Magnetic Storms and Induction Hazards
Total Page:16
File Type:pdf, Size:1020Kb
News: Report Puts Timeline on Cutting Greenhouse Gas Emissions, p. 446 Meeting: Multidisciplinary Monitoring Experiments at Kawah Ijen Volcano, p. 447 What’s on the Web: Antarctic Expedition, Teaching Professional Skills, and More, p. 447 About AGU: Amazon Hack-A-Thon at Fall Meeting, p. 448 Research Spotlight: Magnetic Field Data, Ocean Acidification, and More, p. 452 VOLUME 95 NUMBER 48 2 DECEMBER 2014 Magnetic Storms and Induction Hazards Magnetic storms are potentially hazard- Electrical conductivity in the Earth’s inte- ous to the activities and technological infra- rior ranges from about 10-4 siemens per structure of modern civilization. This reality meter (S/m) in some parts of the upper man- was dramatically demonstrated during the tle to 3 S/m in the ocean. Generally speak- great magnetic storm of March 1989, when ing, electric power grids are susceptible to surface geoelectric fields, produced by the interference from naturally induced geoelec- interaction of the time-varying geomag- tric fields that vary with periods from about netic field with the Earth’s electrically con- 10 to 1000 seconds. Geomagnetic and geo- ducting interior, coupled onto the overlying electric field variation over such periods Hydro-Québec electric power grid in Can- plumbs the Earth’s interior across diffusive ada. Protective relays were tripped, the grid depth and length scales of between about 2 collapsed, and about 9 million people were and 3000 kilometers, but localized conduc- temporarily left without electricity [Bolduc, tivity anomalies can reduce the upper length 2002]. scale to about 50 kilometers. A magnetic storm that was, by some mea- As a subject of natural science, estimat- sures, the most intense ever recorded fol- ing the geoelectric field as a function of geo- lowed a solar flare observed by astronomers graphic location is distinct from the engi- Richard Carrington and Richard Hodgson neering subject of electric currents that in September 1859. Should a storm of simi- flow in power grids in response to geoelec- Tebnad/Dreamstime.com lar intensity occur today, technological sys- tric fields [e.g.,Kappenman , 2001; Pirjola, © tems around the world could be adversely 2002]. Mapping the time-dependent geoelec- Fig. 1. Electric power lines at sunrise. affected. According to some scenarios, the tric field is needed to evaluate the design, future occurrence of a rare “perfect mag- emplacement, and ever- evolving opera- function of time. For magnetotelluric stud- of geomagnetic activity generated by netic storm” might cause widespread failure tion of electric power grids [e.g., Viljanen ies, geoelectric measurements are conven- magnetospheric- ionospheric electric cur- of bulk electric power networks (see Fig- et al., 2012]. Here we focus on the present- tionally made in conjunction with magnetic rents. In principle, empirically parameter- ure 1), with deleterious impact on the Ameri- day challenges and opportunities for study- variometer measurements, with sensors ized maps of the temporal evolution of can economy and security [e.g., Baker et al., ing and quantifying hazardous geoelectric deployed over specific geographic regions ground-level magnetic disturbance can be 2008]. induction. on a temporary basis. The EarthScope constructed by fitting basis function model Public and private agencies have program of the National Science Founda- parameters to ground magnetometer data responded to these findings. Notably, in Geomagnetic Monitoring tion has supported a transportable grid of [e.g., Pulkkinen et al., 2003], thus filling in May 2013, the Department of Energy’s Fed- magnetotelluric sensors across the United the geography between magnetometer sta- eral Energy Regulatory Commission (FERC) Magnetometers around the world, such States. Deployed from 2006 to 2011 in the tions. Time-convolutional filtering of a mag- directed, through Order 779, the North as those that are part of the International Pacific Northwest, the sensor grid is now netic activity map through a conductivity American Electric Reliability Corporation Real- time Magnetic Observatory Network being deployed across the north midwestern model can give a geoelectric field map [e.g., to develop reliability standards to address (INTERMAGNET), record the temporal evo- United States [Schultz, 2009]. Thomson et al., 2009]. Geoelectric model the potential impact of geomagnetic distur- lution of the geomagnetic field Love[ and In contrast, long- term measurements of accuracy can be established by comparison bances on the operation of the bulk power Chulliat, 2013]; an example time series is the geoelectric field at observatories are with geoelectric measurements, and indeed, system. Concerns in the private sector