Space Weather and South America Based on GPS Data Collected During This Geomagnetic Storm
Innovation er and its effects on GPS. This section Monitoring the Ionosphere with GPS also reviews how GPS observables are used to measure TEC. The second sec- tion presents a map of the TEC over North Space Weather and South America based on GPS data collected during this geomagnetic storm. Anthea Coster, John Foster, and Philip Erickson The TEC map clearly illustrates a phe- nomenon known as storm enhanced den- sity (SED), which is driven by process- “Stormy today, clearing up tomorrow.” That may sound like a typical forecast given by your es in Earth’s magnetosphere and is local TV meteorologist, but it could just as well be a forecast of space weather. Here on Earth, associated with large gradients in the high winds, heavy rains, deep snow, and other forms of severe weather can disrupt our daily ionospheric and plasmaspheric TEC. The lives. Conditions on the Sun and in the solar wind, magnetosphere, and the ionosphere can next section presents data from additional also affect our lives through the effects they have on satellites, communications, navigation, sensors that support the GPS TEC obser- and power systems. Scientists are now studying space weather with a wide range of tools to vations and connect the observed SED try to learn more about the physical and chemical processes taking place in the upper atmos- phenomenon with other space-weather phere and beyond. One of these tools is GPS. processes. The final section discusses The signals from the GPS satellites travel through the ionosphere on their way to receivers some specific effects on GPS observa- on or near Earth’s surface. The free electrons populating this region of the atmosphere affect tions that arose from the March 31 storm. the propagation of the signals, changing their speed and direction of travel. By processing the data from a dual-frequency GPS receiver, it’s actually possible to estimate just how many elec- Background trons were encountered by the signal along its travel path — the total electron content (TEC). Space weather — the variability of Earth’s TEC is the number of electrons in a column with a cross-sectional area of one square meter space environment — can adversely affect centered on the signal path. If a regional network of ground-based GPS receivers is used, then the integrity and performance of man- a map of TEC above the region can be constructed. The TEC normally varies smoothly from made systems such as GPS. Solar drivers day to night as Earth's dayside atmosphere is ionized by the Sun's extreme ultraviolet radia- of space weather include solar flares, tion, while the nightside ionosphere electron content is reduced by chemical recombination. coronal holes, and coronal mass ejections But the ionosphere can experience stormy weather just as the lower atmosphere does. Smooth (CMEs). These disturbances are the variations in TEC are replaced by rapid fluctuations, and some regions experience significant- key ingredients of strong geomagnetic ly higher or lower TEC values than normal. In this month’s column, we look at how GPS is storms on Earth. Outbursts of charged being used to study such storms and how it is furthering our understanding of the Earth–Sun particles and electromagnetic energy from environment. — R.B.L. CMEs and solar flares propagate through the solar wind, the tenuous material blow- ing outward from the solar atmosphere. Earth’s magnetosphere, which is the region of space influenced by Earth’s magnet- ic field, shields Earth from most of this erupted material. However, some of the energy from solar disturbances does enter the magnetosphere, disrupting its con- figuration, particle populations, and the important coupling between the outer pace weather and associated distur- received signal. In severe conditions, magnetosphere and the inner layers of Sbances in Earth’s magnetic field can these fluctuations can cause the receiv- Earth’s atmosphere. produce large gradients in the total elec- er to lose lock. Regions of particular importance tron content (TEC) in the mid-latitudes. Until now, the physical mechanisms are Earth’s ionosphere and plasmasphere. For single-frequency GPS users, these that cause these large TEC gradients The ionosphere is the region of free elec- large gradients in the TEC are of con- to form in the mid-latitudes have been trons (plasma) located approximately cern because they can make the ionos- poorly understood. However, by com- 100–2,000 kilometers above Earth’s sur- phere difficult to model and remove, bining data from the global network of face and created by the action of extreme thereby affecting GPS-derived position continuously operating dual-frequency ultraviolet (EUV) sunlight on the gases accuracy. The presence of these gradi- GPS receivers, the development of these of the upper neutral atmosphere. The ents can also affect carrier-phase differ- TEC perturbations can be monitored, plasmasphere is a doughnut-shaped region ential GPS (DGPS) and real-time kine- and our understanding of the physical of low-temperature plasma that co-rotates matic (RTK) applications because the processes involved has been greatly with Earth and is located in the inner- ionospheric term in the observation equa- enhanced. most regions of Earth’s magnetosphere. tions may not cancel, thus making This article discusses the effects of geo- The plasmasphere is the high-altitude unknown ambiguities difficult to resolve. magnetic storms on GPS observations extension of the ionosphere. The inter- In addition, large gradients in the TEC and measurements and focuses on one action of the solar wind, magnetosphere, are frequently associated with ionos- in particular, the March 31, 2001, storm. plasmasphere, and ionosphere during pheric scintillation events that can cause The first of the article’s four sections pre- storm-time conditions is complex and amplitude and phase fluctuations of the sents a brief background of space weath- coupled, and our understanding of the
