Utah State University
DigitalCommons@USU
Physics Capstone Project
Physics Student Research
9-2016
Mysterious mesospheric bores over the South Pole
Christina Solorio
Utah State University
Follow this and additional works at: http://digitalcommons.usu.edu/phys_capstoneproject
Part of the Physics Commons
Recommended Citation
Solorio, Christina, "Mysterious mesospheric bores over the South Pole" (2016). Physics Capstone Project. Paper 41.
http://digitalcommons.usu.edu/phys_capstoneproject/41
This Report is brought to you for free and open access by the Physics
Student Research at DigitalCommons@USU. It has been accepted for
inclusion in Physics Capstone Project by an authorized administrator of
DigitalCommons@USU. For more information, please contact
[email protected].
1 Mysterious mesospheric bores over the South Pole Christina Solorio Mike Taylor 4900 Research Project Abstract. During the winter‐time, the South Pole lies at the center of the giant Antarctic polar vortex which isolates it from the rest of the world. Over the past four years, USU has successfully operated an infrared camera at the US Amundsen‐Scott South Pole Station to investigate dynamics of atmospheric gravity waves within the vortex. Gravity waves are generated when a force disturbs an air packet and buoyant and gravitational forces in turn cause it to oscillate. While gravity waves in general are ubiquitous in the atmosphere, propagating well into the mesosphere and lower thermosphere region (~80‐100 km), one rare type of gravity wave known as a “bore” is characterized by a sharp leading front followed by several distinct trailing waves that grow in number. Bore wave events are observed in tropospheric clouds, but to date observations are rare in the upper atmosphere, especially at high‐latitudes. In this study ranging over 4 winter seasons (April‐August 2012‐2015), we have discovered a surprisingly large number of bores (83 events). Examples of these events are presented together with measurements of their propagation characteristics. These results provide important new information on the generation and propagation of gravity waves within the winter polar vortex. 1. Introduction Earth’s atmosphere is made up of several distinct regions determined by temperature patterns. Figure 1 shows a diagnostic of these altitude layers with respect to temperature, including what typical phenomena occur in each region (Taylor 1986). The mesosphere and lower thermosphere (MLT) region is rich with gravity wave activity. Gravity waves are oscillations generated by the force of gravity when an air packet is displaced. Over the last few decades, gravity waves have been regularly observed in the hydroxyl (OH) airglow layer of the mesosphere (typically ~87 km altitude) using imaging techniques, providing for in‐depth study of gravity wave parameters such as wavelength, velocity, and period (Taylor 1997). As observed by Fritts and Vincent (1987), small‐scale gravity waves (those with periods less than 1 hour) cause significant momentum flux in the MLT region, affecting changes in mean winds and thermal structure in the middle atmosphere (Taylor et al. 1997) Bores were originally observed in rivers as pictured in Figure 2 (Tricker 1965) during seasons of incoming tide and first analyzed by Lord Rayleigh (1908). Large amounts of additional water from the tide overwhelm the outlet of a river, creating an unstable mound of water that eventually breaks into a bore 2 Figure 3. River bore on the Mersey River in England. One of the first documented aerial views of a bore. (Tricker 1965) Figure 1. Vertical temperature profile of the atmosphere. Gravity waves are observed in the airglow layers, typically between 80‐100 km altitude. (Taylor 1986) propagating upstream. Figure 4 shows a basic schematic of this process as explained by Lord Rayleigh. These bore events would later be observed in the troposphere beginning in the late 1970s (e.g. Clarke et al. 1981, Hartung and Sitkowski 2010). A ground‐based image of a tropospheric bore is provided in Figure 3. It would be nearly a century after Lord Rayleigh’s first analysis of bores that these events were observed in the mesosphere. Since the first known observation of a “spectacular” mesospheric bore by Taylor et al. (1995) during the Airborne Lidar and Observations of the Hawaiian Airglow 1993 (ALOHA‐93) campaign, numerous other bores have been reported and studied in the Figure 2. Tropospheric bore as pictured from the ground over the Cape Canaveral, Florida region. (Hartung and Sitkowski 2010) Figure 4. Diagram of a river bore draw by Lord Rayleigh
(1908). The river flows to the right with initial speed u, and
depth l. With the increase of tide, the river later has speed u’
and depth l’. The excess water enters from the right from the
outlet of the river and “folds” over on itself, creating a bore.
