Space Weather. Global Positioning System. Ionospheric Effects. on the. Quarterly. Opinion: Feature: Feature:

Transcription

1 Space Weather Quarterly Volume 5, Issue 2, The International Journal of Research and Applications Ionospheric Effects on the Global Positioning System Opinion: Which model descriptions should be published? Feature: Small research satellites Feature: A new cycle of sunspots

2 Editorial Space Weather Model Descriptions: Which Should Be Published? Since the launch of Space Weather, a number of space weather models of all types have been reported in the journal. One of the more uncertain (and at times controversial) aspects of numerical simulations of the entire Sun-to-Earth environment is for authors (and editors) to determine when such global models and their results are deserving of consideration for publication. Models of solar-terrestrial processes are essential for successful predictions of these processes and of their effects on technological systems. Models can take many forms, from simple empirical expressions of the relationships among two or more physical variables to more detailed physics-based codes of subsets of the solar-terrestrial environment. Numerical simulations of the entire environment from the Sun to Earth s ionosphere and surface now exist. Such simulations generally incorporate modules for the several subsystems of the environment, with each module containing the relevant physics (and chemistry in appropriate cases) that is required. Many codes are in various states of testing, validation, and use in the joint U.S. National Science Foundation/NASA Community Coordinated Modeling Center. Numerical simulations must be considered and evaluated in the context of actual physical measurements. Models can be essential in deriving physical understanding from obtained data. At the same time, data are essential in validating and calibrating models. Therefore, computational simulations considered for publication should ideally report their success, or lack of success, in explaining observations. In particular, reports of failures permit a modeling group and interested outside research groups to better understand limitations of a numerical code and its modules, and thereby spur new modeling insights and developments. Often it appears that modeling teams concentrate on studying some of the 2 S pac e W e at h e r Q ua r t e r ly largest solar-terrestrial events, especially those that involve the release of coronal mass ejections and their transit to Earth. This concentration is natural in that large events are those that can most affect technological systems. While models often do not fully explain all of the data in such large-event studies, these attempts nevertheless illuminate important limitations in the codes and their modules, and therefore point the way to new directions. But which models deserve consideration for publication? The perception of some modelers that Space Weather and its applications-oriented readers favor modeling papers that address extreme events is not in actuality the case. The journal is quite interested in publishing modeling papers that address less severely disturbed solar-terrestrial conditions, but ones that still involve perturbations of the Earth s environment that can produce effects on technical systems. One condition that requires more modeling analysis under numerical simulation is the case where enhanced relativistic electrons are produced at geosynchronous orbit when Earth is hit by an interplanetary corotating interaction region. In such a case, the enhanced relativistic electrons can produce charging of dielectric materials such as shielded electronics and coaxial cables deep inside spacecraft bodies. Modeling, beginning at the Sun and continuing to geosynchronous orbit, is of prime interest for this problem. I invite modelers to consider possible papers on this and other less disturbed but important solar-terrestrial space weather topics. Louis J. Lanzerotti is editor of Space Weather, a distinguished research professor at the New Jersey Institute of Technology, and a consultant at Alcatel- Lucent Technologies Bell Laboratories. Citation: Lanzerotti, L. J. (2008), Space weather model descriptions: Which should be published?, Space Weather, 6, S05003, doi: /2008sw EDITOR Louis J. Lanzerotti, Center for Solar-Terrestrial Research, New Jersey Institute of Technology, Newark; ljl@adm.njit.edu EDITORIAL ADVISORY BOARD Daniel N. Baker, University of Colorado, Boulder; David H. Boteler, Geological Survey of Canada, Ottawa, Ontario; Volker Bothmer, Institut für Astrophysik, Göttingen, Germany; Norma B. Crosby, Belgian Institute for Space Aeronomy, Brussels, Belgium; Richard R. Fisher, NASA Headquarters, Washington, D. C.; Yohsuke Kamide, Kyoto University, Kyoto, Japan; John G. Kappenman, Metatech Corporation, Duluth, Minn.; Joseph E. Mazur, The Aerospace Corporation, Chantilly, Va.; Robert P. McCoy, Office of Naval Research, Arlington, Va.; Hermann J. Opgenoorth, ESTEC, Noordwijk, Netherlands; Robert M. Robinson, National Science Foundation, Arlington, Va.; Howard J. Singer, NOAA Space Environment Center, Boulder, Colo.; Phil Wilkinson, IPS Radio and Space Services, Haymarket, New South Wales, Australia DIRECTOR OF PUBLICATIONS Judy Holoviak MAGAZINE STAFF Publisher Nina Tristani, ntristani@agu.org Manager, News Publications Barbara T. Richman, brichman@agu.org Science Writer/Editor Mohi Kumar, mkumar@agu.org Production Coordinator Faith A. Ishii, fishii@agu.org Editor s Assistant Liz Castenson, spaceweather@agu.org Copy Editor Don Hendrickson, dhendrickson@agu.org Design Consultants Nadine Johnson and Christopher D. Jones Space Weather Quarterly is a digest of selected articles published online in Space Weather: The International Journal of Research and Applications ( Space Weather Quarterly (ISSN ) is published quarterly by the American Geophysical Union, 2000 Florida Ave., NW, Washington, DC 20009, USA POSTMASTER: Send address changes to Member Service Center, 2000 Florida Ave., NW, Washington, DC 20009, USA 2008 American Geophysical Union. Material in this issue may be photocopied by individual scientists and engineers for research or classroom use. For permission for any other use, contact the Editorial Office at spaceweather@agu.org. Change of address: Contact the Member Service Center 8:00 a.m. 7:00 p.m. Eastern time; Tel: ; ; Fax: ; service@agu.org Views expressed in this publication do not necessarily reflect official positions of the American Geophysical Union unless expressly stated.

3 Volume 5 Issue 2, 2008 Contents Departments 2 Editorial: Space Weather Model Descriptions: Which Should Be Published? By LOUIS J. LANZEROTTI Our editor reviews what makes a good space weather model description, and invites modelers to submit papers that address extreme events as well as papers that consider less severely disturbed solar-terrestrial conditions. 4 Editor s Choice By John Goodman Selected new articles on the topic of space weather from AGU journals. Features 6 Small Satellites for Space Weather Research By Therese Moretto and Robert M. Robinson Scientists at the U.S. National Science Foundation discuss how recent advances in sensor and spacecraft technologies are making small satellites a cost-effective means to accomplish key research objectives in space weather. Technical Article 20 Ionospheric scintillation effects on single frequency GPS By r. a. steenburgh, c. g. smithtro, AND k. m. groves Through the examination of several Global Positioning System (GPS) data sets, scientists calculate the probability of scintillation occurrence and determine that scintillation activity is generally greatest during the equinoxes and solar maximum. Cover: Global Positioning System Block IIR-M satellites broadcast a new un-encrypted civilian signal that can be disturbed by adverse space weather. Map: District of Columbia Office of Planning; Satellite: National Executive Committee for Space-Based Positioning, Navigation, and Timing 12 Spring on the Sun: A New Cycle of Sunspots By Irene Klotz The emergence of an oppositely oriented sunspot early this year has renewed debate over the predicted intensity of the next 11-year solar cycle. 15 Space Weather and the Global Positioning System By Anthea Coster and Attila Komjathy As society becomes increasingly dependent on the Global Positioning System (GPS), which can be disrupted by adverse space weather, GPS itself is emerging as one of the premier remote sensing tools to monitor space weather events. Two scientists introduce the journal s open special section, called Space Weather Effects on GPS. Space Weather Quarterly is cosponsored by the U.S. National Science Foundation (grant ATM ) and the International Space Environment Service. Space Weather Quarterly 3

4 Editor s OPINION Choice by Howard J. Singer Selected new articles on the topic of space weather from AGU journals Evolution of the Solar Magnetic Field Geomagnetically Induced Currents Ultimately, the Sun is the source of space weather disturbances that affect us on Earth. And it is only through a better understanding of the structure and evolution of the solar magnetic field on a variety of spatial and temporal scales that there is hope for improving predictive capabilities for major solar disturbances such as X-ray flares and coronal mass ejections. On the topic of how the solar magnetic field evolves from the solar interior and into the solar corona, Archontis [2008] provides a review of the theory and numerical models of solar magnetic flux emergence and examines both recent progress and future challenges. Archontis, V. (2008), Magnetic flux emergence in the Sun, J. Geophys. Res., 113, A03S04, doi: /2007ja ^ First-light image of the Sun from the X-ray telescope on board the Hinode satellite. Archontis [2008] discusses the theory behind the evolution of solar disturbances related to bright areas seen here in the solar corona. JAXA Geomagnetically induced currents can disrupt power transmission lines like those pictured above. ^ Geomagnetically induced currents (GICs) result from rapid changes in Earth s magnetic environment caused by interactions between Earth s magnetic field and solar wind disturbances such as coronal mass ejections and corotating interaction regions. Power lines and pipelines are among the systems that can be disrupted by these induced currents. Kataoka and Pulkkinen [2008] examine the differences in the GIC response to different solar wind drivers and provide a relationship to determine GIC strength from ground-based east-west magnetic field variations. In a related article, Pirjola [2008] examines the practical influence of grounding on GICs in power transmission systems. Kataoka, R., and A. Pulkkinen (2008), Geomagnetically induced currents during intense storms driven by coronal mass ejections and corotating interacting regions, J. Geophys. Res., 113, A03S12, doi: /2007ja Pirjola, R. (2008), Study of effects of changes of earthing resistances on geomagnetically induced currents in an electric power transmission system, Radio Sci., 43, RS1004, doi: /2007rs Terna, Transmission System Operator, Italy 4 Space Weather e Quarterly r

5 Current Research Space Weather articles published online Modeling the Radiation Belt Environment Understanding Earth s radiation belts presents a challenge both to those who operate spacecraft in that hazardous environment and to space weather modelers who attempt to describe the environment and predict its changes. One of the recent physicsbased models that is advancing our ability to portray this important region of space is the Radiation Belt Environment (RBE) model developed by Fok et al. [2008]. In their recent paper they not only describe the importance of including physics such as waveparticle interactions, but they also remind us of what is needed for future improvements. A version of this model is running in real time at Fok, M.-C., R. B. Horne, N. P. Meredith, and S. A. Glauert (2008), Radiation Belt Environment model: Application to space weather nowcasting, J. Geophys. Res., 113, A03S08, doi: /2007ja ^ To better predict when and how satellites in low-earth orbit will be affected by solar flares, scientists from the University of Colorado evaluate solar irradiance variations from both the impulsive and gradual phases of solar flares using the Flare Irradiance Spectral Model. Chamberlin, P. C., T. N. Woods, and F. G. Eparvier (2008), Flare Irradiance Spectral Model (FISM): Flare component algorithms and results, Space Weather, 6, S05001, doi: / 2007SW Researchers investigate the feasibility of providing a real-time ionospheric warning system for Canadian users of differential global positioning systems, based on available real-time data. Skone, S., and A. Coster (2008), Potential for issuing ionospheric warnings to Canadian users of marine DGPS, Space Weather, 6, S04D03, doi: /2007sw Through the examination of several Global Positioning System (GPS) data sets, scientists calculate the probability of scintillation occurrence and determine that scintillation activity is generally greatest during the equinoxes and solar maximum. Steenburgh, R. A., C. G. Smithtro, and K. M. Groves (2008), Ionospheric scintillation effects on single frequency GPS, Space Weather, 6, S04D02, doi: /2007sw Using data collected during a recent substorm, scientists show for the first time a direct connection between the loss of lock on a Global Positioning System receiver and fading caused by auroral precipitation. Smith, A. M., C. N. Mitchell, R. J. Watson, R. W. Meggs, P. M. Kintner, K. Kauristie, and F. Honary (2008), GPS scintillation in the high arctic associated with an auroral arc, Space Weather, 6, S03D01, doi: /2007sw Earth s radiation belts, conceptually drawn here, contain charged particles that can damage spacecraft electronics. Fok et al. [2008] developed a model to characterize the radiation belts, including wave-particle interactions. NASA Living With a Star Scientists show how a three-dimensional kinematic solar wind model can be extended to predict what the Solar Terrestrial Relations Observatory (STEREO) spacecraft might expect in observing large-scale plasma clouds ejected from the Sun. Sun, W., C. S. Deehr, M. Dryer, C. D. Fry, Z. K. Smith, and S.-I. Akasofu (2008), Simulated Solar Mass Ejection Imager and Solar Terrestrial Relations Observatory-like views of the solar wind following the solar flares of May 2003, Space Weather, 6, S03006, doi: /2006sw Full texts of these and other recent AGU journal articles on space weather subjects are available to Space Weather online subscribers under the Technical Articles or Editor s Choice sections. New articles are added every month. Space Weather Quarterly 5

