
系統識別號 
U00260706201715535600 
論文名稱(中文) 
低緯度地區中尺度電離層移行擾動研究 
論文名稱(英文) 
Study of Medium Scale Traveling Ionospheric Disturbance in LowLatitude Ionosphere 
校院名稱 
成功大學 
系所名稱(中) 
地球科學系 
系所名稱(英) 
Department of Earth Sciences 
學年度 
105 
學期 
2 
出版年 
106 
研究生(中文) 
周育霆 
研究生(英文) 
YuTing Chou 
學號 
L46034128 
學位類別 
碩士 
語文別 
中文 
論文頁數 
112頁 
口試委員 
指導教授林建宏 口試委員陳佳宏 口試委員蕭棟元

中文關鍵字 
低緯度電離層
中尺度電離層移行擾動(MSTID)
GPS觀測
自動化判定

英文關鍵字 
low latitude ionosphere
medium scale traveling ionospheric disturbance (MSTID)
GPS observations
autonomous detection

學科別分類 

中文摘要 
中尺度電離層移行擾動(Medium Scale Traveling Ionospheric Disturbance, 簡稱 MSTID)為一種常見的電離層擾動現象，具有100  500公里的波長以及30300公尺/秒的速度。MSTID較常在中緯度地區(如日本)被觀測到，利用日本高密度GPS接收站觀測統計MSTID之特徵，發現在夏季夜晚以及冬季全日有著較高的發生機率，並且MSTID在太陽活動極小期之活躍度有著大於太陽活動極大期的特性。此外MSTID在白天與夜晚也有著不同的傳播特性，白天的MSTID波前方向為東西向、傳播方向為南北方向，而在夜晚則有著西北東南方向的特殊波前並由東北往西南方向的傳播。在過去，因為其擾動波的特性，夜晚的MSTID一般以電離層F層帕金斯電漿不穩定(Perkins Instability)機制解釋其生成，但該機制的不穩定發生率所需時間(數天)遠高於觀測結果(數十分鐘)。近年來，發現電離層散塊E層(Sporadic E, Es)不穩定機制(Es Layer Instability)產生的擾動波前與F層帕金斯電漿不穩定有著相同的排列方向，藉由Es層不穩定機制產生之電場映射至F層，可大幅縮短不穩定發生率所需時間，而與觀測吻合。此外近年來更多的觀測證據顯示MSTID可以傳播至低緯度地區，顯示MSTID相關研究仍有許多可發展空間，特別是低緯度MSTID之研究較無相關文獻，因此本研究將著重於MSTID於低緯度赤道異常地區(台灣地區)之統計分析。為統計三年期間(2013年至2015年)台灣GPS觀測資料，本研究利用互相關(Cross Correlation)及二維快速傅立葉轉換(2DFFT)建立電腦程式自動判斷MSTID傳播之系統，統計結果發現在低緯度地區的春夏季以及冬季全日有著較高的發生機率，其中日間的發生率為夏季較高(72%)，夜間的發生率則為冬季較高(79%)，進一步將MSTID分為南北傳播方向統計發現向北傳播的MSTID常發生在春、夏季，並以夏季日間之發生率較高(57%)，向南傳播的MSTID則通常發生在春、冬季，並以冬季夜間之發生率較高(74%)。統計結果也顯示在太陽活動較小的年份中(2013年及2015年) MSTID事件發生次數多於太陽活動較大的年份(2014年)。
關鍵字：低緯度電離層; 中尺度電離層移行擾動(MSTID); GPS觀測; 自動化判定

