Nighttime medium-scale traveling ionospheric disturbances (MSTIDs) are a commonly-observed structure in the middle latitude ionosphere during the months surrounding the solstices and low solar flux conditions. MSTIDs consist of one or more narrow bands exhibiting uplift in F-region electron density which are elongated from northwest to southeast in the northern hemisphere and southwest to northeast in the southern hemisphere. These electrified structures propagate westward and toward the equator in a direction perpendicular to their long dimension. Previously, this type of MSTID was not thought to occur at magnetic latitudes equatorward of the equatorial anomalies, or approximately ±15° geomagnetic latitude. Here, we present observational evidence from a field-aligned narrowfield imaging system of MSTIDs existing very close to the magnetic equator. We discuss the implications of this observation in terms of the development and propagation of MSTIDs.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
 Nighttime medium-scale traveling ionospheric disturbances (MSTIDs) are a commonly-observed feature in the 630.0-nm O(1D) nightglow emission at middle latitudes during quiet (Kp ≤ 3) geomagnetic conditions, especially at solar minimum [e.g., Mendillo et al., 1997; Garcia et al., 2000; Shiokawa et al., 2003; Martinis et al., 2010]. MSTID structures were initially discovered to be associated with increases in the F-region peak electron density altitude by Behnke , who suggested that the Perkins instability [Perkins, 1973] might be responsible. This instability arises when the nighttime F layer is supported along slanted magnetic field lines by a combination of E × B drifts and neutral winds against gravity, a condition satisfied at mid-latitudes. The growth rate on its own is quite low, leading to the need for some mechanism, such as a gravity wave, to seed the development of the MSTID [e.g., Kelley and Fukao, 1991]. Nighttime MSTIDs of this type are electrified, exhibiting a radially outward electric field, as measured by satellites [e.g., Saito et al., 1995] and radar [e.g., Kelley et al., 2000]. Although the Perkins instability correctly predicts the alignment of MSTIDs with respect to the background magnetic field, it incorrectly predicts the propagation direction. Kelley and Makela  proposed that if a polarization electric field existed along the long dimension of the bands (thereby requiring them to be finite in that dimension), structures generated by the Perkins instability would travel in the observed direction.
 Although the initial observation was made using the Arecibo radar, MSTIDs are now more commonly studied using airglow images [e.g., Garcia et al., 2000; Shiokawa et al., 2003; Martinis et al., 2010] and images of perturbations in the total electron content obtained using dense networks of GPS receivers [e.g., Saito et al., 2001; Tsugawa et al., 2007; Ogawa et al., 2009], both of which allow for the spatial-temporal dynamics of the structures to be observed. MSTIDs appear as one or more alternating dark and bright bands in images of the 630.0-nm emission due to the uplift (descent) of the F layer, which reduces (increases) the intensity of the emission. The 630.0-nm emission can be used as a surrogate for the Pedersen conductivity [Makela and Kelley, 2003], and so the dark bands seen in the images are regions of low conductivity and are subject to the development of an internal polarization electric field [Kelley et al., 2000]. This polarization electric field maps along the magnetic field lines, explaining the conjugacy of MSTID structures observed simultaneously over Sata, Japan and Darwin, Australia [Otsuka et al., 2004].
 The general properties of nighttime MSTIDs are well known from a variety of studies. In the northern (southern) hemisphere, the bands are elongated approximately from northwest to southeast (southwest to northeast) and propagate to the southwest (northwest). The bands have a wavelength typically between 50 and 500 km, periods of 1 to 2 hours, and propagate with a velocity between 50 and 170 m/s [Garcia et al., 2000; Shiokawa et al., 2003].
Shiokawa et al.  presented results from imagers located in Japan taken during 1999 and argued that an equatorward limit of MSTID propagation was around 18° magnetic latitude. More recently, Lee et al.  reported observations from Taiwan showing MSTID propagation to a magnetic latitude of 13° in 2006. Both of these studies suggested that MSTID propagation could be limited by ion drag caused by the high electron density associated with the equatorial ionozation anomalies. This increased ion drag would cause a gravity wave to dissipate before it can seed the MSTID development. Ogawa et al.  presented results showing MSTIDs propagating from north of Japan using radar, GPS, and optical measurements. Thus, the geographic domain of MSTIDs has been widely considered to extend between the equatorial anomalies and the sub-auroral trough. It is important to note that the MSTIDs of interest here are electrified and typically associated with a Perkins-type instability process. Although previous observations of TIDs at equatorial latitudes [e.g., Röttger, 1977; Shiokawa et al., 2006] showed similar waves with similar periods and wavelengths, their propagation direction (in the meridional direction, rather than westward and towards the equator) suggested that they were not the electrified MSTIDs of interest here.
