conditions and substorms. The details of exactly how the particle acceleration occurs remains a fundamental question of auroral physics. It is generally agreed that quasi-static electric fields parallel to Earth's

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS FIGURE 3.3 The aurora viewed from Alaska. Courtesy of Jan Curtis. 1 39 magnetic field are the dominant form of acceleration. Just how these fields are set up and maintained is not known, but it is closely linked to the flow of current upward along the magnetic field. Auroral electrons can also be accelerated by wave processes such as the Alfven wave, a transverse pertur- bation of the background magnetic field by an i educed electric field, which causes plasma to oscillate as if tied or frozen in to the magnetic field. These waves are thought to be launched from the equatorial tail region and to propagate toward the ionosphere. Alfven waves are observed in both the auroral zone and at altitudes well above the auroral acceleration region. Theoretical work also suggests a link between the Alfven waves and the quasi-static fields, in which the waves, by reflecting off the ionosphere, can evolve. Alfven waves are also thought to occur as part of global, field line resonances, in which a standing wave is set up along a field line from one ionosphere to the other. In all of these Alfven wave processes, the perpendicular extent of the wave is a key parameter, but one that has not been measured. Where an aurora occurs, a variety of associated physical phenomena occur along the magnetic field line that threads the ionospheric end where the light emis- sion occurs. Just as electrons are accelerated downward, ions are also accelerated upward to the same energies. Ions are also observed to be heated, predominantly in the perpendicular direction. Owing to the magnetic mir- ror force, which reflects charged particles away from a stronger magnetic field region, these heated ions ac- quire parallel as well as perpendicular energy as they move upward. These field-aligned particle flows are unstable to a variety of plasma waves that are emitted from the auroral regions. In the past decade, two satellite missions, Freja and FAST, along with several sou nd i ng rockets, have el uci- dated many details of this picture of the auroral accel- eration region. These spacecraft exploited high data rates

140 and state-of-the-art instrumentation. Perhaps the most striking result of these missions is the understanding that not only is the downward acceleration of electrons im- portant, but also the upward acceleration of electrons plays a regular role in auroral processes. FAST has made clear just how ubiquitous this upward acceleration of electrons is and that field-aligned current is fundamental to the process. A variety of waves are also associated with this downward current region. This has brought a symmetry to the auroral acceleration process, in which electrons are accelerated both upward and downward, with the current that they carry closing through the iono- sphere. Measurements from FAST have also shown that within the region where electrons are accelerated, the dense background ionospheric plasma is frequently evacuated. The mechanism by which these cavities can be created is not yet known. Other discoveries include the association of ion heating with a variety of low- frequency waves, the observation of solitary wave struc- tures in both upward and downward current regions, and the correlation of high-frequency waves with elec- tron bunching. Despite this I ist of accompl ishments, significant questions remain: How is the quasi-static potential drop that accelerates the electrons distributed along the mag- netic field? How is this potential drop set up and main- tained? What are the perpendicular scales of the Alfven waves that are observed? How are auroral density cavi- ties generated? What role do solitary structures play in auroral physics? Most of these are not new questions, but rather are questions that are not tractable with single spacecraft and exist) ng theories. Several key developments are needed: (1 ~ the rec- ommended ground-based instrumentation, including the AMISR, (2) a multispacecraft mission for auroral sci- ence, and (3) the creation of sophisticated, self-consis- tent plasma models of the auroral region. On the mea- surement side, a dedicated, three- or four-spacecraft mission that allows determination of the evolution of temporal phenomena and spatial distribution (both along and across field lines) of auroral features is needed to make progress on issues such as the structure of elec- tric fields in both quasi-static and wave cases. On the theoretical side, it is time to develop models that can handle an entire auroral field line; the effects of precipi- tating electrons, protons, and heavier ions; and the phys- ics of the cold, dense ionospheric plasma, using nonideal magnetobydrodynamic and two-fluid simula- tions evolving to include two- and three-dimensional particle effects. THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS Ring Current The ring current, consisting of ions ranging in en- ergy from tens to hundreds of keV, embodies sufficient plasma pressure to reduce the magnetic field locally and measurably at equatorial locations on Earth's surface, especially during geomagnetic storm periods of en- hanced convection, u Itimately driven by the solar wind. The contribution of radiation belt particles (1 keV to many MeV) to plasma pressure is negligible, which al- lows them to be treated as test particles in dynamic models, neglecting the fields they produce. During geo- magnetic storms, the ring current can consist primarily of oxygen of ionospheric origin, posing the question, How can eV ions be so quickly accelerated to 100 keV energy on the time scale of enhanced magnetospheric convection (a few hours)? Recent measurements from the Polar and Cluster spacecraft suggest that strong solar wind pressure pulses, which frequently accompany the onset of a geomagnetic storm, play an important role in reconfiguring the dayside cusp region, providing iono- spheric plasma access to the dayside region, where plasma is swept antisunward on reconnected magnetic field lines, increasing the oxygen content of the plasma sheet in the tail (see Figure 3.11. Enhanced convection then brings the oxygen-enriched plasma earthward to form the storm-time ring current. Global measurements of ring current development from the IMAGE satellite, combined with plasma sheet composition measurements of the source population from the planned MMS mis- sion, as well as cusp and nightside auroral zone outflow measurements, wi l l make it possible to trace the energization steps and determine, with solar wind input measurements, which storm triggers are most geoeffective at producing an oxygen-enriched ring cur- rent. Recovery rates of the storm-time ring current, which affects the buildup of relativistic electrons inside geo- synchronous orbit, are species-dependent. Radiation Belts The radiation belts are divided into an inner zone, with peak flux around L = 2 (approximate equatorial distance in units of Earth radii), which comprises pre- dominantly protons produced by cosmic ray interaction with the atmosphere, and a dynamic outer zone of elec- trons whose flux is determined by solar wind control of magnetospheric processes. Figure 3.4 shows the elec- tron variability over most of a solar cycle (bottom panel), as seen by the low-altitude polar-orbiting SAMPEX satel- lite. The most intense fluxes are seen during the declin-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS ing phase of the solar maximum period (1993-1995) and are associated with intervals of recurring high-speed solar wind stream interaction with Earth's magneto- sphere. These high-speed streams map back to solar coronal hole access to the ecliptic plane. Approaching the solar maximum (2000-2001), the flux enhancements are more sporadic and can be identified with CME- driven geomagnetic storms. The solar minimum was quite evident in mid-1996. The black line trace in the bottom panel is a plot of the parameter that character- izes a buildup of the ring current, the average horizontal ~ 2.2 It 6 a 0 5 0 ~ or _' cot o 0 a) Loo 141 component of Earth's magnetic field at the equator. One sees a strong correlation between flux enhancements and this measure of geomagnetic storms, strongest when it is most negative, since the ring current opposes and reduces the magnetic field due to Earth's dipole. Radia- tion belt electron fluxes build up at lower L values dur- ing stronger geomagnetic storms, posing a greater threat to constellations of spacecraft such as the GPS and to the International Space Station during those times. Note that the worst is yet to come in the current solar cycle (23), in terms of both relativistic electron and inner zone SAMPEX/Prolon (] 9-27.4 MeV1/E~ecTron (2-6 MeV), Sunsoof No. and DsT 150 ct) 100 Q~ ct) ~ 50 ~ ~ crew 1.4 in ~ ~0.6 1 ~ ~ 0 7 __ ~ ~ _ it . 40 ~20 oo 2 -20 -40 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1' 2000 2001 ~ ~—MOB ~ ~ ~ _ Rim , _ ~ ~ ~ =:~ l 1 993 1 994 1 995 1 996 1 997 Years 1 998 1 999 4 1 FIGURE 3.4 Thirty-day averaged MeV proton (top) and electron (bottom) fluxes from July 1992 to January 2001, as measured by the SAMPEX satellite in low-altitude polar orbit. Solar cycle variation is evident,with electron fluxes greater during the declining phase of the solar cycle, penetrating to lowest L-value (equatorial radial distance) during large storms characterized by negative Dst geomagnetic activity index. Inner zone proton fluxes maximize at solar minimum (1 996),when cosmic rays have greater access to the inner heliosphere. SOURCE: X. Li, D.N. Baker, S.G. Kanekal, M. Looper, and M.Temerin, 2001, Long term measurements of radiation belts by SAMPEX and their variations, Geophysical Research Letters 28(20): 3827.

142 proton exposure. The latter anticorrelates with the solar cycle, since fluctuations in the solar wind are weaker at solar minimum, allowing cosmic rays greater access to the inner heliosphere and Earth's upper atmosphere. These protons pose the greatest threat to the electronics of spacecraft passing through the South Atlantic Anomaly (SAA) of Earth's magnetic field and are the reason that the Chandra satellite, designed for X-ray and gamma-ray astronomy measurements, has instruments that are not operated during SAA encounters. The planned operation of the satellite was modified on orbit once these effects were encountered. A better under- standing of Earth's energetic particle environment will improve our ability to design future spacecraft missions for both research and appl ications. Terrestrial Gamma-Ray Flashes The observation of terrestrial gamma-ray flashes above thunderstorms constitutes one of the most unex- pected discoveries of the Compton Gamma-Ray Obser- vatory (CGRO) mission. The gamma-ray photon energy extends above 1 MeV, indicating bremsstrahlung radia- tion from >1 MeV electrons, consistent with C.T.R. Wil- son's predictions of upward beams of electrons, now called "runaways," accelerated by thundercloud fields. This discovery provides possibly the first observational evidence for relativistic runaway breakdown, a funda- mental new plasma acceleration process never before observed in nature or produced in the laboratory, which can proceed at much lower electric field levels than conventional runaway breakdown. Runaway accelera- tion may well be an important particle acceleration/ radiation process for astrophysical plasmas. Computer models show that intense transient thundercloud elec- tric fields impose a total potential drop between 20 and 80 km altitude of ~100 MeV, producing highly nonlin- ear runaway avalanche formation of relativistic electron beams over large spatial regions, with gamma-ray flashes emitted between 30 and 80 km altitude. The intense upward-driven relativistic electron beams eventually escape the col I isional atmosphere, enter the radiation belts, and may become trapped as a result of scattering due to plasma wave interactions. EARTH'S ELECTRIC FIELD Earth's magnetic field has been studied for centuries and determines the geometry in which space plasma physics takes place. Its dynamics, however, are con- trolled by Earth's electric field. Techniques for measur- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS ing the electric field are only several decades old and we are just beginning to explore its complexity. Earth is embedded in the Sun's atmosphere, a gigantic magneto- hydrodynamic generator that powers the entire mag- netosphere from high above the planet. These interplan- etary electric fields map down the magnetic field lines, particularly when the interplanetary magnetic field has a southward component, and even reach the strato- sphere. From below, the atmosphere acts as a hydrody- namic generator, creati ng electric fields rangi ng from planetary scales and tidal periods down to turbulent eddies and polarized structures smeller then a kilometer. The associated electric potentials generated map up- ward for vast distances along the highly conducting magnetic field lines. Many of these electric field sources are poorly understood, yet they are a major source of such phenomena as particle energization and the turbu- lence in the ionosphere that can create communication and navigation problems on transionospheric propaga- tion paths. A goal for this coming decade is to increase our understanding of the global electric field and its dynamic and irregular behavior to match our under- standing of the magnetic field. Magnetospheric Electric Fields The complexity of the magnetospheric electric field has become apparent over the last decade. Throughout much of the magnetosphere, the basic picture we have of the electric field is one in which the field is consid- ered to be electrostatic. Since the rarefied plasma in the magnetosphere can be considered to be collision free, the electric field parallel to the ambient geomagnetic field can be shorted out by free charge carriers to first approximation. This implies that the field lines are equi- potentials and that the electric field will therefore map along field lines, scaling inversely with the area of a flux tube or cross section subtended by field lines. The sig- nificance of this mapping lies in the fact that the electric field is connected to the flow velocity of the plasma. The high conductivity in the collisionless plasma implies that the electric field vanishes in the reference frame that moves with the plasma. Thus, in any other reference frame (such as that of Earth), the electric field can be given by the cross product of the background magnetic field and the plasma velocity. In this way, the electric field is directly related to plasma convection in the mag- netosphere. This situation is referred to as the frozen-in condition, since magnetic field lines can be treated as if they move with the plasma convection in this picture, a simplification that breaks down, for example, along au-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS roral field lines. Since the solar wind flows past the magnetosphere, this gives an electric field directed in the dawn-to-dusk direction, which leads to a potential of the order of 100 kV across the magnetopause. Some fraction of this potential drop penetrates into the mag- netosphere and maps along magnetic field lines, caus- ing the convection of the magnetosphere and the poten- tial drop across the polar cap. The magnitude of the potential drop is a good measure of the strength of the convection electric field and of the transfer of energy into the magnetosphere. Across the polar cap, this potential maps to the re- gion of antisunward flow, or equivalently, duskward- directed electric field. At lower latitudes, the electric field (and flow) is reversed, resulting in sunward flow. The region of flow shear is one in which there are field- aligned currents, active aurora, and enhanced iono- spheric conductivity. All of these attest to the dynamic importance of electric fields in the magnetosphere. In- deed, the laminar character of the flow over the polar cap is rarely as smooth as this picture would suggest. Dependencies on interplanetary magnetic field orienta- tion, solar wind dynamics, and magnetospheric activity usually produce a situation that is much more complex. Although the study of electric fields in the polar region has seen many successes in understanding the basic configurations of the polar cap potential structure, the understanding in terms of fundamental theoretical mod- els remains sketchy at best. The key to further under- standing is the increasingly global coverage of polar cap electric field measu remeets from radars and satel I ites. As these measurements give a more complete under- standing of the polar cap configuration, the future holds promise for developing a deeper knowledge, including predictive capability based on external inputs. It is now becoming clear that the large-scale earth- ward convection in the equatorial plasma sheet cannot be characterized simply by mapping the high-latitude ionospheric convection potential along field lines to the equatorial plane (as described above). Rapid magnetic field changes and large inductive electric fields in the midtail plasma sheet decouple motions in this region from those at the ionospheric ends of the flux tube. The long-held view of laminar earthward convection in the plasma sheet is being replaced by a view of disordered small-scale slow flows (which contribute little to earth- ward convection) and transient high-speed flow bursts (called bursty bulk flows, or BBFs) that transport an esti- mated 80 percent of the earthward-directed magnetic flux, energy, and mass. BBFs, which are azimuthally narrow but extended i n length downtai 1, appear to be 143 the normal mode of convective transport in the midtail under all conditions. These flow bursts have been seen as close to Earth as 10 Earth radii (RE). Our detailed understanding of the magnetotail potential patterns and how they evolve in time is still quite primitive. This is largely because electric fields at the equator are much weaker than those at the polar cap (the same cross-field potential is spread over a much larger area) and they encompass a much larger volume of space; a low-alti- tude, polar-orbiting satellite can traverse the polar cap in a matter of minutes and measure electric fields of several to hundreds of millivolts per meter. In the equa- torial plane, even a satellite relatively near Earth for example, one in geosynchronous orbit takes several hours to cover a significant fraction of the magneto- sphere. In the near-Earth plasma sheet, the large-scale con- vection pattern is again well represented by a cross-tail electric field proportional to the solar wind motional electric field, which increases in magn itude with in- creasing magnetic activity. The penetration of the cross- tail electric field into the inner magnetosphere is con- trolled by plasma processes. Currents flowing near the inner edge of the plasma sheet cause the dusk side to charge positive and the dawn side to charge negative. The resulting dusk-to-dawn electric field opposes the convection electric field and acts to shield the inner magnetosphere. "Undershielding" of the convection field can result if the cross-tail potential increases abruptly. This lasts unti I the plasma sheet particles drift sufficiently far earthward to adjust the shielding to an equilibrium level. The converse of this produces "over- shielding." We are left with a picture of a simple cross- tail convection electric field smoothly decreasing in magnitude with decreasing radial distance due to shield- ing effects. This simple convection electric field pattern is modified by Earth's rotation. Dense ionospheric plasma near Earth is dragged along as the result of elec- tron-neutral collisions and co-rotates with Earth. This creates an electric field that extends to a few Earth radii and drives corotation of the entire inner magnetosphere. This combination of sunward and co-rotating convec- tion controls the drift of magnetospheric particles around Earth. Recent results from the Combined Release and Radiation Effects satellite indicate that the large-scale electric field during disturbed times looks very different from the simple models described here. The observed convection during disturbed periods is enhanced in keeping with these simple models but is spatially struc- tured and often reaches maximum values deep in the inner magnetosphere at the location of the ring current.

THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS The low-altitude counterpart of these enhancements pro- duces polarization jets. These jets result when the field- aligned portion of the partial ring-current loop closes in the subauroral ionosphere through a steep conductivity gradient. Intense electric fields are produced that map up field lines to the inner magnetosphere. These strong and structured electric fields alter the drift paths of the ring-current ions that originally created them and pro- duce structuring of the plasmasphere. New instrumentation utilizing improved double probes and other techniques such as the electron-beam- drift technique are now producing improved measure- ments of low-amplitude equatorial fields. It is clear, how- ever, that multiple spacecraft observations are essential to developing a better understanding of the dynamics of the equatorial electric field. Some of these observations will come from the Cluster mission as the satellite sepa- rations are i ncreased. Magnetospheric Mu Itiscale wi l l provide multiple point measurements to fill in the pic- ture. In coordination with these multisatellite observa- tions, auroral imaging is unparal leled in its abi I ity to (1 provide an unrestricted field of view, spatial resolution, sensitivity, and sunlit imaging capability needed to map transient and localized auroral features such as poleward boundary intensifications (the auroral signature of BBFs) and substorm initiation; (2) separate spatial from tempo- ral variability; and (3) investigate dynamical behavior in the global context. These new measurements will lead to fundamental discoveries about the equatorial electric field although we have a crude model of the basic electric field configuration, the structure of the dynam- ics and evolution will only be understood as we begin to get sufficiently global coverage. Mesospheric and ionospheric Effects of Lightning-Driven Electric Fields At any given time, as many as 2,000 thunderstorms are active over the surface of the Earth, providing a global I ightn i ng rate of ~1 00 per second. Lightn i ng d is- charges radiate intense electromagnetic pulses of >20 GW peak power and produce transient quasi-static elec- tric fields of up to ~1 kV per meter at mesospheric alti- tudes, with total potential drop between 20 and 80 km altitude of ~100 MeV. These fields can heat, accelerate, and precipitate electrons in the lower ionosphere, me- sosphere, and the radiation belts. Dramatic recent experimental evidence of strong electrodynamic coupling between tropospheric light- ning discharges (<15 km altitude) and the mesosphere/ lower ionosphere (30 to 100 km) include spectacular luminous optical emissions known as red sprites and elves (Figure 3.5), as well as rapid ionization and con- ductivity changes. Visual accounts of glows in clear air above thunderstorms have appeared in the literature since the 1 9th century, the most vivid accounts being those by air transport pilots. The possibility of "upward" lightning or "lightning to the ionosphere" was seriously considered long before what are now known as sprites were first documented during the past decade. The dis- covery of these elusive lights high in the sky has cap- tured the imagination of the scientific community and the public, with over a thousand articles having ap- peared in newspapers and popu lar magazines world- wide. These new findings raise fundamental questions about the nature of the electrodynamic coupling be- tween thunderclouds and the upper atmosphere. The new observations have been interpreted using several new interaction and coupling mechanisms, in- cluding the heating of the ambient electrons by light- ning electromagnetic pulses or by large quasi-electro- static thundercloud fields and by runaway electron processes. Runaway acceleration and ambient heating processes may influence reaction rates and lead to the prod uction of d ifferent types of ions, potent) al Iy affect- ing mesospheric chemistry and dynamics. The mecha- nisms underlying these spectacular phenomena are just now being uncovered, and their effects on a global scale will need to be assessed during the coming decade. At the very least these I u mi nous phenomena provide a wi n- dow of measurability for the electrodynamics of an A-l-M region not accessible for in situ measurements. The current global electrical circuit models, which consider the fair weather current to be driven by quasi- static thunderstorm electric fields, may need to be sub- stantially revised to account for the sporadic and highly nonlinear component of charge transfer that can be seen in electrical breakdown channels extending from the lower atmosphere to the ionosphere. ~ 1 Electrodynamic Coupling One of the most important roles of Earth's electric field is its ability to couple the different regions of the magnetosphere electrodynamically. It is common to speak of the electric field "mapping" along auroral field lines. Although this mapping of the electric field along field lines is a convenient way of discussing magneto- spheric convection, it is somewhat misleading in a time- dependent situation, since the change in plasma veloc- ity must be associated with forces that accelerate or decelerate the plasma. It is equivalent to say that the

