Convective Ionospheric Storms: A Major Space Weather Problem


  • Jonathan J. Makela,

  • Michael C. Kelley,

  • Odile de la Beaujardiére

  • This article is related to DOI:10.1029/2005SW000145

Some 50 years into the space age, technical societies are deeply committed to the utilization of space. For the military, space is the ultimate high ground from which a variety of surveillance, communications, and navigation systems operate. For industry, the communications and positional/navigational opportunities using space-based systems are virtually unlimited. However, when the plasma between the satellite and the receiver is turbulent, satellite signals scintillate in a manner analogous to the twinkling of starlight as it traverses the turbulent atmosphere, and both communication and navigation systems can be seriously affected.

With the rapidly developing use of space assets comes the realization that ionospheric plasma, through which communications and navigation signals must pass, is not a benign medium. The most severe ionospheric weather occurs within +/-20° of the geomagnetic equator where stored gravitational energy sometimes is released after sunset, depending on the condition of the equatorial ionosphere, to form vast plumes of turbulent plasma. The plumes rise quickly in a manner analogous to thunderstorm convection and, due to their electrical properties, are transmitted rapidly for vast distances north and south along the Earth's magnetic field. Termed convective ionospheric storms (CIS), to emphasize the analogy of the ionospheric process to thunderstorms, these plumes are caught up in the high-speed eastward plasma drift and often last until well after midnight. As a result, a single storm can affect a very large area in its lifetime.

To fully understand the problems posed by CIS, a detailed explanation of what is currently known about this topic is provided here, along with the methods for CIS monitoring. Weaknesses in scientific theory and monitoring capacities are also discussed.

Convective Ionospheric Storms

Within a literature already spanning some 75 years, the term “equatorial spread F” (ESF) has been attached to the phenomenon of CIS and describes a particular instrumental response to these mesoscale disturbances during the transmission and reception of radio pulses. However, the acronym CIS, used in this report, is more general because it reflects a physical process observable by many different techniques.

The morphology of the equatorial ionosphere is quite different from that at other latitudes because the magnetic field is nearly parallel to the Earth's surface. During the daytime, the plasma drifts upward under the influence of an eastward directed electric field. The uplifted plasma then moves along the magnetic field in response to gravity and pressure-gradient forces, producing a fountain effect where plasma moves upward close to the equator and slides down the magnetic field lines on the north and south. Two ionization maxima are formed on either side of the equator, the so-called equatorial Appleton anomalies, which create particularly dense plasma ripe for disturbing satellite signals. The two crests are +/-15° from the magnetic equator.

At night, the electric field is reversed from its daytime value since tidally driven currents cause a negative charge density at dusk and a positive charge density at dawn. The resulting electric field—westward during the night and eastward during the day—moves the plasma up during the day and down at night. However, just before this reversal takes place, the upward drift often increases suddenly, driving the main dense ionosphere, called the F layer, to high altitudes. In addition, production of ions by sunlight ceases after sunset and the conductivity of the lower ionosphere decreases dramatically, reducing its capability to “short out” the electrical potential. This large upward drift and low off-equatorial conductivity, combined with gravity, a decreasing neutral atmospheric density, and a vertical plasma density gradient in the bottomside of the F layer, creates an unstable configuration. If the ionosphere is low in altitude on a given night, the instability usually fades out. But if the postsunset upward surging plasma layer is high enough (at or above 350 km), stored energy can be released explosively, forming the low-density, highly turbulent features that surge upward.

Like a thunderstorm, CIS releases gravitational energy stored in the nighttime ionosphere, making altitude an important factor in the strength of the CIS—the higher the initial height of the layer, the more gravitational potential it accumulates, and the more likely the storm is to be severe. The events usually die out by midnight. Occasionally, if the geomagnetic activity is high enough, CIS can occur after midnight.

Observational Database

The plasma layer is composed of equal numbers of electrons and ions (mostly O+). Photons are emitted when the oxygen atoms recombine with electrons through either a one-step process (radiative recombination) or a more complicated set of ion-chemical reactions. The CIS depletions correspond to a decrease of photon production and appear as a well-defined dark region in photographs. The intensities of these airglow phenomena are subvisual but can be recorded either from the ground or from space using modern, cooled, charge-coupled device (CCD) cameras fronted by narrow filters (a few nanometers in bandwidth) tuned to airglow emissions.

One excellent viewing site for these emissions is the top of the Haleakala volcano on Maui, Hawaii. At this location, the structure and source of the severe weather are imaged in two dimensions by a CCD camera pointed antiparallel to the Earth's magnetic field (Figure 1). In this way, both the initial upsurge of low-density plasma bubbles and their subsequent eastward drift can be readily observed.

Figure 1.

Geometry of and a composite image from a camera on Maui, Hawaii. A series of overhead images are shown, mapped along the magnetic field lines to their ultimate source at the magnetic equatorial plane.

During a severe event, the electric fields which are generated in the low-density regions drive plumes of turbulent, low-density plasma upward to heights of more than 2000 km at the equator. Because of the high conductivity of the plasma parallel to the magnetic field, these equatorial depletions connect electrically to the plasma over Hawaii, causing the Hawaiian plasma to form mirroring features also detectable by CCD cameras. In this manner, airglow emissions are used to trace the development of the event over vast distances.

A Global Positioning System (GPS) receiver collocated with the camera can be used to determine the effect of the plumes on transionospheric radio wave propagation. Low-airglow regions, also called airglow depletions, correspond to the rising plumes, and as the line of sight to one of the GPS satellites moves in and out of the airglow depletions, the scintillation level of the signal (measured by an index called S4) rises and falls dramatically. Just as starlight twinkles as it passes through the turbulent atmosphere, scintillation of the GPS signal demonstrates that the plasma inside the plumes is turbulent (Figure 2).