6 6 S pac e W e at h e r Q ua r t e r ly Small for Space

7 Satellites Weather Research By Therese Moretto and Robert M. Robinson Recent advances in sensor and spacecraft technologies are making small satellites a costeffective means to accomplish key research objectives in space weather. Small- satellite missions can complement existing or planned larger missions by filling gaps in time or coverage, or they can constitute stand-alone missions targeting specific, well-defined science and applications questions. Such missions will also play a crucial role in training the next generation of experimental space scientists, aerospace engineers, and space weather professionals. This article highlights some of the advancements that have been made in recent years that make small- satellite missions both feasible and cost-effective. Several new programs Measuring about 10 centimeters on each side, CubeSat picosatellites weigh at most 1 kilogram. Images courtesy of Nina Tveter/NTNU Info and Alan J. Zuzic. are under way that take advantage of the progress in sensor development, satellite platform design, and launch technology for small satellites. The time is right to integrate these emerging technologies to accomplish key science objectives, particularly in space weather. Throughout this article we use the term small satellite quite loosely to denote a wide range of satellite missions. No clear and universally agreed upon classification of satellites or satellite missions exists. Table 1 shows a rough classification that we shall use here for reference spanning from large satellites to picosatellites based on typical values for satellite mass, cost, and development time frame in each class. However, individual missions often span several classes when considering size, cost, or complexity. We focus here on missions in the microsatellite to picosatellite range. Space Weather Quarterly 7

8 New Launch Opportunities In a recent report by the Assessment Committee for the National Space Weather Program (NSWP; see nswp_ acreport0706.pdf), a key recommendation was for NSWP agencies to investigate the feasibility of using microsatellites with miniaturized sensors to provide cost- effective science and operational data sources for space weather applications. The U.S. National Science Foundation s (NSF) Division of Atmospheric Sciences sponsored a workshop in May 2007 to explore the possibilities and benefits of utilizing small- satellite missions to provide essential measurements for space weather research. In his keynote address at the workshop, the NASA Ames Research Center (ARC) director, Simon Pete Worden, stated his belief that significant science can be done with small satellites at the cost of a few million dollars per mission. He argued that the technology exists, as well as the launch opportunities, and asserted the feasibility and value of executing simple, focused missions of limited duration, in the 2- to 5-year range. Most of the workshop presentations, including Worden s, can be viewed at the meeting Web site, wiki/index.php/ssw. Formerly, the biggest obstacle to routine access to space was the cost of launching to orbit. Owing to the growing demand for lowcost access to space, government agencies and private industries have been striving to explore new technologies and strategies to reduce launch costs while keeping risk of launch failure low. One primary target of these efforts is providing launch opportunities for small satellites, which not only are more amenable to keeping costs low but also offer quicker turnaround times for end-to-end space missions. New launch opportunities include both dedicated flights on low-cost launch vehicles and ride-shares on larger missions. One example that puts these new ideas into practice is the newly created Hawaii Space Flight Laboratory (HSFL) at the University of Hawai i, which is working on establishing a sustainable local integration and launch capability, with support from the Hawaii Space Grant Consortium in partnership with the Pacific Missile Range Facility and the Kauai Test Facility, located at Barking Sands, on the west coast of the island of Kauai. The launch vehicle is Sandia National Laboratories Super Strypi research rocket, Table 1. Satellite Sizes, Costs, and Development Times Class Mass, kilograms Cost, millions of dollars Time, years Large satellite Small satellite Minisatellite Microsatellite Nanosatellite Picosatellite <1 <0.5 <1 which can carry 300 kilograms to low-earth orbit ( LEO). The first two launches are planned for 2009 and 2010, as part of the Low Earth Orbit Nanosat Integrated Defense Autonomous System ( LEONIDAS) program. The purpose of LEONIDAS is to develop a test bed for small satellites created and launched out of Hawaii, with remote sensing experiments and small- satellite technology demonstrations as main focus areas. Payloads for LEONIDAS launches are expected to include microsatellites built by HSFL as well as additional microsatellites and nanosatellites still to be determined. Future plans for LEONIDAS and HSFL include missions/ launches at a sustained rate of at least one per year. Total recurring costs for these launches are expected to be of the order of $9 million, which can be divided among multiple payloads. Plans are to offer this launch service to other university and research customers, including launch opportunities for small secondary payloads at a highly competitive price. In a similar initiative, the U.S. government has established the Minotaur rocket program utilizing Minuteman intercontinental ballistic missiles (ICBMs) as an inexpensive and responsive smallmission launch capability. The price for a Minotaur I launch, which can carry 500 kilograms to LEO, is of the order of $20 million. Recently, the U.S. Air Force TacSat 2 minisatellite, designed to help military commanders gain access to up-to-the-minute satellite data from satellites directly overhead, was launched on a Minotaur I from the NASA Wallops Flight Facility (Wallops Island, Va.) with NASA s smallest satellite to date (the GeneSat 1 nanosatellite, designed for biological research; see as a secondary payload. TacSat 3, with a mission similar to that of TacSat 2, will be launched in the same way in June 2008 together with NASA s second miniature biological space laboratory, GeneSat 2, and a set of three test CubeSat picosatellites (see next section for more information about CubeSat). The TacSat Minotaur launches are under the purview of the U.S. Department of Defense (DOD) Space Test Program (STP). The inclusion of secondary payloads on these launches is the result of dedicated efforts within the STP to utilize excess capacity on the launches they sponsor. It demonstrates strong commitment by the STP not only to fulfill DOD needs but also to offer excess launch capacity to serve other U.S. government needs for small research satellite payloads. Another recent initiative by the STP, aimed at boosting the utilization of excess capacity on large launch vehicles, is the development of the Evolved Expendable Launch Vehicle (EELV) Secondary Payload Adapter (ESPA). ESPA is a ring placed under an EELV primary payload that supports up to six 180- kilogram secondary spacecraft. The first mission to utilize an ESPA was the Space Test Program 1 (STP 1), which launched on an Atlas V rocket in March 2007 from Cape Canaveral. The ESPA ring carried four military research satellites as a secondary payload, two of which are relevant to space weather (one will investigate the causes of communication outages in the upper atmosphere, and the other will investigate plasma irregularities in the upper atmosphere). The satellites were successfully delivered into two different 8 S pac e W e at h e r Q ua r t e r ly

9 low-earth orbits, thus proving the ESPA concept and design. The STP s goal is to include the ESPA ring on as many of its launches as possible, and it is collaborating with NASA on extending the use of ESPAs on NASA EELV launches. Considerable progress in launch vehicle technology for small satellites has also been accomplished in the private sector. The Falcon 1 rocket, developed by the Space Exploration Technology Corporation (SpaceX), has the capacity to carry of the order of 500 kilograms into LEO. Launches are currently offered at a flat rate of $7 million from several launch sites. The company expects to be offering shared launches and launch opportunities for small secondary payloads in the near future. The big issue for ride- sharing, of course, is eliminating the potential risk of damage to the primary payload. This requires careful planning and strong collaboration between payload developers and spacecraft integrators. Ride-sharing has been greatly simplified in recent years because of the development of standardized satellite platforms. Satellite Platforms The development of flexible but standardized, proven spacecraft designs eliminates the need for much of the nonrecurring engineering work typically associated with mission design and implementation. Along with the provision of readily available, cheap commercial-off-the-shelf (COTS) components and subsystems, standardized spacecraft greatly reduce the costs of individual missions while at the same time, making it technically feasible for a much larger community, including student projects, to get involved with spacecraft missions. A recent initiative by the STP to move away from the oneof-a-kind spacecraft philosophy is the design of the Standard Interface Vehicle (SIV). The STP-SIV is designed as an ESPA-class prototype spacecraft to provide a cost-effective platform for utilizing ESPA launch opportunities. Similar concepts to provide fast, cost- effective satellite solutions are being pursued in the commercial arena by United Kingdom s Surrey Satellite Technology Ltd. (SSTL) and the United States s SpaceQuest, Ltd. The latter offers a so-called data-buy concept. In this scenario, scientific instruments are put into orbit by a vendor who provides a spacecraft, launches it, operates it, and collects the data returned by the instrument, while maintaining sole responsibility and ownership of the mission throughout. The vendor recovers the cost of the mission by providing access to the scientific data over the Internet at a predetermined price. Another promising concept is the GeoQuickRide program at NASA s Rapid Spacecraft Development Office at Goddard Space Flight Center. This program offers opportunities to fly scientific instrument packages on government and commercial geostationary satellites. With an estimated 20 launches per year, geostationary satellites constitute an immense potential resource that could be utilized: Most often, these missions are designed with excess capacity of space, mass, power, and data return rates, which means that the only major cost factor involves integrating the small satellite with the launch package. At present, however, Figure 1. The Poly Picosatellite Orbital Deployer (P-POD). This platform can contain three tiny, free-flying CubeSat spacecraft. These CubeSats are launched separately, allowing them to orbit independent of the P-POD and each other. vendors request of the order of $10 million per mission for this service, and as a result it has been done only a few times. The advent of cheaper alternatives for getting research and technology demonstration missions into space will undoubtedly bring this price down in the future. A similar approach is used in the Active Magnetosphere and Planetary Electrodynamics Response Experiment (AMPERE) undertaken jointly by scientists at Johns Hopkins University Applied Physics Laboratory and the Boeing Service Company to develop a new observational space weather capability. Utilizing magnetometer measurements from the attitude control system on board the constellation of Iridium communications satellites, the project will provide near-real-time global estimates of the electric currents flowing into and out of Earth s upper atmosphere. Perhaps the most promising standardized platform for lowcost, student satellite missions is the Poly Picosatellite Orbital Deployer ( P-POD), developed at California Polytechnic Institute ( see Figure 1). The P-POD is a container for three stacked picosatellites named CubeSats, measuring 10 centimeters on each side and weighing at most 1 kilogram (see page 6). P-PODs are designed to accommodate the CubeSat physical layout and design guidelines, taking advantage of a standard, flight-proven deployment system. A thriving student satellite program has developed over the past 4 years based on this CubeSat standard. Through the establishment of standards and collaboration among university groups worldwide, the CubeSat program has aided the construction and Space Weather Quarterly 9

10 launch of 32 student-built picosatellites since With few exceptions, these were all launched outside the United States. Of the order of 40 universities in the United States are involved in the CubeSat program, which has matured much faster than expected. CubeSat projects are usually done on very low budgets, with hardware costs as low as $50,000 and launch costs of a similar magnitude. The success of CubeSat in attracting and exciting students while producing innovation and new technology clearly illustrates the potential for small- satellite projects to play a crucially important role in educating engineering and science students and in doing real science at low cost ( businesstechnology/060830_ cubesats.html). Space Weather Research Using Small Satellites Small- satellite missions can provide valuable measurements that would help answer important outstanding science questions in space weather. Table 2 provides a list of instruments amenable to launch on small satellites. The table indicates the science topics that each instrument would help address and its readiness in a general sense for being flown on small satellites. Readiness is classified as immediate, meaning that such instruments are already suitable for small- spacecraft missions; near-term, indicating that such measurements are likely to be possible from small satellites within the next 5 years; or later, indicating that these measurements are not expected to be possible within 5 years. The readiness estimates consider the smallness of instruments along with their potential to achieve low enough power and data rate requirements. Table 2 clearly demonstrates the great potential for small satellites to have a significant impact in all areas of space weather science. Educational Benefits of Small Satellites A number of successful programs have offered low-cost access to space, including Skylab, Getaway Specials, and Hitchhiker, all supported by NASA between 1973 and Among other accomplishments, these programs have helped attract young engineers into the space industry and contributed to the continuing strength of the U.S. aerospace enterprise. Obviously, there is no substitute for a hands-on approach to learning the skills and experience needed to succeed in the aerospace industry. This realization is also the motivation behind a number of recent initiatives within the DOD; for example, the U.S. Air Force Office of Scientific Research ( AFOSR) has supported the FalconSat program at the Air Force Academy. Under the motto Learn Space by Doing Space, the goal of this program is to have students design, construct, test, launch, and operate a new microsatellite or nanosatellite every 2 3 years. Since the start of the program in 2000, more than 200 cadets have graduated with real satellite experience. The third satellite in the program, the microsatellite FalconSat 3, was successfully launched in March 2007 as one of the satellites of the STP 1 mission. One objective of this satellite is to investigate ionospheric irregularities. Table 2. Potential Space-Based Space Weather Measurements, Their Scientific Objectives, and Their Readiness for Launch on Small-Satellite Missions Instrument Science Readiness a X-ray and gamma ray sensors Ultraviolet (UV) and extreme ultraviolet (EUV) spectrometers High-resolution imagers Solar and Heliospheric Physics particle acceleration irradiance and coronal structure transition region near-term immediate near-term Global imaging solar eruptions later Small Solar Mass coronal mass ejections (very) near term Ejection Imager (SMEI) (CMEs) Energetic particle solar particles immediate detectors Multisatellite solar wind structure immediate Magnetospheric Physics Plasma instrumentation everything immediate? Energetic particle detectors Magnetic and electric fields Global UV imaging Small-scale UV imaging Visible Imaging System (VIS) Wave magnetometers Plasma instrumentation including impedance probes Neutral instrumentation Magnetic and electric fields Global Positioning System (GPS) and radio beacons Photometers, including spectral and imagers acceleration and transport everything magnetosphereionosphere (M-I) coupling M-I coupling M-I coupling electromagnetic ion cyclotron (EMIC), etc. Ionospheric Physics everything neutral dynamics everything electron density, total electron content, irregularities composition dynamics immediate immediate later immediate immediate near-term? immediate immediate? immediate immediate immediate (on the shelf) Dual small satellites drag immediate a Readiness is classified as immediate, meaning that such instruments are already suitable for small-spacecraft missions; near-term, indicating that such measurements are likely to be possible from small satellites within the next 5 years; or later, indicating that these measurements are not expected to be possible within 5 years. Table courtesy of Michael Hesse, NASA Goddard Space Flight Center. 10 Space Weather Quarterly

