A Multi-Spacecraft Approach to Studying Auroral Kilometric Radiation Using the Virtual Wave Observatory

According to the analysis of the radio wave spectrograms captured by the IMAGE, Geotail, Polar, Wind, Stereo-A, and Stereo-B spacecraft, the correlation between the auroral kilometric radiation (AKR) intensity and the auroral electrojet (AE) index have been found to be negatively correlated as well as positively correlated. These findings raised questions against the present notion that AKR intensity and the AE index are positively correlated. Although the original objective of this study was to establish a set of stable threshold values in the auroral AE index for AKR detection, the negative correlations found between the AKR intensity and the AE index made these threshold calculations unattainable. Furthermore, this study was the first attempt to establish a set of stable threshold AE index values associated with AKR using a multi-spacecraft approach provided by the Virtual Wave Observatory (VWO). This approach was essential to ensure that any changes observed in the spacecraft data were due solely to the AKR source. The whole spectrum of AKR was analyzed, normally 80-800 kHz in the satellite spectrograms. The threshold value was determined by comparing AE index plots of AKR events to corresponding satellite spectrograms of the events. However, the poor correlations found through this study indicated that there may be other drivers affecting the AKR intensity other than the strength of the auroral electrojet current. Thus, a future multi-parameter magnetosphere investigation using other geometric indices, such as the Dst index or the Kp index in comparison with the AE index and AKR intensity, could explain how unknown drivers of the AKR intensity and the AE index synergistically caused the inconsistent correlations. The understanding of the magnetosphere conditions, as measured by the AE index, which trigger the AKR emission is crucial to planetary research because it provides a mechanism to remotely sense the state of the Jupiter’s, Saturn’s, Neptune’s, and Uranus’s magnetospheres. Introduction In depth understanding of auroral kilometric radiation (AKR) is very crucial to future planetary research due to its similarities to extraterrestrial non-thermal radiations such as Jupiter’s decametric radiation and Saturn’s saturnian kilometric radiation. These radiation emissions are similar to AKR but are unique in structure based on the dynamics of their magnetospheres and solar wind environments1. In addition, non-thermal radio emissions can be A Multi-Spacecraft Approach to Studying Auroral Kilometric Radiation Using the Virtual Wave Observatory Huy Lam1*, Teresa Petralli-Mallow2, Leonard Garcia3, and Shing Fung4 Student1, Teacher2: Poolesville High School, 17501 West Willard Road, Poolesville, Maryland 20837 Mentor/Senior Programmer3: NASA/GSFC, Greenbelt, MD 2077 Mentor/Research Astrophysicist4: NASA/GSFC, Greenbelt, MD 2077 *Correspondence: [email protected] 51 used to remotely sense the magnetosphere conditions of complex planets such as Jupiter and Saturn. This would benefit future spacecraft launches to these planetary systems to observe high level interest moons within Jupiter and Saturn’s magnetospheres such as Jupiter’s moon, Io, and Saturn’s moon, Enceladus, by ensuring that favorable conditions are present for spacecraft hardware and observation. Furthermore, Jupiter and Saturn are gaseous planets, and using non-thermal radiation emissions to remotely gather information to profile them would be more practical since there is no tangible surface to survey. The sun continually emits a fast-moving stream of plasma known as solar wind. Fortunately, Earth’s magnetosphere acts as a shield that protects life on Earth from the solar wind. Interactions between the solar wind and the magnetosphere create disturbances that can lead to the production of intense non-thermal radio emissions such AKR (Fig. 1)2. Often associated with the auroras, AKR is Earth’s most powerful and well-observed non-thermal radio radiation3,4. The source of AKR is broadly beamed and found near the North and South Auroral Zones on the night side of Earth. When solar winds strike the magnetosphere, electrons travel along Earth’s magnetic field lines and are then driven down to Earth’s auroral electrojet in our high-latitude ionosphere. The influx of electrons causes unstable polarization of the ionosphere which results in AKR emission and aurora borealis generation (Fig. 2)5. Fig. 1. An AKR radio source in relation to a magnetic field line and Earth. Electrons are accelerated downward towards Earth [Louarn, (2006)]5. Fig. 