mesosphere. Overall these events are typically considered rare. In a mathematical analysis of the ALOHA‐93 event, Dewan and Picard (1998) identified the distinct characteristics of a bore as having (1) a bright or dark linear front, (2) a number of small‐scale waves trailing behind this front, and (3) the number of trailing waves 3 growing over time. They further suggested a necessary condition for the formation of a mesospheric bore is the presence of thermal or Doppler duct. Our study turns to the South Pole’s unique atmospheric behavior. Because of Antarctica’s extremely cold temperatures, especially in the winter season, the temperature gradient between the frigid air above the continent and relatively warmer air over its surrounding waters creates what is known as the polar vortex (Posegate 2010). This vortex isolates the South Pole’s atmosphere from the rest of the world. To the best of our knowledge, only one known observation of an Antarctic bore has been reported to date (Nielsen 2006). A
B
2. Methods This study used an advanced mesospheric temperature mapper (AMTM) located at the Amundsen‐Scott South Pole Station with a field of view (FOV) of 144km by 180 km. Images were recorded about every 30 seconds. Data was taken over 4 winter seasons (2012‐2015) for 5 months each (April‐August). Because the South Pole experiences six months of constant darkness in the winter months, we were able to obtain near‐continuous data for a total of about 20 months. Raw images were processed by calibrating them to be oriented such that 0° longitude points up and unwarping so as to account for the wide angle lens. An example of raw versus calibrated and unwarped intensity images is given in Figure 5. Typically images are oriented such that north points up. While this nomenclature does not hold at the South Pole where cardinal directions Figure 5. Data processing method. Image A is an original data image. Image B is the same image after being calibrated and unwarped. Event occurred May 4, 2013. do not apply, it will be used in this report for simplicity in interpreting images. Our primary method for identifying bores was using keograms. In a process described by Taylor et al. (2009), after images have been calibrated and unwarped, a sliver two pixels wide was taken from a central area of each image, first vertically then horizontally. These small sections are pieced together to form two large pictures, called keograms, of how a wave behaves over time, one oriented north‐south and the other 4 east‐west. Because keograms offer a view of wave interactions over time, they can provide additional insight into wave interactions. These keograms were used to identify events with characteristics indicative of a tidal bore, as earlier described. Figures 6 and 7 show a keogram in both intensity and temperature for May 4, 2013. A strong bore event begins around 19:00 UT, showing a bright front, distinct bands, and increasing number of waves as the event progresses. (This is the same event pictured in Figure 5.) 3. Results Over our 4‐year study, 83 frontal events were identified, and 68 of these had measurable wave growth with time. The remainder of events still showed the other bore characteristics of a distinct leading front with several trailing waves. These events were most likely already formed to their full extent before entering the FOV. Figure 8 provides a typical example of a bore shown in both intensity and temperature which took place on May 19, 2012. The image progression from left to right shows a clear increase in number of bands over time. Corresponding intensity and temperature scans for this event are given in Figure 9, plotting the intensity and temperature with respect to position along the image, perpendicular to the wave. Another example of intensity and temperature scans are provided in Figure 9 from an event that occurred on May 4, 2013. Both events show very strong wave patterns in which the leading wave has a higher amplitude than its trailing waves. Figure 6. Close up view of the bore seen in Figure 7 in both intensity and temperature. The wave enters from the north and exits south about an hour and a half later, with very little east‐west movement. Figure 7. Keograms of May 4, 2013. Two pixels were taken from each of ~2500 images taken on this date to provide NS and EW images over time. The upper two keograms show band intensity while the lower two show temperature. A temperature scale is included for reference. The red boxes indicate a wave event that is magnified in Figure 6. 