11 A similar initiative aimed at the broader civilian academic community is the AFOSR s University Nanosat program. With a mandate to create a space cadre of highly trained U.S. university students, the program supports hands-on teaching of systems engineering through design, assembly, integration, test, launch, and in- orbit operation of student-built satellites. The concept is that every other year, about 10 university proposals are selected to receive $110,000 to design and build a satellite over 2 years. So far, the program has led to one mission launch: Two spacecraft of the Three Corner Sat mission, designed to demonstrate stereoscopic imaging, formation flying, and innovative command and data handling, were launched in December However, because of a sensor glitch on board the launch vehicle, the satellites were placed in a lower-than-expected orbit. This caused mission lifetime to run out before science goals were met. Currently, the third mission of the program, NanoSat 3 (designed to demonstrate autonomous high-precision real-time relative navigation using innovative GPS technologies), is in storage awaiting launch in December 2009, and the fourth mission (also aimed at improving navigation accuracy and demonstrating controlled formation flying) is expected to be ready for delivery (i.e., ready for launch) before the end of A New Opportunity Space-based measurements from small satellites have great potential to advance discovery and increase scientists understanding of space weather. Advances in sensor technology and burgeoning launch opportunities are making small- satellite missions more cost- effective. Scientific satellite missions in the $1 10 million range (including launch costs) are possible, and such missions can help fill important observational gaps. Thus, it is important that the cost of missions be kept as low as possible to avoid getting trapped by risk aversion leading to spiraling programmatic complexities and costs. Equally important, such missions play a crucial role in training the next generation of experimental space scientists and aerospace engineers. Effective training of the next generation of aerospace experts has been hampered by the length of time required to plan and execute large space missions. A small-satellite mission that can be designed, built, flown, and operated in 3 4 years is ideal for student participation, permitting them to experience all aspects of the end-to-end mission process. The natural fascination students have for space exploration will be strengthened by the opportunity to see the scientific output of their small-satellite research projects. Regular access to space, provided by small satellites, will maintain creativity and innovation in space science and aerospace engineering and keep a general widespread interest in space. Furthermore, small satellites will inspire the development of new experimental methods and technology. The space weather community will benefit greatly through this new technology. Therese Moretto and Robert M. Robinson are space weather scientists and program directors at the U.S. National Science Foundation, in Arlington, Va. Citation: Moretto, T., and R. M. Robinson (2008), Small satellites for space weather research, Space Weather, 6, S05007, doi: /2008sw Space Weather is online: Visit Attend the Meeting! The 2008 AGU Fall Meeting will provide an opportunity for over 15,000 researchers, teachers, students, and consultants to present and review the latest issues affecting all areas of the Earth and space sciences. Help shape the meeting and submit your session proposal today. For complete Fall Meeting details, including abstract submission, registration, and housing information, please visit: Space Weather Quarterly 11

12 Spring on the Sun: By Irene Klotz Amateur German astronomer Samuel Heinrich Schwabe was searching for a planet inside Mercury s orbit when he made the serendipitous discovery of the Sun s cycle in Scientists later figured out that the cycle supplies the energy that drives space weather. But anticipating the start of a new 11-year solar cycle is a bit like waiting for spring: It s hard to tell sometimes which will be the year s final winter chill. Characterized by a period of low solar activity (solar minimum, meaning there are fewer observed sunspots) followed by a period of high activity (solar maximum, meaning sunspot occurrence is very frequent), cycles are identified by the magnetic polarity of sunspots. In essence, two sunspots in the same solar hemisphere (north or south) will have the same polarity throughout a given sunspot cycle. Further, in one cycle, the polarity of a sunspot in the northern solar hemisphere will be reversed from the polarity of a sunspot in the southern solar hemisphere. Finally, because the Sun flips its magnetic polarity approximately every 11 years, sunspots oriented one way in a given solar hemisphere will be oriented the opposite direction during the next solar cycle. Such an oppositely oriented sunspot was found early this year, indicating the emergence of a new solar cycle. But 3 months later, spots from the old solar cycle reappeared. We have two solar cycles in progress at the same time, said David Hathaway, a solar physicist with NASA. Solar Cycle 24 has begun, but Solar Cycle 23 has not ended. The new cycle is the 24th to be observed since Swiss astronomer Johann Rudolf Wolf latched on to Schwabe s findings and compiled sunspot records to extend observational data sets back to Because the dynamics of the sunspot cycle determines the intensity and timing of adverse space weather affecting Earth, scientists are eager to nail down their next 11-year forecast. Happy New Year! While some folks recovered from their New Year s Eve hangovers, solar scientists pored over pictures and graphs of a dark spot on the face of the Sun. They had been on the lookout for such a speck for months and would know for sure if it was the chosen one by two defining features: First, unlike spots of recent years that clustered around the Sun s equatorial belly, the new tic would form at a higher latitude. Second, it would be magnetically oriented in the opposite direction of previous spots. In the opening days of 2008, scientists became more and more convinced this spot was what they had been waiting for. Spot 10,981 had the two birthmarks: It was located about where you d find Palm Beach, Fla., if it were on the Sun and was magnetically reverse to the last crop (see Figure 1). The only thing off was the timing: It appeared a couple of months earlier than expected. So the space weather forecasters at the National Oceanic and Atmospheric Administration (NOAA) kept watch for a while to make sure before making the big announcement on 4 January. A new 11-year cycle of heightened solar activity, bringing with it increased risks for power grids, critical military, civilian, and airline communications, GPS signals, and even cell phones and ATM transactions, showed signs it was on its way late yesterday when the cycle s first sunspot appeared in the Sun s northern hemisphere, NOAA wrote in a news release. The new cycle had begun. The Need to Predict Sunspot Occurrence Officially, the new cycle is known as Solar Cycle 24, though of course the Sun, which has been around for about 4.5 billion years, presumably had sunspots long before humans appeared and invented tools to see them. NOAA only started numbering the sunspots in Another 17 years passed before the government convened a panel of solar scientists to make an official prediction of what to expect during a cycle s peak years. That panel, and the two that have convened since, forecast how many spots will appear on the face of the Sun and the time of peak occurrence, as the intensity and number of solar storms depend on the number of sunspots. More is at stake than ever before. Our increasingly technologically dependent society is becoming increasingly vulnerable to space weather, said David Johnson, director of the National Weather Service. When the government first began issuing predictions of solar cycle activity in 1989, not too many agencies and entities beyond NASA were particularly interested. The space agency s greatest 12 Space Weather Quarterly

13 A New Cycle of Sunspots Yohkoh/Japanese Institute of Space and Astronautical Science concern was and remains radiation exposure to astronauts in orbit. Solar storms can douse astronauts with more radiation in a single storm than they typically would get in a year. When crew members approach a government-imposed lifetime limit for radiation exposure (a level determined by age, gender, and other factors), they are grounded. NASA mitigates the radiation threat by, for example, not scheduling spacewalks during periods of solar disturbances. Astronauts are better shielded behind their spaceships walls and insulation. NASA also uses space weather forecasts to plan operations for low-earth- orbiting satellites and observatories such as the Hubble Space Telescope. These spacecraft lose altitude due to friction from atmospheric particles; when these particles are heated during periods of intense solar activity, they balloon into space, creating more resistance for orbiting satellites, which slow down and lose altitude. But times have changed. Modern life has become dependent on services and equipment at risk of interruption and breakdown during outbursts from the Sun. People want to know the timing, when it s going to start, and when the peak is going to be, said Joe Kunches, NOAA s space weather forecast and analysis branch chief. Technologies don t even have to be in orbit to be affected. Geomagnetically induced currents can disable power grids on Earth. High radiation levels in the upper atmosphere and enhanced ionospheric currents, which result from solar emissions, disrupt radio transmissions and can force airline operators to reroute flights, particularly those that cross polar regions. Each rerouting costs between $10,000 and $100,000, according to a National Weather Service report. Even more vulnerable are the hundreds of government, military, and commercial satellites and spacecraft orbiting Earth. During a series of storms in October 2003, 59% of NASA s Earth and space science missions were affected, investigators at NASA Goddard Space Flight Center, in Greenbelt, Md., later determined. The storms likely caused the loss of the $640 million ADEOS 2 spacecraft, a Japanese-built satellite that monitored worldwide environmental changes. Last year, a solar flare interrupted the quiescent, waning weeks of Cycle 23 and found a new prey: the ubiquitous Global Positioning System satellite navigation service. The effects were more profound than we expected and more widespread than we expected, said Paul Kintner, professor of electrical and computer engineering at Cornell University. The flare created a freakishly powerful radio burst at its peak, the flare had 20,000 more radio emissions than the entire rest of the Sun that affected nearly every GPS receiver on the sunlit side of Earth. Knowledge of the Sun s power to wreak havoc on critical technological systems makes many scientists wary of the future. We are concerned more severe consequences will occur during the next solar maximum, Kintner said. To prepare for this, predicting the pattern of the next solar cycle is a primary goal. No Consensus By the time the first solar cycle prediction panel met to forecast Cycle 22 in 1989, the Sun already was at peak intensity. The scientists, or course, didn t know that until after the cycle calmed down. The next panel tasked to make a forecast gathered in September 1996, six months, it turned out, after Cycle 23 dawned. Predictions for cycles 22 and 23 fell short even though in both cases the cycles turned out to be already under way at the time the forecast was made. The Sun, it turns out, is a complex beast. It fell on Douglas Biesecker, a solar physicist with NOAA s Space Weather Prediction Center, in Boulder, Colo., to organize a panel to predict the number of sunspots expected during Cycle 24 and the month of peak activity. Unlike past panels, the 11-member team of experts got to work in 2006, two years early, armed with data from dozens of space- and ground-based solar observations, as well as historical records. For the first time, the panel also had computer simulations of the tortured inner workings of the Sun, the convection patterns and rotational structures that flip its magnetic field every 11 years. More than anything, Biesecker wanted the Solar Cycle 24 Prediction Panel to be unanimous and confident in its prediction. Although the scientists present were confident in their assessments, the 11 voting forecasters were divided into two camps. By one measure, the upcoming cycle is going to be a doozy, with frequent and strong geomagnetic storms that will reach a peak in October 2011, with an average of 140 spots, according to certain studies. The other Space Weather Quarterly 13