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS k'687 145 3178 91 45 FIGURE 3.5 Sprite event imaged July 1 3, 1 998, from Langmuir Laboratory in Socorro, New Mexico, during a storm over northern Mexico, 491 km away. Left-hand image was taken with a 50-mm lens-intensified CCD camera, wide field-of-view (FOV) system, while narrow FOV camera (right-hand image) zooms into the sprite with a 1 6-in. Newtonian refracting telescope.The FOV of the telescope in relation to the wide FOV is outlined with a white box on the left-hand panel.This sprite was a huge event associated with a large cloud-to-ground lightning flash of peak current 128.5 kA. It extends from 40 to 90 km altitude and is ~50 km wide. In the wide FOV the sprite appears almost homogeneous, and little structure is evident.The telescopic image reveals that the sprite consists of numerous densely packed glowing filaments. Courtesy of Elizabeth Gerken, STAR Laboratory, Stanford University. motion of plasma causes a shear stress on magnetic field lines tied to the collisionless plasma by almost perfect conductivity, and that the resulting force is responsible for the acceleration of the plasma. In a time-dependent situation, this stress propagates along magnetic field lines as an Alfven wave. Because this wave carries changes in the field-aligned current, this electrodynamic coupling of the magnetosphere induces a current system that per- meates the magnetosphere. The study of these waves, in the ultralow-frequency (ULF) range between 1 mHz and about 5 Hz, is an important aspect of magnetospheric physics in its own right since these waves transmit mo- mentum and energy throughout the magnetosphere. When field-al igned currents encounter the inher- ently conductive ionosphere, they flow across geomag- netic field lines. Pedersen currents, which flow in the direction of the ionospheric electric field, provide an energy sink for the magnetospheric plasma. Since the neutral atmosphere has a much higher mass density than the plasma in the ionosphere, the transfer of momentum to the atmosphere provides a drag force on plasma con- vection in the magnetosphere, much like trying to pull a heavy rock along the ground with a rope. If the convec- tion is strong and long lasting, such as during geomag- netic storms, neutral winds can be accelerated in the direction of convecting ions owing to collisions between the ions and neutrals. Mapping of electric fields along field lines is vio- lated when parallel electric fields develop in the auroral acceleration region. In addition to accelerating auroral particles, these parallel electric fields decouple the mag- netospheric convection from the atmospheric drag, al- lowing convection to proceed with reduced dissipation. New data from satellites such as FAST and Polar have provided new details of these processes, providing strong constraints on theories of this region. These new observations, combined with new theoretical models and numerical simulations of increasing levels of so- phistication, are leading to a period of intense research activity in the physics of auroral acceleration processes. Ionospheric conductivity can be modified by pre- cipitating electrons that produce increased ionization in the ionosphere. This enhances the conductivities of the ionosphere and modifies the current systems that set up the parallel electric fields in the first place. Such interac- tions can give rise to a positive feedback that leads to narrow, very intense current structures. Although au- roral arcs have been measured to extend horizontally less than 100 m, most theoretical scenarios are unable to produce such narrow structures. This ionospheric feedback instability is a promising mechanism for creat-

146 ing such narrow arcs. Indeed, recent observations sug- gest that the formation of narrow arcs is more likely in cases where the background ionospheric conductivity is low, i.e., under non-sunlit conditions. This is consistent with a feedback instability which is stabilized by high conductivity. Because of the electrodynamic coupling along magnetic field lines in the magnetosphere, the auroral ionosphere acts as a screen on which magnetospheric dynamics can be projected. Many observational tech- niques such as radar measurements, measurements from ground magnetometer arrays, and measurements from sounding rockets and low-Earth orbiting satellites- take advantage of the fact that ionospheric convection can be seen as an image of magnetospheric dynamics. An understanding of the coupling between the iono- sphere and magnetosphere as well as the breakdown of such coupling is crucial for the interpretation of such measurements. Moreover, these processes are essential for understanding auroral formation and the flow of plasma through the magnetosphere, and the study of this coupling should be considered a high priority in the next decade. Earth's Atmosphere as a Generator The upper atmosphere is in constant motion. Just as the ocean surges in response to gravitational forces, so- lar heating forces the air to flow across the top of the atmosphere from day to night. This is particularly impor- tant in the thermosphere, where the x rays and EUV from the Sun are absorbed on the dayside, raising the temperature to well over 1000 K. In this height range these in situ tides reach velocities upwards of 200 m/s (720 km/hr), the highest wind speeds in the atmosphere. In fact these winds would flow even faster if the iono- spheric plasma did not act as an electromagnetic break. The wind tries to force an ionized particle across the magnetic field, but the particle is tied by gyration about the magnetic field and so hardly moves and must be struck again and again by the neutral gas to be trans- ported. This ion drag effect can also be described by the electromagnetic force density J x B since the wind in- duces a current when it tries to carry the plasma across the field lines, just as occurs in an electrical generator. If the thermosphere was unbounded, no electrical poten- tial would exist in Earth's frame of reference; instead, electric fields are generated virtually everywhere the wind blows. Electric fields and currents induced by tides were first discovered when it was found that the magnetic THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS field on the surface of Earth varied regularly every day. We now know that propagating tides also surge up from the lower atmosphere, growing dramatically as they en- ter the tenuous upper regions of the atmosphere. Ab- sorption of solar photons by stratospheric ozone and tropospheric water vapor is the source of the dominant, solar-driven diurnal and semidiurnal tides. During the past decade, measu remeets by the U pper Atmospheric Research Satellite (WARS) revealed the seasonal and interannual variability of these global waves along with the existence of other tides that do not propagate with the apparent motion of the Sun. Such so-car led nonmigrating tides may generate significant longitudi- nal variabi I ity and wi 11 modu late the low-latitude winds that generate the electric field. We need to characterize the sources of these tides fully during the next decade. Tides are only the tip of the iceberg as far as atmo- spheric winds are concerned, since planetary waves also occur at lower frequencies, whi le internal gravity waves are very common at higher frequencies. Gravity waves also grow with altitude and become huge in the region collocated with the ionospheric plasma. Currently of great interest is the likelihood that gravity waves induce electric fields in much the same was as tides. If they do, as recent rocket data suggest, they would have a consid- erable effect on ionospheric physics, space weather, and the magnetosphere. The magnetic field lines act as such good conductors that the electric field maps throughout the magnetosphere. For low levels of magnetic activity, Earth's wind field, and the associated electric field it produces, dominates the electrical structure of the inner magnetosphere. VOLATILE WEATHER IN THE UPPER ATMOSPHERE Compared with some other regions of the space environment, many of the processes controlling the ba- sic thermal and dynamic structure of the ionosphere- thermosphere system are reasonably wel I understood. However, fundamental science questions remain, in- cluding the role of ionosphere-thermosphere processes in magnetosphere-ionosphere coupling, the cause of equatorial and midlatitude ionospheric irregularities, and upward coupling of energy and the thermal struc- ture. Equatorial Convective Storms At Earth's magnetic equator, plasma in the F layer of the ionosphere, a region 200-500 km above Earth's sur- face, is supported against gravity by the electromagnetic

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS force.The situation is analogousto one in which a light fluid supports a more massive fluid above it, an unstable equilibrium. The result on many nights is a massive over- turning of the plasma, which can fill more than a billion cubic kilometers with highly turbulent ionized gas or plasma. When this overturning occurs, the low-density regions surge upward at high velocity due to the large electric fields that are generated in the plasma. The pri- mary process operates at large scales, but the large ve- locities involved lead to structure over a range of scale sizes. In situ rocket spectra, coupled with radar observa- tions, show that structure occurs from hundreds of ki lo- meters down to 10 cm, over seven orders of magnitude. The plasma instability analogue of a heavy fluid on top of light fluid has been shown to account for the large- scale upwelling of plasma as the ionosphere attempts to smooth out the sharp nighttime density gradient. Some of the very first computer simulations of space plasmas were built to provide nonlinear analysis of this phenom- enon. Satellites were a valuable tool as well, giving the sort of worldwide coverage only they can provide, and we now have a deep understanding of the causes and seasonal/geograph ic control of th is phenomenon. I n the coming decade, we will begin the next step, moving into the predictive era for this important space weather problem. Crucial to success is fully understanding the equatorial electric field since it will be the single most important parameter in any prediction scheme. The Communications/Navigation Outage Forecast System (C/NOFS) satellite will be dedicated to such measure- ments along with other diagnostic parameters. Its goal is to use the electric field data to predict scinti I rations or flickering of satellite signals, which can disrupt commu- nications. In the future we envision numerous micro- or even nanosatellites dedicated to this endeavor. Surprises in the Midlatitude ionosphere The midlatitude ionosphere has general Iy been thought to be relatively quiescent. Compared with the stormy equatorial zone and the high-latitude zone, this is certainly true. But this is not to say that no structure exists or that space weather effects are unimportant. A major breakthrough in midlatitude studies occurred when charge-coupled device-based all-sky cameras be- came capable of imaging vast portions of the nighttime sky. We found that the midlatitude ionosphere indeed had weather and that we simply did not understand its origin. An example is presented here showing the com- plexity and beauty of the ionization clouds that form 147 and move rapidly across the sky (Figure 3.61. Here the photos are taken in the red-line emission of atomic oxy- gen. This excitation occurs during a two-step process that removes oxygen ions from the ionosphere. The re- action rate depends on both the altitude of the layer and its plasma content and, when combined with other emis- sions or GPS data, can be used to describe the height and content of the layer. Such data may prove invalu- able as we enter an era when prediction of space weather becomes feasible. The inset shows several mea- surements of the total electron content between the ground and GPS satellites. Huge dropouts are seen when the line of sight passes through a dark portion of the i mage. For now we are still in the discovery and character- ization phase for most ionospheric processes, but the pace from discovery to understanding should quicken now that two- and three-dimensional imaging and to- mography can be applied. From a space weather stand- point, one of the worst difficulties involves the effect of the total plasma content between the satellite and the ground or airborne GPS user. This plasma content cre- ates time delays, which translate into navigational er- rors. Ionospheric models are not even close to predict- ing features such as those in the Figure 3.6 image, in which huge gradients in plasma content exist. But we are getting close to the era when mesoscale fair-weather conditions can be modeled. A sizable effort has begun to calibrate simple sensors capable of providing data input analogous to the millions of data points assimi- lated daily by meteorological models. The global auto- mated instrument cluster outlined in Recommendation 1 extends th is concept to i ncl ude al 1-sky i magers, Fabry- Perot i Interferometers, VLF receivers, passive radars, magnetometers, and ionosondes, in addition to power- ful GPS-based systems in a flexible and expandable net- work coupled to fast real-time processing, display, and data distribution capabilities. A second thrust is to vali- date the models using more sophisticated observations, such as those made possible by incoherent scatter ra- dars. Challenges of High-Latitude ionospheric Science Despite 70 years of ionospheric study, it is still im- possible to specify ionospheric composition and density accurately at high latitudes. Unpredicted variability oc- curs on spatial scales of meters to hundreds of kilome- ters and time scales of seconds to hours. This variability appears to be due to poor knowledge of sources, insta- bilities, and transport. Solar ionization by the extreme

148 THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS FIGURE 3.6 Observations of 630.0-nm nightglow emission made with the Cornell All-Sky Imager (CASI) at Arecibo Observatory on February 17, 1998. Each image has been projected onto a map of Earth assuming an emission height of 275 km. Superposed on the top image is the total electron content (TEC) mapped to vertical as measured by satellite-receiver pairs in the GPS constellation. Two satellite-receiver pairs are shown (satellite 10 viewed from St. Croix,Virgin Islands, blue line, and satellite 24 viewed from Isabela, Puerto Rico, green line).The ground tracks shown are for the 350-km pierce point; the satellites move from west to east.The asterisks indicate the location of the satellite at the time of the image.The increase in the 630.0-nm emission is seen to be collocated with an increase in TEC (as shown by satellite 24), while a decrease in intensity is seen to be collocated with a decrease in TEC (as shown by satellite 10). SOURCE: Unpublished figure courtesy of Jon Makela, Cornell University. and far ultraviolet is well known except for variations during solar flares. However, particle sources of ioniza- tion are coupled in from the magnetosphere and known only from individual measurements or statistical studies. Instabilities develop where there are sharp spatial struc- tures in the ionosphere that lead to the breakup of large- scale features and the creation of smaller-scale irregu- larities. Convection transport of large- and small-scale features superimpose unexpected variability on the specification predicted by ionospheric models. Ionospheric convection due to electric fields per- pendicular to the background magnetic field is well un- derstood in principle, but its application in ionospheric models fails to match reality because the sources of

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS variabi I ity of E cannot be prescribed adequately. The electric field is partly imposed externally, by coupling to the magnetosphere and solar wind, and partly internally, through development of polarization charges due to plasma density gradients perpendicular to B. This prob- lem is expected to be solved as coupled ionosphere- magnetosphere models become mature. Large-scale (1,000-km) ionospheric patches are fea- tures of the dark high-latitude ionosphere that persist and travel across the polar cap in a few hours. The source may be the breakup of the tongue of high ioniza- tion entering the polar cap from the sunlit dayside. Iono- spheric models demonstrate that externally applied elec- tric fields that are the result of structure in the solar wind cause the convection pattern to have periods of rapid change, when flow is broken up, causing patches of disconnected plasma. Although such a mechanism is plausible, the distribution of patches cannot be reliably predicted without specification of the solar wind source and its coupling through the magnetosphere. Once the patch has formed, it degrades by instabilities that create meter-scale irregularities in its wake. These irregularities are the source of ionospheric scintillations (scattering of radio signals) that impede communications and naviga- tion. The ionosphere is embedded in the neutral atmo- sphere, which acts as a source of ionospheric mass though ionization, thermodynamic modulation, and dy- namic friction through collisional coupling and dynamic drag. This strong coupling of the ionospheric and neu- tral media is well represented in current models such as those at NCAR and NOAA. In developing an adequate model of the high-latitude ionosphere, the main thrust should be to extend these codes to couple with the magnetosphere and solar wind. We are entering the era when the remaining scien- tific problems in ionospheric physics will be picked off one by one and solved. An important tool for this effort is the Advanced Modular Incoherent Scatter Radar and I idar faci I ity. AMISR wi 11 be a state-of-the-art faci I ity with a phased-array incoherent scatter radar (ISR) as the cen- terpiece. This highly versatile instrument will be ringed by less expensive complementary systems, typically op- tical in nature. The science plan is to target unsolved problems in aeronomy by placing the AMISR in appro- priate geographic locations for 3 to 5 years. The first science goal is to understand the coupling of the neutral atmosphere to the high-speed current-carrying plasma in the auroral region. This coupling involves momentum transfer from the plasma to the neutrals, heating due to currents, composition changes of the thermosphere, and 149 particle impact ionization associated with the aurora, to name a just a few aspects. The first AMISR site wi 11 be in the Fairbanks, Alaska, area to take advantage of already existing instrumentation and the Poker Flat Research Range. Subsequent sites will be decided upon with com- munity input from a scientific advisory panel. Candidate locations include deep in the polar cap, which has never been studied using the ISR technique, and the off-equa- torial zone to study effects of equatorial ionospheric upwelling and downflow along magnetic field lines to adjacent latitudes, which can severely affect communi- cations. One Atmosphere: Upward Coupling of Energy One of the major unresolved issues in a quantitative description of the A-l-M system is the extent of upward coupling of energy and momentum from the lower at- mosphere. The thermal structure of the mesosphere and lower thermosphere provides one of the most vivid dis- plays of the direct dynamical coupling between the up- per and lower atmosphere (Figure 3.7~. The inverse tem- perature gradient with low temperatures in the summer polar mesosphere is due to a gravity-wave drag force that supports strong meridional circulation from the sum- mer pole to the winter pole, accompanied by upwelling and downwelling that in turn result in cooling and warming, respectively. The drag is thought to arise from the breaking up and dissipation of small-scale gravity waves carrying momentum and energy from sources in the lower atmosphere. It is realized by modelers that waves have important effects in Earth's upper atmo- sphere, and that in the absence of measurements they introduce representative parameters to replace the wave sou rces. I mportant effects of these waves must be pa- rameterized in current models of Earth's upper atmo- sphere. The existing schemes incorporate a number of tunable parameters that allow prescribing a realistic set of tropospheric sources. Little is known about longitude variabi I ity or about the processes i evolved with smal 1- scale mixing at the base of the thermosphere (turbo- pause) that lead to global variations. There is an urgent need to better characterize the sources and evolution of gravity waves on the global scale. Owing to the paucity of ground-based observing sites, which inherently pre- cludes measurements over the oceans, this can be ac- complished only by satel lite-borne experiments. Solar atmospheric tides that originate in the lower atmosphere are particularly strong and ubiquitous fea- tures of the mesosphere and lower thermosphere. Upper Atmospheric Research Satellite observations reveal that

1 50 FIGURE 3.7 Illustration ofthe multitude of dynamical and physi- cal processes throughout Earth's atmosphere and ionosphere, including gravity waves, tides, planetary waves, and the quasi- biennial and semiannual oscillations.Courtesy of Rashid Akmaev, CIRES, University of Colorado, Boulder. diurnal tidal amplitudes of the meridional wind exceed 80 to 100 m/s at low latitudes at equinox, with a strong semiannual variation. Very strong variations from the day-to-day to interannual time scales of both the diurnal and semidiurnal tides are clearly seen in the UARS data and are supported by observations from ground-based optical and radar facilities. These perspectives are valu- able but limited. Any satellite provides a global picture of a tide at only two local times on a given day. Continu- ous ground-based observations are needed, but the ex- isting facilities are too few to provide a complete geo- graphic perspective on these inherently global-scale waves. A complete characterization of atmospheric tides and their important effects in the upper atmosphere can only be accomplished by correlative analysis and as- similation of both ground-based and satellite-borne mea- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS surements. Further, tidal diagnostics from a satellite con- stellation are needed to fully resolve the spatial and temporal variability of atmospheric tides. In the equatorial, low-latitude, and mid-latitude ionosphere, a number of investigations have interpreted oscillations in plasma density and electrodynamics as a manifestation of planetary waves in the neutral atmo- sphere with 2-, 5-, 10-, and 16-day periods. Several interpretations have been suggested, including modula- tion of the turbopause, excitation of normal modes in the upper atmosphere, and the modulation of other up- ward-propagating waves including tides and gravity waves. Recent studies also show that between 1 5 and 25 percent of observed ionospheric electron density variations cannot be ascribed to space weather or geo- magnetic activity but are probably attributable to waves of tropospheric origin. The wave drivers can only be resolved with additional satellite measurements that are analyzed in concert with observations from ground- based networks. Comparatively little is known about the interac- tions between these various waves and the impact they have on the upper atmosphere of Earth. There is an urgent need to develop comprehensive observation and modeling programs to untangle the complex web of interacting processes that occur over scales from kilo- meters to thousands of kilometers. It is very likely that many of the outstanding questions about the thermal and dynamic structure of the upper atmosphere will turn out to be linked to the labyrinth of global wave coupling processes. Plasma-Atmosphere Interactions Collisional interactions between ionized and neu- tral particles are a dominant process in the upper atmo- sphere. At h igh latitudes, the i mposed magnetospheric electric fields set the plasma into motion through colli- sions with the dense neutral medium and can drive strong neutral winds. The atmospheric resistance to ion motion gives rise to dissipation in the form of frictional heating, raising neutral and ion temperatures, driving wind surges and upward ion outflow, and changing the global neutral composition and circulation. In regions of large electric fields, anomalous electron collisions due to plasma instabilities can change conventional con- ductivities, and variability or fluctuation in the electric field can induce uncertainties in heating rates by a fac- tor of 2. At midlatitudes the collisional interaction between plasma and the atmosphere is reversed. In the presence

