Building Collaboratories in Space Physics
Space physicists study the interaction of charged particles with electric and magnetic fields in the space environment. The interactions are complex because the motions and distributions of the charged particles are determined by the fields, but at the same time, those distributions and motions affect or determine the electric and magnetic fields. Space physics research has many aspects: (1) fundamental physics, in which the goal of the research is to understand the types of interactions that occur; (2) a phenomenological component aimed at describing and understanding the distribution and behavior of plasmas in settings such as Earth's magnetosphere; and (3) attention to applications, for example, attempts to predict the occurrence and nature of disturbances in Earth's plasma environment in order to prevent interference with or damage to systems such as power grids or Earth-orbiting satellites.
With advances in technology has come an increase in the magnetosphere's influence on human activities. Our communications, weather, surveillance, and test ban verification satellites are stationed deep in an active region of the magnetosphere. Communications at many wavelengths are affected by ionospheric conditions, which in turn are affected by magnetospheric conditions. Every solar cycle since the 1930s has brought electrical power system disruptions caused by disturbances in the magnetosphere, a relationship acknowledged only over the most recent solar cycles. Pipeline corrosion is accelerated by currents induced by rapid magnetic changes. Magnetic prospecting and other activities may be helped or hindered, depending on magnetospheric activity at the time.
Understanding the cause and the nature of magnetospheric disturbances is thus an important endeavor supported by the National Science Foundation and the National Aeronautics and Space Administration, as well as the National Oceanic and Atmospheric Administration, Department of Defense, and Department of Energy, and requiring both ground- and space-based observations of electric and magnetic fields, waves, plasmas, and energetic particles. Interpreting the information gathered generally involves joint collaborative analysis of two or more different data sets. Thus collaborative studies are frequent in space physics, if not generally imperative.
This chapter briefly outlines the types of data collected and the instrumentation, methodology, and techniques of analysis used in the field of space physics; reviews some of the initial collaborative efforts in space physics research and describes additional ongoing collaborative programs; and suggests how a national collaboratory for space physics might benefit researchers and, in turn, what components scientists would regard as basic to a useful collaboratory.
SPACE PHYSICS RESEARCH
Data Collection and Instrumentation
The earliest recorded observations of phenomena now studied by space physicists were sightings of the aurora. Later, when the compass became a precise tool for navigation, short-period (seconds to hours) fluctuations in Earth's magnetic field were recorded and the existence of an ionosphere was
inferred. Early in the 1900s, over-the-horizon radio transmissions proved the ionosphere's existence. The alignment of auroral forms with Earth's magnetic field and the correlation of auroral activity with solar activity deduced in the late 19th century were the first real indications of the existence of a magnetosphere and its control, in some unknown manner, by the Sun. In 1930 Sidney Chapman and A.C. Ferraro developed the first real model of the magnetosphere. They postulated that a plasma (which we now called the solar wind) was emitted intermittently to cause a compression of Earth's magnetic field and a resultant magnetic cavity in the flowing solar wind (which we now call the magnetosphere). By the mid-1950s the plasma density of the magnetosphere had been probed remotely using lightning-generated, very-low-frequency waves. Nevertheless, the beginning of space physics as a discipline was marked by the discovery of the Van Allen radiation belt and the in situ exploration of Earth's magnetosphere and its interaction with the solar wind made possible with the advent of rockets and satellites in the late 1950s and 1960s.
The data gathered in space-based observations are collected by sets of 3 to perhaps 10 different sensors positioned on Earth-orbiting satellites or interplanetary probes. A typical satellite might carry a magnetometer to detect slowly changing magnetic fields and a separate sensor to measure magnetic oscillations; a device to record electric fields and waves; plasma analyzers (often a coordinated set of sensors) to measure the fluxes of charged particles as functions of their mass, energy, and direction of motion; and one or more sensors to measure high-energy charged particles. The data consist of time sequences of the sensors' outputs. When possible, similar or complementary data are collected by other satellites in different locations in space.