14 Figure 1. Sunspot 10,981, the first spot of the Solar Cycle 24 family, shown with sunspot 10,980, a speck associated with Cycle 23. half of the panel expects the Sun to take a bit of a breather, with only roughly 90 spots during August 2012, which would represent solar maximum. Each side is confident its prediction is the correct one. We re all pretty entrenched in our position, Biesecker said. One camp will clearly have egg on its face when this is over. The first test for either prediction is the timing of solar minimum, the finale of Cycle 23. The appearance of sunspot 10,981 in January, which was followed in the last weeks of March by three big sunspots matching the magnetic polarity of the Cycle 23 brood, did nothing to end the debate. Until an uninterrupted sequence of specks with the same magnetic polarization appears, spot 10,981 shows only that a transition is under way. Over the next 2 years, spots from both cycles 23 and 24 will form. The Dynamo Factor The Sun s cycle is generated by the interaction between two plasma flows. The first churns beneath the surface, sending plasma from the poles to the equator, where it rises and streams back to the poles. The second current pulls at the Sun s surface layer, which rotates faster at the equator than it does at the poles. The Sun s massive magnetic field, which runs from pole to pole, gets repeatedly twisted around the equator, periodically creating bright knots of concentrated areas, leading to sunspots, flares, and the most powerful storms in the Sun s repertoire, coronal mass ejections. All these phenomena serve to dissipate the magnetic knots. Charged particles join the streams heading back to the poles, becoming the raw materials for the next solar cycle. We see what the flux is doing at the surface, Biesecker said. The big question is what happens to it once it sinks beneath the surface at the poles. Does it move along like a conveyor belt? Somehow, the residual tension has to be destroyed or sink, or we d see it build up again. NOAA Understanding the inner plasma flows is key to predicting what will happen on the Sun s surface. The plasma traps, redistributes, and reorganizes the solar magnetic fields. By tracing sound waves reverberating inside the Sun, scientists can produce pictures of the interior, in much the same way that ultrasound devices are used to bounce sound waves inside the human body to render images for medical diagnosis or to produce a picture of a fetus. The solar data are compiled into computer models and used to predict the next solar cycle. There are many variants of these so-called dynamo models, but members of the Solar Cycle 24 Prediction Panel fell into two camps: those who believed the models, which favored a stronger cycle and earlier solar maximum, and those who didn t. The models were tested with data from past cycles to see if they could match the actual number of observed sunspots. The results, though impressive, were not enough to convince half the panel the models could predict the future. Everybody believes there is a flow underneath the surface of the Sun, said Biesecker. But is it a fast or a slow process? Those who believe that the cycle will be smaller and weaker follow a more traditional approach to sunspot forecasting, based on statistics. The strength of the Sun s magnetic field and the number of sunspots in one cycle, for example, are key components of predicting the next cycle. At the core of the disagreement is the importance of magnetic fields around the Sun s poles as the previous cycle fades. Late-cycle polar fields are the basis of the method that led to predictions of a weaker Cycle 24. By contrast, the solar physicists who believe that a stronger cycle will manifest look at more data over longer periods of time. The controversy should clear soon: If more Cycle 24 sunspots don t appear by the middle of this year, the solar scientists forecasting a stronger cycle may change their minds. Time Will Tell By the time Cycle 25 rolls around, in 2019 or so, scientists hope there will be a lot less guesswork involved. To prepare for this future need, NASA plans to launch the Solar Dynamics Observatory (SDO) later this year. SDO is designed to determine how the Sun forms and sustains its magnetic field, how it converts magnetic energy into heat and light, what drives the forces behind the solar wind and the release of highly charged particles during flares and other solar storms, and how all these processes affect the solar irradiance the source of virtually all energy that enables and affects life on Earth. SDO, currently targeted for launch in December from Cape Canaveral, Fla., is the debut spacecraft for NASA s Living With a Star initiative. It is a one-way relationship, with the Sun calling the shots, but space weather forecasters hope at least to take away the element of surprise. Irene Klotz is a freelance writer for the American Geophysical Union. Citation: Klotz, I. (2008), Spring on the Sun: A new cycle of sunspots, Space Weather, 6, S05004, doi: /2008sw Space Weather Quarterly

15 Space Weather and the Global Positioning System By Anthea Coster and Attila Komjathy The ability to monitor space weather in near real time is required as our society becomes increasingly dependent on technological systems such as the Global Positioning System (GPS). Certain critical applications including railway control, highway traffic management, emergency response, commercial aviation, and marine navigation require highprecision positioning. As a consequence, these applications require real-time knowledge of space weather effects. In recent years, GPS itself has become recognized as one of the premier remote sensing tools to monitor space weather events. For this reason, Space Weather has opened a special section called Space Weather Effects on GPS. Papers in this section (see the sidebar box on page 17) describe the use of GPS as a monitor of space weather events and discuss how GPS is used to observe ionospheric irregularities and total electron content gradients. Other papers address the implications that these space weather features may have on GPS and on Global Navigation Satellite System (GNSS) District of Columbia Office of Planning (map) and (satellite) Space Weather Quarterly 15

16 operations in general. Space weather impacts on GPS include the introduction of range errors and the loss of signal reception, both of which can have severe effects on marine and aviation navigation, surveying, and other critical real-time applications. To recognize the many important contributions that were made during the pioneering years of the 1980s and early 1990s and to provide some insight as to what can be expected in the years to come, we introduce the GPS special section by reviewing the 25- year history of GPS as an ionospheric monitoring system. This review will help give readers the context they need to understand ideas presented in this special section. Early Use of GPS for Ionospheric Studies The ionosphere is a dispersive medium, meaning that these changes in speed and direction are a function of frequency. GPS was developed by the U.S. Department of Defense with the primary goal of being an all-weather space-based navigation system. GPS system design is heavily dependent on the accuracy of atomic clocks on board the satellites, and the design of this satellite system during the early 1970s took advantage of advances in clock technology. However, the GPS signals must transit the ionosphere to communicate with ground receivers. This transit introduces signal propagation errors because the ionosphere affects the propagation speed and direction of all radio signals (including GPS). In addition, electron density irregularities in the ionosphere can introduce amplitude and phase fluctuations, a process known as scintillation. The ionosphere is a dispersive medium, meaning that these changes in speed and direction are a function of frequency. They are also proportional to the varying electron density along the line of sight between the receiver and the satellite. Thus, the cumulative effect at the receiver is proportional to the total electron content, or TEC, which is equal to the total number of electrons in a column with a cross-sectional area of 1 square meter along the line of sight between the satellite and the receiver. For GPS, the change in propagation speed introduces both a range delay, the equivalent of measuring a slightly longer distance to the satellite than is actually the case, and a phase advance in its observables. To obtain very accurate positions from GPS, this ionospheric delay/advance must be removed. To this day, the ionosphere remains the largest error source in the GPS navigation solution. The GPS system was designed to operate at two frequencies, L1 ( megahertz) and L2 ( megahertz). While the majority of handheld GPS receivers operate only on the L1 frequency, higher-quality GPS receivers, such as those used by the geodetic community, operate on both frequencies and are designed to take advantage of the frequency dependence of the ionospheric terms. If the GPS range and phase can be measured at both frequencies, the ionospheric range delay and phase advance can be computed. For the ionospheric scientist, this also means that the TEC can be measured almost exactly. Ionospheric scientists were quick to recognize that the GPS dual-frequency measurements could be used to measure the ionospheric TEC at multiple locations. As soon as the first experimental GPS satellites were launched (between 1978 and 1985), scientists began using GPS signals to monitor the ionosphere. There were 10 experimental satellites (the GPS Block I) and they formed the GPS Demonstration System, which was designed to test and validate the use of a satellite-based system for real-time global navigation and timing. The early GPS measurements, collected primarily using stand-alone receivers, were compared with ionospheric measurements taken with other instruments, such as Faraday rotation sensors or incoherent scatter radar platforms. GPS was also quickly recognized as a useful instrument for the study of propagation parameters and scintillation. Starting in 1989, the first fully operational GPS satellites were launched. These satellites were launched into a 55º inclination orbit, lower than the 63º inclination of the Block I orbits. Furthermore, the operational satellites had the full suite of signal capabilities, including selective availability (SA) and antispoofing (AS), both of which deliberately degraded the precision of the GPS signals for nonmilitary users. SA involved destabilizing the satellite oscillators and altering broadcast orbit and clock parameters. Typical range accuracies for civilians with GPS SA turned on were around 100 meters. SA was permanently turned off in May Of more importance to ionospheric scientists was the implementation of AS, which was turned on full time in early Essentially, AS involves an encryption of the GPS code on the L2 frequency, and only users with a decryption key are allowed full access to the code. However, cross-correlation techniques were developed that allowed receiver manufacturers to reproduce a degraded GPS L2 signal with significantly less gain than GPS L1. Ionospheric scientists have based most of their GPS TEC measurements on this degraded GPS L2 signal. With GPS modernization, a new civilian signal that is not encrypted is being broadcast on L2. This new signal, called L2C, is available on the new GPS Block IIR-M satellites, which are in the process of being launched. As part of the modernization program, GPS satellites will also be broadcasting another civilian signal at a third frequency, L5 ( megahertz). The first GPS satellite with this capability will be launched in the near future. Issues in the Use of GPS to Measure the Total Electron Content By the mid-1980s, two primary issues were recognized with the use of GPS data to measure ionospheric TEC. Both fall under the category of unknown signal contributions and occur due to either multipath or system hardware differential delays or biases. Multipath delays are propagation phenomena that result in radio signals reaching the receiving antenna by two or more paths. Multipath delays can result in GPS calculating its position erroneously, and 16 Space Weather Quarterly

17 can make the determination of the absolute TEC value extremely difficult. This issue is considerably worse at low elevation angles. GPS multipath issues were described by Bishop et al. [1985, 1994] and remain a concern, although multipath may be mitigated by the use of higher elevation cutoffs and the use of choke-ring antennas. The second significant issue is the additional differential delay between the two GPS frequencies introduced by the receiver and satellite hardware. These differential delays, commonly referred to as the receiver and satellite biases, can be significant and if not removed correctly, can corrupt the GPS TEC measurements. Initially, the community was uncertain how to estimate either the receiver or the satellite biases. Work at the NASA Jet Propulsion Laboratory led to initial algorithms for bias estimation. Other methods were developed using single receivers [and were followed by more powerful techniques based on the global network of receivers. The estimation of satellite and receiver biases remains an area of significant and active investigation within the ionospheric community. New and enhanced techniques have been recently developed that estimate receiver differential biases for all available GPS stations (typically around 1000 sites) on a daily basis. The research community needs more efficient and improved estimation algorithms to properly perform process and quality checks on the large amount of GPS data currently available on a daily basis. International GPS Service The International GPS Service (IGS, now the International GNSS Service) network of GPS receivers played a central role in developing the use of GPS for monitoring the ionosphere. The concept for the IGS was initiated in 1989 by a group of scientists affiliated with the International Union of Geodesy and Geophysics. These scientists recognized that because scientific investigations require the highest-precision data possible, GPS offered tremendous potential for use in Earth science research provided that all errors, including those introduced by signal propagation, are minimized. To understand what is meant by the term highest-precision data, it is simplest to consider how the GPS position estimate is obtained. The GPS estimate of a user s position depends on knowledge of (1) the position of the GPS satellites, (2) the time as measured by the satellite and receiver clocks, and (3) the estimated ranges from different GPS satellites to the GPS receiver. To obtain the highest-precision GPS measurements, all of the above must be also known with the highest precision possible. The GPS community achieves this by estimating the precision ephemerides for the GPS satellites (a precision ephemeris is essentially a table of values that provides an extremely accurate position of a GPS satellite as a function of time) and additional clock information (such as the clock drift rates for the atomic clocks on board the GPS satellites and the receiver s clock offset). A global network of receivers was (and is) needed to best estimate these parameters. The requirement of a global network can be understood in context of the GPS satellite orbits. To improve upon the GPS orbits, data need to be collected at multiple points along the entire orbit. Observations collected in a single hemisphere or country alone can not provide this. With support from numerous scientific organizations, the IGS was founded in 1992, with a network of about 20 geodetic receivers worldwide. At its tenth anniversary, IGS consisted of over 200 actively contributing organizations in more than 80 countries and a global network of more than 350 stations. IGS began around the same time as the World Wide Web. The concept of the World Wide Web was invented in the late 1980s. By the end of 1992, there were over 50 Web servers in the world, many located at universities or other research centers. For the IGS, the World Wide Web enabled the easy transfer of data files and Space Weather Effects on GPS Listed below are the papers included in this special section as of the date of this magazine s publication. These papers (and others that will be published in the future) are available in their entirety at ntent=specialsections&ssid=gps1. Bust, G. S., G. Crowley, T. W. Garner, T. L. Gaussiran, R. W. Meggs, C. N. Mitchell, P. S. J. Spencer, P. Yin, and B. Zapfe (2007), Four-dimensional GPS imaging of space weather storms, Space Weather, 5, S02003, doi: /2006sw Coster, A., and A. Komjathy (2008), Space weather and the global positioning system (GPS), Space Weather, 6, S05DXX, doi: /2008SW Feltens, J. (2007), Development of a new three-dimensional mathematical ionosphere model at European Space Agency/ European Space Operations Centre, Space Weather, 5, S12002, doi: /2006sw Kintner, P. M., B. M. Ledvina, and E. R. de Paula (2007), GPS and ionospheric scintillations, Space Weather, 5, S09003, doi: /2006sw Skone, S., and A. Coster (2008), Potential for issuing ionospheric warnings to Canadian users of marine DGPS, Space Weather, 6, S04D03, doi: /2007sw Skone, S., and R. Yousuf (2007), Performance of satellite-based navigation for marine users during ionospheric disturbances, Space Weather, 5, S01006, doi: /2006sw Smith, A. M., C. N. Mitchell, R. J. Watson, R. W. Meggs, P. M. Kintner, K. Kauristie, and F. Honary (2008), GPS scintillation in the high arctic associated with an auroral arc, Space Weather, 6, S03D01, doi: /2007sw Steenburgh, R. A., C. G. Smithtro, and K. M. Groves (2008), Ionospheric scintillation effects on single frequency GPS, Space Weather, 6, S04D02, doi: /2007sw Space Weather Quarterly 17

