Back to Science for Everyone
The National Space Weather Program (NSWP) defines space weather as the "conditions on the sun and in the solar wind, magnetosphere, ionosphere, and thermosphere that can influence the performance and reliability of space-borne and ground-based technological systems and can endanger human life or health." It further observes that "Adverse conditions in the space environment can disrupt satellite operations, communications, navigation, and electric power distribution grids and lead to a variety of socioeconomic losses [NSWP, 1996] .
While the NSWP clearly involves many areas of basic research, it also includes, by definition, an applied aspect. If the program is to have an impact on space weather's adverse effects on technological systems, the scientific community must learn to understand the real problems that users have. Conversely, users must learn what types of information may become available. Practical solutions to the problems space weather poses depend on a continuing dialogue between both communities.
The National Science Foundation is fostering this interchange by funding meetings that bring together participants from the user, scientific, and forecasting communities. Unlike typical scientific meetings, the focus is not on recent scientific and technical advances. Rather, it is to discuss the interfaces between the three communities. User needs and the priorities assigned by users are an important aspect of determining how the scientific community can best direct its activities.
The first meeting was held October 17–18, 1996, at the Electric Power Research Institute (EPRI) in Washington, D.C. The subject of the meeting was the effect of space events on electric power transmission systems. Disturbances in space can lead to large, rapidly rising ionospheric currents or electrojets which in turn can induce currents in extended conductors near ground level to which electric power transmission systems, cable communication systems, and pipelines can be vulnerable. Such geomagnetically induced currents (GICs) have been observed since 1847, when significant currents in telegraph cables were observed by telegraph operators. In March of 1989, a solar event in space caused a major electric power blackout in Quebec. The same solar event caused transformer damage in New Jersey and was responsible for tripping protective circuits throughout the United States and Canada. A more detailed description of this phenomenon has been provided in numerous articles, [e.g., Kappenman et al., 1997] .
Meeting attendance was limited to stimulate discussion and enhance the exchange of ideas between people who approach problems from different perspectives, as do engineers and scientists. Attendance was about one third engineers employed by electric utility or related companies, and two thirds scientists involved in space weather forecasting, the study of solar, solar wind, magnetospheric, ionospheric, and lithospheric phenomena related to GICs.
The meeting's organizers also invited engineers from a major cable communications company, but they declined to participate because they had not experienced severe problems since 1972 and felt that in terms of their systems the problem was presently under control. The meeting's organizers had also expected greater representation from the electric power industry.
The meeting certainly generated lively discussion. Many areas of interaction across the disciplinary boundaries were identified. Though it is probably true that the topics raised at the meeting were not new, appreciation of their significance was greatly enhanced, many bridges were established between the different groups, and the participants gained greater awareness of some of the requirements and constraints faced by other groups.
Discussions were positive in terms of technical progress. The Sunburst Program, an activity sponsored by EPRI that involves several electric power companies, has directly measured induced currents in power lines over a wide grid. Canadian and Finnish researchers discussed techniques for calculating power line currents from ionospheric currents taking into account current distribution in the ionosphere and spatial variations in ground conductivity. This work helps define a natural interface: if the space scientists can predict ionospheric current distributions and their time dependence, these can be translated to power line effects depending on local ground conductivities and power line circuit parameters. Furthermore, it became apparent that some power company requirements can be met by specifying a likelihood of disturbances in a general area. While reliable predictions are paramount, absolute predictions are not always necessary. This suggests that while the width of ionospheric currents is important, the location of the currents may not have to be specified with the same accuracy.
A strictly empirical algorithm that is not physics-based shows good correlation in relating solar wind parameters to auroral electrojets and ground magnetic field disturbances. With continuous satellite solar wind measurements this is a potential prediction system because of the time delays between solar wind events and terrestrial events. The meeting participants urged that this algorithm be coupled with the theory of transformation of ionospheric to power line currents and tested with the Sunburst observations. This informal recommendation seemed to be based more on the desire for testing an end-to-end prediction system than endorsement of any particular method. Others were encouraged to initiate similar tests.
There is a range of timeframes over which power companies can take preventive steps. One- or two-day warnings would allow maintenance procedures that shut down some facilities to be rescheduled, thus maintaining the full reserve for emergency situations. The costs for this activity would not be high and, therefore, it would require good but not high prediction accuracy. If shorter-timeframe warnings were issued, some protective circuits could be desensitized in seconds. These circuits are used in power grids to isolate problems that are unrelated to GICs but can also be tripped by a secondary reaction to GICs when the GIC magnitude is large but not in itself damaging. These unnecessary circuit trips can cause problems by affecting other parts of the grid. However, because of overall system safety considerations and insurance requirements, this should only be done when high certainty exists and only for a time measured in minutes. Power company participants suggested a series of desired lead times and related accuracies of forecasts that will probably be refined in the future both in terms of need and forecast feasibility.
Much of the discussion also addressed the degree of need within the power industry. How do the problems associated with GICs rate among the myriad of other problems presently requiring attention by power company managers?
In addition to direct statements by several of the power company participants, there were other indications that problems with GICs are currently a low priority. The Sunburst Program, which is the primary focus of power company interest in the GIC problem, has been cut significantly and is only being funded at a rate of $60,000 per year. None of the power companies represented had even one full-time staff person working on the problem. One power company, which suffered multimillion-dollar transformer damage in the 1989 storm, is still not paying much attention to the problem. They concluded that the damage was peculiar to one particular transformer and that others on their system would not be vulnerable. Another power company expressed willingness to accept a recent nonspace-related equipment failure on the basis that it was an event that occurred with a probability of less than once every 50 years. For most individual power companies direct space-generated failures have not been demonstrated to be that frequent.
Other reasons for the low priority power companies give to GICs were discussed. Power company management is presently focusing most of its attention on the upcoming restructuring of the industry, which will allow individual consumers to buy electricity from power companies of their choice. This introduction of competition in a previously highly regulated industry is probably the most significant change in the structure of the power industry in decades. Even without this momentous change, recognition of the importance of space weather effects may have faded because it has been a long time since the last solar maximum. Some concern was raised that competition may lead to an increase in potential system failures from GIC impacts, because it will put a premium on operating at full capacity with minimal safety margins. This effect, should it occur, is not yet being felt. Finally, since society stops functioning during a blackout, the societal cost of a power outage, as estimated by a Department of Energy study, is more than an order of magnitude greater than the direct cost to power companies.
The meeting did not reach a general consensus on how to deal with the lack of industry interest, but it did facilitate a useful exchange of ideas that generated a number of joint technical projects. A follow-up workshop was recommended.—Harry E. Petschek, MHP, Inc., Lexington, Mass.; and William E. Feero, Electric Research & Management, Inc., State College, Pa.
DOE Report HCP/T5103-01, Impact Assessment of the 1977 New York City Blackout, July 1978.
Kappenman, J. G., L. J. Zanetti, and W. A. Radasky, Eos Trans. AGU, 78, 4, 1997.
The National Space Weather Program, The Strategic Plan, Office of the Federal Coordinator for Meteorological Services and Supporting Research, Silver Spring, Md., August 1995.