 In this work, we present for the first time examples of MSTIDs at magnetic latitudes approaching the geomagnetic equator over western South America. This finding is of considerable importance because MSTIDs may be responsible for seeding the post-midnight, quiet-time equatorial plasma bubbles reported from the Pacific sector during the recent deep solar minimum period [e.g., Miller et al., 2009, 2010].
 The Portable Ionospheric Camera and Small-Scale Observatory (PICASSO) is located at the Cerro Tololo Inter-American Observatory (CTIO) near La Serena, Chile (geographic: 30.17 S, 289.19 E; geomagnetic 16.72 S, 0.42 E). The imager is oriented with the optical lines of site approximately parallel to the geomagnetic field in the F region in the configuration proposed by Tinsley . The field of view of PICASSO, projected to an assumed emission altitude of 250 km, is shown in Figures 1 and 2. Based on the International Geomagnetic Reference Field [MacMillan et al., 2003], this field of view covers magnetic inclination angles ranging from −26.0° to −5.5° and declination angles from −7.5° to 4.0°. PICASSO collects 90-second exposures of the 630.0-nm O(1D) dissociative recombination emission line during sun- and moon-down conditions. See Makela and Miller  for complete details.
 The images presented in this paper have been processed as follows. The first step is to spatially register the images using the starfield. The second step reduces noise in the images using an asymptotically optimal blind denoising algorithm developed by Atkinson et al. . Finally, the images are projected into geographic coordinates, assuming an emission altitude of 250 km.
Figures 1 and 2 each illustrate three images of the 630.0-nm emission collected at Cerro Tololo on 20 January 2007 and 27 January 2009, respectively. Note that these are not the only two MSTID events observed by PICASSO since its deployment in August 2006, but are chosen as illustrative examples. A more complete statistical analysis of the MSTIDs observed from this site is outside of the scope of this letter. In both of the cases presented here, the geomagnetic conditions were relatively quiet for 24 hours preceding and following the event. The MSTID band structures appear in the images several hours after local sunset and bear the classical features of MSTIDs. As expected in the southern hemisphere, they are elongated from southwest to northeast, at an angle with respect to the local magnetic field. Both events propagate toward the northwest at a speed of approximately 100 m/s. The width of the band is on the order of several hundred km. These properties are detailed in Table 1. Based on these observed properties, as well as their occurrence during the solstice conditions and low geomagnetic activity, we conclude that these are, indeed, MSTIDs associated with a plasma instability.
Table 1. Properties of Two MSTID Events Observed Using the PICASSO Imaging System at the Cerro Tololo Inter-American Observatory
Dark-Band Width (km)
LT of First Appearance
20 Jan 2007
27 Jan 2009
 The two rather clear events presented in this paper demonstrate that MSTIDs are observable in the airglow at very low geomagnetic latitudes. In both events, the band structure extends all the way to the visible horizon, fading out somewhat due to atmospheric extinction at very low elevation angles of the imager. This corresponds to a magnetic latitude less than 5°. This conclusively demonstrates that MSTIDs can occur significantly closer to the magnetic equator than previously observed.
 The results of Ogawa et al.  show that MSTIDs occur simultaneously over a very large region, rather than being generated in one region only and propagating to other regions to be observed. This would suggest that the structures observed here might be generated locally. However, the conditions studied by Perkins  become increasingly less valid the closer one gets to the magnetic equator. Essentially, the situation studied by Perkins involves perturbing the balance between the downward diffusion of the F layer caused by gravity along a slanted magnetic field line and the upward E × B drift:
where g is gravity, D is the magnetic inclination (dip) angle, 〈νin〉 is the field-line integrated ion-neutral collision frequency, c is the speed of light, E0y is the zonal component of the background electric field, and B is the magnetic field intensity. This equation is equivalent to equation 17 of Perkins . However, close to the equator, the downward diffusion velocity tends to zero due to the near-horizontal magnetic field lines, and the balance represented by the above equation is not set up. This raises an interesting dilemma. If the Perkins instability is ultimately responsible for the development of MSTID structures, it seems unlikely that the MSTIDs observed from CTIO were generated locally, as the magnetic inclination angles range from −24° to the south of PICASSO's field of view to −8° to the north.