In addition, data gathered simultaneously from ground-based instruments are used to examine phenomena such as disturbances in the geomagnetic field (as detected by arrays of ground-based magnetometers), changes in the ionospheric density (as indicated by radar, riometer, and rocket-based observations), enhancements in atmospheric emissions signifying excitation by energetic particles (as shown in images from photometers and all-sky cameras), and activity on the Sun (as shown by, e.g., coronagraphs). Data on particles and fields are now often augmented with images of Earth's atmosphere in the visible, ultraviolet, or x-ray regions of the electromagnetic spectrum, thus enabling space physicists to relate observed high-altitude phenomena to ground-based observations.
The data matrices produced by space- and ground-based instruments thus include many different kinds of measurements often taken at widely separated sites, but often with good (but differing) time resolution. A challenge in analyzing and interpreting the data is to combine and compare them so as to deduce a global picture of the behavior of the magnetospheric system. Combining and analyzing these various data sets and types both require extensive electronic communications, including electronic mail and network transfer of data and text files, between the home institutions (sometimes international) of the many investigators involved.
Methods and Technologies for Data Analysis
Space plasma physicists use a combination of analytical and numerical methods to interpret and understand data, as well as to assist in the planning of observational campaigns. These methods can range from the use of simple linearized equations to model the early time evolution of plasma waves to the use of extensive multidimensional codes that attempt to simulate the full complexity of these nonlinear systems. The simulations can consist of calculations of the locations and motions of many millions of individual charged particles, together with the electric and magnetic fields that they generate, or they can be solutions of simultaneous partial differential equations that describe the plasma as a ''fluid.'' Running a typical simulation requires the use of a supercomputer, although some of the newer workstations can now run some of the smaller codes. The amount of numerical "data" generated is so great that if one were to attempt to keep it all, it would far exceed the capacity of even the largest computers. Even after pruning, the files that are kept necessitate good (reliable and fast) network communications between the
supercomputer and the researcher's home site, so that the "data" can be transferred between the two locations for detailed analysis, manipulation, or graphical display.
Simulations are now becoming a common part of space physics projects because the models are becoming sufficiently realistic to warrant direct comparisons with the measurements. Indeed, one of the greatest challenges in a number of current programs is the synthesis of models and measurements through coordinated displays and analyses. Thus, manipulation and display of diverse data sets from both numerical modeling and a wide range of experiments are a focus for computational and telecommunications activities in space physics.
EXAMPLES OF COLLABORATIVE EFFORTS IN SPACE PHYSICS RESEARCH
Of the initial collaborative efforts in space physics research, four are examined below: (1) the Space Physics Analysis Network, developed to satisfy the need of researchers to be in close and rapid electronic contact with collaborating scientists; (2) Coordinated Data Analysis Workshops, a response to the need to synthesize extensive data sets into succinct scientific results; (3) the Active Magnetospheric Particle Tracer Explorer/Charge Composition Explorer, a scientific mission with a strong, centralized facility for data analysis; and (4) the Sondre Stromfjord Observatory testbed, a new effort that, when complete, will allow scientists to operate their instruments in Greenland remotely from sites in the United States. Also discussed are the Solar-Terrestrial Energy Program, NSF's Geospace Environment Modeling Program, and the International Solar-Terrestrial Physics program, which provide additional evidence of the increasingly important role of collaboration in the space physics community.
The Space Physics Data System, envisioned as an aid for individual researchers and a first step toward building a broad-based system for handling space physics data, is outlined as an approach whose potential utility has been acknowledged within the community of researchers.
Initial and Ongoing Collaborative Programs
Space Physics Analysis Network
In September 1980 members of the space plasma physics science community from more than three dozen institutions met to discuss what steps were needed to provide a more coordinated approach to solving many data access and analysis problems. There was then, and still remains today, a strong need to better utilize existing and diverse space physics databases through collaborations with remotely distributed space scientists. This initial working group recognized that the technology existed to interconnect distributed computer systems that would provide the "enabling environment" for remotely accessing databases at a low cost (Greenstadt and Green, 1981). It was quickly realized that a computer-to-computer communication system could be achieved (which satisfied many of the desired objectives even though significant funding was not available) by making maximum use of existing computers, equipment, and facilities at the remotely distributed sites. Within a year after that initial meeting, the Space Physics Analysis Network (SPAN) became operational with three nodes at locations in Alabama, Texas, and Utah. Almost immediately, many scientists became involved in previously impossible collaborative activities such as simply comparing data on the same time scale (Green et al., 1983; Rees et al., 1986; Sanderson, 1990; Thomas and Green, 1988a,b; and Thomas et al., 1987).