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS of inclined magnetic fields, neutral winds, driven by global pressure gradients, force plasma up and down in the direction of the magnetic field. During space weather events the wind surges, propagating from high latitudes, and circulation changes can drive winds that are many times their normal quiet-time values, which can cause a substantial redistribution of plasma. Winds can push ions to regions of altered neutral composition, either increas- ing or decreasing plasma densities. In addition, the cir- culation changes drive upwelling at high latitudes, caus- ing neutral composition changes that can be transported by the background and storm-induced winds and fur- ther modify the plasma structure at all latitudes. Differing mobilities of electrons and ions in the pres- ence of magnetic and electric fields also give rise to field-al igned currents and vertical electric fields. Tidal waves propagating from the lower atmosphere grow to large amplitudes in the lower thermosphere before they dissipate and, through collisional interaction with the plasma, drive a global current system. During geomag- netic disturbances both E- (~100 km) and F-region winds can be disrupted and decrease or reverse normal quiet- day electrodynamics. At low latitudes, with the near- horizontal magnetic field, the neutral wind-driven elec- tric fields dominate the plasma structure, giving rise to the equatorial ionization crests on both sides of the mag- netic equator and anomalous wind and tempertaure fea- tures. The redistribution of plasma sets up conditions conducive to the generation of equatorial convective storms. We are only just beginning to understand the complexity of plasma atmosphere coupling during so- lar-wind-driven storms, when the neutral wind fields and electrodynamics are disturbed. The Geospace Electrodynamic Connections mis- sion, currently scheduled for launch in 2009, will make systematic multipoint measurements and when com- bined with ground-based observations will delineate and bring to closure our understanding of the key roles the ionosphere and thermosphere play in the Sun-Earth con- nection. This cluster of dipping spacecraft will resolve spatial and temporal variations and energy transfer in the transition region between the magnetosphere and the ionosphere and thermosphere. GEC wi 11 identify the relative importance of key processes, including ener- getic inputs, current systems, ionosphere-thermosphere coupling, and thermospheric waves on varied spatial and temporal scales. Although many of the physical processes i n pi asma- atmosphere interactions are understood, the aspect that is still poorly characterized is the global-scale dynamic response to solar events. There is a pressing need to 1 51 understand the interactions between global-scale neu- tral dynamics, composition changes, plasma density and conductivity, and electrodynamics, particu larly during geomagnetic storms, when variability is at its greatest. Previous single-satellite missions and ground-based ob- servations have provided a valuable but limited per- spective. Remote sensing, such as with the GUVI instru- ment on the recently launched TIMED satellite, has the potential to follow global-scale neutral composition, but simultaneous in situ observations are required to inter- pret and relate the images to the actual conditions pres- ent in the atmosphere. Pairs of satellites can begin to address the important scale sizes in the plasma density and neutral dynamics, but ultimately, multispacecraft missions will be required to track the global-scale dy- namic response. The technology that will enable such nanosat constellations in low-Earth orbit is coming soon and must be a priority in future missions. Photochemistry and Radiative Transfer The ionosphere is created and maintained by ener- getic solar photon irradiance in the extreme ultraviolet and x-ray regions of the spectrum that ionizes a small fraction of the neutral atmosphere. Absorption of solar EUV radiation at other wavelengths also heats the high- est part of the neutral atmosphere. At longer ultraviolet wavelengths, solar radiation penetrates progressively deeper into the atmosphere, and its main effect is photo- dissociation of molecular gases. The deposition of this energy drives a complex cycle of photochemical re- sponse that interacts strongly with atmospheric trans- port. The thermosphere becomes mostly atomic above ~200 km, whi le the lower thermosphere from ~1 00 to 200 km acts as a crucible for photochemical energy deposition and dynamical interchange of atoms and molecules. Solar cycle variation of the solar spectrum increases with increasing energy, from roughly a few percent in the near- to midultraviolet to about double in the ex- treme ultraviolet and by an order of magnitude or more i n the x-ray region. I nstru meets on UARS provided ac- curate measurements of the ultraviolet spectrum adja- cent to visible wavelenths throughout the 1990s. The extreme ultraviolet and x-ray regions were measured during the 1 970s by the Atmospheric Explorer satellites. After a prolonged hiatus, the Solar Extreme Ultraviolet Experiment (SEE) instrument on TIMED began making new measurements in December 2001. Several prob- lems in thermospheric photochemistry remain outstand- ing, leading to a long-standing suspicion of problems

1 52 with the older short-wavelength data. New observations of solar soft x rays (from about 1 to 20 nm) by a combi- nation of rocket measurements and data from the solar x-ray photometer on the Student Nitrous Oxide Explorer (SNOE), a UNEX-class satellite, have found fluxes four or five times higher than the Atmospheric Explorer-era estimates at all levels of activity. SEE is also observing these higher fluxes. Applied to photochemical and gen- eral circulation models, the ionospheric and minor spe- cies problems now appear to be solved. This leads to a general revision of thermosphere/ionosphere variabi I ity, with larger solar cycle variation expected. Other unre- solved aeronomical issues, including discrepancies be- tween measurements and model predictions, must be explored in the context of solar irradiance measurements and variability. It must be remembered that although the solar EUV flux and the upper thermosphere response have been measured in the past, comprehensive, simul- taneous measurements of both the source and the atmo- spheric response are still lacking. Since solar EUV is the dominant and high Iy variable energy and ionization source in the thermosphere and ionosphere, it is im- perative that the relationship between EUV flux and the state of the ionosphere-thermosphere system be assessed quantitatively. The mesopause is particularly amenable to remote- sensing observation of airglow emission layers. Mea- surements of both airglow and auroral emissions by ground-based CEDAR instrumentation have contributed significantly to our understanding of this region of the atmosphere. The TIMED mission will bring even greater advances. TIMED is remotely measuring winds, tem- perature, composition, and cooling rates throughout the mesosphere and lower thermosphere as well as auroral energy inputs, and it also carries solar EUV and x-ray instrumentation to quantify the primary sources of en- ergy. Our understanding of chemical-dynamical cou- pling should vastly improve with input from TIMED mea- surements. But there are fundamental and unresolved issues that TIMED cannot decipher, including the char- acterization of gravity waves, wh ich are the u nderlyi ng drivers of the basic state of the mesopause region and the inherent spatial and temporal tidal variability of the upper atmosphere. Particle Effects in the Middle Atmosphere The particle flux from the Sun and the magneto- sphere represents a large source of energy and ioniza- tion for the lower thermosphere and ionosphere. The energy flux, which varies by two orders of magnitude, is THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS deposited by particles with energies that range from hun- dreds of eV to several hundred MeV. The intensity, spec- trum, and distribution of the precipitation are functions of solar and geomagnetic activity. The mean values also show longer-term dependence on the solar cycle. The particle-induced ionization leads to dissociation of tightly bound N2 molecules and to the formation of the reactive nitrogen compounds NO and NO2. These com- pounds, when transported to lower altitudes, participate in a catalytic cycle leading to destruction of ozone. Con- tinuing research seeks to understand the impact of mag- netospheric and solar energetic particles on the chemis- try and electrodynamics of the middle atmosphere. The production of NO and NO2 due to the precipi- tation of energetic particles is determined by the deposi- tion of energy from particle penetration. The particle energy is absorbed bythe atmosphere through collisions with, and scattering by, atmospheric constituents, which result in either dissociation or ionization of the atmo- spheric constituents. It takes an energy deposition of approximately 35 eV to produce each electron pair in the atmosphere. In addition to the energy imparted di- rectly to the atmosphere during collisions, a certain amount of energy is converted to x rays by the brems- strahlung process. This is associated with the rapid de- celeration of energetic electrons during their penetra- tion i nto the atmosphere, provid i ng add itional ion ization down to altitudes of 20 km. The ion and neutral odd-nitrogen chemistry initi- ated by energetic particles produces secondary elec- trons, which in turn ionize and dissociate the major atmospheric species. In the sunlit atmosphere, the pho- todissociation of NO is important; however, in fall, win- ter, and spring in the polar region, the photolytic reac- tion is negligible and the resulting NO lifetime is sufficient for NO to be transported to the mesosphere and stratosphere. There have been several studies of these effects suggesting that precipitating electrons could lead to the formation of oxides of nitrogen and hydro- gen. These could affect ozone as low as 60 km, and the ion formation rates due to these electrons could domi- nate ion formation rates of other processes down to ap- proximately 40 km. Studies of ion formation rates using electron measurements from geostationary orbit have shown that the fluxes of energetic electrons and hence the ion formation rates are strongly modulated by solar activity. Simulations suggest that precipitating electrons could significantly affect the budgets of both odd nitro- gen compounds and ozone within the stratosphere. This coupling mechanism between solar activity, solar wind structures, the energetic particle populations within the

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS magnetosphere, and the chemical state of the middle atmosphere represents an important linkage that must be understood to assess the natural variations in the global chemical state of the middle atmosphere. MICRO- AND MESOSCALE CONTROL OF GLOBAL PROCESSES The global behavior of the thermosphere, iono- sphere, and magnetosphere is critically dependent on small-scale processes in some key locations. Although it is not intuitively obvious that processes on small scales can have global consequences, there are some that are essential to the existence of large-scale structures as we observe them. It is now understood, for example, that the thermal structure of the mesosphere and its circula- tion patterns are much more dependent on energy and momentum carried by small-scale waves than on global thermal sources such as solar ultraviolet radiation. Simi- larly, kinetic effects due to turbulence and heating are often more effective than classical diffusion in determin- ing the diffusion rate. More dramatically, some struc- tures and boundaries would not exist at all without small-scale processes. The magnetopause is a thin shell at the nose of the magnetosphere that has spatial scales comparable to the ion gyroradius. Auroral arcs would be diffuse and unable to create the observed dramatic, localized effects without small-scale acceleration re- gions that form spontaneous Iy a few thousand ki I ome- ters above Earth and create the kilometer-scale struc- tures. The importance of small-scale processes is well understood, but their simulation in global-scale models is made difficult because they are sub-grid-scale and must be parameterized. Inclusion of properly formulated physical representations of these sub-grid-scale effects is an urgent research requirement. Transfer of Momentum and Energy by Internal Atmosphere Waves Earth's dense and dynamic troposphere is a source of very strong internal waves, which carry energy and momentum upward. To conserve energy in an exponen- tially decreasing atmosphere, these waves grow in am- pl itude and eventual Iy break, deposit) ng thei r energy and momentum in the mesosphere. A remarkable circu- lation pattern is set up, leading to temperatures in the summer polar regions, in full 24-hour sunlight condi- tions, as low as 100 K. This is the coldest fluid in the inner solar system and results in the highest clouds on Earth (see Figure 3.2~. These clouds seem to have be- 1 53 come more frequent over the last 100 years and may be a sensitive marker for global change. Waves also seem to interact with atmospheric tides all over the globe to create winds in the upper mesosphere and lower ther- mosphere in excess of 500 km/hr, which is nearly super- sonic. Models do not come close to explaining these high winds, which create conditions for velocity-shear- driven instabilities of the neutral atmosphere in regions long thought to be quite stable. Understanding the role of medium- and short-period waves in the energy and momentum budget of Earth's upper atmosphere is a cru- cial goal of the next decade. Mesoscale Auroral Features and Their Role in Auroral Oval and Magnetosphere Processes Thermospheric vertical mixing in the auroral oval is driven by both global and local processes. The global process establishes an equilibrium for each constituent. It is driven by the temperature structure of the auroral thermosphere. By this mechanism alone, molecular spe- cies can be elevated to F-region heights when the ther- mospheric temperatures are raised during an auroral substorm (see next section). Vertical wi nds cause local- ized upwellings and downwellings and are much more effective at redistributing the thermosphere during heat- ing events. Coupled regions of upwelling and down- welling tend to neutralize each other on global scales but serve to raise molecular species very rapidly at the center of auroral heating. Upward fluxes of molecular species are significantly enhanced due to the upwelling process, with a resulting increase in molecular concen- trations at F-region heights, affecting the time constants for change i n ionospheric concentration. Recent results from the FAST and Polar satellites have indicated that the narrow-scale plasma structures that produce auroral arcs are confined to altitudes be- low about 6,000 km, where the bulk of auroral accel- eration appears to occur. These results, along with theo- retical considerations, suggest that energy enters the auroral zone at intermediate and large scales and then filaments into narrow structures due to magnetosphere- ionosphere interaction. This filamentation is probably due to a combination of ionospheric conductivity ef- fects and the formation of density holes on a variety of scales in the auroral zone. Understanding these compli- cated interactions is a high priority for future research. Understanding the scale sizes of the parallel electric fields that accelerate auroral particles is an important and active area of research. Models based on the reflec- tion of electrons out of a converging magnetic field pro-

1 54 duce smoothly varying, weak electric fields. On the other hand, auroral zone observations in regions of both upward- and downward-going current indicate that lo- calized regions of weak parallel electric fields are dis- tributed along auroral field lines. Some recent observa- tional and theoretical results suggest that the auroral potential drop may occur as a result of large parallel electric fields, usually referred to as double layers or sheaths. Determining the relative contribution of these various structures would probably require a multispace- craft auroral mission. Convection in the outer magnetosphere is strongly affected by the ionospheric drag caused by collisions with the neutral thermosphere. When parallel electric fields are present, however, magnetospheric convection can become uncoupled from this ionospheric drag, lead- ing to the possibility of fast flow channels. Investigations of magnetosphere-ionosphere coupling and its relation to substorms are still in an early stage; however, it will be a promising area of research in the next 10 years as the dynamics of the magnetosphere as a coupled system becomes better understood. The multiscale dynamics of the auroral zone, in which phenomena from narrow auroral arcs to large-scale magnetospheric convection are clearly interrelated, is a prime example of how smaller-scale processes can have large-scale effects. Connection and Reconnection of Magnetic Field Lines in the Magnetosphere As noted in the section on electrodynamic coupling, the highly conductive magnetic field lines electrody- namically couple different magnetospheric regions. However, this frozen-in condition does not hold every- where, and its violation in localized regions has a strong impact on magnetospheric dynamics. For example, a strict application of the frozen-in condition implies that the magnetopause is an impenetrable boundary since infinitely conductive plasmas cannot penetrate or dif- fuse across field lines. However, a continuous transport of mass, momentum, and energy from the solar wind into the magnetosphere is observed. To understand this transport has become one of the most important chal- lenges i n magnetospheric physics. Magnetic reconnection of oppositely directed mag- netic field lines usually takes place in strong current sheets such as those that occur at the magnetopause and in the tail. Observations in the magnetosphere have shown that the orientation of the interplanetary mag- netic field is a major factor in controlling geomagnetic activity, ionospheric convection, the structure of the THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS magnetic field, and the dynamics of plasma flows at the magnetopause. Accordingly, magnetic reconnection has been widely accepted as the main mechanism for cou- pling the solar wind to the magnetosphere-ionosphere system. At Earth's magnetopause, reconnection results in the formation of open magnetic field lines, leading to exchange of mass along open flux tubes and an efficient transfer of solar wind momentum and energy by mag- netic stresses. Nightside reconnection is considered to be critical for releasing magnetic energy stored in the tail. The breakdown of the frozen-in condition is often described in terms of the electron behavior as described by a generalized Ohm's law. In the recent GEM re- connection challenge, magnetic reconnection was stud- ied in a simple current sheet configuration under a speci- fied set of initial conditions. These conditions were modeled by different codes, ranging from fully electro- magnetic particle codes to conventional resistive mag- netobydrodynamic codes. The rate of reconnection ob- tained from the different models was found to be about the inflow speed, which depends on external magnetic field strength and density, and to be insensitive to the mechanism for breaking the frozen-in condition. The observations indicate that the coupling of the solar wind and the magnetosphere by reconnection is controlled by global parameters rather than localized microinstabilities or other kinetic effects. Observations also show that magnetic shear is not the only factor that controls the dynamic processes of the solar wind-mag- netospheric interaction. Larger-scale causes and effects of reconnection are important since most measurements sample the external region around the reconnection site. Both case studies and statistical studies based on numer- ous in situ observations provide evidence for large-scale quasi-steady reconnection and patchy intermittent re- connection. Because observations in the narrow region where magnetic field and plasma disconnect are diffi- cult to obtain, observational tests of the theoretical pre- dictions of reconnection have focused on locations outside this diffusion region. Comparison of the observa- tions with the testable predictions given by theory and simulation provides valuable information for reconnec- tion research. Over the next decade, a systematic effort will be needed to establish and develop a fundamental theoreti- cal understanding of the dynamics of magnetic recon- nection. The lack of a complete dynamical model for interactions at the magnetopause has led to a number of controversies in magnetopause physics and our under- standing of substorms. In addition to considering the

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS localized breakdown of the frozen-in condition, atten- tion also needs to be focused on the reconfiguration of the plasma after reconnection has taken place. Observa- tional studies that can refute or verify the results from such theory and simulations are clearly necessary. Multi- point spacecraft missions such as Cluster 11 and Mag- netospheric Multiscale are required to separate space and time variations and to simultaneously sample the plasmas in the inflow and outflow regions of a recon- nection site. A direct interaction between the theoretical predictions and analysis of data from spacecraft and grou nd-based i nstru meets wi 11 be needed to obtai n a complete understanding of reconnection processes. Magnetic reconnection is a universal process that also occurs in phenomena such as solar flares, coronal mass ejections, the galactic dynamo, and astrophysical accretion disks, as well as in laboratory plasmas. How- ever, the magnetosphere is the only cosmic plasma where reconnection processes can be observed in con- siderable detai 1. U nderstandi ng of the reconnection pro- cess in the magnetosphere will be a significant step to- ward the understanding of related phenomena occurring in astrophysics, solar physics, and laboratory plasma physics. DYNAMICS OF GEOMAGNETIC STORMS, SUBSTORMS, AND OTHER SPACE WEATHER DISTURBANCES There are a number of different modes of geomag- netic activity, including magnetic storms and substorms. A magnetic storm occurs when merging of the interplan- etary magnetic field with Earth's magnetic field causes deep and intense circulation of the magnetospheric plasma, building up the energy content of the ring cur- rent to unusually high levels. The close relationship be- tween strong southward i nterpl anetary magnetic field that lasts for 3 hours or more and large magnetic storms is a consequence of the central role of convection. Substorms, on the other hand, are impulsive events that unload energy stored in the magnetotail. A period of southward IMP triggers en tranced recon nection at the dayside magnetopause, decreasing magnetic flux on the dayside and adding magnetic flux (and thus stored mag- netic energy) to the magnetotail. Substorms return mag- netic flux to the plasma sheet and dayside magneto- sphere from the magnetotail. They can occur multiple times within a magnetic storm period but also several ti mes per day even i n the absence of storm activity. The nature of the coupling between storms and substorms is an unresolved issue that is receiving much attention. 1 55 It should be noted that storms and substorms are not the only dynamic occurrences in the magnetosphere. Other features that have been identified i ncl ude pole- ward boundary intensifications and associated plasma sheet flow bursts, the magnetospheric response to solar wind dynamic pressure enhancements ("dynamic pres- sure disturbances"), and steady magnetospheric convec- tion intervals. Poleward boundary intensifications, which are enhancements in the aurora that occur at the poleward edge of the auroral oval, can strongly affect the entire plasma sheet and auroral oval. Dynamic pres- sure disturbances lead to dramatic increases in the mag- netopause, tail, field-aligned, and ionospheric current systems. They also lead to large enhancements in essen- tially all energetic particle populations within the mag- netosphere, in auroral particle precipitation, and in large-scale reconnection rates along the open-closed field line boundary. Periods of steady enhanced convec- tion (e.g., convection bays) lead to large increases in fluxes in the inner plasma sheet and to large and long- lasting enhancements in morningside auroral electron precipitation. We have known for some time that the nature of an effective solar wind driver depends on the type of geo- physical response in question. For example, high-speed solar wind streams with embedded magnetic field fluc- tuations produce strong radiation belt enhancements and prolonged substorm activity but very weak mag- netic storm responses. However, the A-l-M system is not a passive element responding in a prescribed way to the solar wind. The state of the system is a function of cou- pling and feedback processes determining the effective- ness of solar wind forcing. Over the next decade much work will be done on the underlying physical processes and coupling mechanisms that alter the geoeffectiveness of solar wind drivers and possibly modify the distribu- tion of solar wind energy within the A-l-M system. It is expected that fresh insights into the interpretation of statistical data sets and of correlations between mag- netic and solar wind indices will result and that previ- ously unappreciated geoeffective elements in the solar wind will be identified. Future progress in understand- ing the response of the magnetosphere, ionosphere, and upper atmosphere to the energy inputs from the Sun and solar wind requires an integrated approach that views the A-l-M as a tightly coupled system. Sources and Losses of Radiation Belt Electrons Earth's radiation belts are situated in the closest re- gion of near-Earth space, which hosts the majority of