英文摘要 
Study of Medium Scale Traveling Ionospheric Disturbance in LowLatitude Ionosphere
Author’s Name: YuTing Chou
Advisor’s Name: Charles C. H. Lin
Department of Earth Sciences, National Cheng Kung University
SUMMARY
This thesis statistically studies the medium scale traveling ionosphere disturbance (MSTID) using the network of ~100 GPS receivers in Taiwan. Previous studies showed that MSTID with wavelengths of 100~500 km and phase velocities of 30300 m/s is commonly observed in the middle latitude ionosphere. The daytime MSTID with wavefronts aligned in eastwest direction propagates in northsouth directions, while the nighttime MSTID with the wavefronts aligned in the northwestsoutheast (NWSE) direction propagates along northeastsouthwest (NESW) direction. It is more frequently occurred in nighttime than in daytime and during solar minimum than maximum. To statistically study the MSTID features in the low latitude equatorial ionization anomaly region of ionosphere, an autonomous MSTID detection algorithm is developed for detection of its occurrences and direction of propagation by using both the twodimensional fast Fourier transform (2DFFT) and the cross correlation analyses. By applying the autonomous algorithm on the GPS observations around Taiwan (Geographic 20°N30°N, 115°E125°E) during 20132015, statistical analyses of MSTID is then performed. Results show that, the daytime MSTID appeared most frequently in local summer (72%, average of 20132015) and the nighttime MSTID appears more frequently in local winter with appearance rate of 79% (average of 20132015). The occurrence rate of northward propagation (57%, average of 20132015) is higher than the southward propagation (45%, average of 20132015) in daytime during summer, while the occurrence rate of southward propagation (74%, average of 20132015) is higher than the northward propagation (48%) in nighttime during winter. The statistical results in this thesis show that the nighttime MSTID in low latitude ionosphere is quite different from that in the middle latitudes, where MSTID is considered to be stronger in summer than in winter. The occurrence rate of MSTID in the low latitude ionosphere may as well be related to the coupled Es layer and Perkins instability that is highly related to the wind shear in the Eregion altitude. The wind shear variations derived from the empirical NRL horizontal wind model (HWM) are applied to interpret the seasonal dependences of MSTIDs propagation and occurrence. It suggests that the background wind might play an important role affecting the MSTID occurrence.
Key words: low latitude ionosphere; medium scale traveling ionospheric disturbance (MSTID); GPS observations; autonomous detection
INTRODUCTION
The traveling ionospheric disturbances (TIDs) are known to be triggered by the gravity waves launched by disturbances due to magnetospheric and lower atmospheric forces. Among TIDs, the disturbances with the horizontal wavelength of 100300 km are classified as the medium scale TIDs (MSTIDs) [c.f. Hunsucker, 1982; Otsuka et al., 2013]. MSTID is commonly observed, especially by GPS total electron conten (TEC) observations, in the middle latitudes and the past statistical studies show that it has high occurrence rate in nighttime in local summer [Tsugawa et al., 2007; Kotake et al., 2007]. Additionally, the recent study reported by HernándezPajares et al. (2012) show that the northward MSTIDs appear more frequently than the southward MSTIDs in summer, while the southward MSTIDs appear more frequently than northward MSTIDs in winter.
The generation mechanism of MSTID was previously explained various mechanisms during daytime and nighttime. The daytime MSTIDs with wavefronts aligned in eastwest direction propagate in northsouth directions and are explained by the gravity wave effects. On the other hand, the nighttime MSTIDs with the wavefronts aligned in the northwestsoutheast (NWSE) direction propagate along northeastsouthwest (NESW) direction and are explained by coupling of Perkins and Es (sporadic E) layer instabilities, as the Perkins instability has very low growth rate alone. The Es layer instability is mainly produced by the effect of wind shear in ionosphere E region. The wind shear produces the polarization electric fields in the Es and the electric fields are mapped along magnetic field lines to the F region to accelerate the Perkins instability.
Previous studies showed that MSTID is most commonly observed in the middle latitude ionosphere with the wavefronts in the northwestsoutheast (NWSE) direction, wavelengths of 100~500 km, phase velocities of 30300 m/s and propagation direction of northeastsouthwest (NESW). Because most of the statistical studies were in the midlatitudes, but the characteristics of MSTID might be related to the location for various local times and seasons. Especially for the lowlatitude ionospheric where the dynamics is much more complex, as there are several disturbance sources from lower atmosphere could affect the lowlatitude ionosphere.
This thesis studies the medium scale traveling ionosphere disturbance (MSTID) statistically at the lowlatitude equatorial ionization anomaly (EIA) region of ionosphere using the network of ~100 GPS receivers in Taiwan during 20132015. To statistically study the MSTID features, an autonomous MSTID detection algorithm is developed for detection of its occurrences and direction of propagation by using both the twodimensional fast Fourier transform (2DFFT) and the cross correlation analyses.
METHODOLOGY
We develop an autonomous MSTID detection algorithm to automatically detect the MSTID over Taiwan by using the twodimension fast Fourier transform (2DFFT) and cross correlation method. The algorithm is applied on the GPS data around Taiwan (Geographically 20°30°N, 115°125°E) to statistically analyze MSTID occurrence and propagation directions in the low latitude ionosphere EIA during 20132015. Due to the limited coverage of GPS observation in the zonal direction around Taiwan, the northsouth propagation direction is addressed and the eastwest direction is omitted.