42 GPS World May 2003 www.gpsworld.com Innovation
TEC resolution of ambiguities difficult. (A 80 140 TEC unit is approximately equivalent to 0.162 meters of range delay at L1, or 1 meter of delay at L1 is equivalent to 6.159 60 120 TEC units). Gradients in the TEC can be caused 40 by traveling ionospheric disturbances, bubbles, plumes, streams, or sharp bor- 100 ders in the TEC. Small structures with- 20 in the ionosphere — frequently associ- ated with large-gradient regions — can 0 80 cause scintillation. For GPS users, scin- tillation is observed as amplitude and phase fluctuations in the received sig- –20 nal. Severe scintillation effects in either 60 amplitude or phase can cause a GPS
Geodetic latitude (degrees) –40 receiver to lose lock. Large gradients in TEC and scintilla- 40 tion are primarily associated with the –60 equatorial and polar regions. However, large gradients in TEC and scintillation –80 20 can be observed at mid-latitudes 220 240 260 280 300 320 340 during moderate to severe geomagnetic Geodetic longitude (degrees) disturbances. Because the majority of FIGURE 1 A map of total electron content determined by a network of GPS receivers on GPS users are located at mid-latitudes, March 31, 2001, at 19:30 UTC shows the plume of storm-enhanced density stretching the disturbances caused by large geo- across the Great Lakes and into Canada. magnetic storms can potentially affect the average GPS user. many processes involved is continual- pheric range delay/advance is (to first The March 31, 2001, geomagnetic ly evolving. order) a function of both the TEC along storm was preceded by two coronal mass For GPS users, the impact of space the line of sight from the receiver to the ejections and an X-class solar flare (a weather can usually be attributed to dis- satellite and the inverse carrier frequency major sudden eruption of energy) that turbances in the ionosphere as well as squared. By computing the difference in had occurred two days earlier. On the the plasmasphere, which in turn can the range measurements made at two day of the storm, charged particles cause degradation in range measure- frequencies, the TEC is measured. reached Earth and induced disturbances ments and in severe circumstances, loss GPS currently operates at 1575.42 in the geomagnetic field. The planetary of lock by the receiver of the GPS signal. MHz (L1) and 1227.6 MHz (L2). Additional geomagnetic index, Kp, is a global mea- As GPS signals propagate through the factors that must be accounted for in this sure of the disturbance of the geomag- ionosphere, the propagation speed and calculation are the system instrumental netic field by auroral ionospheric cur- direction of the GPS signal are changed delays between the L1 and L2 signals rents. It is based on a quasi-logarithmic in proportion to the varying electron den- caused by both the satellite and receiv- scale. Values of 4 and greater indicate sity along the line of sight between the er hardware. These values must be solved moderately disturbed conditions. On receiver and the satellite. The accumu- for and removed from the measurement this day, the Kp index was at disturbed lated effect, by the time the signal arrives to determine accurate TEC values from levels (greater than 5) throughout the at the receiver, is proportional to the inte- GPS data. By combining data from mul- day and into the first part of the fol- grated TEC, the number of electrons tiple GPS ground stations, one can pro- lowing day. It reached a value of 9 (“9 in a column stretching from the receiv- duce maps of the TEC as a function of minus” is close to its maximum, indi- er to the satellite with a cross-section- latitude and longitude. These maps can cating severely disturbed geomagnetic al area of one square meter. This in turn be used to monitor the gradients in the conditions) between 3:00 and 9:00 UTC. affects the GPS range observable: a delay TEC that develop during time periods of is added to the code measurements and geomagnetic storms. TEC Storm Maps an advance to the phase measurements. Knowledge of these TEC gradients is We have combined dual-frequency data To achieve very precise positions from important to various GPS users. When from more than 125 ground-based GPS GPS, this ionospheric delay/advance a GPS signal encounters large gradients receivers to construct maps of the ver- must be taken into account. in TEC, the ionospheric error in the range tical TEC as a function of latitude and One can take advantage of the very measurement is difficult to model and longitude during the storm. Each map same ionospheric delay/advance that remove (required for single-frequency represents the state of the ionosphere causes problems for the majority of GPS GPS users), or in the case of differential averaged over approximately five min- users. By measuring this delay, prop- GPS, it cannot be canceled out. For DGPS utes. Sequentially combining these maps erties of the ionosphere can be inferred, or RTK users, differences as small as two allows us to produce a time history of and these properties can be used to mon- TEC units over the baseline (one TEC TEC perturbations in two dimensions. itor space-weather events. The ionos- unit is 1016 electrons/meter2) can make Figure 1 shows a single TEC map asso-
