Pages 1037-1042


M. Brittnacher1 and G. K. Parks1 and J. F. Spann2 and G. A. Germany3

1 Geophysics Program, University of Washington, Seattle, WA 98195, USA
2 Space Sciences Laboratory, Marshall Space Flight Center, Huntsville, AL 35812, USA
3 Optical Aeronomy Laboratory, University of Alabama, Huntsville, AL 35899, USA


The Ultraviolet Imager (UVI) experiment on the Polar spacecraft has been operating since 21 March 1996, and is returning high quality global images of auroral activity. This article presents first images of global aurora from the UVI experiment. We discuss the capabilities of the UVI instrument and how the data can be used to study substorm physics. A preliminary survey of 80 days of UVI images indicates that the dayside auroral activity is intense and dynamic, and occurs as frequently as the nightside activity. These features are demonstrated using the images obtained on 9 April 1996 when the UVI observed simultaneous dayside and nightside activities.


Global auroral imaging in ultraviolet wavelengths provides important information on solar wind- geomagnetic field interactions (Kaneda et al., 1977; Frank et al., 1981; Meng and Huffman, 1984; Anger et al., 1987). This article presents first results from the UVI experiment (Torr et al., 1995) on the Polar spacecraft. Polar was successfully launched on 24 February 1996 into an 86 ° inclination orbit, with an apogee of 9 RE (geocentric distance in Earth radii) in the northern hemisphere and a perigee of 1.8 RE. The period of Polar is about 18 hours. The UVI experiment is one of four instruments mounted on a despun platform which can be oriented along one axis. The other experiments are an X-ray camera (Imhof et al., 1995), a visible wavelength camera (Frank et al., 1995), and a particle instrument (Blake et al., 1995) that images particles in and near the loss cone.

A unique feature of the UVI camera is the newly developed narrow-band, far ultraviolet filters (Zukic et al., 1993) that are flown in space for the first time. Figure 1 shows the measured characteristics of the four UVI filters on Polar determined before launch. The instrument counts (in the CCD) per photon flux as a function of wavelength, measured relative to the nominal operating gain level during ground calibration, indicates the sensitivity and spectral bandwidth of each filter. The 130.4 ± 4.0 nm and 135.6 ± 6.0 nm filters are designed to detect separate atomic oxygen emission lines, and the 140-160 nm (LBHs) and 160-180 nm (LBHl) filters are designed to detect the Lyman-Birge-Hopfield molecular nitrogen emissions. By distinguishing the two atomic oxygen emissions and the longer and shorter portions of the LBH band we are able to study and model the auroral and dayglow emissions more accurately than was previously possible.

Fig. 1. The characteristics of the four filters determined by ground measurements. The calibration scale is in terms of counts (in the CCD) per photon flux. The absolute scale is valid for the instrument gain level 13 with the aperture door open.

The total field of view (FOV) of UVI is 8° and the single pixel resolution is approximately 0.04°. A single pixel projected to 100 km altitude from apogee is approximately 40x40 km (however, the Polar spacecraft has an unresolved wobble that degrades the pixel resolution to approximately 1x10 pixels). Typical image acquisition times are 18.6 and 36.8 seconds. Since the UVI can stare at the polar region continuously for 6 to 8 hours, an entire auroral substorm sequence can be observed without interruption. Information on changes in auroral features is obtained with timescales that are less than or comparable to many of the dynamic timescales of solar wind-geomagnetic field interactions.


The four images in Figure 2a-d, which were acquired during a 5 minute period on 9 April 1996, are presented as an example of the qualitative difference between what is observed in each of the four filters. The false-color scale for each of the images is calibrated in units of photon flux at the UVI aperture. Geographical gridlines are shown in order to compare the relative location of the day/night terminator. Local noon is just below the center of the right side of each image. The terminator has a small tilt from the vertical and is located near 90 ° N latitude. In the 130.4 nm image (Figure 2a), the terminator appears to be shifted anti-sunward from the position seen in the LBH images (Figures 2c and 2d). In the 135.6 nm image (Figure 2b), the terminator is slightly anti-sunward of the LBH location. The outlines of the auroral oval are clearly seen in the LBHs, LBHl, and 135.6 nm images, even on the dayside, but are less distinct in the 130.4 nm image. These images show that leakage of solar emissions into neighboring filters is minimal, corroborating the sharp long-wavelength cutoff in the LBH and oxygen filters as determined during ground calibration. It is apparent from these images that the UVI can clearly distinguish the dayside aurora from the dayglow emissions.