5 Figure 8. Progressive images of a NW traveling bore showing clear wave generation. There is one bright band present in the first images, three clear bands in the second images, and four clear bands with the beginning of a fifth in the third image. Because temperature images tend to be somewhat grainy providing for noisy temperature scans, each temperature scan was averaged to create a smoother, more intelligible curve. With our current processing software, this process was trivial for waves propagating north, east, south, or west, but for bores traveling anywhere in‐between it was more involved. Figure 10 illustrates the two averaging methods using an especially interesting night of data when two bores were observed simultaneously. A sequence of three images of these bores is provided to show the wave progression. The first entered the FOV from the south and traveled nearly due‐north, while the second entered from the northwest and propagated southeast. Figure 9. Intensity and temperature scans for bores that occurred May 19, 2012 and May 4, 2013. Both show strong wave patterns in both signatures. For the northward wave, a vertical scan was taken across the whole image where the bands were most clear. To average the temperature 6 Figure 10. Progression of a unique “double bore” event. One wave travels north and the other southeast. Temperature scan averaging for the northward wave was done using a vertical and horizontal average method using software from the crosshair shown in the first image. Horizontal information in this case was ignored. Averaging for the southeastward wave was done by drawing five lines, three of which are shown in the third image, and averaging their values. While this was done with temperature images, intensity images are shown here for ease of understanding. values of each position along the image, the scan “line” was extended to a width of 10 pixels, and the value of a position along the line was averaged with the other 10 pixels adjacent to it. Thus, each temperature value that made up that scan was an average and some of the noise from the image was eliminated from the scan. The first image in Figure 10 highlights the narrow area used for this scan. Figure 11. Intensity and temperature scans for the double event in Figure 10. A similar process was used for the SE‐traveling wave, but was done manually. To create a non‐
vertical or –horizontal scan line it had to be drawn in by selecting a point before and after the wave bands. Five of these scan lines were drawn, each one pixel up and over from the line before it, thus maintaining the same scan distance by using the same length of line, and assuring that each point along the scan was averaged with those at the same position along the line. Three of these lines are shown in the third image in Figure 10. Each value on the line was then averaged with the others and the final average temperatures were used in the scan. 7 Table 1. Parameters for nine exceptional bores. When compared with average values in Table 2, the average of these nine events is higher. Table 2. Ranges and averages of parameters for all 83 events. Parameters from left to right are wavelength (λ), direction of propagation (DOP) with 0° corresponding to N, velocity (v), period (τ), duration, number of total waves in an event (W), and rate of wave generation (Ẇ). Another averaging attempt was made for non‐
vertical or –horizontal events by rotating the image until the wave appeared vertical or horizontal and then using the first method to automate the averaging. While the software was able to accommodate the rotated images, the temperatures deviated significantly from expected values. Following the initial mesospheric bore observation by Taylor et al. during the ALOHA‐