2. The auroral electrojet current in Earth’s upper ionosphere is depicted. The points represent the magnetometer stations. NLK, OND, ANT, and RUG are relay stations connecting to the magnetometer stations that process the AE index data. [Space, Telecommunications, and Radioscience Laboratory, (1997)]. The Journal of Experimental Secondary Science January 2012 Volume 1 Issue 3 Twelve ground data stations, located in Earth’s northern polar regions, measure the strength of the auroral electojet current and represent the strength of the current with the auroral electrojet (AE) index. As a result, the strength of magnetosphere disturbances due to solar winds in the auroral regions can be measured by the AE index. Voots et al. [1977]6 concluded that the AKR intensity was a reliable indicator of the AE index through statistical analysis using the IMP-6 satellite. This notion has been adopted since then and applied in research. For example, Kurth and Gurnett [1988]7 developed a proxy to derive the AE index value from the AKR intensity which led to the development of the “AKR index” still used today, and Kunamoto et al. [2005]3 established seasonal and solar cycle correlations between the AKR intensity and the AE index for the “AKR index”. With the notion that the AKR intensity was a reliable indicator of the AE index, the original objective of the study was to establish a set of stable threshold AE values for the AKR emission by using a novel multi-spacecraft approach with datasets, mainly spectrograms, provided by the freeware Virtual Wave Observatory (VWO)( http://vwo.nasa.gov). The threshold value is one of the factors that can be used in sensing magnetosphere conditions needed for AKR emission. In addition, this study was the first attempt to establish a threshold AE value for the AKR emission using the VWO. Before the VWO, the act of gathering and comparing multi-spacecraft datasets of a magnetospheric event was encumbered by time scale differences and cross database searching. The study of AKR using more than one satellite required simultaneous download access to multiple satellite databases and data navigation tools. The VWO seeks to solve this problem by connecting to all satellite databases and acting as a data delivery service. By using the VWO, datasets were quickly gathered and compared to the AE index from multi-spacecraft databases with a single search engine. The multi-spacecraft approach ensured that any simultaneous changes seen in the spectrograms were due to changes in the source and rather than the orbit or instrumentation of the particular spacecraft. Furthermore, this approach also greatly improves the spatial coverage of the AKR source7. Daily spectrograms from the following spacecraft were analyzed: Geotail, Wind, IMAGE, Polar, Stereo A, Stereo B, and Cluster. Corresponding AE values from the daily spectrograms were plotted to correlate with the observed AKR activity. The approximate threshold of AE value was determined to be below a certain AE value when AKR emission is not observed. Because Green et al. [2004]8 concluded that the AKR frequency and intensity vary seasonally as a function of dipole tilt of the Earth, seasonal data was recorded. Unexpectedly, the AKR intensity and the AE index were found to be negatively correlated which disputed findings by Voots et al. [1977]6 and put previous AKR intensity and AE index value studies into question including: Green et al. [2004]8, Kurth and Gurnett [1988]7, Kunamoto et al. [2005]3, Kurth et al. [1998]7, and Benson and Akasofu [1984]4. In addition, the results revealed that there is a great fluctuation in the threshold values. Huy Lam, Teresa Petralli-Mallow, Leonard Garcia, and Shing Fung Page 2 of 7 52 Materials and Methods Spacecraft Orbital Alignment: Data was collected from Geotail, Polar, Wind, IMAGE, Stereo A, Stereo B, and Cluster spacecrafts as spectrograms using the VWO Data Query. These spacecraft orbit Earth with elliptical trajectories with Earth as a focus. Geotail was operational from 1992-2007, Wind was operational from 1994-present, Polar was operational from 1996-1997, IMAGE was operational from 2000-2005, and Stereo A and Stereo B were only close enough to Earth to observe AKR from late 2006-early 2007. The freeware SSCWeb (http://sscweb.gsfc.nasa.gov) orbit viewer was used to determine and document at what months the spacecraft had their orbital apogee in the northern hemisphere. The documented months were then searched for spacecraft datasets containing closely similar AKR signatures (Fig. 3 below) during the times when all spacecraft were in the northern hemisphere (to make sure the spacecraft were viewing the same