Within 8 years, SPAN grew rapidly to include several thousand computer nodes in the United States and was extended to Japan and many countries in Europe and South America. The network relied explicitly and to an unprecedented degree on the active involvement of its users.
A scientific oversight group, the Data Systems Users Working Group, guided the entire effort, meeting approximately every 9 months (see, for example, Baker et al., 1984). The actual operation of
the network required extensive volunteer labor from the institutions on the network, which were also represented in the users working group. This approach linked the users and developers in a tightly coupled feedback loop and enabled the network to meet many of the users' highest-priority needs, greatly enhancing many NASA and non-NASA space science programs (Green and King, 1986; Thomas and Green, 1987; and Winterhalter, 1986).
SPAN succeeded far beyond the dreams of its initial developers. Its success was recognized both by participants and by those responsible at NASA for electronic communication. Eventually, the SPAN effort was given a more permanent home as part of NASA's contribution to and participation in the Internet. The new NASA Science Internet currently preserves the original SPAN functionality and has added TCP/IP connectivity.
Coordinated Data Analysis Workshops
The concept of the Coordinated Data Analysis Workshop (CDAW) arose following the highly successful data-gathering phase of the International Magnetospheric Study from 1976 to 1978. The CDAW's purpose was to bring together all available magnetospheric data for specific periods of time to determine how a particular magnetospheric process worked. Nine CDAWs have been held to date. Recent workshops have focused on global-scale solar-terrestrial physics problems requiring diverse data sets and a variety of modeling skills. CDAWs in general have involved a broad cross-section of the space physics community (especially the solar wind-magnetosphere-ionosphere community). The initial phase of a CDAW effort is selection of the problem to be addressed; relevant data are then identified and collected from multiple (as many as 10) spacecraft and from scores of ground-based facilities. These data are then sent to the National Space Science Data Center (NSSDC) at Goddard Space Flight Center to be placed into a common data format. The resulting databases have often included literally hundreds of individual data sets for selected analysis intervals. Those who contributed data are then invited to gather at a common site and jointly analyze aspects of the problem at hand.
Early CDAWs were "paper" workshops—participants simply brought data records plotted on a common time scale for comparison with other data. The more recent workshops have utilized a common database accessed by interactive computer systems used during the workshops, held normally at the NSSDC but also at other locations such as Stanford University and Toyokawa, Japan. The face-to-face workshops, themselves a key element of the overall CDAW process, have had as few as a dozen participants (in splinter CDAW meetings) or as many as 100 participants in the major CDAW meetings. After the workshops, the participants return to their home institutions to prepare joint oral and written papers on their results. Recently this process has been aided by the dissemination of the CDAW databases on CD-ROM.
A major aspect of CDAWs has been the cooperative sharing of expertise. Such sharing has been facilitated by the face-to-face workshop format, which allows scientists who sometimes have very different views of the physics involved to discuss alternative interpretations of the same data sets. Perhaps an even more significant aspect of CDAWs has been the open sharing of data (prior to publication) obtained from many different instruments. The sharing aspect of CDAWs has been greatly aided by the adoption of "rules of the road" as given in Appendix C.
One of the major difficulties associated with CDAWs is the large overhead cost involved in assembling the infrastructure needed for the collaboration, including the assembling of the data and the coordinating of meetings and people. Thus far fewer workshops have been held than the space physics community would find most beneficial.