Fig. 2a-d. Four images, obtained with the (a) (upper left) 130.4 nm, (b) (upper right) 135.6 nm, (c) (lower left) LBHs, and (d) (lower right) LBHl filters over a five minute period on 9 April 1996, demonstrate the differences seen by narrow-band separation of the spectral emissions. The images are overlaid by geographic latitude circles in 10 ° increments and longitude lines every 45 ° from 0 °. The "N"-shaped structure in the upper right of all four images is an image artifact.

A sequence of four UVI images taken near apogee on 9 April 1996 is shown in Figures 2e-2h. These images are representative of other images obtained by the UVI during the first 80 days of operation when the geomagnetic activity was moderate. One of the purposes of this article is to provide relevant information about these images and to discuss how the UVI data can be used for studies of large-scale auroral and magnetospheric dynamics. The images, which are plotted in geographic coordinates, demonstrate that at apogee the circular FOV enables viewing of the Northern hemisphere above 60 ° N latitude, and during moderate geomagnetic conditions the UVI covers the entire auroral oval. The sunlit portion of Earth is toward the lower right side of the images. Although the day-night terminator runs through the auroral oval at the time these images were acquired, the dayglow (seen as a blue-green background) is relatively suppressed compared to the bright auroral intensity.

This sequence of images, taken within a period of 45 minutes with the LBHs filter, illustrates the dynamics of the dayside aurora and the onset of an auroral substorm on the nightside. The false-color scale is the same for all four images in order that relative differences in intensity may be observed. Observation of solar wind plasma by the SWE experiment (Ogilvie et al., 1995) and the interplanetary magnetic field (IMF) by the MFI experiment (Lepping et al., 1995) aboard the Wind spacecraft were used to characterize the solar wind during this sequence of images. The solar wind speed was 375 ± 15 k/s and the Wind spacecraft was located about 80 RE upstream of Earth. A propagation time of about 18 minutes was accounted for in all correlations between Wind data and UVI observations.

The first image taken at 1357 UT (Figure 2e) shows intense activity in the dayside auroral oval (over North America and Greenland) and weak activity on the nightside (over Siberia). Although it appears that the nightside emissions are weaker than the dayside, the intensity of the dayside oval includes the contribution from airglow. In order to make an accurate comparison of dayside and nightside auroral precipitation, it will be necessary to remove the portion of the emissions that are produce by airglow. Two distinct regions of precipitation in the morning and afternoon sectors, with a minimum in the emission level near noon, can be seen. The latitudinal extent of the westward (morning) region is broader than the afternoon region and the maximum intensity is greater.

Fig. 2e-h. Substorm sequence at (e) 1357 UT (upper left), (f) 1436 UT (upper right), (g) 1439 UT (lower left), and (h) 1444 UT (lower right). The images are overlaid by continent boundaries and the same geographic gridlines as in Fig. 2a-d.

The next image taken at 1436 UT (Figure 2f) shows the dayside auroral activity continued and intensified. The eastward portion of the active region of the dayside aurora has shifted equatorward, and the brightest region was in the afternoon sector. This dayside intensification increased while the nightside auroral activity continued to subside. The nightside auroral oval is faint but still discernible. The first indication of an onset of auroral activity on the nightside occurred three minutes later at 1439 UT (Figure 2g). This onset, near local midnight, coincided with a small but noticeable decrease of pre-existing auroral structures on the dayside. The IMF BZ component declined from 6 nT to about 0 nT over the 25 minute period before onset and, for the ten minute period prior to onset, was 0 ± 0.5 nT. Near the time of onset the IMF BZ component went southward but fluctuated between 0 and -1 nT. The IMF BY component was -6 ± 1 nT for the previous 20 minutes. By inspection of the IMF data, an obvious trigger for the substorm is not apparent.