 Thus, it would appear more likely that the MSTIDs observed in our images must be generated farther poleward of the observing region, eventually propagating into our field of view. This would seem to be supported by the fact that the MSTID in the images presented here appear later in local time, typically close to local midnight. That is, enough time must elapse between their generation farther poleward and their appearance close to the magnetic equator. For the examples presented here, the MSTIDs first appeared approximately three hours after local sunset (see Table 1). Taking a drift velocity of 100 m/s at an azimuth angle of 300° results in a projected velocity of approximately 50 m/s in the equatorward direction. Thus, if the observed MSTIDs developed immediately after sunset, they could have propagated at most 540 km, or about 5° in latitude. The magnetic inclination angle 5° south of this location is D = −33.5° and the Perkins stability requirement may be satisfied. This would put the growth region near El Leonicto, Argentina, where MSTIDs have been observed previously [Martinis et al., 2006]. However, even if the Perkins instability is active at this location, the growth rate would be quite small, an oft-cited drawback of the Perkins mechanism.
 This modest growth rate for the Perkins instability would have to be augmented by an increase in the effectiveness of a seeding mechanism. Gravity waves have been suggested as one such mechanism [e.g., Kelley and Fukao, 1991; Miller et al., 1997]. Both Shiokawa et al.  and Lee et al.  argued that the increase in electron density near the regions associated with the equatorial ionization anomalies would increase the ion drag, dissipating gravity waves before they could potentially seed the development of MSTIDs. As the background electron density decreases, the time constant for ion drag reduces and it is more likely that MSTIDs could be seeded by gravity waves, perhaps overcoming the reduction in the Perkins growth rate. This would suggest that our results, and those for MSTID-seeded EPBs presented by Miller et al.  might be specific to the deep solar minimum and the resultant reduced electron densities. This may explain why no reports of MSTIDs at low latitudes were made for the previous solar minimum period.
 Although it had been noted in previous optical studies [e.g., Kelley and Makela, 2001] that MSTIDs were finite in this direction, it is plausible that the MSTID structures may simply be significantly elongated in the perpendicular-to-k direction. The expanded view of nighttime MSTIDs provided by the dense GPS networks in North America show that the bands can extend as long as 2,000 km [Tsugawa et al., 2007]. Thus, it is plausible that the structures being presented here were actually seeded further poleward and simply elongate to the magnetic equator. Whether the structures develop fully elongated or elongate over time as they propagate towards the magnetic equator cannot be differentiated from our observations from this single site. Verification of this hypothesis will require observations in regions to the south and east of CTIO.
 Alternately, Tsunoda and Cosgrove  have proposed a mechanism in which the F region and E region are coupled and the electric fields initiated by current-driven Hall polarization in sporadic E layers would cause positive feedback. The growth rate of this process is larger than that provided by the Perkins instability alone [Cosgrove et al., 2004]. However, these studies were performed under midlatitude conditions and magnetic field geometries, so it is unclear at this time if this coupling mechanism would be effective for the geometries suggested by our observations of MSTIDs close to the magnetic equator (especially the larger spatial separation and differing magnetic field inclinations of the F and E layers).
 In this paper, we have presented evidence of MSTID band structures approaching the magnetic equator. This is significantly closer to the equator than had been previously observed. We suggest that this may be a result of the deep solar minimum during which the observations were obtained. However, it is not yet clear if the traditional view of the development of the MSTIDs controlled by either the Perkins instability or a coupled E–F layer mechanism can explain their occurrence this close to the magnetic equator and further theoretical and modeling work is needed to completely understand how MSTIDs can occur this close to the magnetic equator. Nevertheless, these observations of the occurrence of these structures at near-equatorial latitudes lends credence to the hypothesis of Miller et al.  that these structures, and more specifically their internal polarization electric field, are a plausible seed for the post-midnight, quiet-time equatorial plasma bubbles observed during the most recent deep solar minimum.
 Work at the University of Illinois at Urbana-Champaign was supported by the National Science Foundation through grant ATM-06-44654. ESM acknowledges support from a National Science Foundation CEDAR post-doctoral award AGS-0924914. ERT was supported by the National Science Foundation through AGS-0946902.