1 56 commercial and military spacecraft in orbit around our planet. While the particle populations in this region have been measured since the dawn of the space age, the physical mechanisms underlying some of the most fun- damental properties of the belts are not yet understood. In particular, we do not know (1 ) the primary means by which the radiation belt electrons are accelerated to energies of hundreds of keV to many MeV, (2) the rela- tive roles of different types of waves in the loss of the particles via resonant wave-particle interactions, or (3) why some geomagnetic storms result in flux enhance- ments while others do not. While theoretical models of radiation belt dynamics do exist, they are not yet estab- lished enough to allow accurate, reliable predictions of dynamical response to solar and geomagnetic activity. From the point of view of basic plasma physics, the radiation belts constitute the A-l-M region within which mechanisms and effects of wave-particle interactions are man ifested i n thei r richest variety. Th is region harbors a variety of plasma waves that can resonantly interact with the particles, scattering them in energy and pitch angle and alternately accelerating them or causing them to preci pitate out of thei r trapped orbits. The waves i n- volved range in frequency from mHz to the electron gyrofrequency (kHz), and the underlying physical pro- cesses are typically highly complex, requiring a combi- nation of macroscale and microscale modeling. The ionosphere and the solar wind are two plasma reservoirs which are the ultimate sources of radiation belt electrons. Since the temperature of both of these plasmas is <10 eV, significant acceleration must occur for radiation belt electrons to attain energies up to sev- eral MeV. The physical mechanisms by which electrons are accelerated are not yet understood. The identifica- tion of the primary acceleration mechanisms is needed for a quantitative understanding of radiation belt dy- namics and for a proper assessment of radiation effects on satellites. While there is general agreement that elec- tron acceleration must involve the violation of one or more of the adiabatic invariants (approximate constants of cyclotron-, bounce-, and drift-motion), candidate ac- celeration mechanisms under consideration include ra- dial diffusion and scattering by wideband electric or magnetic field fluctuations, drift resonance with ULF (mHz) waves, and resonance with waves near the elec- tron gyrofrequency. At present, we do not know (1 ) the wave modes that can resonate with the particles over the very wide range of particle energies involved, (2) the locations of the primary acceleration regions and the importance of the plasma parameters and propagation conditions for efficient acceleration, or (3) the time THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS scales for wave-driven acceleration. Resolving these questions will require simultaneous and multipoint mea- surements of waves and particles before, during, and after storm-time acceleration. The primary loss processes for trapped radiation belt electrons consist of precipitation resulting from pitch angle scattering of the particles via wave-particle inter- actions and/or Cou lamb col I isions. Cou lamb col I isions with the dense atmosphere dominate at low altitudes, while wave-particle interactions are generally believed to be the agents of loss at higher L-shells. The waves that are responsible include whistler-mode waves (named for their descending frequency vs. time at a fixed point away from the source), generated in the magnetosphere and by lightning discharges, as well as byVLF transmit- ter signals. However, the relative roles of different waves and the regions over which they are dominant are not yet known. Quantitative understanding of loss processes is important for modeling the relaxation and quiet time structure of the belts. Second, quantification of electron loss during disturbed times sets constraints on the mini- mum amount of acceleration required. Electron loss is thus taking on a new significance in the effort to under- stand the processes responsible for electron accelera- tion. Third, precipitating radiation belt particles can pen- etrate to altitudes of between 40 and 70 km, where they can affect the chemistry and dynamics of the lowest regions of the A-l-M system. Assessment of the global effects of such precipitation must be based on a quanti- tative understanding of electron loss rates under differ- ent conditions. Response of the Ring Current Observations of the response of the magnetosphere to strong driving by interplanetary magnetic clouds have revealed new features of ring-current dynamics and cou- pl i ng that wi I I be a focus of research efforts i n the next decade. IMCs can have time scales for southward IMF as long as 12 hours and a slow, smooth rotation of south- ward to northward IMF as the cloud moves past Earth. Large solar wind dawn-to-dusk electric fields associated with these southward interplanetary magnetic fields are mapped down along magnetic field lines across Earth's polar cap and magnetotail. These electric fields are as- sociated with a strong convection that draws plasma from the magnetotail deep into the inner magnetosphere. Ions moving into the inner magnetosphere are energized as they drift across electric equipotentials to form the bulk of the storm-time ring-current populations. Thus the magnitude of the cross-tail electric potential is an

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS indicator of the energy available for accelerating ring- current ions. The ring-current ions generate an electric field that shields the inner magnetosphere from the con- vection electric field. In the time-dependent case, there can be overshielding or undershielding of the convec- tion field. Evidence for overshielding has been found in images of a shoulder feature in the outer boundary of the plasmasphere by the IMAGE spacecraft. As a consequence of the long time scales associated with the IMC, plasma sheet ions, which move on open drift paths into the inner magnetosphere, are not cap- tured on closed drift paths but make one pass through the inner magnetosphere before being lost at the day- side magnetopause. This produces an asymmetric ring current, with up to 90 percent of the particle energy flowing along open drift paths during the buildup of the ring current, referred to as the main phase of a storm. When the solar wind east-west electric field weakens, open drift paths are converted to closed, trapping the ring current in the inner magnetosphere. This trapped ring current is symmetric, requiring no ionospheric clo- sure current, and decays slowly through collisions with neutrals and the ionized plasmasphere and through scat- tering of ions in the fields of plasma waves. New ener- getic neutral atom (ENA) images of the ring-current struc- ture and evolution from the Polar and IMAGE missions clearly reveal the asymmetric open drift paths of the main-phase ring current and its gradual conversion to a symmetric form as the IMP turns northward. Calcula- tions of global ring-current loss rates using ENA images document the rapid loss lifetimes associated with the asymmetric ring current. Clearly ring-current intensity and evolution are a function of both the injection strength, as represented by the large-scale electric field, and the source strength, which includes solar wind and ionospheric contributions. New information from recent superstorms indicates that the injection strength as measured by the convec- tion potential is not simply a function of the solar wind east-west electric field strength, but is modified signifi- cantly by the A-l-M system. This surprising result indi- cates that the A-l-M system actively limits the amount of solar wind energy input. The situation is further compli- cated by variations in plasma sheet density, which may determine the relative strength of superstorms when the convection potential has become saturated. Clearly the structure and dynamics of the large-scale electric field and its impact on the inner magnetospheric populations must be a major focus of magnetic storm and inner magnetospheric research in the coming decade. It is also clear that ion source strength is important 1 57 in determining ring-current intensity, composition, and dynamical variation during magnetic storms. Respond- ing to strong solar wind driving, the plasma sheet can become enriched in ionospheric ions and structured in the radial and azimuthal directions, affecting the ring- current strength and subsequent decay. If a superdense plasma sheet population moves into the inner magneto- sphere in response to a storm-time convection electric field, it has a major impact on the strength of the ring current being formed at this time. A sharp decrease in plasma sheet density can trigger ring-current decay even in the presence of strong solar wind driving. A major focus of the next decade wi 11 be on understanding the dynamical variations processes responsible for loss of magnetotai I plasma and refi 11 ing (from both ionospheric and solar wind sources) during the elevated magnetic activity associated with storms. Coupling of Magnetotail Dynamics to the ionosphere Earth's magnetotail couples to the ionosphere at high latitudes. A dynamic, visual reminder of this coupling is provided by auroras in the Northern and Southern Hemi- spheres. In the past decade, progress has been made in understanding pieces of this coupling, much of which occurs at the boundary of the plasma sheet and the low- density region of the tail called the lobes, the plasma sheet boundary layer (PSB L), shown in Figure 3.1 a. Al- though this region was identified more than 15 years ago, many early observations were made by equatorial spacecraft at large radial distances. From such measure- ments, the boundary was shown to have spiky, large- amplitude electric fields, significant plasma flow, includ- ing bursty bulk flows, counters/reaming electron and ion beams, enhanced broadband electrostatic noise, and field-aligned currents. Many of the same features are also observed at low altitude, such as field-aligned particle beams, field- aligned currents, and large-amplitude electric fields. More recently, observations from the Polar spacecraft have confirmed the connection between the PSBL in the tail and the ionospheric auroral zone. These observa- tions have included detailed comparison of plasma dis- tribution functions, which show evidence of auroral ac- celeration. There is also evidence for field-aligned current systems and large electric fields. The large elec- tric fields are located predominantly at the poleward edge of the high-latitude plasma sheet observed by Po- lar and appear to be correlated with substorm activity. The electric and magnetic fields have also recently been

1 58 shown to carry significant energy flux directed toward Earth, which powers the aurora at ionospheric heights. To date, the existing studies of the magnetotail-iono- sphere connection are mainly statistical, with a few case studies. The circumstantial case for this coupl ing is ex- cellent, but there remain many questions about the de- tails of the process. Just how electromagnetic energy flux is generated in the tail and converted to particle energy is not known. How the field-aligned currents are generated in the tail and how they connect to the au- roral zone is understood only qualitatively. NASA's ex- isting and planned multi-spacecraft missions will shed light on these processes; however, most of the envi- sioned missions involve a small number (between 2 and 5) of spacecraft at relatively small separations. To more clearly tie down the connection between the magneto- tail and ionosphere, spacecraft at high and low altitudes along roughly the same field line are needed. Within existing data sets, few conjunctions exist, reducing the likelihood of clear signatures for the coupling processes. A multispacecraft magnetosphere-ionosphere coupling mission, combined with suborbital and ground-based measurements, is within the cost range of NASA MIDEX and suborbital resources and NSF resources. The planned MagCon mission will provide much improved temporal and spatial resolution giving global coverage of the tail region. Storm-Plasma Sheet Coupling One of the most i mportant dynam ical processes i n the magnetosphere is the substorm cycle, which pro- duces intense electric fields and energized plasma dis- tributions moving earthward and tailward. New infor- mation that bears on the relationship between storms and substorms has recently been obtained by ENA imag- ing. A comparison between ENA images of isolated and storm-time substorm plasma injections reveals impor- tant differences between these types of events. Storm- time ion injections are more prolonged and suggest that a strong, quasi-steady, cross-tail electric field is a cause of the transport rather than the periodic stretching and collapsing of the magnetic field, which is characteristic of the substorm cycle. In addition, the magnetospheric region participating in the storm-time ion injections is much broader, encompassing most of the nightside in- ner plasma sheet region. Both ENA images and ring- current models suggest that the enhanced convection electric field during storm activity moves ion injections deep into the inner magnetosphere, where they are strongly energized to form the storm-time ring current. THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS In the absence of strong convection, during isolated substorms, these injections have essentially no impact on ring-current formation but act to enhance the cross- tail current at large radial distances. We already know that long intervals of strong con- vection are necessary to produce an intense ring cur- rent. Variable substorm electric fields may play an im- portant role in diffusing energetic ions inward to form the high-energy tail of the ring-current distribution but are not responsible for the major portion of ring-current energization. It appears that substorms modify the plasma distributions in the inner plasma sheet, creating high-temperature source populations for the ring cur- rent. Even more dramatic modifications to the plasma sheet density occur in association with severance and loss of the outer portion of the plasma sheet, which sometimes occurs. This substantive loss of magnetotail plasma is important to magnetotail energetics. Cold and Hot Plasma Interactions at the Plasmasphere-Ring Current Interface Innovative imaging techniques are already reveal- ing new details of the strong coupling between hot and cold plasma populations in the inner magnetosphere. These details have implications for ring-current decay; plasma wave generation, propagation, and damping; and thermal plasma energetics and dynamics. Ring-cur- rent precipitation loss is enhanced on occasion in a region coincident with the dusk bulge, and plasma- spheric tails (also called drainage plumes) are formed in response to changes in the global convection pattern (Figure 3.8~. An array of troughs and bite-outs in the thermal plasma structure appear to be signatures of the structures in the global electric field produced by ring- current penetration electric fields. Final Iy, observational evidence exists that large-scale electric fields associated with partial ring-current closure and ring-current shield- ing effects dramatically alter the duskside thermal plasma dynamics. The structuring of the plasmasphere has important consequences for the growth and propa- gation of a variety of plasma waves. These waves are believed to be important in producing loss of ring cur- rent and radiation belt ions and electrons as well as in redistributing energy among different plasma species. The ring current is a major heat source for the plas- masphere. However, the temperature structure of the plasmasphere is not related in a simple way to magnetic activity. The details of this relationship are important for global models of the inner magnetosphere. The contri- bution of ring-current ion precipitation to the heating

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 59 FIGURE 3.8 Plasmasphere and auroral oval as viewed by EUV instrument on the IMAGE satellite, showing plasma on dusk side (left) swept toward magnetopause (top) during a period of enhanced magnetospheric convection electric field. Image taken on August 11, 2000, 1755 UT. Courtesy of B.R. Sandel, University of Arizona. and ionization of the neutral atmosphere (at subauroral- and midlatitudes) and to alterations in the minor species chemistry is not yet understood. Subauroral ion precipi- tation zones intensify and move equatorward as mag- netic activity increases. This ion precipitation is a time- varying mixture of oxygen and proton components, which have very different interactions with thermal ion and neutral populations. The rates of volume heating and ionization of the neutral atmosphere by precipitat- ing ring-current oxygen ions can equal or exceed day- time solar heating and ionization rates. Precipitating pro- tons produce lower heating rates for comparable characteristic energies and energy fluxes, but even higher ionization rates, which peak at lower altitudes. In certain regions, protons have been shown to make an important contribution to E-region ionization. SOLAR VARIABI LITY AN D CLI MATE Solar Variability and Middle Atmospheric Chemistry The middle atmosphere, extending from about 10 or 15 km altitude upward to about 90 km, is a region whose chemistry is dominated by minor constituents, and especially by ozone. Ozone (O3) iS formed through dissociation of O2 by solar ultraviolet (UV) radiation, producing atomic oxygen which reacts with O2 again to form ozone. The concentration of ozone is determined

1 60 by a balance between its production rate and its loss rate through photodissociation, chemical reactions, and transport processes. Photodissociation of ozone by solar UV radiation has two important effects: It is the princi- pal source of heat for the middle atmosphere, and it protects Earth's surface from UV radiation, which can destroy proteins and nucleic acids such as DNA. Indeed it is widely thought that life could not have emerged from the protective ocean until the ozone concentration had developed sufficiently to allow life to exist on land. Destruction of ozone by catalytic chemical reac- tions has become the focus of worldwide interest in recent years, since the halogen constituents involved (mainly chlorine and bromine) are almost entirely an- thropogenic in origin. Variations in solar UV radiation, however, also produce changes in ozone concentration, and the magnitude of these changes remains poorly un- derstood. An additional complication is that changes in ozone concentration also cause changes in the strato- spheric temperature distribution and consequent changes in the atmospheric transport of ozone. This feedback can also influence the dynamics of the upper troposphere, as has been shown by simulations using general circulation models. In these circumstances it is vitally important that we understand the chemistry of ozone and the other minor constituents of the middle atmosphere in detail, and in particular the variability that is likely to occur. Changes in ozone production will take place primarily as a result of variabi I ity i n sol ar UV red iation i n the 1 70- to 242- nm wavelength range. The variation in this wavelength range over the 1 1 -year solar-activity cycle has been esti- mated using data from spacecraft missions, particularly the Upper Atmosphere Research Satellite, but a knowl- edge of possible variation on longer time scales with high spectral resolution is required for a ful I understand- ing. To a large extent this will be accomplished by the Solar Radiation and Climate Experiment mission, cur- rently schedu led for launch in 2003 and also intended as a component of the NPOESS series, a merger of DOD and NOAA monitoring capabilities. Solar Irradiance Variability and Climate Since the Sun is the ultimate driver of our climate system, it is reasonable to suspect that there might be a link between solar variability and the changes in climate that are known to have taken place in the distant and recent past and are probably continuing to take place at the present time. Despite many claims of correlations between solar-activity indicators and climate variables, THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS however, the existence of such a link has been hard to prove. Part of the difficulty has been identifying a physi- cal mechanism to couple solar variability to the lower atmosphere and the surface of the Earth. In the case of variability in the Sun's luminosity (total irradiance), the difficu Ity is showing that the variations are large enough to have a significant effect on the surface climate. Vari- ability in the Sun's irradiance in the ultraviolet is sub- stantial on a decadal time scale however, the radiation is largely absorbed in the middle atmosphere and its influence on surface climate is not completely clear. Interest in variations in the Sun's total irradiance (the solar "constant") was sparked by two developments: the need to identify the natural sources of climate vari- ability so that the anthropogenic impact of greenhouse- gas emissions could be estimated, and the discovery that total irradiance does indeed vary on time scales at least as long as the 11-year solar-activity cycle. The cli- matic effects of variations on shorter time scales are probably negl igible, wh i le the variations over the 1 1 - year time scale that have been identified seem too small to have a significant impact. There is an increasing amount of evidence from paleoclimate studies and iso- topic measurements, however, that total irradiance prob- ably varies at longer time scales and may have been an important factor in past climate changes. Since such variations can either weaken or amplify the anthropo- genic effects, and thereby increase or reduce the time available to address these effects, it is of the utmost importance to estimate their magnitude. Part of the variation in the Sun's irradiance takes place in the ultraviolet region of the spectrum, which is responsible for both the formation and the destruction of ozone in the stratosphere. The relative UV variations on the 11-year solar-activity time scale are an order of magnitude larger than the variations in total irradiance, and recent modeling work has shown that the former can have a significant effect on tropospheric dynamics and hence on climate. Presumably, total irradiance variations at longer time scales would be accompanied by UV variations of larger magnitude, which could amplify climatic effects. This possibility is relevant to the prediction of future climate change, and further studies are needed. Solar irradiance measurements can only be carried out from spacecraft, since absorption and scattering by clouds and the atmosphere make ground-based mea- surements unreliable. Since long-term measurements are needed, the instruments will have to be replaced from time to time. The inevitable differences in abso- I ute Cal i bration of d ifferent i nstru meets make it essen-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 61 tial that successive instruments have a period of over- have not yet shown a clear trend, and further observa- lap. tional and modeling efforts are clearly needed. Frontier Issues While changes in solar irradiance provide a direct means of changing the surface climate, other less direct mechanisms have been suggested, and further observa- tional and modeling studies are needed to test their fea- sibility. The basis for most of these mechanisms is solar- induced variability in cloud formation and cloud properties. Si nce clouds have a major i nfl uence on cl i- mate and the terrestrial radiation budget, they could provide a powerful means of amplifying the effects of solar variability, if a connection to solar variability can be shown to exist. The flux of galactic cosmic rays striking Earth varies in intensity over 11-year intervals, in antiphase with the solar magnetic activity cycle, as a result of interactions between the cosmic ray particles and the interplanetary magnetic fields. It has frequently been suggested that the ionization produced by these particles could influence the nucleation of cloud particles, which would serve to couple solar activity to weather and climate. An alterna- tive suggestion is that changes in the global electric field influence the freezing of supercooled water droplets and the release of latent heat. While observations have oc- casionally been taken as support for these mechanisms, the mechanisms remain controversial. If they are real, solar-induced changes in cloudiness are likely to pro- duce regional rather than global changes, in contrast to the global effects of variations in irradiance. The societal impacts of an improved ability to predict regional cli- mate changes are important, however, and further study is justified. Space Climate Recent work has pointed out the likelihood of an anthropogenic effect on the upper atmosphere and iono- sphere, mainly as a result of the enhanced CO2-induced cooling by radiation to space. The subject is still in its infancy, but the predicted changes could have a signifi- cant societal impact by changing the density of the me- sosphere and thermosphere. This would in turn affect the decay of satellite orbits and, possibly, the global circulation in these upper regions. The changes could also affect the radio-propagation characteristics of the ionosphere. Since the anticipated changes are long term, "space climate" is an appropriate name for the range of effects. Observations of upper-atmosphere parameters MAGNETOSPHERIC, IONOSPHERIC, AND ATMOSPHERIC PROCESSES IN OTHER PLANETARY SYSTEMS The physical and chemical processes that control the atmospheres and plasma environments of the plan- ets and their satellites are essentially the same through- out the solar system, but they are manifested in very different ways owing to differing heliocentric distances, planetary sizes, atmospheric compositions, and intrinsic magnetic fields. A comparative approach to understand- ing the processes controlling atmospheres, ionospheres, and magnetospheres throughout the solar system can be very effective. We have learned much about the neutral and plasma environments of the planets and satellites in our solar system over the past four decades, but our knowledge of these environments remains far from com- plete. Over the next decade a systematic effort will be needed to address the most basic deficiencies in this knowledge. Structure and Dynamics of the Upper Atmospheres and ionospheres of the Planets and Satellites The structure and dynamics of the upper atmosphere (mesosphere, thermosphere, and exosphere) and iono- sphere of a planet are strongly affected by inputs of energy and momentum from both the lower atmosphere, below it, and the magnetosphere and solar wind above. Our knowledge of planetary atmospheres has advanced considerably over the past three decades yet lags far behind our knowledge of the terrestrial system. The structure of the ionospheres of Venus and Earth is known rather well, but our knowledge of the other ionospheres in our solar system is limited. In situ measurements have been made in the upper atmospheres and ionospheres of Venus, Earth, Mars, and comets. Ionospheres have also been detected remotely with the red io occu Itation technique at Jupiter, Saturn, Uranus, and Neptune and at the satellites lo, Europa, Ganymede, and Titan. How- ever, this technique only provides total electron density, and measurements have not been made of other impor- tant quantities such as ion composition and tempera- tures. At Mars, retarding potential analyzers on the Vi- king 1 and 2 landers each provided a single set of ion density profiles for the major ion species, and the Mars Global Surveyor measured the total neutral density, the magnetic field, and superthermal electron fluxes.