The final image, 1444 UT (Figure 2h) shows further brightening of the nightside region which has now expanded in both longitude and latitude. The auroral oval is now active at nearly all local times, including the evening sector. However, three distinct regions of precipitation that may map to different regions of the magnetosphere can still be distinguished. The nightside auroral activity continued to intensify and expand westward for ten more minutes before it began to subside (not shown). Note that the nightside auroral activity occurred while the dayside aurora continued to be dynamic and active. We are currently studying what relationships exist, if any, between the dayside and nightside auroral activities. Preliminary examination of the UVI data thus far indicates that the dayside auroral activity is a common occurrence and can be very dynamic and intense. Dayside auroral forms have been observed previously by Meng and Huffman (1984) and Elphinstone et al. (1989, 1992). Activity near the 14 MLT sector that is comparable to discrete arc precipitation on the nightside during auroral-breakup conditions was previously observed in Viking images (Lui et al., 1987; 1989). In addition to the dayside and nightside auroral activities, very faint precipitation occurring in the polar cap region can be seen along the terminator.

As mentioned above, the dayside auroral activity occurred throughout this day. No obvious correlation between the IMF and the dayside auroral activity was observed in this preliminary analysis. Quantitative methods will be used in future studies to determine whether the dayside, or more localized, auroral precipitation is strongly correlated with IMF parameters, an observation that has been reported based on ground scanning photometer measurements near noon (Sandholt, 1990).


One of the main objectives of the Polar UVI camera is to obtain quantitative information from which important auroral parameters can be derived. These parameters include the characteristic energy and energy flux of auroral precipitation at all local times. The N2 LBH transitions are particularly suited to this task since they are electric dipole forbidden and the main excitation mechanism is electron impact (Germany et al., 1994a). The energy flux of precipitating electrons can be derived from the photometric brightness recorded in the images obtained with the LBHl filter. Since the longer wavelength LBH emissions are only mildly absorbed by O2, the brightness in Rayleighs is linearly related to the incident energy flux. The observed longer wavelength LBH emissions are also insensitive to the characteristic energy and the energy distribution (Germany et al., 1994a). The characteristic energy can be derived from the fact that the cross- section for O2 absorption is much greater for the shorter wavelength LBH emissions and that the penetration depth for auroral electrons increases with the energy. The photometric brightness of the images taken with the LBHs filter has an inverse dependence upon the characteristic energy of the electrons. This means that auroral precipitation with a higher characteristic energy actually appears less bright in the image compared with softer precipitation. The characteristic energy can be inferred by calculating the ratio of the LBHs to LBHl brightness measured in consecutive images. Here we are making the assumption that the time scale for significant changes in the aurora averaged over a region on the order of 100x100 km is longer than a few image acquisition periods (about 2 minutes). The 135.6 nm atomic oxygen emission is similar to the short wavelength LBH emissions and will be used as a separate diagnostic to determine characteristic energies. Since the ionospheric concentration of atomic oxygen is much more variable than N2, the variation in the observed ratios of 135.6 nm and LBHs to LBHl intensities can be used as a further diagnostic of the variation in the high altitude oxygen concentration. Determination of the characteristic energies and energy fluxes of auroral electrons will enable us to study, for example, the sources of dayside precipitation and how much energy is deposited on the dayside, relative to other local times. These quantities will be used as input parameters into models of the global atmosphere-magnetosphere-ionosphere system (Rees et al., 1995), ionospheric conductivities (Germany et al., 1994b) and the influence of the aurora on the composition of the neutral atmosphere (Germany et. al., 1990).


There are still many unresolved questions about the aurora and its connection to the dynamics of the solar wind-geomagnetic field interaction. What is the source of the dayside precipitation and is there a relationship between the dayside and the nightside auroral activities? Are all substorms initiated on closed field lines? What mechanisms are responsible for substorms and what is the role of the geomagnetic tail and the plasma sheet? These are some of the important questions that need to be studied in the quest for understanding how the solar wind mass, momentum and energy are transported into the magnetosphere.

The UVI observations can provide a useful framework for studying these problems. A systematic investigation of the UVI images in correlation with high quality data obtained by particle and field experiments on Polar as well as data from ground-based instruments and various other spacecraft (Wind, Geotail, Interball, etc.) should guide researchers to a deeper understanding of magnetosphere-ionosphere dynamics and insight into what physical mechanisms are involved.


The UVI camera was designed by Dr. M. Torr and the filters by Dr. M. Zukic. The UVI research is supported in part by a grant from the National Aeronautics and Space Administration, NAG 5-3170, to the University of Washington.