AKR event) and downloaded. Establishing AE Index Threshold Values: Comparisons were made between the AKR intensity and the AE index. All frequencies within the bandwidth of the AKR emission (normally 80-800 kHz) were analyzed. AKR was considered “On” when at least two spectrograms showed an AKR signature and was considered “Off ” when at least two spectrograms in our datasets showed no AKR signatures (Fig. 3). Times when AKR was “Off ” or “On” was then determined using the spacecraft time scale. Numerical, one minute resolution AE index data from the Kyoto Data Index Service (http:// wdc.kugi.kyoto-u.ac.jp/aedir/index.html) for the day and times in which AKR emission occurred was then downloaded. This AE index data was then plotted and aligned in time with spacecraft spectrograms. Datasets were considered to be positively correlated if the AE index values increased and decreased together with the AKR intensity. Datasets were considered to be negatively correlated if the AE index values increased and the AKR intensity decreased or if the AKR intensity increased and the AE index values decreased. A threshold value, designated by a red line (as shown in Fig. 4, 6, 8, 10), was then determined and recorded only for positively correlated datasets. This threshold was established by locating a point in the AE index plot where AKR no longer emits and connecting that point with another point in the AE index plot where AKR reemits. AKR emission was not expected when the AE index values were below the threshold value. Poor correlations made threshold values indeterminable. The number of magnetometer stations that registered AE index measurements for each dataset was then recorded. To account for the seasonal variation in AKR intensity and frequency discovered in Green et al. [2004]8 and Kunamoto et al. [2005]3, the thresholds were organized by season. Statistical Analysis: The results (threshold values) were planned to be standardized and statistically analyzed. However, statistical analysis was not able to be accomplished because the threshold values were found to be unstable due to the unexpected finding associated with poor correlation between the AKR intensity and the AE index. Figure 3. AKR signature recorded by the Polar spacecraft showing “On” and “Off ” AKR. The Journal of Experimental Secondary Science January 2012 Volume 1 Issue 3 Huy Lam, Teresa Petralli-Mallow, Leonard Garcia, and Shing Fung Page 3 of 7 53 Results Although several datasets were analyzed, the two most outstanding datasets of positively correlated AKR intensity and AE index values (Fig. 4,5,6,7) and two datasets of negatively correlated AKR intensity and AE index values (Fig. 8,9,10,11) are presented and discussed in detail. The spectrograms displayed the intensity of AKR emission as a function of frequency (Hz or KHz) and time in hours (Universal Time (UT)). In these spectrograms, AKR exhibited a distinctive oscillating structure. Spectrograms and AE index plots were stacked vertically in time to show changes in the AKR intensity with relation to changes in the AE index, measured in nanoteslas (nT). AKR signatures were defined as “On” and “Off ” in each figure and vertical lines show the portions in the AE index plot where the AKR was “On” and “Off ”. The threshold was then determined. This was indicated by a red line. In addition to these datasets, several others are also included to further illustrate the negatively correlation between the AKR intensity and AE index. Table 1 documented the dates of other studied cases of positively correlated and negatively correlated datasets with the number magnetometer station detection. Table 2 showed fluctuations in the calculated thresholds for positively correlated datasets. Furthermore, the observations associated with the usage of the multi-spacecraft approach and the VWO are discussed in this study. Fig.4. Spacecraft observations from Geotail, Wind, and Polar with AE index plot on 9/10/1996 showing positively correlated AKR intensity and AE index values. The indicated threshold is 275 nT. Circles show the advantage of using a multispacecraft approach. Wind observes AKR “Off ”, but Geotail and Polar observe weak, not disappearance of AKR. Even with positively correlated datasets, minor anomalies for short periods of time such as between 0500-0600 UT exist. The multi-spacecraft approach allowed this study to greatly reduce the spacecraft bias that could have occurred if datasets from only one spacecraft had been used. An example illustrating the strength of this approach was shown in Fig. 4, when Wind observed an AKR gap from 06:00-07:00 UT (red circles in Fig.4). If only the Wind spacecraft