44 GPS World May 2003 www.gpsworld.com Innovation ciated with TEC gradients. Large gra- 65 dients in the TEC are clearly apparent over the continental United States from approximately 16:00 to 22:00 UTC, with 60 the peak gradients in the TEC observed between 19:00 and 19:30 UTC. As shown 55 in the maps, between 17:00 and 22:00 Millstone Hill ISR UTC a plume of TEC moved northward and westward out of the larger region of 50 enhanced TEC (TEC greater than 100
TEC units) in the southeast corner of the 45 United States. In Figure 1, the TEC plume is observed from the New England coast across the Great Lakes region and into 40 central Canada. The GPS map of this TEC plume provides a two-dimensional pic- Geodetic latitude (degrees) 35 ture of the ionospheric phenomenon SED, GPS TEC > 50 TECu which was initially identified in the early 1990s using observations from the 30 Millstone Hill Observatory’s (Westford,
Massachusetts) incoherent scatter radar. 25 The data used to produce these maps 240 250 260 270 280 290 300 included data from both the International Geodetic longitude (degrees) GPS Service and the Continuously FIGURE 2 Intense sunward ion flux was measured by the Millstone Hill incoherent scat- Operating Reference Stations (CORS). ter radar at 19:30 UTC on March 31, 2001. The color coding indicates a range of flux The CORS network is coordinated by the values from 1013 to 1015 electrons/meter2/second. The red contours indicate the area U.S. National Geodetic Survey. The data in the GPS-measured total electron content (TEC) map where TEC values exceed 50 were accessed by means of publicly avail- TEC units. able data archives on the World Wide Web. The line-of-sight TEC values were turbed equatorial anomaly (defined below) observed over Florida in North America converted to vertical TEC values using a move farther north (over Florida), the and, correspondingly, over Argentina and simple mapping function and were asso- conjugate extent of this enhanced TEC Chile in South America. According to this ciated to an ionospheric pierce point lat- is also moving farther south (over Argentina model, one would also anticipate deplet- itude and longitude, assuming the ionos- and Chile). ed TEC values close to the geomagnet- phere to be compressed into a thin shell A large geomagnetic storm produces ic equator, although we lack access to at the peak ionospheric height of 350 kilo- two separate effects. The first occurs when GPS receivers in this region to confirm meters. GPS satellite and receiver bias- strong-penetration eastward electric fields this assumption. es were estimated and removed from the redistribute plasma in the equatorial ionos- The second effect results from a com- data. Maps of TEC were prepared at 5- phere, magnifying what is commonly plex interplay between electrodynamic minute intervals. The vertical TEC has referred to as the Appleton, or equatorial, processes in the inner magnetosphere been binned in 2 2–degree latitude/lon- anomaly. The Appleton anomaly can be and the ionosphere that leads to the devel- gitude bins and color coded between described as an electron density deple- opment of the subauroral polarization 20 and 140 TEC units. No smoothing was tion that develops at the geomagnetic stream (SAPS). The SAPS electric fields used; the high level of detail is primarily equator along with regions of enhanced are established along the outer bound- attributed to the persistence of a well- electron density that typically peak at ary of the plasmasphere, known as the organized connected structure and to the approximately 15–20 degrees north and plasmapause, and are directed radially large quantity of data processed. south geomagnetic latitude. During a large outward at the equator, which, tracing To emphasize the global nature of these geomagnetic storm, strong eastward elec- along magnetic field lines, maps to a pole- space-weather phenomena, Figure 1 shows tric fields near the equator lift up equa- ward direction at higher latitudes. The data from both hemispheres. Isocontours torial plasma (due to E B forces in which SAPS electric fields overlap the plas- of geomagnetic latitude and longitude at E is the electric field vector and B is the masphere, stripping away its outer layer the shell height illustrate the connection magnetic field vector) to higher than nor- and redistributing previously equatorial between magnetically conjugate loca- mal altitudes. Eventually, the uplift from plasma to a higher latitude. The plume