We apply the highpass filter on the raw GPSTEC data to extract data with periods less than 1hour. The filtered TEC are then interpolated to the fixed grid as function of time (30 s resolution) and latitude (0.25° resoution). As the period of MSTID is normally in between 2060 min, the maps of TEC perturbations are organized hourly. It is noted that we select continuous data last more than 40 minutes for each latitudes of the map and the data gap less than 1hour are substituted by zero value. Then, the timelatitude TEC map is further analyzed by using the 2D FFT and the frequencywavenumber matrix is obtained. The positive (negative) wavenumber represents northward (southward) propagation of the disturbances. The northward (southward) propagation of MSTID map could be obtained by performing the inverse 2D FFT on the data of positive (negative) wavenumber. Finally, we perform cross correlation calculation on the each of the northward/southward MSTID maps. The cross correlation is performed for each of the latitudes in respect to its adjacent latitudes (higher and lower neighboring latitudes). The cross correlation is performed in every other latitudes, and totally 18 sets of cross correlation coefficients will be obtained for one MSTID map. If 7 out of 18 sets of cross correlation coefficients has a value larger than 0.55, the MSTID event is then identified.
The abovementioned autonomous detection algorithm is performed for all GPSTEC data during 20132015. There are two statistical definitions are applied in this study:
1. Statistic1: We count the MSTID events detected by each of GPS satellites and calculate the occurrence rate by dividing the total number of detected MSTID events by the total observable satellites within the 1hour period.
2. Statistic2: We count the MSTID events occurred in the specific spatial area (geographic 2030oN, 115125oE) during the 1hour period. The monthly average of occurrence rate is calculated in every 2hour period.
In general, Statitic2 is more suitable to identify the statistical results of the MSTID phenomena, and we discuss the results based on Statistic2.
RESULTS AND DISCUSSIONS
By applying the twodimension fast Fourier transform (2DFFT) and cross correlation method to the algorithm on the GPS observations in the equatorial ionization anomaly region around Taiwan (Geographic 20°N30°N, 115°E125°E) during 20132015, statistical analyses of MSTID is performed. The results show that, MSTID appears more frequently in local winter during nighttime with appearance rate of 79%, and it is most frequently appeared in daytime of summer (72%). The occurrence rate of northward propagation (57%) is higher than the southward propagation (45%) during daytime in summer and the occurrence rate of southward propagation (74%) is higher than the northward propagation (48%) during nighttime in winter. The statistical results of the nighttime MSTID in lowlatitude ionosphere is quite different from that in the midlatitudes, where MSTID is considered to be stronger in summer than in winter. Additionally, the number of MSTID event seems to be related to the solar activity showing the smallest occurrence rate of MSTID on 2014 where the Sun is more active than 2013 and 2015.
Our results also show that the daytime occurrence rate is higher than nighttime during summer, whereas the nighttime occurrence rate is higher than daytime during winter. According to the empirical horizontal wind model2014 (HWM14), the northward meridional background wind around ~200300 km altitude is weaker in summer and becomes stronger in winter. The weaker northward wind in summer may allow more upward propagation of gravity waves and the stronger wind in winter may filter out more gravity waves to prohibit its upward propagation. The background wind filtering of the gravity wave could explain the larger/smaller daytime occurrence of MSTID during summer/winter, as the daytime MSTID is more related to the gravity wave.
The nighttime MSTID movements may be related to the movement of Es layer instability triggered by the Eregion wind shear. According to HWM14, the results show that the wind shear around Eregion (100110 km altitudes) propagates southward/northward with time and its velocity coincides with MSTID speed. It suggests that the movement of MSTID may be related to the movement of the wind shear.
CONCLUSION
The main findings of the statistics for the MSTID occurrences, direction of propagations and the background wind effects are summarized as follows.
Statistical results of the MSTID occurrence near Taiwan region in 20132015:
1. The high occurrence rates (>70%) appear in both daytime and nighttime for spring and summer during 20132015.
2. The least occurrence rate (<60%) occurs during autumn (SeptemberNovember) in 2014 and 2015, expect for 2013 with a high occurrence rate (>80%).
3. The high occurrence rates (>80%) appear in nighttime for winter during 20132015 but the high occurrence rates also appear in daytime of 2013.
4. The daytime occurrence rate is higher than nighttime during summer, except July 2015. The nighttime occurrence rate is higher than daytime during winter solstice months, for the entire three years statistics.
5. The number of MSTID event seems to be related to the solar activity showing the smallest occurrence rate of MSTID on 2014 where the Sun is more active than 2013 and 2015.
Statistics of northward and southward propagations:
1. The northward propagation of MSTID occurs more frequently in spring and summer. Especially, the number of northward event decreases substantially in daytime during winter.
2. The southward propagation of MSTID occurs more frequently in spring and winter. Especially, the number of event increases obviously in nighttime during winter.
The background wind effects according to HWM14 simulations:
1. Gravity wave filtering could explain the daytime results for larger occurrence rate in summer than in winter.
2. The movements of the wind shear around Eregion (100110 km altitudes) that is important to the Es Layer Instability are in agreement with the movement of MSTID.