dataset was studied, it could have been concluded that from 06:00-07:00 UT, AKR was “Off ”. However, there was strong indication of AKR emission observed by both Geotail and Polar during this time window. Reducing spacecraft bias ensured that the changes in spacecraft spectrograms were due to solely the AKR source and not the orbital trajectory or the instrumentation of the spacecraft. In addition, multi-spacecraft datasets from Geotail, IMAGE, Wind, Polar, Stereo-A, and Stereo-B were quickly retrieved and organized by the VWO for analysis. This study demonstrated the power of the VWO in cross dataset wave data retrieval and the importance of the VWO in future research. Thresholds of Positively Correlated Datasets. Voots et al. [1977]6 concluded that the AKR intensity and the AE index were positively correlated. As a result, much research has been based on the notion that AKR intensity was a reliable indicator for the AE index, for example, Green et al. [2004]8, Kurth and Gurnett [1988]7, Kunamoto et al. [2005]3, Kurth et al. [1998]9, and Benson and Akasofu [1984]4. Fig. 4 and Fig. 6 highlighted most prominently a positive correlation between the AKR intensity and the AE index. As a result, threshold AE index values for AKR were effectively established on these datasets. Although the threshold values were effectively established, based on the theory developed by Voots et al. [1977]6, Table 2 showed that the values varied greatly across multiple data sets, including values within their respective seasons. These fluctuations indicated that there exists other influential driver(s) of the AE index, not just AKR. To further understand these fluctuations, in order to establish a set of stable threshold values, additional studies need to be conducted. A multi-parameter magnetospheric state study is the next step to examine how the magnetosphere as a whole affects AKR emission as well as the auroral electrojet current and therefore the AE index. An additional study of the AE index trends using data from months prior to 09/10/1996 and 06/05/1996 may also help future studies understand to what extent magnetosphere conditions prior to the emission of AKR could affect the AKR emission. Negatively Correlated Datasets: In contrast to the theory developed by Voots et al. [1977]6, the analysis of the results also indicated that the AKR intensity was, in many cases, negatively correlated with the AE index. Fig. 8 and Fig. 10 highlighted most prominently how the AKR intensity and the AE index were negatively correlated. In some gaps, AKR was “Off ” but AE values surpassed the threshold value (Fig. 8). In others, AKR was “On” but AE values remained well below the threshold value (Fig. 10). Voots et al. [1977]6 concluded that these observed negative correlations were caused by the position of the spacecraft and the lack of magnetometer station spatial coverage, not the AKR source or the AE current. However, the negatively correlated AKR and AE index results challenge these conclusions. Voots et al. [1977]6 claimed that large AE index values corresponding with little AKR intensities or “Off ” AKR was because the IMP6 satellite was on the dayside of Earth. Due to a cutoff propagation effect of AKR, Voots et al. [1977]6 explained that dayside satellites could not detect nighttime AKR at low latitudes which accounted for high AE index values but no AKR. However, the analysis of the data showed that on 06/04/2002, IMAGE and Geotail were on the night side of Earth when they observed the AKR gap between 02:00-04:30 UT (Fig. 8). On the same day, Wind was at a high latitude (high z coordinate value) and also observed the same AKR gap from 02:00-04:30 UT (Fig. 8, 9). Similar AKR structures in spectrogram plots from IMAGE, Geotail, and Wind confirmed that the AKR gap was due only The Journal of Experimental Secondary Science January 2012 Volume 1 Issue 3 Huy Lam, Teresa Petralli-Mallow, Leonard Garcia, and Shing Fung Page 4 of 7 54 Fig. 5. This is a Cartesian coordinate system representation of spacecraft trajectories on 09/10/1996 from 0000-0600 UT with Earth as the origin (Re is Earth radii and the Sun is in positive X direction). Because the Wind spacecraft is much farther away from Earth than Geotail and Polar, Wind has its own orbital plot to prevent scale distortion. All spacecraft are were good positions to observe northern AKR and were most likely viewing the same AKR emission. This was confirmed in Fig. 4 where all spacecraft observed similar AKR structures. Fig. 6. Spacecraft observations from Geotail, Polar, and Wind with AE index plot on 06/05/1996 showing positively correlated AKR intensity and AE index values. The