18 products. More important, the World Wide Web helped GPS researchers organize their standardized products and services, such as the Receiver Independent Exchange Format (RINEX) files. The importance of the IGS service to developing the use of GPS for ionospheric monitoring purposes and for studying the structure of worldwide space weather events cannot be overstated. The IGS service provides high-quality data in a standard format that are freely available and easily accessible through the World Wide Web to all scientists at With the establishment of the IGS network, ionospheric scientists began investigating the use of producing maps of TEC based on the global network of receivers. The ionospheric working group of IGS was established in Currently, four Ionospheric Associate Analysis Centers (IAACs) contribute with their rapid and final vertical TEC (VTEC) maps to the IGS products. The four IAACs are the NASA/California Institute of Technology Jet Propulsion Laboratory, the Center of Orbit Determination in Europe (CODE), the European Space Agency (ESA), and the Technical University of Catalonia (UPC). IAACs compute the global distribution of TEC independently using different models. As an IGS final product, the four independently derived VTEC maps are combined for the IGS community. This global TEC information is used for purposes such as calibration of single-frequency GPS receivers and altimeters, and investigations of the global temporal and spatial behavior of ionospheric TEC. Data from about 1000 GPS receivers are currently available on a daily basis to monitor the temporal and spatial variability of the global ionosphere. These receivers include networks, such as the Continuously Operating Reference Stations (CORS), in addition to the IGS network of receivers. Algorithms have been developed to process all of these data sets in a time-efficient manner, enabling daily monitoring of the quiet and storm-time ionosphere that affects satellite-based radio navigation systems such as GPS, the Russian Global naya Navigatsionnaya Sputnikovaya Sistema (GLONASS), and Galileo, a new European satellite navigation system that is currently under development. Real-Time Ionospheric Monitoring Systems Real-time ionospheric monitoring systems based on GPS data were first developed in the early 1990s. In 1994, work began on a larger networked real-time system called the Wide Area Augmentation System (WAAS). WAAS was developed jointly by the U.S. Department of Transportation (DOT) and the Federal Aviation Administration (FAA) with the goal of becoming the future primary means of civil air navigation. WAAS is designed to augment GPS without WAAS, clock drift, satellite orbit errors, and ionospheric effects including disturbances create undesirable error and uncertainty in the GPS signal. WAAS is needed to meet the very strict civil aviation requirements for integrity, accuracy, availability, and continuity. WAAS is based on a network of ground-based GPS reference stations in the North American continent that continuously measure ionospheric slant delays (the line of sight range delays between the GPS receiver and the satellite). All of these data are transferred to master stations, where the integrity of the system is assessed and the measurements are combined to compute ionospheric vertical delays (the delay due to the ionosphere looking directly overhead) for a virtual set of ionospheric grid points (IGPs) that is spaced over much of North America. By using a bilinear interpolation algorithm, WAAS avionics use the IGPs to estimate the adjustments needed to correct for receiver-to-satellite ionospheric delays. Although WAAS was the first GPS-based augmentation aviation system in operation, similar systems are currently in development in, for example, Europe, Japan, and India. Collectively, these systems are called Satellite Based Augmentation Systems (SBAS). Ionospheric Storm Detection Prior to 2000, the density of GPS receivers in the IGS network was relatively sparse, forcing scientists to integrate GPS TEC data into ionospheric models to produce continuous TEC maps. In 2001, the Massachusetts Institute of Technology Haystack Observatory was the first group to make use of all available GPS data to produce strictly data-driven plots of the TEC using no underlying Data from about 1000 GPS receivers are currently available on a daily basis models to smooth out gradients. Because of this lack of smoothing, the narrow plumes of storm-enhanced density (SED) that form over the United States during geomagnetic storms could clearly be observed in the GPS TEC maps, first reported by Coster et al. [2001]. In a ground-breaking paper, Foster et al. [2002] linked the narrow plumes seen in GPS TEC observations as measured from the ground with the plasmaspheric plumes as seen from NASA s Imager for Magnetopause-to-Aurora Global Exploration (IMAGE) satellite in space at 8 Earth radii. This last observation clearly demonstrated that a dense network of ground-based receivers could play a significant role in measuring the coupling between the ionosphere and the magnetosphere. GPS TEC maps produced by this method have now been widely disseminated throughout the atmospheric research community and have become one of the standard means to study the effects of geomagnetic storms. Space-Based Ionospheric Measurements Ground-based GPS receivers allow for good data coverage over land, but not over the oceans. As proposed by Hajj et al. [1994], putting GPS receivers in space is one way of addressing this lack of coverage. The idea of using GPS receivers in space to sense properties of the atmosphere grew out of earlier work in the remote sensing of planetary atmospheres in the 1970s, the atmospheres of Mars, Venus, and Jupiter were probed using the technique of radio occultation. Planetary occultation is when a smaller astronomical body passes behind a larger astronomical body, wholly obscuring 18 Space Weather Quarterly

19 its view. Similarly, radio occultation can be thought of as when the line of sight to a satellite is obscured by a planet, or in the case of GPS, the Earth. For GPS, a satellite in low-earth orbit (LEO) tracks the signal from a GPS satellite. The LEO satellite is typically in an orbit between 500 and 800 kilometers, and GPS satellites are in nearly semisynchronous orbits (they travel around the Earth approximately twice a day) at an altitude of approximately 20,200 kilometers. From the perspective of the LEO satellite, the GPS satellite rises or sets several times a day. As the occultation occurs, the signal that is measured from the GPS satellite is refracted, or bent, by differing amounts as it propagates through different layers of the atmosphere. By measuring the amount of refraction as the GPS satellite is rising or setting, scientists are able to reconstruct properties of the different layers of the atmosphere (e.g., relative humidity and temperature profiles in the troposphere, and electron density in the ionosphere). The Global Positioning System/Meteorology (GPS/MET) experiment was designed as a proof of concept mission to demonstrate that GPS signals occulted by the Earth s atmosphere could be used to measure properties of our atmosphere and ionosphere. The GPS/MET mission lasted from 3 April 1995 to March 1997 and was highly successful. Numerous ionospheric studies were reported based on GPS/MET data. The success of GPS/MET spawned a number of other radio occultation missions to provide ionospheric measurements over oceanic regions, including the Argentine Satelite de Aplicanciones Cientificas-C (SAC-C), the U.S.-funded Ionospheric Occultation Experiment (IOX), and Germany s Challenging Minisatellite Payload (CHAMP). The joint U.S./Taiwan Constellation Observing System for Meteorology, Ionosphere and Climate (COSMIC; a new constellation of six satellites, nominally provides up to 3000 ionospheric occultations per day (Figure 1). COSMIC measurements have been provided officially for public use since 28 July Researchers have begun integrating COSMIC-derived TEC measurements with ground-based GPS TEC data and assimilating these data into models (such as JPL s Global Assimilative Ionospheric Model (GAIM), originally sponsored by the U.S. Department of Defense) so that three-dimensional global electron density structures and ionospheric drivers can be estimated. The large amount of space-based measurements provided by the current COSMIC mission have made this possible. Future Directions The advent of real-time global ground- and space-based GPS measurements is expected to revolutionize the accuracy of ionospheric specification, nowcast, and forecast. Recently, the NOAA Space Weather Prediction Center developed a new data assimilation product that characterizes ionospheric TEC over the United States. In the next 10 years, the real-time characterization of the global ionosphere is expected to become a standard product. This characterization will rely heavily on data from GPS measurements, but it will be enhanced by the real-time measurements from other sensors. Figure 1: A day s worth of COSMIC soundings (green) compared to existing soundings from weather balloons (red) shows how spatial coverage of the ionosphere over the oceans is dramatically increased through COSMIC. This constellation nominally provides up to 3000 ionospheric occultations each day. Illustration by Bill Schreiner, UCAR. As technology advances, societies of tomorrow are expected only to increase their need for highly accurate communications and navigation systems. Through collecting new data and finding new ways of analyzing ground- and space-based GPS data to minimize signal propagation errors, scientists and operators will be sure to meet these future needs. The full text of this feature article, including references, is available online at Anthea Coster works on measuring and understanding the geophysics of adverse space weather at the Massachusetts Institute of Technology s Haystack Observatory in Westford, Mass. Attila Komjathy is a space weather scientist at the NASA/California Institute of Technology Jet Propulsion Laboratory in Pasadena, Calif. References Bi shop, G. J., J. A. Klobuchar, and P. H. Doherty (1985), Multipath effects on the determination of absolute ionospheric time delay from GPS signals, Radio Sci., 20, Bi shop, G., D. Coco, P. Kappler, and E. Holland (1994), Studies and performance of a new technique for mitigation of pseudorange multipath effects in GPS ground stations, in Proceedings of the 1994 ION National Technical Meeting, pp , Inst. of Navig., Fairfax, Va. Co ster, A. J., J. C. Foster, P. J. Erickson, F. J. Rich (2001), Regional GPS mapping of storm enhanced density during the July 2000 geomagnetic storm, in Proceedings of ION GPS 2001, pp , Inst. of Navig., Fairfax, Va. Fo ster, J. C., P. J. Erickson, A. J. Coster, J. Goldstein, F. J. Rich (2002), Ionospheric signatures of plasmaspheric tails, Geophys. Res. Lett., 29(13), 1623, doi: /2002gl015067, Ha jj, G. A., R. Ibanez-Meir, E. R. Kursiniski, and L. J. Romans (1994), Imaging the ionosphere with the Global Positioning System, Int. J. Imaging Syst. Technol., 5, Citation: Coster, A., and A. Komjathy (2008), Space weather and the Global Positioning System, Space Weather, 6, SD6001, doi: /2008sw Space Weather Quarterly 19

20 SPACE WEATHER, VOL. 6, S04D02, doi: /2007sw000340, 2008 Ionospheric scintillation effects on single frequency GPS R. A. Steenburgh, 1 C. G. Smithtro, 2 and K. M. Groves 3 Received 21 May 2007; revised 26 November 2007; accepted 26 December 2007; published 3 April [1] Ionospheric scintillation of Global Positioning System (GPS) signals threatens navigation and military operations by degrading performance or making GPS unavailable. Scintillation is particularly active within, although not limited to, a belt encircling the Earth within 20 degrees of the geomagnetic equator. As GPS applications and users increase, so does the potential for degraded precision and availability from scintillation. We examined amplitude scintillation data spanning 7 years from Ascension Island, U.K.; Ancon, Peru; and Antofagasta, Chile in the Atlantic/American longitudinal sector as well as data from Parepare, Indonesia; Marak Parak, Malaysia; Pontianak, Indonesia; Guam; and Diego Garcia, U.K. in the Pacific longitudinal sector. From these data, we calculate percent probability of occurrence of scintillation at various intensities described by the S 4 index. Additionally, we determine Dilution of Precision at 1 min resolution. We examine diurnal, seasonal, and solar cycle characteristics and make spatial comparisons. In general, activity was greatest during the equinoxes and solar maximum, although scintillation at Antofagasta, Chile was higher during 1998 rather than at solar maximum. Citation: Steenburgh, R. A., C. G. Smithtro, and K. M. Groves (2008), Ionospheric scintillation effects on single frequency GPS, Space Weather, 6, S04D02, doi: /2007sw Introduction [2] The term scintillation is used to describe the fluctuation of electromagnetic wave intensity brought about by various phenomena. A familiar example is the scintillation of visible light from stars by the Earth s atmosphere. Scintillation of GPS signals begins after sunset at the geomagnetic equator when plasma depletions or bubbles develop in response to ionospheric instability. The bubbles grow as high as 1500 km, diffusing north and south along geomagnetic field lines. This phenomenon has been described in detail by Kelley [1989] and more recently by Sultan [1996] and Eccles [1998]. The work of de La Beaujardiere et al. [2004] summarizes what is currently known and unknown about the processes leading to the development of the instability and the resulting depletions. Electromagnetic waves transmitted from satellites are diffracted as they pass through these bubbles. The amplitude and phase of the resulting signal can fluctuate wildly at the receiver. Models and measures have been created [see, e.g., Fremouw et al., 1980; Rino, 1979a, 1979b], and evaluated [Beach et al., 2004] to describe the effect of scintillation on electromagnetic waves. A comprehensive 1 Space Weather Prediction Center, Boulder, Colorado, USA. 2 Air Force Institute of Technology, Wright-Patterson Air Force Base, Ohio, USA. 3 Air Force Research Laboratory, Hanscom Air Force Base, Massachusetts, USA. treatment of scintillation can be found in the work of Wheelon [2001, 2003]. [3] We determined the number of GPS satellites affected by scintillation using coarse acquisition (C/A) code transmissions on the L1 frequency ( MHz). We found that nocturnal low-latitude ionospheric scintillation can disrupt single frequency GPS receivers enough to significantly degrade precision or render GPS unavailable. This phenomenon occurs in a region that encompasses a third of the Earth s surface. The spatial and temporal characteristics of scintillation have important implications as the number of GPS users and applications continues to increase. 2. Background and Approach [4] In this section we begin by discussing the evolution of ionospheric conditions leading to scintillation and summarizing some of what is known about the temporal and spatial characteristics. A brief analysis of a typical night of scintillation at Ascension follows. Finally, the procedures used to analyze the data are presented Equatorial Plasma Depletions Lead to Scintillation [5] Before examining the implications of scintillation, we briefly review its origins. In the E-region of the ionosphere, tidal winds drive an eastward dynamo current in the sunlit hemisphere [Kelley, 1989, chap. 3]. The resulting 20 Space Weather Quarterly