Figure 2.

Disruption of GPS signals as they traverse a CIS event documented by an all-sky camera on Maui. As the line of sight to the satellite, indicated by the small box along the satellite trajectory in the top panels, intercepts an airglow depletion, the scintillation index (S4) rises dramatically and the signal is too noisy to measure the TEC (note the data dropouts in the bottom panel).

Ionospheric radars, which are analogous to atmospheric weather radars, have been used to study these plumes for several decades. Initially, the radar echoes come from a narrow height range in a region where the plasma density gradient is upward. However, when the stored energy of this system is released explosively, the low-density, highly turbulent features are seen surging upward in the radar echoes.

Radar and imager data have provided considerable information on this phenomenon, but in situ data from rockets and satellites have also been very important. In particular, rockets and satellites definitively proved that the upward surging plumes are regions of low-density plasma. Campaigns in Peru, Brazil, and the Marshall Islands, involving many launches of sounding rockets such as CONDOR and EQUIS and satellites such as Atmospheric Explorer, Dynamics Explorer, and those of the Defense Meteorological Satellite Program, have been invaluable in understanding CIS.

Most important for the future, a joint project between the Air Force Research Laboratory and the Department of Defense Space Test Program, called the Communication/Navigation Outage Forecasting System (C/NOFS), has a mission of predicting CIS and will soon provide detailed, three-dimensional images of these storms. The project will gather the most complete set of in situ and ground-based data ever attempted. These data sets include measurements of electric and magnetic fields, neutral winds, photon emissions, Doppler radar images, GPS and lower-frequency observations of scintillations, GPS occultation observations, tomographic maps of the ionosphere, plasma density and its fluctuations, and total electron content [see de la Beaujardiére, 2006].

Remaining Problems in the Field

Although space weather scientists have a solid grasp of the underlying physics of CIS, understanding the day-to-day variability of this phenomenon has proven difficult. The major challenge is to understand the physical processes that lead to the formation of plasma irregularities in the ionosphere and to identify the mechanisms that trigger or inhibit plasma instability.

Meeting this challenge requires accurate modeling of the parameters that are part of the instability growth rate. The electric field, whether caused by neutral wind (the dynamo electric field) or by magnetospheric and solar wind phenomena, is probably the most important of these parameters. Since the height of the ionosphere immediately after sunset is a major factor in determining whether or not CIS will occur, knowing the eastward electric field component, which drives the plasma upward, is vital to our predictive capability. The electric field measurements obtained by the C/NOFS satellite will provide crucial information on this key parameter.

Other, more vexing issues involve lower atmospheric tides and waves, which create initial disturbances and the electrical conductivity of the low-latitude ionosphere in contact with the rising ionosphere at sunset, which may or may not short-circuit the electric fields. The physics-based modeling of plasma bubbles—how they are born, how they evolve in time and space, and how they die—also needs considerable improvement.

Today space weather scientists are building assimilative physics-based models like those used in weather forecasting, but currently the models are starved for data. The meteorological community uses hundreds of balloon soundings that are made twice a day around the world in their assimilative models. The ever-growing numbers of GPS receiving stations on the ground and on orbiting spacecraft are beginning to provide ionospheric data similar in density to the balloon soundings in meteorology. For example, the two GPS signals transmitted on every satellite can be combined to give a measure of the total number of electrons between the satellite and the receiver. Total electron content (TEC) information will be one of the key data streams assimilated in the models now being developed for predicting severe space weather.

Other instruments will provide excellent corollary information to the C/NOFS satellite. For example, images looking down on the ionosphere from a low-Earth orbit, such as those obtained by the Global Ultraviolet Imager (GUVI) on NASA's Thermosphere-Ionosphere-Mesosphere Energetic and Dynamics (TIMED) satellite, have proven valuable in tracing plumes as well as the high-density, high-airglow regions of the Appleton anomaly, also called equatorial arcs, which are symmetrically located about the magnetic equator (Figure 3). Eventually, a set of ionospheric imagers in geostationary orbit is hoped to be established. These imagers would be analogous to tropospheric weather satellites, providing continuous information on the ionosphere, as well as on the structures within.

Figure 3.

Ground-based and GUVI observations of a CIS event on 22 September 2002. The two bright bands in the GUVI image, labeled equatorial arcs, are images of the dense, bright Appleton anomalies. The slanted dark strips that periodically cross the image are airglow depletions seen from orbit. The inset to the top right is a ground-based image from Hawaii in which the airglow depletions line up quite well with two of the GUVI strips, verifying that the orbital data were not instrumental in origin.

After the C/NOFS satellite is launched, a major worldwide effort will take place to enter the space weather forecasting age. With the satellite measurements, the global electric field at low latitudes—a key parameter—will become available to assimilate into models, analogous to lower atmospheric weather forecasting. Initially, the focus will be on the South American region, as it is the most heavily instrumented sector and includes the large radar in Jicamarca, Peru, which is heavily supported by the U.S. National Science Foundation. Subsequent campaigns will then be carried out in the western Pacific and Indian Ocean regions, which also have a considerable amount of ground-based infrastructure. With new experimental and modeling tools, scientists worldwide are eager to finally and fully understand and predict convective ionospheric storms.


  • Jonathan J. Makela is an assistant professor in electrical and computer engineering at the University of Illinois at Urbana-Champaign.

  • Michael C. Kelley is a professor of electrical and computer engineering at Cornell University, Ithaca, N. Y.

  • Odile de la Beaujardiére heads the Space Plasma Disturbance Specification and Forecast Section of AFRL at Hanscom AFB, Mass.