tions in the North and South American the E B force is overcome by the force of low-latitude plasma is then advected sectors. Latitude contours at 10-degree of gravity, which pulls the plasma down rapidly westward and poleward by the E intervals from 20 to 60 degrees north and along magnetic field lines. B drift developed by the strong SAPS south are shown. For example, 35 degrees These effects combine to deposit equa- electric field. The result of this com- latitude and 290 degrees longitude are torial plasma farther north and south plex sequence of events is the space-weath- magnetically conjugate to 60 degrees than normal, enhancing the off-equa- er SED plume extending across the Great latitude and 305 degrees longitude. The tor density concentrations that exist when Lakes into Manitoba as illustrated in series of maps illustrates that as the the ionosphere is quiescent. The enhance- Figure 1. enhanced TEC values of the storm-per- ment explains the very large TEC values www.gpsworld.com GPS World May 2003 45 Innovation FIGURE 3a, b, c Data collected at 6 approximately 21:20 UTC on March 31, 2001, by the IMAGE satellite and 4 by the ground-based GPS network show the correlation of observed storm-time features: (a) IMAGE satel- 2 lite extreme ultraviolet image of
plasmasphere and plasmaspheric 0 plume; (b) equatorial projection of storm-enhanced density (SED) fea- ture and plasmapheric plume; (c) –2 GPS total electron content map IMAGE EUV plasmapause showing SED feature. 3a –4 21:30 UTC TEC > 30 TECu
–6 –4 –2 0 2 4 6 8 TEC 40 3b Sunward distance (Earth radii)
70 35 singly ionized helium (He ) in the plas- masphere. Ionized atomic hydrogen (H ) 60 30 — a proton — is the principal ion with- in the plasmasphere, which accordingly is sometimes called the protonosphere. 50 25 Singly ionized helium accounts for 3–30 percent of the plasmaspheric material. 20 The brightness of the resonantly scat- 40 tered 30.4-nanometer feature is direct- ly proportional to the He column abun- 15 dance. Images generated from data 30 acquired at apogee can therefore show Geodetic latitude (degrees) 10 the structure of the entire plamasphere. Figure 3a shows an image taken at 21:21 20 UTC on March 31, 2001, which can be 5 interpreted as follows: The Sun is on
10 the right, and Earth’s shadow extends 0 200 220 240 260 280 300 320 through the plasmaphere on the left. The 3c Geodetic longitude (degrees) bright ring near the center is the northern aurora. The feature of note is referred to as a plasmaspheric drainage plume, which Magnetospheric Connections pheric density and plasma E B veloc- is composed of plasmaspheric material Data from other sensors support our inter- ity inside the SED plume. Figure 2 pre- seen stretching sunward. This plasmas- pretation of the GPS TEC observations sents an ISR measurement of the ion flux pheric drainage plume represents plasma and also establish connections with Earth’s measured at the same time as the GPS that is eroded from the outer plasmas- magnetosphere and plasmasphere. TEC measurement shown in Figure 1. phere during storm-time conditions The Millstone Hill incoherent scat- Ion flux is a product of both electron den- described earlier. ter radar (ISR) operated continuously sity and velocity. In Figure 2 the flux is The data in Figure 3b represent the mag- during the March 31, 2001, storm. The color coded, with red indicating the largest netic equatorial projection of the plasma- Millstone Hill radar is located at a mid- values. Of note is the intense sunward pause and the plasmaspheric drainage geomagnetic latitude near 55 degrees. ion flux observed across the region cor- plume seen in the IMAGE EUV obser- Throughout the storm period, the radar responding to the SED plume. The ISR vations along with the 30 TEC unit con- used a full south-to-north elevation angle data indicate that the SED plume, mapped tour line of the GPS map shown in Figure scan to monitor ionospheric features over by the GPS data, is rapidly advecting sun- 3c. These data are projected into the equa- a wide range of latitude. Each of these ward, in agreement with the scenario pre- torial plane using simple dipole mapping. scans was repeated for 45 minutes and sented earlier. The observations from the ground using provided observations of temperature, The second set of ancillary data asso- GPS clearly maps to the plasmaspheric electron density, and electron and ion ciated with the GPS TEC map was plume as viewed from space. SED is there- velocity as a function of altitude. Using collected by NASA’s Imager for fore the ground-based signature of the the Millstone Hill ISR electron density Magnetopause-to-Aurora Global Explo- erosion of the outer plasmasphere caused measurements, we can compute the TEC ration (IMAGE) satellite. This satel- by strong storm-time subauroral polar- up to 800 kilometers. These data are dis- lite, with an apogee altitude of 7.2 Earth ization electric fields. The ground-based cussed later in the article. radii, can image Earth’s plasmaphere by GPS receivers are in fact observing and Radar azimuth scans to the west were using sunlight at a wavelength of 30.4 being affected by a magnetospheric space- also made that sampled both the ionos- nanometers resonantly scattered from weather phenomenon.