21 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Table 1. Temporal and Spatial Characteristics of Equatorial Plasma Bubbles and Scintillation Based on Remote Sensing and in Situ Data a Characteristic Value Temporal Characteristics Solar/seasonal sensitivity parallels solar cycle but no short term solar flux dependence; worst at equinox Geomagnetic sensitivity no correlation with Kp or Dst Individual bubble duration 1:20 h Interval between formation 2:40 h Number of bubbles d Daily cycle begins 1h after sunset, peaks 2200L, dissipates by sunrise Spatial Characteristics Latitudinal behavior worst under Appleton Anomaly crests, less at geomagnetic equator Longitudinal extent of 1000s km activity Longitudinal EPB interval km Horizontal velocity m/s Vertical velocity m/s, 40% reach m/s Vertical extent km Trajectory drift eastward, tilt westward with height Decay fossil bubbles form as spread F and upward drift ends Density variation up to 10 2 lower plasma density inside EPB a Compiled from Schunk and Nagy [2000], Dandekar and Groves [2004], Immel et al. [2003], and K. Groves (personal communication, 2006). eastward electric field is perpendicular to the geomagnetic field lines, which in turn are nearly horizontal at the geomagnetic equator. The resulting E B drift drives the plasma upward. The plasma then diffuses along B field lines under the influence of gravity and pressure gradients. The plasma accumulates north and south of the dip-equator. The accumulation of plasma poleward of the geomagnetic equator is known as the Appleton Anomaly. [6] As Earth rotates into darkness, dramatic changes take place which can lead, ultimately, to scintillation. At the boundary between day and night (the terminator), the eastward electric field increases in response to increasing neutral winds; polarization charges within conductivity gradients at the terminator enhance the eastward electric field about an hour after sunset... [de La Beaujardiere et al., 2004] before becoming westerly. The upward E B drift intensifies, raising the F layer. This is called the prereversal enhancement. [7] Photoionization is greatly diminished after sunset and the F region is quickly eroded from below as recombination takes place. The combination of lift and bottomside erosion results in steep upward density gradient forces at the base of the F layer, directly opposed by gravity. This configuration is described by Rayleigh- Taylor instability, in which a heavier fluid is supported by a lighter fluid. If perturbed, the fluid turns over and creates low-density structures (bubbles) that rise through the denser fluid, and in the American sector, drift eastward. In the equatorial ionosphere, they are known as equatorial plasma bubbles (EPBs) although their shape more closely resembles elongated tubes. [8] EPBs have been studied using remotely sensed [see, e.g., Immel et al., 2003; Cervera and Thomas, 2006; Dandekar and Groves, 2004], in situ [Burke et al., 2004a, 2004b; Gentile et al., 2006a, 2006b; Su et al., 2006] and multipoint observations [Burke et al., 2003]. Some characteristics of EPBs and scintillation are shown in Table 1. Table 1 indicates scintillation at GPS frequencies in equatorial regions is typically a nocturnal phenomenon beginning about an hour after local sunset and lasting until dawn. The magnitude of scintillation is greatest during the equinoxes and during the solar maximum. Scintillation is more pronounced at the Appleton Anomaly crests. We examined GPS scintillation with particular attention to these temporal and spatial characteristics. Amplitude scintillation is characterized by the S 4 index, the ratio of the standard deviation of the signal intensity to its mean and can be written: pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi hi S 4 ¼ 2 i hi 2 i hii [9] Alternatively, Davies [1990] presents a model of the S 4 index based on physical parameters: S 4 ¼ Bl 3 ZLK sec 2 chdn 2 e i where B is a parameter that depends on geometrical factors, fundamental constants, etc., l is the wavelength of the carrier, Z and L are the layer thickness and height, respectively, c is the angle between the vertical and the ray, and hdn e 2 i is the mean square deviation of the electron number density. As well as directly linking S 4 to electron number density, this equation shows S 4 increasing as the path nears the horizon. S 4 values are typically below 0.05 during quiet periods and can exceed 0.7 during extreme scintillation. [10] The Nakagami probability density function relates S 4 values to the fade depth in db [Fremouw et al., 1980], a more familiar and operationally applicable quantity. At the 99th percentile, S 4 values of 0.3, 0.5, and 0.8 correspond to fade depths of 3.6, 6.9, and 13.7 db, respectively. [11] This means, for example, given S 4 = 0.3, a randomly sampled value of fade depth has a one percent chance of exceeding 3.6 db. The S 4 values, 0.3, 0.5, and 0.8 represent moderate, high, and extreme levels of scintillation respectively, although the particular adjectives used to describe severity vary. [12] Anderson et al. [2004] found an interesting relationship between S 4 measured at low-latitude stations and the ð1þ ð2þ Space Weather Quarterly 21

22 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Table 2. SCINDA Site Characteristics a Location Latitude, deg Longitude, deg Years Elevation, m Dip Angle, deg Ancon Antofagasta , Ascension Marak Parak Guam Pontianak Parepare Diego Garcia a Negative signs indicate south latitude or west longitude. Magnetic data from International Geomagnetic Reference Field (IGRF) and are approximately centered on the data period. magnitude of the E B drift velocity: If the E B drift velocity is less than 20 m/s between 1830 and 1930 LT, then S 4 will be less than 0.5 [representing a fade depth of 6.9 db at the 99th percentile], 85% of the time and if E B drift is greater than 20 m/s, S 4 will be greater than 0.5 about 90% of the time. [13] To conduct our investigation, amplitude scintillation data spanning 7 years were collected from eight GPS receiver sites encircling the Earth at low latitudes. Data was collected by a NovAtel 11-channel narrow correlator receiver described by Van Dierendonck et al. [1996]. This receiver has the ability to track multiple GPS satellites through strong scintillation. Signals were sampled at 50 Hz and for each minute an S 4 value was computed and recorded. The data was detrended and noise corrections applied. The elevation angle dependence of S 4 was not removed. Since the data were all processed in the same way, the elevation angle dependence appears similarly throughout the data. From a user perspective, the retention of the elevation angle dependence presents a more complete picture of the potential impacts of scintillation. [14] The characteristics of the receiver sites are shown in Table 2. The stations were grouped by longitudinal sector and ordered by increasing magnetic dip angle. Ancon, Antofagasta, and Ascension fell in the Atlantic and Americas sector, hereafter referred to as AA. The remainder of the stations fell in the Pacific and Indian Ocean sector, hereafter referred to as PIO. Ancon and Marak Parak lie close to the geomagnetic equator as shown by their small dip angles. Conversely, Ascension and Diego Garcia lie farthest from the geomagnetic equator Scintillation During One Evening at Ascension [15] Figure 1 shows the (top) elevation and (bottom) S 4 values recorded each minute for two GPS satellites, identified as PRN 15 and 17, during the evening of day 289 at Ascension Island in Each GPS satellite is assigned a particular Pseudo Random Noise (PRN) code which is used as an identifier. At 1956 UT, S 4 values approach 0.8, indicating extreme scintillation, prior to data gaps affecting PRN 17. In Figure 1, S 4 values are truncated at 1.4; values above 1.4 are considered questionable. The combination of high S 4 values and high elevation angle suggest scintillation caused the outages. At 2117 UT, another gap is observed in conjunction with high S 4 values. However, this gap begins at an elevation below 20 where the Figure 1. (top) Elevation and (bottom) S 4 values recorded each minute for two GPS satellite links, identified as PRN 15 and 17, during the evening of day 289 at Ascension Island in The UT hour is indicated on the abscissa. At high elevations, dropouts typically result scintillation, as indicated by the high S 4 values. At elevations below 20, multipath (the reception of a GPS signal from several directions) is suspected. 22 Space Weather Quarterly

23 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 2. The average number of satellite links with S 4 > 0.3, 0.5, and 0.8, respectively (solid), on day 289 at Ascension Island, UK in 2001, averaged over 15 min and rounded. The UT hour is indicated on the abscissa. The dashed line shows the total number of GPS satellite links during the same period. The values were calculated by collecting 15 min of 1-min data from the receiver and taking the averages for each S 4 bin. influence of multipath complicates interpretation. Multipath occurs when a signal is received from several directions because of reflections from low-level obstructions [Hofmann-Wellenhof et al., 2001]. [16] As a precautionary measure, researchers have excluded data below elevations up to 40 [Thomas et al., 2004], although 15 seems more common [see, e.g., Datta- Barua et al., 2003]. Data below 15 were not used for this research. During the data gaps, no S 4 values can be recorded. [17] In general, only a small number of satellites are affected simultaneously by scintillation. This is a consequence of the geometry of the GPS constellation, the scale of the plasma depletions, and the relative motion of the plasma depletions and the satellites. In this example, PRN 15 lags PRN 17 by approximately 15 and is slightly west. The S 4 data suggest that as PRN 17 exits the plasma irregularity, PRN 15, following close behind, enters the irregularity and data gaps begin to appear. This example demonstrates how satellites with similar orbits can be affected in rapid succession. [18] Typically observations from four satellites are used to obtain a GPS navigation solution; receivers can routinely acquire signals from more than ten simultaneously. The geometry of the GPS constellation and its relation to the receiver site, as well as the characteristics of the site and equipment, determine the number of satellites available. The position of the constellation relative to Earth changes throughout the year causing a fluctuation in the number of available satellites. This occurs because the GPS ground track repeats every sidereal day, the time it takes the Earth to rotate once on its axis relative to inertial space (the stars), not every solar day. Since the sidereal day is shorter than the solar day by 4 min, a given satellite will become available 4 min earlier each day. This results in a shift in the constellation over the course of a year. [19] Figure 2 shows the average number of satellites whose signals are received with S 4 > 0.3, 0.5, and 0.8, respectively (solid line), during the evening of day 289 at Ascension Island, UK in The dashed line shows the total number of satellites whose signals were received during the same period. The values were calculated by collecting 15 min of 1-min data from the receiver and rounding the averages for each S 4 bin. For 15 min between 2115 UT and 2130 UT, 80 percent have S 4 > 0.5. [20] For 90 min spanning 2115 UT to 2245 UT, between 60 and 80 percent of available satellites at Ascension are experiencing S 4 > 0.3. In general, the number of satellites simultaneously received with S 4 > 0.8 is low. This is consistent with outage statistics showing a direct relationship between the maximum S 4 reported and the variance of S 4 before, during, and after an outage [Steenburgh, 2007]. In other words, a minority of available satellites typically experiences extreme scintillation. [21] The geometry of the GPS constellation relative to the receiver determines the precision of the position estimate. This influence is quantified by the dilution of precision (DOP), which links measurement and position errors. In practice, a DOP value of 1 is very good and 10 is generally considered very poor, although the relative badness depends upon the particular circumstances. Variants of DOP include geometric DOP (GDOP), vertical DOP (VDOP), and horizontal (HDOP). We limited our focus to HDOP which is useful for military applications such as search and rescue and for civilian applications such as deep sea drilling. A thorough description of DOP calculation can be found in the work of Hofmann-Wellenhof et al. [2001]. [22] Scintillation influences DOP by disrupting GPS transmissions and causing the receivers to lose lock, thereby reducing the number of satellites which are available to create a navigation solution. Depending upon the geometry and number of satellites available, the impact on HDOP of losing a GPS satellite can range from bothersome to catastrophic. Recall from section 2.2 that data below 15 were not used for this research in an attempt to limit contamination of the data by multipath effects. It is important to note that even in the absence of scintillation, this exclusion will result in higher HDOP values than might be obtained otherwise, assuming no degradation from multipath Data Processing [23] Our aim was to determine the number of GPS satellites exceeding S 4 thresholds 0.3, 0.5, and 0.8 at the 50th, 75th, and 95th percentiles as a function of solar cycle, season, and time of day and to determine the resulting effect on HDOP. The percentiles in this case represent a 50, 25, and 5 percent chance of the number of satellites exceeding a given S 4 threshold will be higher than the Space Weather Quarterly 23