1 62 The lower atmospheres of most planets, and per- haps a few satellites such as Titan, can deliver signifi- cant inputs of energy and momentum into the upper atmosphere via the upward propagation and breaking of internal gravity waves. On Mars it appears that gravity wave production, and consequent upper atmosphere heating, is associated with dust storms that sometimes occur in the lower atmosphere near the surface. Neutral density measurements by the accelerometer onboard the Gal i lea probe provided evidence for gravity wave propa- gation into the upper atmosphere of Jupiter, but the source mechanism is not known. Energy inputs into up- per atmospheres also come from the external plasma environment such as magnetospheres. Auroral emissions are one manifestation of such inputs. Missions are needed to study the structure and dynamics of planetary atmospheres and ionospheres and their coupling to the lower atmosphere and to the magnetosphere or solar wind. The Nozomi mission to Mars should provide some of this information, but it remains to be seen what shape the spacecraft and instruments will be in after the unan- ticipated 4-year flight delay before arrival at Mars. The ESA's Mars Express mission will also study Mars, but further missions will be required at both Venus and Mars to unravel the neutral dynamics and its role in A-l-M coupling. Such missions should includeVenus and Mars climate orbiters (VCO and MCO) and a Mars Thermo- sphere ionosphere Dynamics Orbiter (MTIDO). An ap- propriate mission to Jupiter, such as a Jupiter polar or- biter or a Jupiter auroral orbiter, should also be considered; it might emphasize the complex A-l-M is- sues at auroral latitudes. A mission to Jupiter's satellite Europa has been suggested; any spacecraft sent to Europa should include ion and neutral mass spectrom- eters in its instrument complement in order to character- ize the composition of the atmosphere and ionosphere. Such measurements would also provide valuable infor- mation on the surface composition of this important sat- ellite. Among ongoing missions, the Cassini mission should provide much information about the ionosphere of Titan and some information about Saturn's iono- sphere. Characterization of the Structure, Dynamics, and Composition of the Planetary Magnetospheres Planets with sufficiently large intrinsic magnetic fields, such as Mercury, Earth, Jupiter, Saturn, Uranus, and Neptune, and Jupiter's satellite Ganymede, possess magnetospheres that act as obstacles to the solar wind. By contrast, the external plasma environment (e.g., solar THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS wind) interacts rather directly with the upper atmo- spheres and ionospheres of bodies that have small in- trinsic magnetic fields, such as Venus, Mars, Titan, and comets. In this case, "induced" magnetospheres form in which the external magnetic field is enhanced and drapes around the body. The terrestrial magnetosphere has been the subject of extensive study over the past 40 years or so by dozens of Earth-orbiti n g sate I I ites, i n c I ud i n g Dyn am i cs Exp I orer, ISEE, Geotail, Polar, and IMAGE. The magnetospheres of Mercury, Jupiter, Saturn, Uranus, and Neptune have been explored by the Mariner, Pioneer, Voyager, and Galileo missions. The Ulysses and Cassini spacecraft also made measurements within the magnetosphere of Jupiter, and the Cassini mission will provide extensive data on Saturn's magnetosphere. Our knowledge of Jupiter's magnetosphere has ad- vanced greatly as a result of the Galileo mission and earlier missions, although many questions concerning its structure and dynamics remain. Recent global MHD models have helped to put the data returned from these various missions into context. Among the many out- standing questions concerning this magnetosphere are the following: What are the relative roles of planetary rotation and the solar wind in driving the dynamics? Where are the locations of reconnection regionts) on the magnetopause? What are the acceleration mechanisms) for the observed energetic particles? What causes the energetic particle bursts coming from the magnetosphere but observed outside it? What are the identity and na- ture of the particle populations responsible for the Jo- vian aurora? What is the role of the ionosphere (and ionospheric electrical conductivity) in the A-l-M inter- action? What are the mechanisms for the interaction of the magnetospheric plasma with the various satellites (e.g., the lo plasma torus)? One mission that would fill many of the gaps left from earl ier missions, particularly in the inner magnetosphere and high-latitude regions, would be a Jupiter polar orbiter or a Jupiter Auroral Orbiter (lAO). Planetary auroras are observable manifestations of atmosphere-ionosphere-magnetosphere coupling and occur when energetic charged particles from an exter- nal pi asma envi ran ment i nteract with a neutral atmo- sphere. Very often, auroral particle precipitation is a sign of field-aligned electrical currents, which result from dynamical/MHD stresses somewhere in the A-l-M sys- tem. Jupiter provides a particularly interesting example of a planetary aurora. The total auroral power is a few times 1013 watts, and the auroral emissions are observed in the x-ray, ultraviolet, visible, infrared, and radio re-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS "ions of the spectrum. For example, observations of au- roral ovals by the Hubble Space Telescope (MST) reveal a complex spatial and temporal morphology (Figure 3.9~. A JAO-type mission and regular remote monitoring at all wavelengths (e.g., by HST and other faci l ities) wi l l be needed to understand this aurora. Mercury has an intrinsic magnetic dipole moment but only a thin neutral exosphere, and it has a small magnetosphere that interacts with the solar wind. Al- most all our information on this unique magnetosphere comes from instruments onboard the Mariner 10 space- craft. It is time that we explore this magnetosphere more thoroughly and look for phenomena such as sub- storms. The proposed Messenger mission is a start in this direction. Electrodynamical Coupling Processes at Weakly Magnetized Planets and Bodies The solar wind is able to interact directly with the atmosphere and ionosphere of a planet or object pos- 1 63 sessing only a weak intrinsic magnetic field. Examples of weakly magnetized bodies include Venus, Mars, PI uto, comets, and satel I ites such as lo and Titan. For the satellites, the relevant external plasma flow is associated with the parent planet's magnetosphere. A thorough re- view of the solar wind interaction with weakly magne- tized planets and bodies can be found in the report by the Panel on Solar Wind and Magnetosphere Interac- tions, but a concise review of this topic will also be given below. The sol ar wi nd i Interaction with Ven us was most recently studied from 1978 through 1992 by the Pioneer Venus mission. The instrument complement on the Pio- neer Venus Orbiter was well designed, and much was learned about th is i Interaction. However, sign if icant gaps in our knowledge remain, such as the cause of the mag- netic flux ropes observed in the ionosphere by the mag- netometer and the cause of ion outflow on the nightside. No mission is currently planned forVenus. Although many missions to Mars have been under- taken over the past three decades, we still do not have a comprehensive picture of the solar wind interaction with FIGURE 3.9 Jupiter's aurora as seen with the Hubble SpaceTelescope. Emission is evident from a main auroral oval,which is thought to be associated with field-aligned electrical currents originating in the outer part of the Jovian magnetospere. Auroral emission is also present at lower latitudes at the magnetic footprints of the satellites lo, Europa, and Ganymede and is evidence for the electrodynamical coupling of these satellites with the Jovian atmosphere. Courtesy of John Clarke, Boston University.

1 64 this planet. All the NASA missions lacked comprehen- sive i nstru ment packages that cou Id have provided co- ordinated information on the upper atmosphere, iono- sphere, and plasma environment. Only recently have we even learned (from the magnetometer onboard the Mars Global Surveyor) that strong localized, crustal magnetic anomal ies exist on Mars that apparently create mini-magnetospheres, affecting the interaction with the sol ar wi nd. The solar wind interaction with comets is domi- nated by the mass-loading process, in which neutrals distributed over a large volume of space are ionized by solar radiation and contribute heavy cometary pickup ions to the solar wind flow. Halley's Comet was inten- sively studied by many spacecraft (and especially the European Giotto spacecraft, which approached to within 600 km of the nucleus). The U.S. International Cometary Explorer (ICE) spacecraft flew through the plasma tail of comet Giacobini-Zinner. Remote observations of com- ets continue to be an important source of information on these primitive objects. For example, the ROSAT and Chandra observatories have recently seen soft x rays coming from a large number of comets. It has been demonstrated that these x rays are produced by the charge transfer of solar wind heavy ions with cometary neutrals. Currently, the only approved cometary mission with instruments relevant to A-l-M is ESA's Rosetta mis- slon. The solar wind interaction with Pluto might be cometary in nature, but nothing is really known about it. Any mission to Pluto now being contemplated should include an instrument capable of measuring pickup ions in the vicinity of Pluto and learning their composition. Other interesting objects with weak intrinsic mag- netic fields and interesting A-l-M effects associated with external plasma interactions include the Galilean satel- I ites lo, Europa, Ganymede, and Cal I isto and Saturn's largest moon, Titan. The jovian satellites were studied by the Gal i lea mission, as wel I as by the earl ier Voyager mission and by remote observations, but much remains to be learned. As noted above, any spacecraft sent to Europa should be equipped with ion and neutral mass spectrometers to characterize the atmosphere and iono- sphere. Saturn's magnetospheric plasma strongly interacts with Titan's upper atmosphere and ionosphere, perhaps helping to drive the complex hydrocarbon chemistry in the lower ionosphere that contributes to haze formation. Most of the important A-l-M issues at Titan will be ad- dressed by the Cassini mission starting in 2004. THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS 3.3 SOCIETAL IMPACT OF SPACE WEATHER The maturity of the atmospheric-ionospheric-mag- netospheric discipline has made possible a close con- nection between A-l-M science and applications for the benefit of society. The application of space physics and aeronomy to societal needs is now referred to as space weather. Whi le relevant activities have been pursued by NOAA and DOD for many years, the science commu- nity has only recently become more closely connected with this pursuit, since the inception of the National Space Weather Program i n 1 995 and NASA's Livi ng With a Star program. The NSWP is a multiagency endeavor to understand the physical processes, from the Sun to Earth, that result in space weather and to transition advances in science into operations in order to assist users affected by the space environment. This transitioning of knowledge has been an obstacle for many years and is still a major challenge. The scientific community should be encour- aged to bridge the gap between the science and user needs by developing operational products. It is im- portant that resources be available to help scientists pre- pare their models and data for transitioning to op- erations, pursue rigorous model val idation, ensure robustness of computer codes, and provide the docu- mentation essential for efficient and appropriate use of their models and data. Similarly, resources must be avail- able to the commercial and defense agencies, such as NOAA and DOD, so they can transition advances in modeling and data and put to use the new understand- ing of the physics of space and the upper atmosphere. In the area of policy research, it is important that cost- benefit analyses of the societal impacts of space weather be conducted, along with annual updates to keep pace with rapidly changing technological systems. NASA's new LWS program is an excellent opportu- nity to provide measurements and develop models that will clarify the relation between sources of space weather and the ultimate impact of space weather on society. Those space weather phenomena that most di- rectly affect life and society include radiation exposure both in space and at commercial airline altitudes; com- munications and navigation errors and outages; changes in the upper atmosphere that affect satellite drag and orbital decay; radiation effects on satellite electronics and solar panels; and power outages on the ground due to geomagnetically induced currents, to name a few.

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS The LWS science architecture team has proposed an end-to-end approach to understanding how solar vari- ability affects life on our planet. In the A-l-M area, a geospace missions network is being defined. The net- work will include geosynchronous-transfer-orbit satel- lites to study the radiation environment across L-shells, essential to understanding the rapid buildup of outer zone electron fluxes and the access and trapping of so- lar energetic protons in the inner magnetosphere. The network will also include low-Earth orbiting spacecraft with higher inclination, designed to study both thermo- sphere-ionosphere effects and precipitation of energetic particles from the radiation belts, and, finally, a high- inclination elliptical remote-sensing spacecraft to pro- vide global imaging relevant to current systems in the high-latitude ionosphere that affect power outages through GlCs. LWS will also contribute to missions-of- opportunity flights, leveraged programs, instrument de- velopment, and the development of end-to-end models that emphasize boundaries and linkages, above and be- yond the scope of existing SR&T grants. The Panel on Atmosphere-lonosphere-Magnetosphere Interactions en- dorses these goals in its recommendations, which are set forth in Chapter 6, and stresses the need to develop solar and geospace missions in parallel. The goals of space weather modeling and observa- tion are multifaceted. The first is the production of a suite of accurate climatological models describing the mean conditions, together with an estimate of the range of states as a function of conditions such as season, and solar and geomagnetic activity. Such models would be a valuable tool for spacecraft design for instance, to esti- mate the expected environmental extremes of radiation belt particles on a polar-orbiting spacecraft. A second goal is to capture the "weather" of the system that is, the minute-to-minute or hour-to-hour changes in the space environment. Space weather can be divided fur- ther into specification and forecast, the latter being an extreme test of our physical understanding of the com- plete solar-terrestrial system. It is imperative that we embrace all available techniques to achieve these goals. For instance, we must adopt the advanced data assimi- lation techniques that have been the mainstay of con- ventional weather forecasting for many years. We must endeavor to develop comprehensive satellite and ground-based observing systems to provide measure- ments to drive the data assimilation models. It is impor- tant that research and operations work together to achieve maximum benefit from new multispacecraft ob- serving systems. For instance, every effort must be made to disseminate measurements from science missions for operational use in real time. An excellent example of 1 65 this has been the availability of solar wind data from NASA's Advanced Composition Explorer (ACE) space- craft. We should strive to make this the norm, thereby maximizing the use of scarce community resources. It is also important that a system be developed to allow data from existing operational satellites, such as DMSP and NOAA-POES, to be downloaded in real time rather than once per orbit. The 2-hour delay in orbit-by-orbit satel- lite communication reduces the value of the data for operational use. The maturity of our understanding of the physics of the upper atmosphere enables us to move beyond quali- tative science to a true quantitative description of the environment. Such an advance involves developing ap- propriate metrics to measure the ability of models or theories to make predictions. When we develop or im- prove empirical models such as the International Refer- ence ionosphere, it is no longer sufficient to specify the parameters themselves instead, we must strive to char- acterize the variability of the system under different con- d itions. The societal impacts of space weather are broad, affecting communications, navigation, human health, power distribution, and satellite operations (Figure 3.101. Space weather is of international concern, and other nations are pursuing parallel activities that can be lever- aged through col laborations. The European Space Agency and individual nations within ESA have pro- grams for studying space weather. For example, the ESAls 2001 report on space weather1 articulated possible ele- ments of a future space segment for an operational, ser- vice-oriented, European space weather system. Like- wise, Japan has a solar cycle's worth of energetic particle data from the polar-orbiting Akebono satellite to con- tribute to radiation belt studies, and comparisons be- tween YoLkoh x-ray images of the Sun during the last solar cycle and images that will be obtained from the new x-ray imager to be flown on the GOES spacecraft will provide an important cycle-to-cycle and long-term baseline on x-ray input to the atmosphere. A brief re- view of the principal space weather effects follows. COMMUNICATIONS Communications can be separated into near-ground and space-based systems; both are affected by the space ~ ESA. 2001. Space Weather: Concurrent Design Facility Study Re- port, CDF ~1 (A), December. Available online at <http://www.estec. esa. n l/wmwww/wma/spweather/esai n itiatives/>.