Active Magnetospheric Particle Tracer Explorer/Charge Composition Explorer
In the fall of 1984, the Active Magnetospheric Particle Tracer Explorer/Charge Composition Explorer (AMPTE/CCE) spacecraft was launched into Earth orbit in the fall of 1984 as part of a three-spacecraft Explorer mission to study plasma interactions in naturally occurring plasmas and in those created in space through the release of chemicals. It carried five scientific instruments that made measurements of the electric and magnetic fields and plasmas in Earth's magnetosphere. The CCE was designed to measure the ions of barium and lithium that entered the magnetosphere from releases in the solar wind and the magnetotail. This unique experiment indicated that barium was being released from a large portion of Earth's surface.
The complementarity of the spacecraft experiments (the instruments each gave quite distinct measurements), the relatively small size of the science team (˜ 20 to 30 scientists), and the management of the science mission from a single institution provided an important opportunity to develop a centrally located but remotely accessed data analysis system that fostered scientific collaboration. This central facility produced raw data from spacecraft telemetry, provided these data and information about spacecraft position to all remote institutions, and processed and distributed survey data products. Remote institutions accessed the data through dedicated telephone lines and SPAN, allowing individual scientists to further process data both at the central facility and at their home institutions. Thus, those with access to the database could work with others without incurring the overhead costs involved in contacting many institutions.
This mission was successful because it provided almost effortless access to a reasonably well funded, well-managed central data facility, as well as access to a relatively small number of dedicated scientists who understood individual instrument performance and were interested in scientific collaboration, and the freedom to analyze data remotely using tools available at home institutions rather than depend solely on the standard analysis routines provided by the central facility.1
Sondre Stromfjord Observatory Testbed
Several existing ground-based observatories, such as the Sondre Stromfjord Observatory in Greenland and the numerous stations in Antarctica used to study upper-atmosphere and space physics, offer only limited access because they are located in remote regions of the world. The space physics community and other researchers would benefit enormously if these instruments could be operated remotely via computer networks, thus improving access to real-time data.
The Sondre Stromfjord Observatory in Greenland is currently being used as a testbed for enabling collaborative research. The facility has many attributes that make it an excellent choice for a collaboratory project: it is remote, the user community is distributed and manageable in size, and it already has a modest networking infrastructure in place.
The real-time viewing of data obtained at the Sondre Stromfjord Observatory will allow scientists to perform experiments that previously required on-site decision making, allowing those not present at the remote observatory to respond, for example, to rarely occurring geophysical phenomena such as solar proton events, which affect ozone depletion and represent a hazard to aircraft flying on transpolar routes. Also, the use of computer networks will enable more experimenters to be involved simultaneously in research at the remote facility, thus enhancing decision making and productivity. The possibility of remotely adapting preplanned experiments to existing geophysical conditions is also an exciting prospect.
Solar-Terrestrial Energy Program
The worldwide community of solar-terrestrial scientists has embarked on an exciting and intellectually rewarding project: to understand quantitatively the linkages from the Sun through the interplanetary medium and into the depths of the surrounding geospace. The variety and complexity of the physical processes involved in these linkages have challenged our ability to understand the total system. Now, through a concerted global effort, the Solar-Terrestrial Energy Program (STEP) has begun to use remarkable new observational tools and modeling capabilities to achieve an unprecedented comprehension of our solar-terrestrial system. STEP was approved by the Scientific Committee of Solar-Terrestrial Physics in 1986 and launched in 1990; the International Council of Scientific Unions gave its formal endorsement in 1987 and recently extended the program through 1997.
The main scientific goal of STEP is to advance the quantitative understanding of the coupling mechanisms responsible for the transfer of energy and mass from one region of the solar-terrestrial system to another; the main practical goal is to improve the ability to predict the effects of the variable components of solar energy and mass flows on the terrestrial environment, on technological systems in space and on Earth, and on the biosphere.
A well-coordinated ground- and space-based observing program is essential to accomplish these goals. Basic in situ measurements will be obtained by the various spacecraft missions approved by the Inter-Agency Consultative Group as a cooperative project of the world's four major space agencies. In parallel with these efforts, STEP will coordinate the use of specially designed ground-based instruments and aircraft, balloon, and rocket experiments and will promote theory development, modeling, and simulation studies on an international scale.