46 GPS World May 2003 www.gpsworld.com Innovation 6 6
4a 4b 4 4
2 2
0 0
–2 –2
–4 –4
Geomagnetically quiet Geomagnetically disturbed L1 single-difference residual ionospheric delay (meters) residual ionospheric delay L1 single-difference L1 single-difference residual ionospheric delay (meters) residual ionospheric delay L1 single-difference –6 –6 0 4 8 12 16 20 24 0 4 8 12 16 20 24 Time (UTC hours) Time (UTC hours)
FIGURE 4a, b Illustration of the effect of geomagnetically dis- and Chatham, Massachusetts: (a) geomagnetically quiet time turbed conditions on cancellation of the ionospheric delay over (March 26, 2001); (b) geomagnetically disturbed time (March the approximately 200-kilometer baseline between Westford 31, 2001).
Storm-Time Ionospheric Effects ple of loss of lock in data collected by the contribution is accounted for by the plas- The average GPS user is primarily inter- Chatham (CHT1) GPS receiver. These masphere. The second effect of TEC dis- ested in storm-time phenomena as they data are shown in Figure 5, which com- tribution involves the design of the map- affect particular GPS receivers. So how pares the number of satellites tracked ping function used to map or project the do gradients in the TEC affect GPS users? on both L1 and L2 on March 31, 2001, measured GPS line-of-sight TEC to an The first example of these effects shows to the number of satellites tracked five estimated vertical TEC. The majority the large gradients that appear over an days earlier. As many as seven GPS satel- of mapping functions used in GPS pro- approximately 200-kilometer baseline lites were not in track during the first cessing are based on a single-layer approx- between GPS receivers at Westford and time period of geomagnetic activity, imation that assumes an ionospheric peak Chatham, Massachusetts. Figure 4 shows between 5:00 and 7:00 UTC on March 31 altitude of between 350 and 450 kilo- the residual ionospheric delay in meters (see Figure 4b). meters. at the L1 frequency (1575.42 MHz) on The last observation of note involves The amount of TEC above 800 kilo- a quiet day (March 26, 2001) contrasted the measured distribution of TEC dur- meters was measured by comparing the with the residuals observed on a geo- ing geomagnetically disturbed conditions. Millstone Hill ISR’s estimate of TEC in magnetically disturbed day (March 31, TEC distribution affects the use of GPS the ionospheric E and F regions with the 2001). Each color represents a different data in two ways: The first involves the GPS TEC measurement. The Millstone GPS satellite, and the value of differen- incorporation of GPS TEC measurements Hill ISR is sensitive to the electron den- tial delay at L1 measures the difference in ionospheric models, which typically sity only up to altitudes of 800–1,000 kilo- in the measured line-of-sight TEC (con- assume that 80–90 percent of TEC is below meters, whereas the GPS TEC measure- verted to units of meters at L1) at the two 1,000 kilometers. The additional TEC ment measures out to 20,000 kilometers. receivers to a particular GPS satellite. DGPS and RTK applications require Further Reading • “Storm-Time Plasma Transport at Middle the ionospheric term to cancel out to cor- and High Latitudes,” by J.C. Foster, Journal For an introduction to space weather, see rectly resolve carrier-phase ambiguities. of Geophysical Research, Vol. 98, 1993, pp. Very large differential delays, as much as • “A Primer on Space Weather,” by the Space 1675–1689. Environment Center, National Oceanic and 4 meters at L1, were observed between • “Ionospheric Signatures of Plasmaspheric Atmospheric Administration, Boulder, several of the satellites on March 31, 2001. Tails,” by J.C. Foster, P.J. Erickson, A.J. Coster, Colorado,
48 GPS World May 2003 www.gpsworld.com Innovation 2 80 ) 2 1 70
0 60 GPS TEC (< 20,000km) –1 50 electrons/meter
–2 16 40 –3 30 –4 Geomagnetically –5 20 disturbed period
tal electron content (10 10 Difference in number of satellites tracked in number Difference –6 MH ISR TEC (<800km) To –7 0 0 4 8 12 16 20 24 15 16 17 18 19 20 21 22 23 24 Time (UTC hours) Time (UTC hours)
FIGURE 5 A comparison of the number of satellites in track by a FIGURE 6 TEC measured out to 20,000 kilometers by GPS and GPS receiver at Chatham, Massachusetts, on March 31, 2001, to TEC measured by the Millstone Hill incoherent scatter radar out the number of satellites in track on March 26, 2001, illustrates to 800 kilometers showed a large increase in the electron con- the difficulty the receiver had tracking satellites during a period tent in the upper ionosphere and plasmasphere during the when the geomagnetic field was significantly disturbed. March 31, 2001, storm.