1 66 THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS FIGURE 3.10 Space weather hazards, including space, air, and ground effects. Space weather describes events in space that affect Earth and our technology. Severe solar eruptions can cause disturbances that dramatically affect both the ionosphere and magnetosphere. Courtesy of L. Lanzerotti. environment but in quite different ways. Near-ground communication systems (ground-to-ground and aircraft- to-ground) utilize high frequencies (3-30 MHz) and use the ionosphere actively as a reflector to propagate a radio signal from transmitter to receiver. Satellite com- mun ications transmit at h igher frequencies (VH F. U H F. L-band, etc.), where it is irregularities in the ionosphere that are most likely to disrupt the fidelity of the radio signal. For high frequencies, two main space environment factors affect the propagation: absorption of the signal i n the D-region and changes in the reflecting properties of the ionosphere in the F-region (near 300 km altitude). Absorption is the process by which the energy of radio waves is converted into heat and electromagnetic noise through interactions between the radio wave, iono- spheric electrons, and the neutral atmosphere. Most of the absorption occurs in the ionospheric D-region (50- 90 km altitude), where the product of the electron den- sity and the electron/neutral collision frequency attains a maximum. Within this region the neutral density is relatively constant over time, so variations in the local electron density drive the total amount of absorption. The electron density is a function of many parameters and normally varies with local time, latitude, season, and stage of the solar cycle. These natural changes are predictable and affect absorption only moderately at the lowest HE frequencies. Much more significant changes in the electron density, and therefore the absorption strength, are seen in response to solar x-ray flares (the classic short-wave fade) and solar proton events (the classic polar cap absorption), both of which change the lowest usable frequency. Routine monitoring of the so- lar x-ray flux by the GOES series of operational space- craft provides a reasonable specification of the current condition of absorption on the dayside of Earth. Fore- casting an x-ray flare a day or two in advance would be an ideal situation for HE operators, but even a short- term forecast of a few hours presents a huge challenge for solar physics research. Likewise, specifying the spa- tial extent and intensity of polar cap absorption during a solar proton event is possible with existing polar orbit- ing and geostationary monitors, but a reliable forecast poses a similar challenge given our current understand-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS ing of solar and space physics. With the ever-present need for commercial airlines to maintain good commu- nication links, particularly with the increasing number of polar routes, there has been a renewed emphasis on accurate predictions of HF propagation conditions. The reflect) ng properties of the F-region ionosphere are influenced by another space weather event: the geo- magnetic storm. During a storm the high-latitude iono- sphere is impacted directly, but the energy in jected into the upper atmosphere drives winds and composition changes in the neutral atmosphere that produce both increases and decreases in F-region electron density across the whole globe. These changes in the iono- sphere affect the maximum usable frequency. At present, HF users rely on climatological models of the ionosphere to choose which frequency to use along a particular propagation path and have little or no fore- cast capability. Day-to-day variability is difficult to cap- ture, and the more severe disruptions during a storm are not yet adequately quantifiable. Advances in combin- ing models and data using data assimilation techniques are in progress and hold the best hope for improved specification of the F-region conditions; they also may allow at least a short-term forecast (6-12 hours). The main driver of this activity is the promise of new global satellite observing systems such as the Constellation Observing System for Meteorology ionosphere and Cli- mate (COSMIC), or the Geospace Missions Network, wh ich is a component of NASA's LWS program. Grow- ing networks of ground-based Global Positioning Sys- tem receivers complement the satellite observations and are essential to providing global coverage for specifica- tion and initialization for short-term forecasts that make use of physical models. Many military and civilian systems depend on HF and UHF communications and will continue to do so for the foreseeable future. Localized conflicts in future military confrontations very likely will require secure communications for locating downed pilots. Complex operations may be delayed if ionospheric conditions are threatening. The airline industry uses the HF band ex- tensively for communications. Communications systems at higher frequencies that are less susceptible to space weather effects are also used by airlines, but these addi- tional capabilities are not available at all locations, par- ticularly in the northern polar region. Near the poles, the ionosphere becomes strongly modified during large solar proton events. At the highest latitudes, these ener- getic protons arriving from the Sun have direct access to the ionosphere, where their energy is deposited. The HF communications systems are unable to operate well at 1 67 these times, so airlines must be rerouted to lower lati- tudes, at substantial cost to the airlines and ultimately to the consumers. The use of imaging holds great promise for tracking space weather disturbances, much as weather systems on Earth are now tracked from geosta- tionary orbit. The Drug Enforcement Agency uses over-the-hori- zon (OTH) radars in its battle against smuggling. These systems were important in cold war defense against cruise missiles and continue to play a huge role in the surveillance of regions outside the United States. They are now used to detect airplanes and ships from the velocity information encoded in the HF signals as they bounce off the ionosphere. Since the system uses HF waves that refract from the bottom of the ionosphere, a range of space weather effects can disrupt it. Allies such as Australia use OTH systems for wide-area surveillance in Asia and are keenly aware of space weather. Ground-to-satellite communication signals are only weakly affected by the ambient ionospheric densities. However, there are times, at low latitudes and in the auroral oval and polar cap, when plasma irregularities are predominant. Irregularities diffract even the very- high-frequency communication signals, leading to scin- tillation in both the amplitude and phase of the waves. Severe scinti I ration can result in complete loss of the radio transmission. Forecasting the occurrence of iono- spheric irregularities and scintillations is a major chal- lenge, but physical understanding of the system is ad- vancing rapidly. Renewed efforts in this area hold distinct promise for the future. The Air Force, with NASA's help, has embarked on an ambitious satellite program called the Communications/Navigation Outage Forecast System to predict these events and the scintilla- tion they cause. The Navy also has a sizable program in ionospheric characterization and prediction. NAVIGATION H igh-frequency satel I ite-based navigation systems, of which the GPS is an example, rely on propagating signals through the ionosphere from ground to satellite. Ionospheric i rregu rarities are therefore just as i mportant for these navigation applications as they are for satellite communication. Precise positioning can also be affected by the total electron content of the ionosphere, due to changes in the refractive index of the medium. As the navigation signals pass through the plasma the signal is retarded by the medi um, so the arrival ti me of the signal, and hence the calculation of position, is compromised. If the ionosphere is invariant, this correction can be

1 68 easily defined, but the highly dynamic nature of the system introduces an uncertainty for single-frequency GPS users. Commercial airlines, for instance, that rely on single-frequency GPS navigation systems require accurate specification of the current conditions. The Federal Aviation Administration attempts to make cor- rections to TEC climatology by using the Wide Area Augmentation System (WAAS) over the continental United States. Of particular concern are local gradients in electron content, which are difficult to capture in the coarse observi ng grid of the WAAS system. Specifyi ng and forecasting the state of the TEC has the potential to greatly advance the adoption of new data assimi ration techniques and observing programs. Ionospheric mod- els and the science on which they are based must con- tinue to improve if a WAAS-type system is to be able to capture these steep gradients in electron content. The GPS was invented for military purposes but it now has widespread application in both military and civilian uses. The ability to pinpoint one's position is of great value, and any outage or degradation due to space weather can be a threat to life and military advantage. The combination of GPS and cell phone technology and links to the 91 1 system hold great promise for safety and emergency situations. Low-frequency ground-based navigation systems, of which loran is an example, are normally unaffected by the ionosphere. However, during solar x-ray and proton events, the reflecting height of the radio wave is low- ered, introducing an ambiguity between the directly propagated signal and the signal reflected from the iono- sphere (known as the sky wave). The util ity of the system is compromised, requiring the system to be turned off for the duration of the event. ELECTRIC POWER ISSUES On March 13 and 14, 1989, one of the largest geo- magnetic storms on record plunged Quebec Province, Canada, and its 6 million inhabitants into darkness for over 9 hours and cost the utilities millions of dollars. In this case, as well as others like it, geomagnetically in- duced currents resulting from disturbances in the space environment have been the cause of power outages and damage to electric power generation and distribution systems. More recently, the smaller but still significant geomagnetic storm on July 15 and 16, 2000, caused voltage variations and tripping of protective devices at many locations in the United States. During large geomagnetic storms, rapidly time-vary- ing electric fields and currents that are often co-located THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS with the auroral oval intensify and expand to lower lati- tudes and broaden in area to expose large portions of Earth's surface to this destructive source of energy. Given enough warning, power companies can take protective measures to reduce effects on their systems by better management of connections and power transfer among grids, by deferring maintenance, by using a more con- servative load margin, and by adjusting protective de- vices. During the current solar cycle, the power industry may have become more vulnerable to space environ- ment disturbances because of regulatory changes that have resulted in increased utility connectivity and smaller operating margins. At the same time, new re- search, observations, and models are beginning to con- tribute to improved warnings and alerts of geomagnetic storms provided by the NOAA Space Environment Cen- ter, and private industry products are also being offered for use by the power utilities. Primary among the improvements are observations from the ESA/NASA SOHO spacecraft that provides so- lar observations of so-called halo events, or Earth-di- rected coronal mass ejections, that often result in geo- magnetic storms in a few days. However, we do not yet have the scientific understanding necessary to interpret these observations to skillfully predict the magnitude and timing of a disturbance reaching Earth, especially with the long lead times of several days that are desired by the power utilities and others affected by solar distur- bances. Another important observation is from the NASA Advanced Composition Explorer (ACE) spacecraft that is located at the L1 Lagrangian point between Earth and the Sun, about 1 percent of the way to the Sun. NASA and NOAA, along with collaborators from many nations worldwide that are tracking ACE, work together to make these data available in real time for the detection of geoeffective solar wind conditions and input to mag- netospheric models that forecast geomagnetic activity. These data are critical for providing short-lead-time warnings, about one-half to one hour, of impending geo- magnetic storm conditions. After completion of the ACE mission, it is imperative to continue these measurements with new solar-wind-monitoring measurements for both research and space weather operations. There are challenging opportunities to improve the magnetosphere-ionosphere-upper atmosphere models that use solar and solar wind data as input to determine the location and severity of geomagnetic disturbances. Coupled models of the entire Sun to upper atmosphere system are needed to test our understanding of solar- terrestrial processes and to improve our ability to fore- cast geoeffective disturbances. Finally, globally distrib-

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS uted ground-based magnetometer measurements are needed to validate model predictions and to provide nowcasts of geomagnetic d istu rbances with i mproved spatial resolution. ASTRONAUT, AIRLINE, AND SATELLITE HAZARDS The impacts of space weather on society are also increasing rapidly as we maintain a permanent presence in space, as our economy and lifestyle depend more and more on satellite technology, and as high-latitude airline flights become more frequent. Astronauts are subject to considerable radiation even in low-Earth orbit. When the International Space Station (ISS) changed its orbit to accommodate Russian launches, it moved to higher lati- tudes, where it is more susceptible to solar energetic protons, and it entered the dangerous portion of the radiation belt. As detailed in the 2000 NRC report Radia- tion and the Internationa/ Space Station, astronauts could receive a significant radiation dose (reaching allowable limits) during ISS construction due to large solar proton events that may occur during space walks. This study emphasized that the radiation risk to astro- nauts could be greatly reduced by employing data and models to adjust scheduled space walks. Missions to Mars will be very dangerous indeed, since solar flares give very little time to take safety precautions and the radiation levels could be life threatening. Space walks for construction and maintenance of the ISS can also be somewhat dangerous during geomagnetic storms as the radiation belt fluxes increase. Even polar airplane flights at times have high radiation levels. Although Earth's atmosphere still provides substantial protection from solar radiation, at high latitudes during large solar proton events, the radiation dose to airline passengers and crews becomes enhanced. Similarly, our satellites in space are vulnerable to a wide variety of environmental effects. Low- to medium- energy electrons create an electrical charging layer on the surface of spacecraft that can discharge and damage satellite systems. High-energy electrons can penetrate deep into the satellite's components, including onboard computers, and create electrical discharges that can cause problems ranging from minor anomalies to total failure of the spacecraft. Similarly, solar energetic pro- tons and heavy nuclei can penetrate deep into a space- craft's components and disrupt satellite operation. In addition, some spacecraft rely on the relatively steady magnetic field within the magnetosphere to maintain their orientation and control their momentum. During times when the solar wind is enhanced, the protective 1 69 shield of Earth's magnetic field can be compressed to within some satellite's orbits, leaving the spacecraft out- side the protective magnetosphere and unable to main- tain normal control. SATELLITE DRAG AND COLLISION AVOIDANCE Satellites in low Earth orbit are significantly affected by atmospheric drag. During a magnetic storm the at- mosphere is heated and expands. The dynamic nature of the energy input gives rise to the launch of neutral density waves, and the momentum forcing at high lati- tudes can lead to deep neutral density holes. In the past, after intense dynamic events, NORAD lost track of many space objects and had to laboriously reacquire them over the following days and weeks. Our prediction of such events could greatly assist in keeping track of all the space debris in orbit. Day-to-day neutral density perturbations contribute the largest error in the position determination of space objects and debris. With accurate knowledge of neutral density the uncertainty windows on predicted position could be significantly reduced and so limit the need for ISS to maneuver to avoid col I isions. 3.4 EXISTING PROGRAMS AND NEW INITIATIVES Our understanding of the A-l-M system is being ad- vanced through a number of vigorous programs, ranging from international and multiagency programs to smaller scale programs. Current examples are NASA's Interna- tional Solar-Terrestrial Physics (ISTP) program, the TIMED Solar Terrestrial Probe, the IMAGE MIDEX mis- sion and the SAMPEX and FAST Small Explorer (SMEX) satellites; the National SpaceWeather Program; the NSF- initiated SHINE, GEM, and CEDAR programs and their international counterparts; NOAA's space weather pro- grams; the International Space Environment Services program; and numerous DOD activities. These programs have supported satellite and ground-based monitoring instruments and the related data analysis and theory and modeling efforts. Through these targeted programs and the critically important base programs funded by NSF, NASA, NOAA, and the DOD, much important scientific progress has been made, helping us to focus our re- search effort on high-priority issues. The progress that

1 70 has been made and the important new questions that have been addressed are serving to excite and to moti- vate the next generation of young scientists who will lead our future explorations of the solar-terrestrial envi- ronment. Our progress has also led to direct application of research models and data for operational space weather prediction and new efforts to model the Sun- Earth system from a systemic and holistic perspective. While progress is being made, however, there are still very few areas where our space weather modeling capa- bilities have sufficient accuracy to influence economic decisions being made by private industry and the gov- ernment. Future programs and interagency activities wi 11 continue to refine our understanding of the A-l-M sys- tem and should lead to an improved and quantitative understanding of the complex coupling between the many different physical regimes extending from the Sun to Earth. Future progress in understanding the response of the magnetosphere, ionosphere, and upper atmosphere to the energy inputs from the Sun and solar wind as well as to forcing from the lower atmosphere requires a new approach that goes beyond the ISTP program's method of tracing the flow of energy through geospace with single-point observations. Future progress requires a sys- tems approach that views the A-l-M as a tightly coupled system with microscale drivers and feedbacks that modify the first-order global response. We need a more detailed view of the dynamical behavior of boundary regions in geospace such as the dayside magnetopause (to view the entry of solar wind energy and momentum), the nightside inner plasma sheet (to view the current disruption/acceleration boundary region), the midtail region (to view the nightside, near-Earth, magnetic merg- ing site that releases stored magnetotail energy), and key coupling regions between the ionosphere and mag- netosphere (to view the closure of currents through the low-altitude ionosphere). In these regions, microscale processes are known to control the global system be- havior. Clusters of satellites flying in close formation can resolve the dynamical response of the regions (by mea- suring gradients, curls, and divergences of important quantities) as well as distinguish spatial from temporal variations. In effect, they give us a space- and time- resolved view of the dynamical behavior of key regions rather than of the single-point, basic-state quantities measured in past missions. To maximize the information gained from these mis- sions, creative coordination of approved programs across agencies (e.g., NSF, NASA, NOAA, ONR, and DOD) is needed to place the system's microscale drivers THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS in the context of the solar wind energy input that they mediate and the global system response that they en- gender. Such a cluster in equatorial orbit would give us a global knowledge of the zonal electric field at the magnetic equator the single most important parameter needed for predicting space weather disturbances there. New, movable, state-of-the-art ground-based radars planned by NSF in conjunction with existing sophisti- cated arrays provide a powerful network for sensing the global magnetospheric configuration as mapped onto Earth's upper atmosphere. These radio-wave systems must be augmented by relocatable optical systems as well. New instrumentation on the NOAA geosynchronous satellites allows continuous monitoring of the solar disk in x rays to view oncoming explosive events. The launch i n 2004 of STEREO, part of NASA's Livi ng With a Star program, will allow stereo viewing of explosive events leaving the solar surface. Other elements of LWS in- clude a multisatellite radiation belt mission and a mis- sion to study magnetosphere-ionosphere coupling. The deployment of an operational geosynchronous global ionospheric imager by ONR is under discussion. DOD- operated Los Alamos National Laboratory geosynchro- nous satellites provide important multipoint monitoring of plasmas entering the inner magnetosphere from the magnetotail, the source populations for the ring current and radiation belts. New instrumentation on the DOD Defense Meteorological Satel I ite Program satel I ites wi 11 provide snapshots of the auroral oval that can be in- versed to provide energy input and conductivity esti- mates. We also need assured access to large-aperture telescopes capable of collecting laser light scattered back to Earth from metallic atom layers. These lidar systems are the only reliable tools for ground-based study of the 80-120 km component of the atmosphere; they can be augmented by rockets and ground-based imagers to provide the data we need to understand this important interface zone between the atmosphere and space. A systems view of the solar wind-A-I-M interaction motivates a search for new ways of providing snapshots of global quantities rather than single-point measure- ments in key regions of the A-l-M system. The IMAGE mission is the first to use new technologies to provide global snapshots of the changing spatial configuration of the ring current and plasmasphere. Imaging of auroral emissions, which are the signatures of electrodynamic coupling between a planet's magnetosphere and iono- sphere, remains the only realistic means by which to examine the global dynamics of such extended systems

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS and is an excel lent tool for pi aci ng other observations i n the global context. A vigorous development program for new grou nd- and space-based measu remeet tech n iq ues and i nstru mentation must be part of the ongoi ng effort. These include both miniaturization of instrumentation for smal I satel I ite constel rations and new remote-sens- ing instruments that provide snapshots of large-scale re- gions from space or ground observatories. The systems view requires an enhancement of ef- forts aimed at developing global theoretical models of the Sun-Earth system, including the simultaneous devel- opment of new software technologies for efficient use of parallel computing environments. Of particular impor- tance is the development of adaptive grid technologies for models that address the large range of spatial and temporal scales characteristic of the global system struc- ture and response. However, the A-l-M system is not simply multiscale, but it also requires inclusion of addi- tional physical processes in boundary layers, tran- sitioning between weak and strong magnetic field regions and coupling ionized and neutral populations across substantial gradients in density, flow velocity, energy, pitch angle, and species for example, along magnetic field lines. Single-particle drift motion, which is species-dependent, is required to resolve features of the ring-current buildup and recovery in a geomagnetic storm not contained in single-fluid MHD codes. Like- wise, parallel electric fields are an essential feature of the codes that capture auroral-arc-scale transverse struc- ture, an important part of magnetosphere-ionosphere coupling at high latitudes, which mediates the large- scale conductivity of the ionosphere, in turn affecting global MHD structure. A final example of where additional physical pro- cesses must be included in boundary regions is the decoupling of electron and ion motion, so-called Hall currents, which must be taken into account in any code that accurately computes the reconnection rate near an x-point of magnetic field line merging. Yet the global flows into and out of the merging region are well de- scribed by single-fluid MHD codes, as has been demon- strated by extensive data comparison. Hal I MH D, multifluid, and particle codes that incorporate gradient- curvature drifts and parallel electric fields are examples of computational approaches that include the subgrid physics needed to accurately describe the non- Maxwellian features of the A-l-M system. GEM's Geospace General Circulation Model efforts and the multiagency-sponsored Coordinated Community Mod- eling Center rapid prototyping activity are examples of important ongoing programs aimed at maximizing the 1 71 usefulness of and access to global modeling efforts by the scientific community. NSF's highly successful SHINE, GEM, and CEDAR programs and the recent coordination of these groups into Sun-to-Earth analysis campaigns highlight the need to focus this broad range of expertise on issues involved in coupling between and among the Sun, the solar wind, the magnetosphere, and the ionosphere/atmosphere. To facilitate collaboration between scientific communities that rarely come together and among scientists, space- weather forecasters, and the user communities, new information technologies are needed that provide seam- less virtual interactions. The ongoing Scientific Commit- tee on Solar Terrestrial Physics (SCOSTEP) Solar Terres- trial Energy Program Results, Applications, and Modeling Phase (S-RAMP) space weather campaign, which seeks to draw together all of these communities in analyzing an international set of observations, is a large-scale, ongoing experiment in the uses of virtual collaboration technologies. NSF's information technol- ogy initiatives should be utilized as much as possible to develop important collaboration technologies in sup- port of such major community analysis efforts. 3.5 TECHNOLOGIES FOR THE FUTURE DATA ASSIMILATION The many civi I fan and mi I itary users of transiono- spheric and subionospheric communication channels and navigational systems are in need of physics-based ionosphere models that are driven by data assimilation. By the end of the decade we may have access to the requisite quantity of data, but we must aggressively test and validate these tools as they become more sophisti- cated. A goal of the National Space Weather Program is to complete and test data assimilative models similar in design to those of the meteorological community. Suc- cessful prediction of ionospheric disturbances in the equatorial zone seems well within our capability in this decade. At middle latitudes, long thought to be quies- cent, recent results show that fundamental physics ques- tions remain unanswered. At high latitudes, many chal- lenges remain. With large and distributed data sets being collected by multiple agencies through various techniques and