have shown in Figure 2(a). As part of NSWP, the much more sparsely distributed in geogra- a geoelectric modeling project can inform motivated reinsurance companies to com- USGS magnetic observatory network pro- phy. The Japanese Meteorological Agency evaluations of the adequacy of the existing mission related assessments of risk. duces high-quality magnetometer data for has supported 1-second measurements of ground magnetometer network. In the end, As part of an interagency project coor- real-time nowcasting of magnetic storm con- the geoelectric field at its three magnetic estimates of the storm time geoelectric field dinated under the auspices of the U.S. ditions [Love and Finn, 2011]. Similarly inte- observatories since 1983; see Figure 2(b). need only be accurate enough to make oper- National Space Weather Program (NSWP), grated geomagnetic monitoring projects are Geoelectric measurements at observatories ational hazard assessments and to enable the U.S. Geological Survey (USGS), the supported in other countries, such as in Can- are supported by Germany’s GeoFor schungs- mitigation of deleterious effects. National Oceanic and Atmospheric Admin- ada, the United Kingdom, and Japan. Zen trum, the British Geological Survey, and istration (NOAA), and NASA are working Most national magnetic observatory net- France’s Institut de Physique du Globe de Space Weather Prediction together to improve regional assessments works are relatively sparse in geographic dis- Paris. Analog geoelectric measurements and real- time operational estimation of tribution. For example, the USGS operates were supported at the Tucson magnetic The space weather that drives magnetic natural induction hazards that can be real- only six stations in the lower continental observatory from 1932 to 1942, but other- storms originates at the Sun. Abrupt ejec- ized at the Earth’s surface during magnetic United States, which is sufficient to resolve wise, very little long-term geoelectric moni- tions of concentrated plasma from the solar storms. geomagnetic activity on a continental scale toring has taken place in the United States. corona can be detected using telescopes on and to assess the general dynamic state of However, the USGS is considering a proj- the NASA–European Space Agency Solar The Natural Hazard the magnetosphere- ionosphere system. More ect for long-term 1-second resolution geo- and Heliospheric Observatory (SOHO) detailed analyses can exploit data from var- electric monitoring at a few of its magnetic spacecraft. Coronal mass ejections typi- The geophysical quantity that directly iometer magnetometer networks that are observatories. cally take about 2 days to traverse the Sun– interferes with the operation of electric operated for specialized space weather proj- Earth distance; energetic ejections can cross power grids is the geoelectric field. It is gen- ects [e.g., Viljanen et al., 2004; Yumoto et al., Modeling and Mapping the distance in as little as 18 hours. If an erated in the Earth’s interior through geo- 2012]. ejection is directed toward the Earth, then magnetic induction driven by magnetic Although some geoelectric monitoring is NOAA’s Space Weather Prediction Center activity originating overhead in the magneto- Geoelectric Measurements important, it is challenging to directly use (SWPC) will issue a prediction of the com- sphere and ionosphere. geoelectric measurements to estimate induc- mencement time and intensity of a magnetic Direct measurement of the geoelec- tion hazards across continental or even storm. From an upstream orbit between the tric field is conceptually simple: The volt- small-scale regional geography. Storm time Sun and Earth, NASA’s Advanced Compo- BY J. J. LOVE, E. J. RIGLER, A. PULKKINEN, age between a pair of electrodes, planted geoelectric fields realized at one site are not sition Explorer (ACE) spacecraft monitors AND C. C. BALCH straight into the ground, is measured as a always well correlated with those at another solar wind conditions, with transmitted data site, a difficulty that is partly due to the arriving at Earth 15 to 45 minutes in advance localized complexity of lithospheric electri- of an oncoming coronal mass ejection. cal conductivity [e.g., McKay and Whaler, Within the magnetosphere and above the 2006] and the relatively high conductiv- ionosphere, several satellites provide in situ ity of ocean water compared to the litho- magnetic field monitoring of the Earth’s sur- sphere. Therefore, a priority for induction rounding space environment. hazard science is the development of three- Improved predictions of space weather dimensional models of Earth conductivity require improved monitoring and physics- that have a spatial resolution similar to the based modeling of the heliosphere and spatial scale of regional