Anger, C. D., S. . Babey, A. Lyle Broadfoot, R. G. Brown, L. L. Cogger, et al., An ultraviolet imager for the Viking spacecraft, Geophys. Res. Lett., 14, 387, 1987.

Blake, J. B., J. F. Fennell, L. M. Friesen, B. M. Johnson, W. A. Kolasinski, et al., Comprehensive energetic particle and pitch angle distribution experiment on Polar, Space Sci. Rev., 71, 531, 1995.

Elphinstone, R. D., and J. S. Murphree, Dayside auroral forms as observed from above by the Viking UV imager, in Report of the workshop on ionospheric signatures of cusp, magnetopause and boundary layer processes, ed. by T. Rosenberg, Univ. of Maryland, 1989.

Elphinstone, R. D., J. S. Murphree, D. J. Hearn, L. L. Cogger, P. T. Newell, and H. Vo, Viking observations of the UV dayside aurora and their relationship to DMSP particle boundary definitions, Ann. Geophysicae, 10, 815, 1992.

Frank, L. A., J. D. Craven, K. L. Ackerson, M. R. English, R. H. Eather, and R. L. Carovillano, Global auroral imaging instrumentation for the Dynamics Explorer mission, Space Sci. Instr., 5, 369, 1981.

Frank, L. A., J. B. Sigwarth, J. D. Craven, J. P. Cravens, J. S. Dolan, et al., The visible imaging system (VIS) for the Polar spacecraft, Space Sci. Rev., 71, 297, 1995.

Germany, G. A., M. R. Torr, P. G. Richards, and D. G. Torr, The dependence of modeled OI 1356 and N2 Lyman Birge Hopfield auroral emissions on the neutral atmosphere, J. Geophys. Res., 95, 7725, 1990.

Germany, G. A., M. R. Torr, D. G. Torr, and P. G. Richards, Use of FUV auroral emissions as diagnostic indicators, J. Geophys. Res., 99, 383, 1994a.

Germany, G. A., M. R. Torr, P. G. Richards, and D. G. Torr, Determination of ionospheric conductivities from FUV auroral emissions, J. Geophys. Res., 99, 23297, 1994b.

Imhof, W. L., K. A. Spear, J. W. Hamilton, B. R. Higgins, M. J. Murphy, et al., The Polar ionospheric X-ray imaging experiment, Space Sci. Rev., 71, 385, 1995.

Kaneda, E. et al., Vacuum aurora television camera, in Proceedings of the twelfth international symposium on space technology and science, Tokyo, Japan, 233, 1977.

Lepping, R. P., M. H. Acuna, L. F. Burlaga, W. M. Farrell, J. A. Slavin, et al., The Wind magnetic field investigation, Space Sci. Rev., 71, 207, 1995.

Lui, A. T. Y., D. Venkatesan, G. Rostoker, J. S. Murphree, C. D. Anger, et al., Dayside auroral intensifications during an auroral substorm, Geophys. Res. Lett., 14, 415, 1987.

Lui, A. T. Y., D. Venkatesan, and J. S. Murphree, Auroral bright spots on the dayside, J. Geophys. Res., 94, 5515, 1989.

Meng, C.-I., and R. E. Huffman, Ultraviolet imaging from space of the auroral under full sunlight, Geophys. Res. Lett., 11, 315, 1984.

Ogilvie, K. W., D. J. Chornay, R. J. Fritzenreiter, F. Hunsaker, J. Keller, et al., A comprehensive plasma instrument for the Wind spacecraft, Space Sci. Rev., 71, 55, 1995.

Rees, M. H., D. Lummerzheim, and P. G. Roble, Modeling of the atmosphere-magnetosphere-ionosphere system MAMI, Space Sci. Rev., 71, 691, 1995.

Sandholt, P. E., M. Lockwood, T. Oguti, S. W. H. Cowley, S. C. Freeman, et al., Midday auroral breakup events and related energy and momentum transfer from the magnetosheath, J. Geophys. Res., 95, 1039, 1990.

Torr, M. R. et al., A far ultraviolet imager for the international solar-terrestrial physics mission, Space Sci. Rev., 71, 329, 1995.

Zukic, M., D. G. Torr, J. Kim, J. F. Spann, and M. R. Torr, Filters for the International Solar Terrestrial Physics Mission far-ultraviolet imager, Opt. Engineering, 32, 3069, 1993.