1 72 coveri ng d iverse regions i n geospace, new i Information technologies are required to improve data access from a single coordinating site, to allow remote searching of these distributed data sets, and to simplify data assimila- tion into global models in real time and for postevent analysis. I Improvements i n meteorological weather forecast- ing have demonstrated the utility of adopting sophisti- cated data assimilation techniques. It is essential thatthe space physics and aeronomy communities learn, further develop, and implement these methods for both opera- tional and scientific use. The potential benefit of new observing systems can come to fruition only if maxi- mum use is made of the data; this, in turn, will happen only if a comprehensive data assimilation program is developed. This is not a trivial task, and the effort in- volved shou Id not be underestimated. Assimi ration mod- els can also provide the tools to visualize a wide net- work of data and provide guidance on when and where to target observations to ensure efficient use of resources. Data assimilation is the optimal combination of data with the physical understanding embedded in physical models. It is distinct from data synthesis, where a small number of observations are used to adjust a model out- put. There are numerous data assimilation methods avai fable in meteorology and oceanography that can be applied to the space environment. SPACECRAFT AND INSTRUMENT TECHNOLOGY There are several areas of technological develop- ment that are needed to implement A-l-M objectives on future missions efficiently. The needed technology splits into two areas: spacecraft subsystems and instrumenta- tion. Both areas require active funding by NASA to fos- ter improvement. It should be emphasized that the best return on NASA's investment will come from a peer- reviewed competition open to universities and industry as well NASA centers. In the case of spacecraft sub- systems, the return will be immediate and far-reaching. Improvements in telemetry, command and data han- dling (CDH), attitude control, and power systems can be directly transferred to private industry, where they will make American spacecraft more competitive in a global economy. Instrumentation development has a less direct impact, but the innovations that come from better scien- tific instruments often lead the way for incorporating new technology into spacecraft subsystems. It is im- portant to realize that developments in these areas must support traditional, highly instrumented spacecraft as THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS well as smaller, more simply instrumented spacecraft (micro- and nanosatel I ites). Future NASA space science missions will increas- ingly rely on a multispacecraft approach, as is amply discussed in this report and those of the other panels. For such missions to be achieved at reasonable cost, spacecraft subsystems must become more efficient in their use of mass and power. For every scientific space- craft built in recent memory, the majority of mass and power go to spacecraft subsystems, not instrumentation. To improve the performance of these systems, NASA will need to foster research and innovation. Develop- ment of highly power-efficient CDH systems with good flexibility and increased-capacity mass memories with small impact on spacecraft resources will be critical to these missions. NASA is currently working on high-effi- ciency thruster systems. Multispacecraft missions can require frequent station keeping to maintain optimum spacecraft positioning; more efficient thrusters can have a direct impact on spacecraft size and mission longevity. This work should be continued and expanded. Both power generation and power storage improve- ments are needed. Although solar cells have become more efficient, development of still higher efficiency solar cells should be encouraged. For planetary mis- sions to the outer solar system, radioisotope thermal generators (RTGs) are needed. The development and use of RTGs is a politically sensitive issue, but if NASA is to continue exploration of Jupiter and beyond, these power sources must be developed so that the public is satisfied that they are safe. Without safe, politically ac- ceptable RTGs, exploration of the solar system and be- yond will be significantly limited. Battery development should also be encouraged. By decreasing the mass re- quirements for a given amount of stored power, smaller, more efficient spacecraft can be constructed. Multispacecraft missions will also place significant new demands on telemetry reception capacity. Much of this reception capacity is currently contained within the Deep Space Network (DSN). However, the DSN, as pres- ently configured, consists of relatively large, expensive receiving antennas. Recent experiences with Cluster and other missions suggest that DSN is stretched very thin. The next generation of multispacecraft missions will not fly at extremely large distances from Earth; most are envisioned to be within its magnetosphere. This sug- gests that NASA should consider augmenting the DSN with arrays of smaller, less-expensive antennas that can handle missions that stay within 20-30 RE of Earth. If each of the main DSN stations (Canberra, Goldstone, Madrid) had two, three, or four smaller dishes, then the

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS upcoming multispacecraft missions could have their te- lemetry reception offloaded onto these smaller dishes, freeing the larger ones for planetary or other low-signal- level missions. As part of making such a system efficient, automation of receivers must be considered to reduce operating costs. NASA should develop options for man- aging data reception to find solutions that will give effi- cient and sufficient capacity. On the instrumentation side of new technology, the key is to provide funding to scientists to develop the instrumentation. This support must be for the develop- ment of new basic technologies and materials as well as specific instrumentation designs. As an example, basic research in magnetoresistive materials may lead to new, highly efficient magnetometers. However, we also need novel designs for electrostatic detector optics to improve the efficiency of detection of low- and mid-energy ion mass analyzers. In the current environment, instru- mentation innovation does occur, but it tends to be spo- radic, and nonuniversity groups that have internal over- head return tend to be favored, because they can internally finance modest development efforts. By pro- viding steady, regular funding for instrument develop- ment that is openly competed for and peer-reviewed, along with support for suborbital and UNEX-class flight opportu n ities, NASA wi I I al I ow researchers to cou nt on continuing funding for their efforts to come up with new detector designs that wi 11 measure more accurately and more efficiently. An important part of the development of new, com- pact systems for both instrumentation and spacecraft is the microelectronics that is at the heart of these systems. Central to microelectronics is the development of mod- ern parts with good radiation tolerance. In particular, NASA must foster the development of red-hard micro- processors, programmable gate arrays, and other digital and analog electronic components so that the United States can remain competitive with the rest of the world. 3.6 RECOMMENDATIONS Future research in atmosphere-ionosphere-magneto- sphere science must prominently support projects, theo- ries, and models that address the three-dimensional, dynamic behavior of the coupled A-l-M system. Crucial to understanding dynamic, complex geophysical phe- nomena such as magnetic reconnection, auroral pro- 1 73 cesses, and electrodynamic ionosphere-thermosphere coupling are measurements from multiple platforms (e.g., the recently launched four-satellite Cluster 11 mis- sion and the planned Magnetospheric Multiscale mis- sion). Also critical to achieving such understanding are advances in the area of numerical simulation, including the development of mature coupled ionosphere models and the incorporation in global models of proper physi- cal representations of sub-grid-scale effects. Future measurements and models must pay even greater attention to these essential aspects of near-Earth space. The overarch i ng goal s are these: 1. To understand how Earth's atmosphere couples to its ionosphere and its magnetosphere and to the atmo- sphere of the Sun and 2. To attain a predictive capability for those pro- cesses in the A-l-M system that affect human ability to live on the surface of Earth as well as in space. We currently have a tantalizing glimpse of the physi- cal processes controlling the behavior of some of the individual elements in geospace. We must now address cross-cutting science issues, which include · the instantaneous global system response of the A-l-M system to the dynamic forcing of the solar atmo- sphere. For example, how does the magnetosphere limit solar wind power input, manifest in saturation of the polar cap potential ? How do the neutral atmosphere and the ionosphere respond to sudden and long-term changes on the Sun? In view of the multiple temporal and spatial scales we must understand · the role of micro- and mesoscale processes in controlling the global-scale A-l-M system. The exchange of mass, momentum, and energy between the geophysical domains (e.g., connection of solar wind plasma at the magnetopause, ionospheric outflow, up- ward propagation of electromagnetic and mechanical energy from the lower atmosphere) is a key element in the coupled A-l-M system. It is now imperative that we understand · the degree to which the dynamic coupling be- tween the geophysical regions controls and impacts the active state of the A-l-M system.

1 74 The Sun is now recognized as one of the important factors in global change. Accordingly, we must resolve · the physical processes that may be responsible for the solar forcing of climate change. These critical science issues thread the artificial bound- aries between the discipl ines, but within each discipl ine, important science questions remain. For example, Earth's outer magnetosphere, acting as a powerful par- ticle accelerator, is often populated by a surprising de- gree of relativistic electrons that pose a radiation hazard to space-based systems. It is important that we deter- mine · the origin of the multi-MeV electrons in the outer magnetosphere and the cause of the pronounced fluctuations in their intensity. In the thermosphere and ionosphere, one of the funda- mental science issues that must be resolved is to deter- mlne · the balance between internal and external forc- ing in the generation of plasma turbulence at low lati- tudes. To accomplish our goals we note that simultaneous, multiplatform remote-sensing observations of the A-l-M system as well as in situ measurements are urgently needed in order to specify the many interconnected dy- namic, thermodynamic, and composition variables. As our understanding of the complexity of the thermo- sphere, ionosphere, and magnetosphere grows, so does the requirement to capture observations of these mul- tiple facets of the coupled media. In the next decade, NASA should give highest prior- ity to multispacecraft missions such as Magnetospheric Multiscale (MMS) (Box 3.1), Geospace Electrodynamics Connections (GEC), Magnetospheric Constellation (Mag- Con), and Living With a Star's geospace missions, which take advantage of adjustable orbit capability and the advancing technology of small spacecraft. Missions that involve large numbers of simply instrumented space- craft are needed to develop a global view of the system and should be encouraged. NSF, for its part, should sup- port extensive ground-based arrays of instrumentation to give a global, time-dependent view of this system. Ground- and space-based programs should be coordi- nated as, for example, is being done in the Thermo- sphere-lonosphere-Mesosphere Energetics and Dynam- THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS ics (TIMED)/CEDAR program to take advantage of the complementary nature of the two distinct viewpoints. NASA, NSF, DOD, and other agencies should encour- age the development of theories and models that sup- port the goal of understanding the A-l-M system from a dynamic point of view. Furthermore, these agencies should work toward the development of data analysis techniques, using modern information technology, that assimilate the multipoint data into a three-dimensional, dynamic picture of this complex system. Funding for the NASA Supporting Research and Technology (SR&T) pro- gram should be doubled to bring the proposal success rate up from 20 percent to the level found in other agen- cies. SolarTerrestrial Probe (STP) flight programs should have their own targeted postlaunch theory, modeling, and data analysis support. MAJOR NSF INITIATIVE Simultaneous, multicomponent, ground-based ob- servations of the A-l-M system are needed in order to specify the many interconnecting dynamic and thermo- dynamic variables. As our understanding of the com- plexity of the A-l-M system grows, so does the require- ment to capture observations of its multiple facets. The proposed Advanced Modular Incoherent Scatter Radar (AMISR) (Box 3.2) will provide the opportunity for coor- dinated radar-optical studies of the aurora and coordi- nated investigations of the lower thermosphere and me- sosphere, a region not well accessed by spacecraft. Initial location at Poker Flat, Alaska, will allow coordi- nation of radar with in situ rocket measurements of au- roral processes. Subsequent transfer to the deep polar cap will enable studies of polar cap convection and the mapping of processes deeper in the geomagnetic tail. 1. The National Science Foundation should extend its major observatory component by proceeding as quickly as possible with Advanced Modular Incoherent Scatter Radar (AMISR) and by developing one or more lidar- centered major facilities. Further, the NSF should begin an aggressive program to field hundreds of small auto- mated instrument clusters to allow mapping the state of the global system. Ground-based sensors have played a pivotal role in our understanding of A-l-M science and must continue to do so in the coming decade and beyond. Anchored by a state-of-the-art, phased-array scientific radar, the $60 million AMISR is a crucial element for A-l-M. A distributed array of instrument clusters would provide

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 75 The Magnetospheric Multiscale (MMS) mission is a multispacecraft Solar Terrestrial Probe to study magnetic reconnection, charged particle acceleration, and turbulence in key boundary regions of Earth's magnetosphere. These three processes which control the flow of energy, mass, and momentum within and across plasma boundaries occur throughout the universe and are fundamental to our understanding of astrophysical and solar system plasmas. Only in Earth's magnetosphere, however, are they readily accessible for sustained study through the in situ measu remeet of plasma properties and of the electric and magnetic fields that govern the behavior of the plasmas. But despite four decades of magnetospheric research, much about the operation of these fundamental processes remains unknown or poorly under- stood.This state of affairs is in large part attributable to the limitations imposed on previous studies by their dependence upon single-spacecraft measurements,which are not adequate to reveal the underlying physics of highly dynamic, highly structured space plasma processes. To overcome these limitations, MMS will employ four co-orbiting spacecraft, identically instrumented to measure electric and magnetic fields, plasmas, and energetic particles.The initial parameters of the individual spacecraft orbits will be designed so that the spacecraft will form a tetrahedron near apogee.Thus configured, the MMS ~cluster"will be able to measure three-dimensional fields and particle distributions and their temporal variations and three-dimensional spatial gradients with high resolution while dwelling in the key magnetospheric boundary regions,from the subsolar magneto- pause to the high-latitude magnetopause, and from the near tail to the distant tail. Adjustable interspacecraft separa- tions from 10 km up to a few tens of thousands of kilometers will allow the cluster to probe the microphysical aspects of reconnection, particle acceleration, and turbulence and to relate the observed microprocesses to larger-scale phenom- ena. MMS will uniquely separate spatial and temporal variations over scale lengths appropriate to the processes being studied down to the kinetic regime beyond the approximations of MHD. From the measured gradients and curls of the fields and particle distributions, spatial variations in currents, densities, velocities, pressures, and heat fluxes will be calcu- lated. In order to sample all of the magnetospheric boundary regions, MMS will employ a unique four-phase orbital strategy involving carefully sequenced changes in the local time and radial distance of apogee and, in the third phase, a change in the inclination of the orbit from 10 degrees to 90 degrees. In the first two phases, the investigation will focus on the near- Earth tail and the subsolar magnetopause (Phase 1; 12 RE apogee) and on the low-latitude magnetopause flanks and near- Earth neutral line region (Phase 2; apogee increasing from 12 to 30 RC). In Phase 3, MMS will use a lunar Gravity assist to C, , ~ - -, achieve a deep-tail orbit with apogee at 120 RE and to effect the inclination change to 90 degrees. In this phase, MMS will study plasmoid evolution and reconnection at the distant neutral line. In the final, high-inclination phase, perigee will be increased to 10 RE and apogee reduced to 40 RE on the night side, and the MMS cluster will skim the dayside magneto- oause from pole to pole, sampling reconnection sites at both low and high latitudes. The nominal MMS mission has an operational duration of 2 years.While some mission-enhancing technologies such as an interspacecraft ranging and alarm system are desirable, no new mission-enabling technologies are required for the successful accomplishment of the MMS science objectives. MMS is a mission of both exploration and understanding. Its primary thrust is to study on the mesa- and microscales the basic plasma processes that transport, accelerate, and energize plasmas in thin boundary and current layers the processes that control the structure and dynamics of the magnetosphere. With sensitive instrumentation and variable spacecraft orbits and interspacecraft spacing, MMS will integrate for the first time observations and theories over all geomagnetic scale sizes, from boundary layer processes that operate at the smallest scale lengths to the mesoscale dynamics that couple solar wind energy throughout the Earth's space environment. The major science goals of the MMS mission include an understanding of the following: Reconnection at the magnetopause at high and low latitudes, Reconnection in the magnetotail and the associated magnetotail dynamics, Plasma entry into the magnetosphere, Physics of current sheets, Substorm initiation processes, and Cross-scale coupling between micro- and mesoscale phenomena.

1 76 THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS The Advanced Modular Incoherent Scatter Radar (AMISR) is a state-of-the-art phased-array incoherent scatter radar (ISR).This highly versatile instrument will be ringed by less expensive complementary systems, typically optical in nature. The science plan is to target unsolved problems in aeronomy by placing AMISR in appropriate geographic locations for periods of 3 to 5 years.The first science goal is to understand the coupling between the neutral atmosphere and the high- speed current-carrying plasma in the auroral oval.This interaction involves electrodynamic forcing via momentum transfer from the plasma to the neutrals,Joule heating due to the currents that flow, composition changes of the thermosphere, and particle impact ionization associated with the aurora, to name just a few aspects.The first AMISR site will be in the Fairbanks area to take advantage of existing instrumentation and the Poker Flat Rocket Range. Subsequent sites will be decided on the basis of community input by a panel of research scientists. Candidates include a location in the deep polar cap,which has never been studied using the ISR technique,and a location in the off-equatorial zone to study development of the ionospheric anomaly and its severe effects on communications systems. The full AMISR will have three faces, each of which is a phased-array ISR capable of pulse-to-pulse beam swinging.The system will provide measurements of electric fields, ion and electron temperatures, electron density, ion composition, and neutral winds in the meridian plane.Three faces will allow a very wide area to be studied from a single location. Alterna- tively, the faces can be deployed separately since each is in its own right a very powerful system. A complementary set of optical- and radiowave-based sensors will accompany the deployment of the AMISR and extend its capabilities. The design of the AMISR is completed and a prototype element has been constructed and tested successfully. Once the project is approved, a first face can be constructed in about 2 years. Subsequent faces will be online in about the same time scale.The total cost is $60 million, including the associated additional instrumentation. the high temporal and spatial resolution observations needed to drive the assimilative models, which the panel hopes will parallel the weather forecasting models we now have for the lower atmosphere. Much of the neces- sary infrastructure for such a project has already been demonstrated in the prototype Suominet, a nationwide network of simple Global Positioning System/meteorol- ogy stations linked by the Internet. The proposed pro- gram would add miniaturized instruments, such as all- sky imagers, Fabry-Perot interferometers, very-low- frequency receivers, passive radars, magnetometers, and ionosondes in addition to powerful GPS-based systems in a flexible and expandable network coupled to fast, real-time processing, display, and data distribution ca- pabilities. Instrument clusters would be sited at universi- ties and high schools, providing a rich hands-on envi- ronment for students and training with instruments and analysis for the next generation of space scientists. Data and reduced products from the distributed network would be distributed freely and openly over the Internet. An overall cost of $100 million over the 1 O-year plan- ning period is indicated. Estimated costs range from $50,000 to $150,000 per station depending on the in- struments to be deployed. Adequate funding would be included for the development and implementation of data transfer, analysis, and distribution tools and facili- ties. Such a system would push the state of the art in information technology as well as instrument develop- ment and miniaturization. Extendi ng the present radar-centered upper atmo- spheric observatories to include one or more lidar-cen- tered facilities is crucial if we are to understand the boundary between the lower and upper atmosphere. Fortunately, a number of military and nonmilitary large- aperture telescopes may become avai fable for transition to lidar-based science in the next few years. Highest priority would be given to a facility at the same geo- graphic latitude as one of the existing radar sites. NASA ORBITAL PROGRAMS The Explorer program has since the beginning of the space age provided opportunities for studying the geo- space environment just as the Discovery program now provides opportunities in planetary science. The contin- ued opportunities for University-Class Explorer (UNEX) (Box 3.3), Small Explorer (SMEX), and Medium-Class Explorer (MIDEX) missions, practically defined in terms of their funding caps of $1 4 mi l l ion, $90 mi l l ion, and $180 million, allow the community the greatest creativ- ity in developing new concepts and a faster response time to new developments in both science and technol- ogy. These missions also provide a crucial training grou nd for graduate students, managers, and engi neers.