Crucial to the success of STEP are the dedicated information and data systems that allow scientists from all participating countries to improve communications among themselves and facilitate the coordination and standardization of measurements, as well as the interchange and analysis of data. These systems include those operating at satellite situation centers and coordinated data-handling facilities, at new ground-based observation situation centers and computer simulation centers, and during coordinated data analysis workshops. These facilities should provide sufficient support at present for the active programs, but the problem of accessing and studying data from older missions remains. Toward the end of the STEP mission, when the data acquisition phase is complete, this problem will apply also to currently active programs.
Geospace Environment Modeling Program
Geospace encompasses the regions from Earth's upper atmosphere to the Sun. The Geospace Environment Modeling (GEM) program at NSF is an effort to study the near-Earth portion of geospace ranging from the lower ionosphere to the environment where Earth interacts with the solar wind. This space plasma environment, called the magnetosphere, is an electrodynamic system that links Earth's atmosphere with the local astrophysical system; thus magnetospheric physics as a discipline lies at the intersection of Earth-system science and modern astronomy.
The purpose of GEM is to enable research, especially collaborative research, on the dynamical and structural properties of geospace that will lead to the construction of a geospace general circulation model with predictive capability. GEM pursues this goal by supporting ground- and space-based observational research, as well as theoretical research, in modeling of the geospace environment. A geospace general circulation model is analogous to general circulation models of the lower atmosphere and will require extensive interdisciplinary and intradisciplinary research.
The strategy for achieving the GEM goal is to undertake a series of campaigns involving theoretical and observational research focusing on particular aspects of the geospace environment such as the magnetotail and substorms, global plasma models, cusp signatures, and electrodynamic coupling.
The first phase of GEM, which is now in progress, focuses on the magnetospheric cusp and boundary layers. The next phase of the program, scheduled to begin in 1994, focuses on the magnetotail and substorms. For each campaign, working groups of scientists are formed to analyze the data and develop the computer-based visualization and simulation programs needed to study given phenomena. These programs will be developed in a modular fashion with the intention of building a flexible yet robust research model of near-Earth geospace—the most highly coupled of geospheres—that will be available to other members of the scientific community.
International Solar-Terrestrial Physics Program
The International Solar-Terrestrial Physics (ISTP) program is the major research effort in space plasma physics for the decade of the 1990s. It began in the early 1980s as a NASA four-spacecraft mission called Origins of Plasma in Earth's Neighborhood (OPEN); the goal was to monitor the flow of mass, momentum, and energy from the solar wind through the magnetosphere into the upper terrestrial atmosphere. Since that original OPEN concept, the program has evolved in many ways. Japan took over responsibility for one of the spacecraft (GEOTAIL), and the European Space Agency has put two of its satellite programs (SOHO and CLUSTER) into the overall ISTP mission set. Moreover, Russia is also planning to coordinate some of the former-Soviet Union's independently planned satellite missions, such as INTERBALL, with ISTP, thus leading to well over a dozen highly instrumented spacecraft that will contribute to ISTP. This international collaboration is being coordinated by the four major space agencies through the Inter-Agency Consultative Group.
ISTP is unique in the degree to which ground-based measurements and theory will play a central role. Furthermore, the observations most critical for collaborative studies, whether collected by space-or ground-based instruments, will be placed into a "key parameter" database that will be accessible almost instantaneously to all participating scientists, who will share a huge, common database of high-quality measurements, analysis and display tools, and special models. The ISTP program will thus provide what will be tantamount to a continuing Coordinated Data Analysis Workshop environment for analysis of space physics data.
ISTP will continue throughout the 1990s as new spacecraft are launched. However, in the period from 1993 to 1996 there will be unique opportunities to assemble global geospace data sets in conjunction with particularly intense data collection campaigns and large-scale analysis efforts. The various world space agencies are planning now, through a series of workshops, the optimum ways to acquire, analyze, and disseminate the ISTP data. It is expected that the data systems now being set in place will suffice to facilitate the required collaborative studies during the active phase of the ISTP program. However, it is not certain how the infrastructure support needed in the post-project period will be provided.