24 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Table 3. Summary Plot Data Combinations Showing Years Used and Grouping of 1-min Data a Solar Cycle Seasonal Diurnal Ancon Antofagasta Ascension Marak Parak Guam Pontianak Parepare Diego Garcia One minute 15-min block all 15-min block data Weeks and number obtained. For example, if the analysis reveals four satellites exceed an S 4 of 0.3 at the 95th percentile, then under the same geophysical conditions on a different day, there is only a 5 percent chance of observing more than four satellites with S 4 > 0.3. Thus the combination of an S 4 threshold of 0.3 at the 95th percentile represents the most conservative estimate of scintillation impact. We processed the data as follows: [24] 1. A 15 mask was applied to all data to reduce the possibility of multipath contamination. [25] 2. Times were adjusted to be relative to local sunset and wrapped (see below). [26] 3. S 4 values from 1 h before sunset to 7 h after sunset were assembled based upon the timescale being examined as indicated in Table 3. [27] 4. Counts were taken of the number of S 4 values exceeding the 0.3, 0.5, and 0.8 thresholds for a given period. [28] 5. For each S 4 threshold, the data were binned according to S 4 thresholds and the 50th, 75th, and 95th percentiles calculated for each bin and rounded to integer values. all and a For the solar cycle comparison, individual years were analyzed. For the other comparisons, the data from all the years shown was combined. One-minute data were grouped into 15 min blocks or simply combined. For the seasonal comparison, all weeks were treated individually. For the solar cycle and diurnal comparison, weeks and (centered on the equinoxes) were used. [29] The UT stamps were converted to local sunset relative times for easy comparison between sites. For some locations, data spanned UT midnight; in other words, a single night s scintillation was spread over 2 UT days. [30] Consequently, there was a portion of data in the early morning hours on a given UT day, a large gap during daylight, then more data beginning in the UT evening. Data prior to sunrise on a given UT day was wrapped around to the end of that UT day by adding 24 h to the local sunset relative times. [31] Table 3 describes how the data were assembled. For the solar cycle comparison, individual years were analyzed. One minute data for weeks and (straddling the March and September equinoxes) were combined and sorted in to 15 min blocks and analyzed as described above. With a minimum of four days of data required for each of the 6 weeks and, assuming data from at least four satellites each minute in every 15 min block, a minimum of 1440 observations were used per block. The years used for comparison were chosen to provide one year of data during the peak year(s) of the solar cycle, either 2000 or 2001 and one off-peak year. The particular choices were driven by a desire to balance the availability of data during a particular year with a desire to choose years with similar levels of solar activity (as indicated by smoothed monthly F10.7 cm radio flux). [32] For seasonal comparisons, 1-min data from the years shown in Table 3 were combined and analyzed for each seven day period. Again, a minimum threshold of 4 d of data were required in any period. Assuming data from at least four satellites for each minute of the 8 h for which calculations were made, a minimum of 7680 observations were used for each 7-d period. As before, years were chosen to balance the availability of data during a particular year with a desire to choose years with similar levels of solar activity. [33] For diurnal comparisons, we again used weeks and The one minute data were sorted into 15 min blocks as in the solar cycle case. A minimum of 1440 observations were used for each block. [34] In addition to the statistics described above, HDOP was calculated at one minute intervals using all available satellites, regardless of the S 4 value. Satellites with S 4 exceeding thresholds were incrementally removed and new HDOPs determined. Satellite positions (ephemeris) are required to determine HDOP. We used almanac data, a coarse but readily available form of ephemeris, to calculate HDOP as follows: [35] 1. Almanac data were compiled for the particular day and times for which observational data were available. [36] 2. If almanac output indicated a visible satellite that did not appear in the data, that satellite was assigned an S 4 value of [37] 3. HDOP was calculated using all satellites, regardless of S 4 and using only satellites with S 4 < 0.8, 0.5, and 0.3, respectively. In this analysis, an HDOP value of 1 was assigned when less than four satellites were available. 24 Space Weather Quarterly

25 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 3. Number of GPS satellite links with S 4 > 0.3 at the 95th percentile during the solar maximum (dashed red line) and off-solar maximum (solid blue line) for the Atlantic and Americas longitudinal sector. The hours after local sunset are indicated on the abscissa. The years used for comparison are shown in Table 3. The health-bit included in the Almanac data was not considered [38] 4. All times were adjusted to be sunset relative and data were wrapped as before. 3. Results and Discussion [39] In this section we present our results for the solar cycle, seasonal, and diurnal comparisons, examine the effect of latitude and longitude, and conclude by presenting the HDOP results. The results were grouped by longitudinal sector. We chose to display the 95th percentile for S 4 > 0.3, since these values were sensitive enough to reveal the characteristics we sought to examine. At lower percentiles and higher S 4 values, the trends were similar, but of a lesser magnitude Solar Cycle Behavior [40] Recall from Table 1 the production of EPBs and resulting scintillation mirrors the solar cycle. Figure 3 shows the number of GPS satellites received with S 4 > 0.3 at the 95th percentile during high (dashed red line) and low (solid blue line) solar activity in the AA longitudinal sector. The stations are listed from top to bottom by proximity to the geomagnetic equator. The hours after local sunset are indicated on the abscissa. The years used for comparison are shown in Table 3. The increase in scintillated satellites at Ascension (Figure 3, bottom) during the peak of the solar cycle is most dramatic; the number of scintillated satellites doubles and represents over 80 percent of the available satellites. The number of scintillated satellites at Ancon (Figure 3, top) increased to a lesser extent. The difference in solar cycle response between Ascension and Ancon can be attributed to differences in latitude. Recall from Table 1, scintillation is more pronounced at anomaly crest stations like Ascension and less so at stations near the geomagnetic equator like Ancon. Conversely at Antofagasta (Figure 3, middle), the number of scintillated satellites is higher during the offsolar peak year (1998) for the first 3 h of the evening. [41] This result is not entirely without precedent. Using in situ satellite data, Su et al. [2006] found for the longitudinal sector in which Antofagasta is located, EPB activity was greater during the solstices of 2003 than those of 2000, although the solar activity was greater in We limited our comparisons to the equinoxes yet found more severe scintillation in 1998 at Antofagasta than in 2000, although the solar activity was lower in [42] Results were more consistent in the PIO sector, shown in Figure 4. Again the stations are listed from top to bottom by proximity to the geomagnetic equator. In all cases, the number of satellites affected by scintillation increased in the peak solar year from the off-peak year. [43] The most dramatic increases occurred at Parepare and Diego Garcia, nearer the anomaly crest. At Parepare, the number of affected satellites represents over 85 percent of those available (not shown) for 3.5 h and for a total of 1.25 hours during the evening includes all available satellites Seasonal Behavior [44] Figures 5 and 6 show the results of a seasonal comparison using 3 years of data (Table 3). Like the earlier figures, the values on the ordinate represent the number of available satellites experiencing S 4 > 0.3 at the 95th percentile. The abscissa, however, now indicates the week number. The results were smoothed by applying a 5-week boxcar average to the raw output. References to spring and fall are with regards to the Northern Hemisphere seasons. [45] Figure 5 shows the results for the AA sector. The smoothed peaks are near, but not precisely centered on, weeks and ; Ancon (solid red line) peaks earliest at week 7 and the others peak at week 10. The autumnal peaks for all stations occur after the equinox, from weeks 42 to 45, with Antofagasta peaking latest. The seasonal variability is greatest at Antofagasta (dashed blue line). Taking into account the total number of available satellites at Ascension during the spring peak, approximately 65 percent are experiencing S 4 > Space Weather Quarterly 25

26 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 4. As in Figure 3 but for the Pacific longitudinal sector. [46] In the PIO sector (Figure 6), the spring peak is delayed for all stations but Diego Garcia. Marak Park (dash-- dot green line) peaks at week 22, the remaining sites between weeks 15 and 17. In the fall, the greatest lag occurs at Diego Garcia (solid red line) which reaches the maximum on week 43. Parepare (solid brown line) exhibits dual peaks at week 37 and 43. The greatest seasonal variability is found at Parepare and Pontianak. Like Ascension in the AA sector, roughly 65 percent of the available satellites at Parepare are experiencing S 4 levels above Diurnal Behavior [47] Recall from section 2.1 and Table 1 the diurnal cycle of scintillation was characterized by onset about an hour after local sunset, a peak around 2200L, and dissipation by Figure 5. Number of GPS satellite links with S 4 > 0.3 at the 95th percentile over 1 year. The week number is indicated on the abscissa, with the equinoxes between weeks and weeks The years of data combined for each site to create this image are shown in Table 3. The results were smoothed by applying a 5-week boxcar average to the raw output. The seasonal variability is greatest at Antofagasta (blue dashed line). The seasonal peaks are located near the equinoxes, more closely aligned with the spring equinox (weeks ) than with the fall equinox (weeks ). 26 Space Weather Quarterly

27 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 6. As in Figure 5 but for the Pacific longitudinal sector. The seasonal peaks seem to be more closely aligned with the fall equinox than with the spring equinox. The spring onset at Marak Parak is delayed beyond week 20. sunrise. Figure 7 shows the diurnal behavior for the AA sector. The ordinate indicates the number of available satellites with S 4 >0.3 at the 95th percentile as a function of hours after local sunset. Figure 8 shows the same for the PIO sector. In the work of de La Beaujardiere et al. [2004], it is noted that EPBs at the anomaly crest stations arrive 60 to 90 min later than the geomagnetic equator stations. They attribute this to... a finite upwelling speed of the irregularities at the magnetic equator. This behavior is evident in the AA sector (Figure 7) where the ramp-up of activity occurs about 3 h after local sunset at Ascension, while the increase at Ancon and Antofagasta occurs within 2 h. [48] In the PIO longitudinal sector (Figure 8), however, the lag is 15 to 30 min between Marak Parak and Parepare, Figure 7. Number of GPS satellite links with S 4 > 0.3 at the 95th percentile over one evening. The hours after local sunset are indicated on the abscissa. The delayed onset of scintillation at Ascension occurs because the plasma bubbles formed at the geomagnetic equator take time to diffuse along geomagnetic field lines to the anomaly crests. Space Weather Quarterly 27

28 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 8. As in Figure 7 but for the PIO. about half that in our AA sector comparison. Geographically, Ascension and Ancon are latitudinally separated by 472 km and Marak Parak and Parepare are separated by 1143 km. The differences in magnetic inclination, however, are and 22.5, respectively. Thus, Ancon and Ascension are nearly twice as far apart geomagnetically as Marak Parak and Parepare, so the onset delay is twice as great for Ancon and Ascension. As in earlier examples, up to 80 percent or more of the available satellites were experiencing scintillation in excess of the threshold Longitudinal Comparison [49] We obtained slightly higher numbers of scintillated satellites in the PIO sector than the AA sector for all comparisons. This is in contrast to results obtained by Burke et al. [2004a] and later by Gentile et al. [2006a] who studied the occurrence of equatorial plasma depletions detected by Defense Meteorological Satellite Program (DMSP) vehicles. However, Gentile et al. [2006a] note this may be a consequence of spatial and temporal differences in sampling between the DMSP satellites and the groundbased receivers HDOP [50] HDOP calculations were made using the technique described in section 2.3. Figure 9 shows the annual results for Ascension, during October through December 2001, assuming no scintillation and using only the satellites whose PRN identifiers were recorded by the receiver. The ordinate indicates day of year; the abscissa indicates the hours after local sunset. HDOP values are color coded. Recall from section 2 that HDOP = 1 is considered very good. HDOP 10 are lumped into the infinite category colored red. The white spaces indicate missing data. Gray shading indicates observations from less than four satellites were recorded by the receiver and a value of 1 was assigned, as described earlier. Note the gray areas in this example only last a minute and are scattered throughout the figures. In general, HDOP values are between one and three most of the time. The striations running from the upper left of the figure to the lower right reflect fluctuations of HDOP created by the change in the GPS constellation s relationship to Earth as noted in section 2.2. [51] Figure 10 shows the results after removing all satellites with an S 4 > 0.5 at the 95th percentile, meaning there is only a 5 percent chance that a greater number of satellites would have an S 4 > 0.5. Recall from Table 1 that scintillation tends to peak around 2200L (local time); we have chosen to display the data from 4 to 7 h after sunset so that 2200L appears on the left side of the figure, between hours four and five. In Figure 10 the range of HDOP values shown increases dramatically. Additionally, hours pass during which fewer than four satellites are available. In section 2 we noted that observations from four GPS satellites are typically used to generate a navigation solution. In the event that observations were received from fewer than four satellites, HDOP was not calculated. The striations that were evident in Figure 9 are not as pronounced in this figure. The tremendous temporal variability, which remains a significant forecast challenge, is easily seen. [52] We chose to display HDOP results when satellites with S 4 > 0.5 at the 95th percentile were removed. At S 4 > 0.3, even more of the figure was shaded gray. The opposite was true for S 4 > 0.8. Similarly, if the percentile was lowered to the upper quartile or median, the results were less dramatic, since the number of satellites allowed to exceed the given S 4 threshold was higher. The removal of 28 Space Weather Quarterly