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 77 Small spacecraft missions can be extremely productive scientifically and can also provide a fertile training ground for students of science and engineering. NASA has attempted to establish several new lines of small-end missions, including the UNEX mission line in space science. Of course, the Small Explorer (SMEX) program (at a larger scale size and cost) has been a remarkable success, and the smaller sounding rocket and balloon programs have in the past been immensely rich and rewarding programs. In carrying out the Student Explorer Demonstration Initiative (STEDI) program,the Universities Space Research Association set an excellent tone for how to manage small missions. Appropriate levels and numbers of reviews were employed and key types of help were provided to STEDI teams, as needed. It has been widely acknowledged that small-spacecraft missions can provide a profound educational experience for university students. It has been from the ranks of such highly trained students that many present-day principal investiga- tors of NASA space science missions have emerged.To have a future space science program with strong experimental content, the United States must ensure that student training continues to be a high priority. This demands that small, focused spacecraft missions be available to the university research community,which in turn means that an ample number of spacecraft payloads must be made available to researchers. The NASA UNEX program was generally viewed as a direct successor to the STEDI program. However, UNEX has been effectively cut from future NASA budgets. It is regrettable that this program and the opportunities afforded by the STEDI concept will not be available to university scientists for research and educational opportunities. Moreover,the stresses that apparently continue to occur in the sounding rocket and balloon programs of NASA suggest that the suborbital program also is very limited in the access to space it gives for young scientists and engineers and as a hands-on training ground for them. See A Space Physics Paradox for further discussion.3 It would seem that NASA has identified larger-spacecraft missions as its primary focus of attention and funding.This means that very small, Pl-class spacecraft missions are not a high priority for it. NASA and other agencies could serve the university community in a most beneficial and effective way if they would offer low-cost launch possibilities to university groups. This would allow the community to revivify the UNEX program, establish appropriate small-spacecraft launch capabilities, strengthen the engineering and science education program, and fully develop this nation's small satellite program potential. In carrying out these steps, the agencies would perform an immense service for university researchers throughout the nation. At a cost of ~$20 million per mission and with launches once or so per year, the program would make very modest resource demands. NRC. 1994. A Space Physics Paradox: Why Has Increased Funding Been Accompanied by Decreased Effectiveness in the Conduct of Space Physics Research? National Academy Press,Washington, D.C. Imager for Magnetopause-to-Aurora Global Exploration (IMAGE), launched in March 2000,is an example of a highly successful MIDEX mission; it was preceded by the first two ongoing SMEX missions, Solar Anomalous and Magnetospheric Particle Explorer (SAMPEX) and FastAuroral Snapshot Explorer (FAST), launched in 1992 and 1996, which have provided enormous scientific re- turn for the investment. The Aeronomy of Ice in the Mesosphere (AIM) SMEX was recently selected for launch in 2006. The UNEX program, after the great suc- cess of the Student Nitric Oxide Explorer (SNOE), launched in February 1998, has effectively been can- celled. This least expensive component of the Explorer program plays a role similar to the sounding rocket pro- gram, with higher risk accompanying lower cost and a great increase in the number of flight opportunities. An increase in funding to $20 million per mission with one launch per year would make this program viable with modest resources. 2. The SMEX and MIDEX programs should be vigor- ously maintained and the U N EX program should quickly be revitalized. The Solar Terrestrial Probe (STP) line of missions defined in the NASA Sun-Earth Connection (SEC) Road- map (strategic planning for 2000 to 2025) has the poten- tial to form the backbone of A-l-M research in the next decade. The missions that are part of the current pro- gram includeTIMED, launched in February 2002, Solar- B, Solar Terrestrial Relations Observatory (STEREO), MMS, G KC, and MagCon. After TIMED, launched in February 2002, the next A-l-M/STP mission, MMS, is in the process of instrument selection for a 2009 launch.

1 78 The STP cadence, with one A-l-M mission per decade (TIMED was significantly delayed), has fallen behind the NASA SEC Roadmap projections. 3. The panel heartily endorses the STP line of missions and strongly encourages an increase in the launch ca- dence, with GEC and MagCon proceeding in parallel. The A-l-M research community has very success- fully utilized the infrastructure developed within the ISTP program. The integration of the data from spacecraft and ground-based programs beyond those funded by the ISTP project itself such as those of NOAA, LANE, and the DOD have contributed substantially to our under- standing of the global system. Comparisons between the Sun-Earth system and other Sun-planet or stellar-planet systems provide im- portant insights into the underlying physical and chemi- cal processes that govern A-l-M interactions. Improved understanding of A-l-M coupling phenomena such as planetary and terrestrial auroras would benefit from such an approach. 4. The Sun-Earth Connection program partnership with the NASA Solar System Exploration program should be revitalized. A dedicated planetary aeronomy mission should be pursued vigorously, and the Discovery pro- gram should remain open to A-l-M-related missions. NASA SUBORBITAL PROGRAM The NASA Suborbital program has produced out- standing science throughout its lifetime (Box 3.4~. Many phenomena have been discovered using rockets, rock- oons, and balloons, and many outstanding problems brought to closure, particularly when space-based fa- cilities are teamed with ground-based facilities. These phenomena include the auroral acceleration mecha- nism, plasma bubbles at the magnetic equator, the charged nature of polar mesospheric clouds, and mono- energetic auroral beams. This program continues to gen- erate cutting-edge science with new instruments and data rates that are more than an order of magnitude greater than typical satellite data rates. Both unique alti- tude ranges and very specific geophysical conditions are accessible only to sounding rockets and balloons, particularly in the campaign mode. Many current satel- lite experimenters were trained in the Suborbital pro- gram, and high-risk instrument development can occur only in such an environment. To accomplish significant training, it is necessary that a graduate student remain in THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS a project from start to finish and that some risk be ac- ceptable; both are very difficult in satellite projects. The high scientific return, coupled with training of future generations of space-based experimenters, makes this program h igh Iy cost-effective. The sounding rocket budget has been level-funded for over a decade, and many principal investigators are discouraged about the poor proposal success rate as well as the low number of launch opportunities. The sounding rocket program was commercialized in 2000; in this changeover, approximately 50 civil service posi- tions were lost and the cost of running the program increased. Approved campaigns were delayed by up to a year, and it is not yet clear whether the launch rate will ever return to precommercialization levels. Effectively, commercialization has meant a significant decline in funding for the sounding rocket program. An additional concern is that, as currently structured i.e., with a fixed, 3-year cycle for all phases of a sounding rocket project funding is not easily extended to allow gradu- ate students to complete their thesis work, because it is generally thought that such work should fall under the SR&T program, already oversubscribed. The rocket pro- gram has a rich history of scientific and educational benefit and provides low-cost access to space for uni- versity and other researchers. Further erosion of this pro- gram will result in fewer and fewer young scientists with experience in building flight hardware and will ulti- mately adversely affect the much more expensive satel- lite programs. 5. The Suborbital program should be revitalized and its funding should be reinstated to an inflation-adjusted value matching the funding in the early 1980s. To fur- ther ensure the vibrancy of the Suborbital Program, an independent scientific and technical panel should be formed to study how it might be changed to better serve the community and the country. SOCIETAL IMPACT PROGRAM The practical impact on society of variations in the A-l-M system falls into two broad categories: the well- established effects of space weather variations on tech- nology and the less clear yet tantalizing influence of solar variability on climate. The societal impacts of space weather are broad commun ications, navigation, human radiation hazards, power distribution, and satel- lite operations are all affected. Space weather is of in- ternational concern, and other nations are pursuing par- allel activities, which could be leveraged through

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 79 NASAL Suborbital program provides regular, inexpensive access to near-Earth space for a broad range of space science disciplines, including space plasma physics, astronomy, and microgravity. The program has been extremely successful throughout its history, consistently providing high scientific return for the modest funding invested. Phenomena from auroral physics to supernovae have been investigated with sounding rocket and balloon experiments, and many key discoveries in these fields have come from this program. For the science of lower-altitude regions such as equatorial ionospheric irregularities and mesospheric physics, this program provides the only access because these regions are too high for airplanes and too low for satellites. Moreover, the use of new, advanced instrumentation concepts coupled with data rates that exceed those of satellites by a factor of more than 10 have provided measurements of a resolution and quality that are simply unobtainable elsewhere.This science is exciting and central to NASAL mission. The program also has a spectacular track record in training future scientists and engineers. More than 300 Ph.D.theses have been based on rocket data alone.This has provided a regular flow of technically adept individuals who have first- hand experience in building space flight instrumentation. Indeed, most of today's successful satellite experimenters were trained in this program.The experience that students receive developing, flying, and interpreting data from instruments they themselves have built is only possible within this program.This is an unparalleled learning experience that satellite programs cannot match because they are too risk-averse and span too long a period for a graduate student to be involved from start to finish. Despite this tremendous track record in which the Suborbital program has continually demonstrated its scientific validity (made clear through a series of reviews over the past decade), funding for the program has seriously eroded. Its conversion to a government-owned, contractor-operated program, coupled with loss of many civil servant positions, has left the program severely underfunded for operations. Additionally, funding for scientific investigations has remained stagnant, resulting in a significant decline in the number of funded investigations over the past 15 years.There is -treat .~ , .~ .~ .~ , , .~ . . . . . . . . . . . . . . concern among the scientific community that NAbA management does not deem the program sutticlently Important to restore and protect its funding.This attitude must be changed and the program must be restored to a healthy level that will allow it to continue to play its important scientific and student training roles in which it is so uniquely effective.SeeA Space Physics Paradox for further discussion.3 NRC. 1994. A Space Physics Paradox: Why Has Increased Funding Been Accompanied by Decreased Effectiveness in the Conduct of Space Physics Research? National Academy Press,Washington, D.C. collaboration. The role of solar variability in climate change remains an enigma, but it is now at least being recogn ized as i mportant to ou r u nderstand i ng of the natural as opposed to anthropogenic sources of cli- mate variabi I ity. 6. To maximize the societal impact of studies and knowledge of the A-l-M system, the study of solar vari- ability both of its short-term effects on the space ra- diation environment, communications, navigation, and power distribution and of its effect on climate and the upper atmosphere should be intensified by both mod- eling and experimentation. NASA's Living With a Star program, as defined by the Science Architecture Team report, should be imple- mented, with increased resources for the geospace com- ponent. The share of resources required for the Solar Dynamics Observatory, already defined before the start of LOOS, has resulted in an unbalanced portfolio. Missions such as the National Polar-orbiting Opera- tional Environmental Satellite Systems (NPOESS) and Solar Radiation and Climate Experiment (SO RCE) should provide vital scientific data for monitoring long-term solar irradiance, and NPOESS should provide iono- sphere and upper atmosphere observations to fill gaps in measurements needed to understand the A-l-M system. An L1 monitor should be a permanent facility, to provide solar wind measurements crucial to determin- ing the response of the A-l-M system to its external driver. The National Space Weather Program should be strengthened and used as a template for interagency cooperation. International participation in such large scope programs as LWS and NSWP is essential.

1 80 NASA's new Living With a Star program can, over the next decade, provide substantial new resources to address these goals. It is crucial that there be overlap between the geospace and solar mission components of LWS for the A-l-M system to be studied synergistical Iy, that resources be adequate for the geospace compo- nent, and that theory, modeling, and the comprehensive data system that will replace the ISTP infrastructure be defined at the outset, as called for in the Science Archi- tecture Team report.2 The National Space Weather Pro- gram, a multiagency endeavor establ ished in 1 995, ad- dresses the potentially great societal impact of the physical processes from the Sun to Earth that affect the near-Earth environment in ways as diverse as terrestrial weather. The program specifically addresses the need to transition scientific research into operations and to assist users affected by the space environment. Such multiagency cooperation is essential for progress in pre- dicting response of the near-Earth space environment to short-term solar variabi I ity. Several potential mechanisms for a solar variability- climate connection have been suggested: (1 ) changes in the Sun's total irradiance or luminosity, which is the basic driver of the climate system; (2) changes in spec- tral irradiance, particularly in the UV, which drives the chemistry and dynamics of the middle atmosphere and has been shown by modeling studies to influence the dynamics of the troposphere; and (3) the possible influ- ence of cosmic-ray and electric-field variations on cloud nucleation, which could significantly modify Earth's ra- diation balance. 7. The NOAA, DOE/LAN L, and DOD operational spacecraft programs should be sustained and DOD launch opportunities should be utilized for specialized missions such as geostationary airglow imagers, auroral oval imagers, and neutral/ionized medium sensors. The interagency cooperation established in the NSWP is outstanding and is a model for extracting the maximum benefit from scientific and technical pro- grams. It has also been effective at bringing together different scientific disciplines and the scientific and op- erations communities. Interagency cooperation has 2NASA, Living With a Star, Science Architecture Team. 2001 . Report to the Sun-Earth Connection Advisory Subcommittee, August. Avail- able on I i ne at <http://lws.gsfc.nasa.gov/docs/lws_sat/sat_report2.pdf>. THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS worked well in the AFOSR/NSF Maui Mesosphere and Lower Thermosphere Program, and it has been key to the success of the NOAA GOES and NPOESS programs of meteorlogical satellites with space environment moni- toring capabilities. International multiagency coopera- tion has been very successful for the ISTP program, which involves U.S., European, Japanese, and Russian space agencies. Global studies require such inter- national cooperation. The panel recognizes that much more science can be extracted by careful coordination of ground- and space-based programs. MAXIMIZING SCIENTIFIC RETURN Funding for NASA Supporting Research and Tech- nology, including guest investigator studies and focused theory, modeling, and data assimilation efforts, is essen- tial for maximizing the scientific return from large in- vestments in spacecraft hardware. Supporting Research and Technology While spacecraft hardware projects are concen- trated at relatively few institutions, the NASA SR&T program is the primary vehicle by which independent investigations can be undertaken by the broader com- munity. Likewise, NSF helps individual investigators to carry out targeted research through its Division of Atmo- spheric Sciences base programs SH I N E, CEDAR, and GEM. Such individual Pl-driven initiatives are the most inclusive, with data analysis as wel I as theoretical efforts and laboratory studies, and often lead to the highest science return per dollar spent. The funding for such program elements falls far short of the scientific oppor- tunities, with the current success rate for submitted NASA SR&T proposals being 10 to 20 percent. Further- more, limited available SR&T funds have been used for guest investigator participation in underfunded STP-class flight programs. Without adequate MO&DA funding for NASA orbital and suborbital programs, the SR&T budget intended for targeted research on focused scientific ques- tions has been utilized to support broader data analysis objectives. 8. The funding for the SR&T program should be in- creased, and STP-class flight programs should have their own targeted postlaunch data analysis support. 9. A new small grants program should be established within NSF that is dedicated to comparative atmo- spheres, ionospheres, and magnetospheres (C-A-I-M).

PANEL ON ATMOSPHERE-IONOSPHERE-MAGNETOSPHERE INTERACTIONS 1 81 The comparison of the Sun-Earth system to other Sun-planet systems can provide unique insights into how atmo- spheres, ionospheres, and magnetospheres (A-l-M) respond to solar inputs.The physical and chemical processes control- ling these responses manifest themselves very differently at each solar system body,yet are essentially the same at a basic level.AII the techniques that have been used so successfully to understand solar-terrestrial physics (e.g., modeling,ground- based and remote observations, and in situ measurements) need to be applied to other planets and bodies, so that the study of solar-planetary relations becomes the natural extension of the terrestrial space weather effort.Achieving this goal will require several elements, including NASA planetary missions dedicated to A-l-M goals (or including significant A-l-M capabilities),a Discovery program in which A-l-M missions are included,and a grants program within NSF that is dedicated to comparative atmospheres, ionospheres, and magnetospheres (C-A-I-M). A new C-A-I-M grants program at NSF would play a key role in addressing the interdisciplinary issues needed to understand and relate A-l-M processes throughout our solar system, or even at other stellar systems. Such a grants pro- gram would provide much needed resources for analysis of both past and future data sets (from ground- or space-based observatories, or from in situ missions), modeling and data interpretation related to A-l-M objectives, telescope time, special meetings devoted to terrestrial-planetary issues, and the nonmission research support needed to encourage C-A-I-M science activities in the community. Such a grants program would need about $5 million per year in order to adequately develop and explore the linkages between the terrestrial and planetary manifestations of atmosphere- ionosphere-magnetosphere physics. A new C-A-I-M grants program at NSF (Box 3.5) such programs as NASAls ISTP and its Sun-Earth Con- would allow the techniques that have been applied so nections Theory Program, the AFOSR/ONR Multidisci- successfully to A-l-M processes at Earth (modeling, plinaryUniversityResearch Initative program, NSFSci- ground- and space-based observations, and in situ mea- ' - ' ' ^ ' ' surements) to be used to understand A-l-M processes at other planets. Such a comparative approach would im- prove understanding of these processes throughout the sol ar system, i ncl ud i ng at Earth. Presently, a modest $2 million Planetary Science program at NSF covers all of solar system science (except for solar and terrestrial stud- ies), with only a small fraction going to planetary A-l-M research. Theory, Modeling, and Data Assimilation Theory and modeling provide the framework for in- terpreti ng, understand) ng, and visual izi ng diverse mea- surements at disparate locations in the A-l-M system. There is now a pressing need to develop and utilize data assimi ration techniques not only for operational use in specifying and forecasting the space environment but also to provide the tools to tackle key science questions. The modest level of support from the NSF base pro- grams (CEDAR, GEM, SHINE) and NASA SR&T has been inadequate to build comprehensive, systems-level mod- els. Rather, individual pieces have been built and first stages of model integration achieved with funding from , , , ence and Technology Center programs, and the multiagency support for such efforts as the Community Coordinated Modeling Center. Such programs enable the development of theory and model ing infrastructure, including models to address the dynamic coupling be- tween neighboring geophysical regions. Their value to the research community is clearly their provision of longer-term funding, which has been essential to devel- oping a comprehensive program, outside the purview of SR&T. 10. The development and utilization of data assimila- tion techniques should be enhanced to optimize model and data resources. The panel endorses support for theory and model development at the level of the NASA Sun-Earth Connections Theory Program, the AFOSR/ ONR MURI program, NSF Science and Technology Center programs, and the multiagency support for such efforts as the Community Coordinated Modeling Center (CCMC). Support should be enhanced for large-scope, integrative modeling that applies to the coupling of neighboring geophysical regions and physical pro- cesses, which are explicit in one model and implicit on the larger scale.

182 The preceding science recommendations are grouped into three cost categories and prioritized in Table 3.1. Equal weight is given to STP and LWS lines, as indicated by funding level. Small programs are ranked by resource allocation, while the Advanced Modular Incoherent Scatter Radar is the highest-priority moder- ate initiative at lower cost than others. BIBLIOGRAPHY GEM documents. Available online at <http://www- ssc.igpp.ucla/gem/\A/elcome.html>. National Aeronautics and Space Administration (NASA), Office of Space Sciences. 2000. Strategic P/an. NASA, Wash i ngton, D.C. NASA. 2000. Sun-Earth Connection Roadmap, Strategic Planning for 2000-2020. NASA, Washington, D.C. NASA, Living With a Star, Science Architecture Team. 2001. Report to the Sun-Earth Connection Advisory THE SUN TO THE EARTH AND BEYOND: PANEL REPORTS Subcommittee, August. Available online at <http:// Iws.gsfc. nasa.gov/docs/lws_sat/sat_report2.pdf> National Research Council (NRC). 2000. Radiation and the Internationa/ Space Station: Recommendations to Reduce Risk. National Academy Press, Washington, D.C. . NRC.1994. A Space Physics Paradox: Why Has Increased Funding Been Accompanied by Decreased Effectiveness in the Conduct of Space Physics Research? National Academy Press, Washington, D.C. National Science Foundation (NSF). 1995. National Space Weather Program: The Strategic P/an, FCM- P30-1995. Office of the Federal Coord i nator for Meteorology, Silver Spring, Md. NSF.1997. National Space Weather Program: The Implementation P/an, FCM-P31 -1997. Office of the Federal Coordinator for Meteorology, Silver Spring, Md. NSF. 1997. CEDAR Phase 111 Document. Available onl ine at <http://cedarweb.hao.ucar.edu/ index.html>.