Space Physics Data System: a Collaboratory of One
Novel ideas for and approaches to scientific investigation in space physics, as in most disciplines, generally start with the efforts of a single, highly motivated individual, who then, when appropriate, enlists and encourages the participation of his/her peers in the investigation. In space physics, this participation—at both the individual-and multiple-investigator level—generally takes the form of amassing and analyzing data from many different instruments carried on one or more spacecraft.
Certain boundary conditions are thus demanded of a system that supports space physics data analysis. The first set of requirements exists at the individual-scientist level and can be called a "collaboratory of one." Before individuals can participate in a larger collaboratory effort, they must first have their own well-developed ideas that they can communicate to their peers. Thus individual researchers must have ready access to multiple data sources and computational platforms whose use
should not require labor-intensive programming, extensive and arcane methods for locating and transferring data, or detailed knowledge of how the data were initially acquired. An ideal collaboratory of one, the so-called Space Physics Data System, should allow a space physicist to:
Search a global mega-database to locate all data available;
Transparently transfer the actual data without knowing the details of the transport methodology;
Convert the data to scientific units, without having to program, by using a set of generic software tools tied to a standardized, self-documenting file system;
Display and manipulate the multiple data sets via a set of standardized, X-windows-based tools either locally or via network; and
Easily generate value-added analysis and display software tools in a standardized paradigm that allows further use of the derived data sets.
A collaboratory of one that enables such actions would allow individual investigators to use their limited resources for actual scientific analysis as opposed to dealing with "computerese."
Such a Space Physics Data System would facilitate the next step to a joint, multiperson collaboratory requiring a system to accommodate the sharing of derived products and ideas. It is well within the capabilities of current workstation and network technology. In the future, as multimedia technology becomes widespread, sharing of a richer form of ideas and analysis will become possible through the sharing of graphics, text, video, and sound. Thus the collaboratory of one is a necessary first step in the process of developing a global Space Physics Data System involving multiple sources of data and their use at geographically distributed sites. A movement toward such a system has begun within the community with active encouragement from NASA, but to date these efforts have not reached even the pilot project stage.
NEW OPPORTUNITIES PROVIDED BY A COLLABORATORY FOR SPACE PHYSICS
A collaboratory for space physics would provide researchers with a number of opportunities to make new and better use of data acquired by space-based, suborbital, and ground-based experiments. Currently funded investigators could use on-line directories, user-friendly access interfaces, and network file transfer to obtain correlative data on the state of the Sun, solar wind, or ground geomagnetic indices more simply than can be done today. Associated data from various instruments on the same spacecraft could be obtained through a project data request and dissemination system interfacing with the collaboratory network.
A more important use of a collaboratory would be for group studies of global problems. By definition, global studies require data from many instruments and many locations. Such data are invariably of many different types and formats and are difficult for any one individual to assemble and display. A collaboratory could provide access to facilities, support staff, computer hardware, and specialized software to organize and document data so that they could be quickly accessed and displayed by study participants. Case history studies of complex phenomena such as magnetic storms or magnetospheric substorms are an example of potential group efforts. The assimilation of multiple data sets into self-consistent maps of various parameters is another example. In magnetospheric physics the Assimilative Mapping of Ionospheric Electrodynamics model does mapping using the equations that govern ionospheric currents, field-aligned currents, electrical conductivity, and electric fields.
A collaboratory could also be used as a tool in planning multi-instrument, multi-platform campaigns. Specialized tools such as geographic information systems, software for predicting when spacecraft will be at key observing locations in the magnetosphere, and empirical magnetic field models could be brought together in a common computer system. These could be combined with the best available empirical or theoretical models of a phenomenon to predict when and where certain types of observations could be made. The use of particular instruments could be planned, telemetry tracking organized, and specialized ground measurements scheduled.
A similar application would be the capability to provide for a quick response to unexpected or transient phenomena. For example, the development of a particularly active region on the Sun often leads to very large magnetic storms that can severely damage or even destroy spacecraft and ground systems. A collaboratory might be used by a group of experts to quickly bring together data that could be used to predict the occurrence of such activity and perhaps anticipate possible effects.