29 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 Figure 9. HDOP using all available satellites. HDOP values are indicated by the color bar and plotted as a function of the day of year versus the hours after local sunset. The striations running from the upper left of the figure to the lower right reflect variation of HDOP created by the halfsidereal day orbital period of GPS satellites. all satellites with S 4 values exceeding a particular threshold assumes that satellites experiencing that level of scintillation become unavailable simultaneously and are unable to reacquire lock. This is the most pessimistic assumption; some satellites may be intermittently available during a given minute. Similar HDOP analysis can be carried out for the remaining stations, and the behavior mirrors that of the S 4 examples, with more dramatic Figure 10. HDOP after removing all satellites with S 4 >0.5. Gray shading indicates HDOP was not calculated because less than four satellites were available below the S 4 threshold. These outage periods can last an hour or longer. Space Weather Quarterly 29

30 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 results near the anomaly crests using low S 4 values and high percentiles. 4. Conclusion [53] We have examined 7 years of GPS data for eight low latitude scintillation monitoring stations. We found the following: [54] 1. During weeks and , scintillation of GPS signals was higher during solar maximum when compared to an off-peak year in both the AA and PIO longitudinal sectors. However, Antofagasta in the AA sector exhibited an off-peak maximum in activity. Su et al. [2006] found a similar off-peak maximum in EPB activity during solstitial periods but not during the equinoxes. [55] 2. In general, smoothed seasonal peaks in GPS scintillation occurred before the March equinox and after the September equinox in the AA sector, and after both equinoxes in the PIO sector. In both sectors, up to 65 percent of the available satellites experience S 4 > 0.3. [56] 3. Diurnal activity onset at the anomaly crest stations was delayed between 30 and 60 min from the onset at the geomagnetic equatorial stations, with longest delays associated with the greatest differences in magnetic inclination between the stations. [57] 4. Activity in all cases was slightly higher in the PIO sector than the AA sector. This result was opposite of results found using DMSP satellite data to track plasma depletions. The conflict may be attributed to spatial and temporal sampling differences [Gentile et al., 2006a]. [58] 5. For Ascension in 2001, HDOP generally varied between one and three during the fall. However, when GPS satellites with S 4 > 0.5 were removed from the calculations, navigation solutions became impossible for long periods of time, and the remaining HDOP values ranged from one to greater than 10. [59] Our results support recent studies [Rama Rao et al., 2006; Cervera and Thomas, 2006], which found similar temporal and spatial scintillation characteristics using smaller data sets in India and the Pacific sector, respectively. Our work also raises interesting questions. Why was there a greater degree of scintillation in the off-peak year at Antofagasta during the equinox weeks? Is the same suppression mechanism identified by Su et al. [2006] at work? As in other cases, are sampling differences between ground based GPS receivers and satellite sensed plasma depletions to blame? What contributes to the longitudinal differences in the magnitude of scintillation activity and in the onset behavior? [60] Caution is warranted when attempting to apply these results. The receivers in our study were specifically designed to track through scintillation and operated from a fixed location. Differences in receiver design and operation can impact the S 4 values calculated as well as the ability of the receiver to acquire and track the GPS satellites through scintillation. Additionally, mobile operations or less than optimal siting further complicate the picture. [61] Additional insight may be gained by examining patterns of scintillation-induced GPS outages. Keep in mind our study did not address the S 4 data that was missing because scintillation rendered a particular GPS satellite unavailable. What are the effects on the statistics if we attempt to fill in the gaps with proxy S 4 values when scintillation is the suspected cause of missing data? [62] Would such an endeavor result in better agreement between ground-based scintillation measurements and in situ satellite measurements of plasma depletions? The amount of data from the SCINDA network continues to grow. These ground-based observations will eventually be supplemented by data from the Communication and Navigation Outage Forecast System (C/NOFS) satellite [de La Beaujardiere et al., 2004]. Such data is essential to increasing our understanding of GPS scintillation and fostering our ability to accurately model and forecast the phenomenon. Like the amount of data, the community of GPS users is growing. GPS applications range from pigeon racing to precision aircraft approaches. The number of GPS satellites affected by scintillation, and the impacts on availability and precision, warrant continued attention, particularly on the eve of the next solar maximum. [63] Acknowledgments. The authors would like to thank Ron Caton and Charlie Carrano of AER Inc., as well as William Bailey, Dave Kaziska, John Raquet, and Chuck Leakeas of the Air Force Institute of Technology, for their assistance and comments. We would also like to thank the reviewers for their helpful suggestions. References Anderson, D. N., B. Reinisch, C. Valladares, J. Chau, and O. Veliz (2004), Forecasting the occurrence of ionospheric scintillation activity in the equatorial ionosphere on a day-to-day basis, J. Atmos. Sol. Terr. Phys., 66(17), , doi: /j.jastp Beach, T. L., T. Pedersen, M. J. Starks, and S.-Y. Su (2004), Estimation the amplitude scintillation index from sparsely sampled phase screen data, Radio Sci., 39, RS5001, doi: /2002rs Burke, W. J., C. Y. Huang, C. E. Valladares, J. S. Machuzak, and L. C. Gentile (2003), Multipoint observations of equatorial plasma bubbles, Tech. Rep. AFRL-VS-HATR XCAFRL-VS-HA, Air Force Res. Lab., Hanscom Air Force Base, Mass. Burke, W., C. Huang, L. Gentile, and L. Bauer (2004a), Seasonallongitudinal variability of equatorial plasma bubbles, Ann. Geophys., 22, Burke, W. J., L. C. Gentile, C. Y. Huang, C. E. Valladares, and S. Y. Su (2004b), Longitudinal variability of equatorial plasma bubbles observed by DMSP and ROCSAT-1, J. Geophys. Res., 109, A12301, doi: /2004ja Cervera, M. A., and R. M. Thomas (2006), Latitudinal and temporal variation of equatorial ionospheric irregularities determined from GPS scintillation observations, Ann. Geophys., 24, Dandekar, B. S., and K. M. Groves (2004), Using ionospheric scintillation observations for studying the morphology of equatorial ionospheric bubbles, Radio Sci., 39, RS3010, doi: /2003rs Datta-Barua, S., P. H. Doherty, S. H. Delay, T. Dehel, and J. A. Klobuchar (2003), Ionospheric scintillation effects on single and 30 Space Weather Quarterly

31 S04D02 STEENBURGH ET AL.: IONOSPHERIC SCINTILLATION EFFECTS ON GPS S04D02 dual frequency GPS positioning, paper presented at ION GPS/ GNSS 2003, Inst. of Navig., Fairfax, Va. Davies, K. (1990), Ionospheric Radio, 590 pp., Peter Peregrinus, London. de La Beaujardiere, O., et al. (2004), C/nofs: A mission to forecast scintillations, J. Atmos. Sol. Terr. Phys., 66(17), Eccles, J. V. (1998), Modeling investigation of the evening prereversal enhancement of the zonal electric field in the equatorial ionosphere, J. Geophys. Res., 103, 26, ,719. Fremouw, E. J., R. C. Livingston, and D. A. Miller (1980), On the statistics of scintillating signals, J. Atmos. Terr. Phys., 42, Gentile, L. C., W. J. Burke, and F. J. Rich (2006a), A global climatology for equatorial plasma bubbles in the topside ionosphere, Ann. Geophys., 24, Gentile, L. C., W. J. Burke, and F. J. Rich (2006b), A climatology of equatorial plasma bubbles from DMSP , Radio Sci., 41, RS5S21, doi: /2005rs Hofmann-Wellenhof, B., H. Lichtenegger, and J. Collins (2001), Global Positioning System: Theory and Practice, 382 pp., Springer, New York. Immel, T. J., S. B. Mende, H. U. Frey, L. M. Peticolas, and E. Sagawa (2003), Determination of low latitude plasma drift speeds from FUV images, Geophys. Res. Lett., 30(18), 1945, doi: /2003gl Kelley, M. C. (1989), The Earth s Ionosphere, Plasma Physics and Electrodynamics, Academic, San Diego, Calif. Rama Rao, P. V. S., S. Gopi Krishna, K. Niranjan, and D. S. V. V. D. Prasad (2006), Study of spatial and temporal characteristics of L-band scintillations over the Indian low-latitude region and their possible effects on GPS navigation, Ann. Geophys., 24, Rino, C. L. (1979a), A power law phase screen model for ionospheric scintillation: 1. weak scatter, Radio Sci., 14, Rino, C. L. (1979b), A power law phase screen model for ionospheric scintillation: 2. strong scatter, Radio Sci., 14, Schunk, R. W., and A. F. Nagy (2000), Ionospheres: Physics, Plasma Physics, and Chemistry, Cambridge Atmos. and Space Sci. Ser., Cambridge Univ. Press, Cambridge, U.K. Steenburgh, R. A. (2007), Holes: Ionospheric scintillation, GPS and imputation, Master s thesis, Air Force Inst. of Technol., Wright- Patterson Air Force Base, Ohio. Su, S., C. Cho, C. Liu, and H. Ho (2006), Meridional wind effect on the seasonal/longitudinal distribution of equatorial density irregularities inferred from ROCSAT-1 observations, Eos Trans. AGU, 87(36), Jt. Assem. Suppl., Abstract SA43C Sultan, P. J. (1996), Linear theory and modeling of the Rayleigh-Taylor instability leading to the occurrence of equatorial spread F, J. Geophys. Res., 101, 26, ,891. Thomas, R. M., M. A. Cervera, A. G. Ramli, Effendy, P. Totarong, K. M. Groves, and P. J. Wilkinson (2004), Seasonal modulation of GPS performance due to equatorial scintillation, Geophys. Res. Lett., 31, L18806, doi: /2004gl Van Dierendonck,, Q. Hua, P. Fenton, and J. Klobuchar (1996), Commercial ionospheric scintillation monitoring receiver development and test results, paper presented at 52nd Annual Meeting, Inst. of Navig., Fairfax, Va. Wheelon, A. D. (2001), Electromagnetic Scintillation, vol. 1, Geometric Optics, Cambridge Univ. Press, New York. Wheelon, A. D. (2003), Electromagnetic Scintillation, vol. 2, Weak Scattering, Cambridge Univ. Press, New York. K. M. Groves, Space Vehicles Directorate, Air Force Research Laboratory, 29 Randolph Road, Hanscom Air Force Base, MA , USA. C. G. Smithtro, Engineering Physics, Air Force Institute of Technology, 2950 Hobson Way, Wright-Patterson Air Force Base, OH , USA. (christopher.smithtro@afit.edu) R. A. Steenburgh, 2d Weather Squadron/Operating Location-P, Space Weather Prediction Center, W/NP9 325 Broadway, Boulder, CO 80305, USA. (robert.steenburgh@noaa.gov) CHAPMAN CONFERENCE Fall 2008 Chapman Conferences Chapman Conference on Lakes and Reservoirs as Sentinels, Integrators, and Regulators of Climate Change 8 10 September 2008 Lake Tahoe, Incline Village, Nevada, USA Chapman Conference on Shallow Mantle Composition and Dynamics Fifth International Orogenic Lherzolite Conference September 2008 Mount Shasta Resort, Mount Shasta, California, USA Chapman Conference on Biogeophysics October 2008 Portland, Maine, USA Abstract Submission Deadline: 21 July 2008 Chapman Conference on Organic Matter Fluorescence October 2008 University of Birmingham, Edgbaston, Birmingham, UK Abstract Submission Deadline: 4 August 2008 Chapman Conference on Atmospheric Water Vapor and Its Role in Climate October 2008 Kailua-Kona, Hawaii, USA Abstract Submission Deadline: 11 July 2008 Chapman Conference on Universal Heliophysical Processes (IHY) November 2008 Savannah, Georgia, USA Abstract Submission Deadline: 12 September 2008 Chapman Conference on Physics of Wave-Mud Interaction November 2008 Amelia Island, Florida, USA Abstract Submission Deadline: 8 August 2008 For complete details on all Chapman Conferences, including housing information, scientific programs, schedules and more, please visit is posted as details become available. Space Weather Quarterly 31

32 STAY CONNECTED TO THE LATEST IN SPACE WEATHER Subscribe to Space Weather: The International Journal of Research and Applications and join an online forum for the latest news and research findings in the emerging field of space weather. Engineers, scientists, systems designers, industry managers, and government leaders look to Space Weather for: Peer-reviewed research on engineering and scientific advancements Government agency initiatives What s happening in the commercial sector Plus editorials, special sections, and opinion articles from across the space weather community. Editor Louis J. Lanzerotti, New Jersey Institute of Technology Visit and select Free Articles for a sampling of what Space Weather has to offer. If you are interested in submitting an article or to sign up for a free quarterly print digest of the journal Space Weather Quarterly, please also visit the site. For individual or institutional subscriptions, contact service@agu.org.