Haystack Observatory and heads MIT’s Figure 6 As shown in , during the time peri- of the figures incorporating the GPS Millstone Hill Observatory. He holds a od of the SED phenomenon, greater than data. The authors also wish to acknowl- Ph.D. in physics from the University of 50 percent of the TEC at Millstone Hill edge Dr. Bill Sandel for supplying IMAGE Maryland and has 30 years’ experience in was observed to be above 800 kilometers plasmaspheric data and its interpreta- studies of Earth’s ionosphere, magnetos- (the measured GPS TEC value was 65–70 tion and Dr. Jerry Goldstein for his phere, space-weather effects, magnetos- TEC units compared with 20 TEC units analysis of the IMAGE plasmaspheric phere/ionosphere coupling phenomena, and incoherent scatter radar techniques. measured by the ISR). boundary. The Lincoln Laboratory portion of the Philip Erickson received his B.S. and Summary work described in this article is spon- Ph.D. in electrical engineering from GPS, when data are processed in a way sored by the United States Air Force Cornell University in Ithaca, New York. that emphasizes the spatial and tempo- under Air Force Contract AF19628-00-C- In 1995, he joined the staff of the ral variability of the overlying medium, 0002. Opinions, interpretations, conclu- Atmospheric Sciences Group at MIT’s is a powerful tool for studying space- sions, and recommendations are those Haystack Observatory. At Haystack, he is involved with the scheduling, software, weather phenomena whose effects of the authors and are not necessarily and hardware subsystems that operate directly affect the North American endorsed by the United States Air Force. Millstone Hill, one of four American lon- continent. The SED TEC plumes, seen in The Millstone Hill Observatory is gitude sector megawatt–class incoherent GPS data, map directly into the dra- supported by a cooperative agreement scatter radars. Erickson conducts studies matic plasmaspheric plumes observed between the National Science Foundation of topside light ion behavior and E-region in the EUV images collected by NASA’s and the Massachusetts Institute of irregularities using Millstone Hill data. IMAGE satellite. The GPS TEC maps Technology. support the explanation of SED as an “Innovation” is a regu- ionospheric signature of the erosion Anthea Coster is a 20-year member of the lar column featuring of the outer plasmasphere by electric technical staff at Massachusetts Institute discussions about of Technology (MIT) Lincoln Laboratory, fields. Continued monitoring of space recent advances in where she has served as the atmospheric weather using GPS will further enhance scientist at the Millstone Hill Satellite GPS technology and our understanding of the dynamic Tracking Radar. She currently holds a its applications as well processes taking place in the ionosphere joint appointment with the Atmospheric as the fundamentals of and their effects on communications and Sciences Group at the MIT Haystack GPS positioning. The navigation, electric power grids, and other Observatory. In 1991, she and her cowork- column is coordinated by Richard technological systems. ers developed the first real-time ionos- Langley of the Department of Geodesy pheric monitoring system based on GPS, which is now an integral part of the MIT and Geomatics Engineering at the Acknowledgments Radar Calibration System. Coster received University of New Brunswick, who This article is based on a paper pre- her B.A. in mathematics from the appreciates receiving your comments sented at The Institute of Navigation’s University of Texas at Austin and her M.S. and topic suggestions. To contact him, National Technical Meeting held in and Ph.D. in space physics and astronomy see the “Columnists” section on page 2 Anaheim, California, January 22–24, 2003. from Rice University in Houston, Texas. of this issue. John Foster is associate director for The authors wish to acknowledge atmospheric sciences at the MIT Catherine Laurin for generating several www.gpsworld.com GPS World May 2003 49