93 campaign (2005), Dewan and Picard provided an explanation for mesospheric undular bores based on atmospheric ducting. Their analysis has served as a foundation for expected results regarding bores since its publication. Multiple bore observations have been documented over the last two decades, providing observational comparison for Dewan and Picard’s predictions. 4. Discussion Table 1 gives a summary of the wavelength, velocity, and period parameters of 9 particularly strong bore events while a summary of parameter ranges and averages for all 83 events is given in Table 2. With a couple exceptions which will be discussed, our findings overall correspond very well to those of other bore studies. While Dewan and Picard predict a wave generation rate of 3 waves per hour as an upper limit, our South Pole events had a somewhat higher average generation rate of 5.3, with 4‐8 waves/hr being most typical. A histogram of occurrence of growth rates is given in Figure 12. Our range of generation rate was much broader than expected, with a maximum of 15.0 8 Figure 12. Histogram of frequency of observed wave generation rates, with a clear preference of 4‐8 waves per hour. Figure 13. Histogram of frequency of how many total waves occurred in a single event. Figure 14. Histogram of observed directions of propagation showing a strong SE tendency. waves/hr. Furthermore, ~13% of calculated generation rates were 10 waves/hr or higher. Bores observed in northern China and at Bear Lake Observatory, Utah reported by Li et al. (2013) and Seo (1998) respectively, had generation rates less than 3 waves/hr, conducive to Dewan and Picard’s prediction. However, a study by Nielsen et al. (2006) looking at one individual event over Halley Station in Antarctica also observed a higher generation rate than predicted at 6.6 waves/hr. Our bore observations showed noticeably more waves per event than previously reported. The Halley bore seen by Nielsen et al. (2006) reached a total of about 12 waves. An in‐depth study of mesospheric waves was done by Fechine et al. (2005) including 64 bores observed at equatorial latitudes over São João do Cariri, Brazil. In this study they found a range of 2‐12 waves in their series of events, with 2‐6 waves per event being most common. While we found an average of 9 waves per event, 15 events (~18%) had greater than 12 waves, with a couple observations as much as twice the highest reports in the past. A histogram of the frequencies of number of waves per event is given in Figure 13. These higher wave observations may be due to using keograms as a method of identifying bores because keograms often show additional bands that are not obvious in data. Also, events with a greater number of waves are most noticeable in the keograms, which is a likely reason all our reported events had at least 3 trailing waves behind the lead wave. Another interesting finding was that these bores exhibited a strong preference for southwest 9 propagation as shown in Figure 14. This is the same predominant direction other non‐bore events from the South Pole were found to travel. While river bores travel upstream, it seems that mesospheric bores at the South Pole tend to travel with typical gravity wave flow. Plots of wave growth over time were created for several events such as the two shown in Figure 15. A preliminary observation of these plots suggests that the relationship between number of waves and time is not exactly linear, but for most events appeared to have very slight negative concavity as seen in the August 3, 2012 plot. One exception however was seen in the July 3, 2015 graph where the points appear to have an almost exponential trend. The range of wavelengths we observed at the South Pole (10‐37 km) corresponded almost exactly to bore wavelengths reported by Fechine et al (10‐40 km). In addition, our average wavelength was very similar to that of the ALOHA‐93 bore. Likewise, our velocity and period data were also very comparable to other reports. The only possible discrepancy might be that a small number (~5) of our observations had periods slightly higher than so far reported for other bores. They are still well‐within usual small‐scale gravity waves period however. In their equatorial study, Fechine et al. observed a strong majority of bores in the hours before local midnight. In our South Pole study where observations were made 24 hours a day, we observed no apparent trend in time of occurrence throughout the day. A histogram of times of observation is provided in Figure 16. Figure 15. Plots of number of waves over time for two particular events. Most of these plots had trends like those on Aug 3, 2012 (above), but July 3 (below) had a unique trend. Figure 16.