indicated threshold value is 50 nT. Fig. 7. This is a Cartesian coordinate system representation of spacecraft trajectories on 06/05/1996 from 1200-1800 UT with Earth as the origin (Re is Earth radii and Sun is in positive X direction). Because the Wind spacecraft is much farther away from Earth than Geotail and Polar, Wind has its own orbital plot to prevent scale distortion. Although Wind is in the southern hemisphere, it was far away enough from Earth to view the northern AKR source. This was confirmed in Fig. 6 where Wind observed similar AKR structures compared to Geotail and Polar which were in the northern hemisphere. Fig. 8. Spacecraft observations from IMAGE, Wind, and Geotail with AE index plot on 06/04/2002 showing negatively correlated AKR intensity and AE index values. Circles show abnormally high AE index values well after AKR disappeared. The Journal of Experimental Secondary Science January 2012 Volume 1 Issue 3 Huy Lam, Teresa Petralli-Mallow, Leonard Garcia, and Shing Fung Page 5 of 7 55 to the source, not the spacecraft. The AE index values during this time interval increased sharply past 300 nT when there was no AKR detected by all spacecraft. This indicated that the explanation from Voots et al. [1977] was incomplete since the high increase in the AE index values with the corresponding lack of AKR emission observed by the spacecraft was not due to a cutoff propagation effect. Instead, it was due to the source of the AKR and the AE current. Furthermore, Voots et al. [1977]6 claimed that small AE index values corresponding with high AKR intensities or “On” AKR was attributed to low spatial coverage by magnetometer stations. In the dataset collected on 03/25/1996, 10 out of 12 magnetometer stations recorded AE values of the event, well enough to effectively cover the AKR emission (Table 1). In this dataset, an AKR gap existed between 10:15-11:20 UT (Fig.10). Although only Polar was in the northern hemisphere at this time, Geotail and Wind were close to the equator and far enough away to view the northern AKR. Despite the differences in observation points, the existence of AKR and its structures in Geotail’s and Wind’s spectrograms were very similar to Polar’s, which confirmed that all spacecraft were looking at the northern AKR. In agreement with the theory from Voots et al. [1977]6, the AE values dropped below the threshold within this time frame. However, the AE values continued to be well below the threshold while all spacecraft observed high levels of AKR from 11:20 UT onward (Fig. 10). This indicated that the explanation from Voots et al. [1977]6 was incomplete since the low observed values in the AE index corresponding with the high AKR intensities was not due to inadequate magnetometer station coverage. Instead, it was due to the source of AKR and the AE current. Fig. 9. This is a Cartesian coordinate system representation of spacecraft trajectories on 06/04/2002 from 0000-0600 UT with Earth as the origin (Re is Earth radii and Sun is in the positive X direction). Because the Wind spacecraft is much farther away from Earth than Geotail and IMAGE, Wind has its own orbital plot to prevent scale distortion. All spacecraft were in good positions to observe northern AKR and were most likely viewing the same AKR emission. This was confirmed in Fig. 8 where all spacecraft observed similar AKR structures. Fig. 10. Spacecraft observations from Geotail, Wind and Polar with AE index plot on 03/25/1996 showing negatively correlated AKR intensity and AE index values. Circle shows abnormally low AE index values well after AKR reappearance. Fig. 11. This is a Cartesian coordinate representation of spacecraft trajectories on 03/25/1996 from 0900-1300UT with Earth as the origin (Re is Earth radii and Sun is in the positive X direction). Because the Wind spacecraft is much farther away from Earth than Geotail and Polar, Wind has its own orbital plot to prevent scale distortion. Although Geotail and Wind were in the southern hemisphere, they were low enough in latitude and far enough away from Earth to view the northern AKR source. This was confirmed in Fig. 10 where Wind and Geotail observed similar AKR structures compared to Polar which was in the northern hemisphere. Dates in which AE and AKR are positively correlated (mm/dd/yyyy) # AE Magnetometer Station detection out of 12 (Fig. 4)9/10/1996 10 10/12/1996 10 3/6/2002 10 4/7/2001 11 6/6/1997 11 (Fig. 6)6/5/1996 10 1/3/1997 11 12/23/2006 9 Table 1. Other studied cases with magnetometer station detection information. This ensured that spatial coverage for AE value recording was sufficient to conclude that changes in AE values were due to