A collaboratory might also serve as a repository and distribution system for software. Many data formats, data access routines, data analysis systems, and specialized transformation utilities that are developed at public expense are potentially of great value to others. A collaboratory could provide a directory to such software as well as a repository for it and its accompanying documentation.
It is conceivable that a collaboratory could be used for electronic publishing, or at least as an initial step in this direction. Today a central location such as the National Space Science Data Center could be used to receive formatted documents and graphics. Standards would have to be defined and a means provided for translating among the various common formats. Initially an electronic publishing center might be used only for documentation of the data for group studies and software descriptions. Eventually it might serve as a testbed for a broader vision of electronic publishing and literature distribution.
A collaboratory in space physics could be of use to scientists at small universities and colleges who might, for example, use a bulletin board to advertise their interest in cooperating on a research project of a given type. Interested researchers at a funded institution could respond via the bulletin board. Eventually the participants might exchange data through the collaboratory and use the hardware, software, and tools of the collaboratory to analyze the data.
Another possibility is that after the primary research has been done by the initial investigator, the databases developed for specialized studies could be written to CD-ROMs along with appropriate software to access and display the data. Such collections, which might also include electronic versions of the papers produced in the studies, could be used to carry out additional research. Alternatively, teachers or professors at various institutions could make use of the data and accompanying papers to develop problems and exercises for classes in science and mathematics. Yet another possibility is to use a collaboratory to provide a source of general information in science. A collaboratory might have a public bulletin board or user interface that could provide access to general information about space science.
COMPONENTS OF A COLLABORATORY INFRASTRUCTURE
The possible uses of a collaboratory for space physics suggest a number of components that would serve to enable efficient collaboration between two or more space physicists and even "collaborations of one" allowing a single investigator to synthesize data from a number of different sources. Above all, a collaboratory infrastructure for space physics must allow for easy and rapid access to a multitude of different types of data from numerous instruments on a number of different space- and ground-based platforms. In addition,
The system must provide for the education of its users and minimize the learning curve required to become proficient; easily accessible help services (including a telephone number) should be provided.
Standards are a cornerstone of a national collaboratory infrastructure. Even though it is not necessary that all data be in identical format, it is necessary that the data be translatable into a compatible format for seamless integration into data-handling and analysis systems. Standards for user interfaces, network protocols, and so on are fundamental to a collaboratory.
It is important that users of a collaboratory adopt "rules of the road" to protect the originators of data and ideas and to assure that collaborations are carried out in fairness to all participants and contributors, and that data, analysis tools, and ideas are used correctly.
Given the dependence of the space physics community on space-borne instrumentation and remote land-based observatories and facilities, it is essential that agreed-upon methods be developed for the remote operation of these instruments and facilities, especially in support of rapid-response campaigns of coordinated observations.
Networking is the backbone of the collaboratory infrastructure, because it provides the digital communications required for the sharing of data and ideas. Networking services needed include electronic mail, file transfer, remote log-on and execution, database management, teleconferencing, "whiteboard" or electronic "scratchpad" capability, shared access to common graphical displays, and so on. The networking services must all meet basic levels of performance with respect to quality, reliability, and response time, and higher levels of performance should be worked toward.
Electronic mail, a fundamental networking service required by a collaboratory, deserves special mention. Needed is the capability to transfer compound documents, that is, those with special characters and formatting as well as graphics, in a standardized form that is compatible with common word processing and desktop publishing tools. Support services in the form of directories, yellow pages, and so on are also needed.
Computer-supported cooperative work tools are important for access to shared resources such as data and computing resources.
Common access to software libraries is an important component of a national collaboratory. It is especially important that shared software be reliable, portable, and unambiguous as to its use.
The collaboratory infrastructure and its analysis tools should be directly available for supporting the education of space physicists and, more generally, students in the physical sciences.
The infrastructure must be affordable so that (a) it will be built and (b) it will not displace funding otherwise earmarked for space physics research. The collaboratory infrastructure must be regarded by the space physics community as a set of services well worth their support and as a fundamental pan of their suite of research tools, or else it will be considered a funding burden.