Histogram of time of occurrence for all bore events. Unlike Fechine et al. (2005), we saw no distinct trend. Additional plots of wave parameters and event frequency are provided in the appendix. 10 5. Conclusion References After an in‐depth study of over 80 events in 4 austral winters, we determine that mesospheric bores are not as rare as previously believed. In summary, some particularly interesting finding from this study are Dewan, E. M., and R. H. Picard (1998), Mesospheric bores, J. Geophys. Res., 103(D6), 6295–6305, doi:10.1029/97JD02498. Clarke, R. H., Smith, R. K., & Reid, D. G. (1981). The Morning Glory of the Gulf of Carpentaria: An Atmospheric Undular Bore. Monthly Weather Review,109, 1726‐1750. Fechine, J., Medeiros, A. F., Buriti, R. A., Takahashi, H., & Gobbi, D. (2005). Mesospheric bore events in the equatorial middle atmosphere. Journal of Atmospheric and Solar‐Terrestrial Physics, 67, 1774‐1778. doi:10.1016/j.jastp.2005.04.006 Fritts, D. C., & Vincent, R. A. (1987). Mesospheric Momentum Flux Studies at Adelaide, Australia: Observations and a Gravity Wave–Tidal Interaction Model. J. Atmos. Sci. Journal of the Atmospheric Sciences, 44(3), 605‐619. doi:10.1175/1520‐0469(1987)0442.0.co;2 Hartung, D. C., & Sitkowski, M. (2010). An atmospheric undular bore along the eastern Florida coast. Weather, 65(6), 148‐150. doi:10.1002/wea.538 Li, Q., Xu, J., Yue, J., Liu, X., Yuan, W., Ning, B. . . . Younger, J. P. (2013). Investigation of a mesospheric bore event over northern China. Ann. Geophys. Annales Geophysicae, 31(3), 409‐418. doi:10.5194/angeo‐31‐409‐2013 Nielsen, K., Taylor, M. J., Stockwell, R. G., & Jarvis, M. J. (2006). An unusual mesospheric bore event observed at high latitudes over Antarctica. Geophys. Res. Lett. Geophysical Research Letters, 33(7). doi:10.1029/2005gl025649 Posegate, A. (2010). Observing Weather at the Bottom of the World: The South Pole. Weatherwise, 63(5), 28‐34. doi:10.1080/00431672.2010.503840 Rayleigh, L. (1908). Note on Tidal Bores. Proc. R. Soc. London, 81, 448‐449. doi:10.1098/rspa.1908.0102 Seo, Seon‐Hee. All‐sky Measurements of the Mesopheric "Frontal Events" from Bear Lake Observatory, Utah. Diss. Utah State U, 1998. Print. 1. Rates of wave growth can be much higher than originally predicted. 2. Bores were identified with a greater number of bands than have previously been observed. 3. South Pole bores typically travel in the same direction as usual gravity waves. 4. Overall, keograms are a highly successful means of identifying bore events and often provide more insight on bore activity than images alone. Wave parameters in this study were for the most part comparable to bore studies conducted previously, and we look forward to pooling our information with others’ to form greater knowledge of these mysterious bore events. This study further provides insight on atmospheric dynamics of the winter polar vortex. Acknowledgements We would like to thank Yucheng Zhao and P.‐
Dominique Pautet for their contributions to this research effort. This study was supported by the National Science Foundation in conjunction with the Antarctic Gravity Wave Instrument Network under NSF grants #1045356 and #1143587. 11 Taylor, M. J., Observation and analysis of wave‐like structures in the lower thermospheric nightglow emissions, Ph. D. thesis, Univ. of Southampton, Southampton, UK, 1986. Taylor, M. J., Turnbull, D. N., & Lowe, R. P. (1995). Spectrometric and imaging measurements of a spectacular gravity wave event observed during the ALOHA‐93 Campaign. Geophys. Res. Lett., 22(20), 2849‐2852. doi:10.1029/95gl02948 Taylor, M. J., Pendleton, W. R., Clark, S., Takahashi, H., Gobbi, D., & Goldberg, R. A. (1997). Image measurements of short‐period gravity waves at equatorial latitudes. J. Geophys. Res., 102(D22), 26283‐26299. doi:10.1029/96jd03515 Taylor, M. J., Pautet, P., Medeiros, A. F., Buriti, R., Fechine, J., Fritts, D. C., . . . Sabbas, F. T. (2009). Characteristics of mesospheric gravity waves near the magnetic equator, Brazil, during the SpreadFEx campaign. Ann. Geophys.,27(2), 461‐
472. doi:10.5194/angeo‐27‐461‐2009 Tricker, R. A. R., Bores, Breakers, Waves, and Wakes, Elsevier, New York, 1965. 12 Appendix Histograms of other wave parameters including period, velocity, wavelength, and event frequencies sorted by both month and year are provided for additional insight. The velocity plot shows a slight bimodal trend with modes from 30‐40 and 50‐60. From event frequency plots we can see a clear pattern of greater bore observations in late winter, particularly July and August. We also observe a slight drop in total bore events in the most recent two years. 13