論文目次 
摘要 I
Extended Abstract III
誌謝 IX
目錄 X
圖目錄 XII
表目錄 XVIII
第1章 緒論 1
1.1 電離層簡介 1
1.1.1 電離層分層 1
1.1.2 電漿漂移 3
1.1.3 赤道噴泉效應 5
1.2 RayleighTaylor Instability 8
1.3 中尺度電離層移行擾動之生成 9
1.4 MSTID在中緯度地區之統計結果 17
1.5 研究動機與目的 20
第2章 觀測與分析方法 21
2.1 GPS觀測原理 21
2.2 濾波器 27
2.3 使用之分析工具 29
2.3.1 Cross Correlation 29
2.3.2 二維快速傅立葉轉換(2DFFT) 31
2.4 Horizontal Wind Model 33
第3章 資料處理及分析步驟 34
3.1 分析資料步驟及流程圖 34
3.2 GPS資料處理 36
3.3 分析方法之演進 39
3.3.1 互相關(Cross correlation)步驟說明 39
3.3.2 二維快速傅立葉轉換(2DFFT)及互相關(Cross Correlation)步驟 51
第4章 研究結果 66
第5章 結果與討論 88
第6章 結論 108
參考文獻 110

參考文獻 
[1] Arras, C., Jacobi, C., and Wickert, J.: Semidiurnal tidal signature in sporadic E occurrence rates derived from GPS radio occultation measurements at higher midlatitudes, Ann. Geophys., 27, 25552563, doi:10.5194/angeo2725552009, 2009.
[2] Arras, C., J. Wickert, G. Beyerle, S. Heise, T. Schmidt, and C. Jacobi (2008), A global climatology of ionospheric irregularities derived from GPS radio occultation, Geophys. Res. Lett., 35, L14809, doi:10.1029/2008GL034158.
[3] Drob, D. P., et al. (2008), An empirical model of the Earth's horizontal wind fields: HWM07, J. Geophys. Res., 113, A12304, doi:10.1029/2008JA013668.
[4] Drob, D. P., J. T. Emmert, J. W. Meriwether, J. J. Makela, E. Doornbos, M. Conde, G. Hernandez, J. Noto, K. A. Zawdie, S. E. McDonald, et al. (2015),
An update to the Horizontal Wind Model (HWM): The quiet time thermos sphere, Earth and Space Science, 2, 301–319, doi:10.1002/2014EA000089.
[5] Ding, F., W. Wan, G. Xu, T. Yu, G. Yang, and J. Wang (2011), Climatology of mediumscale traveling ionospheric disturbances observed by a GPS network
in central China, J. Geophys. Res., 116, A09327, doi:10.1029/2011JA016545.
[6] HernándezPajares, M., J. M. Juan, J. Sanz, and A. AragónÀngel (2012), Propaga tion of medium scale traveling ionospheric disturbances at different latitudes and solar cycle conditions, Radio Sci., 47, RS0K05, doi:10.1029/2011RS004951.
[7] Hedin, A. E., et al. (1991), Revised Global Model of Thermosphere Winds Using Satellite and GroundBased Observations, J. Geophys. Res., 96(A5), 7657–7688, doi:10.1029/91JA00251.
[8] Kotake, N., Otsuka, Y., Ogawa, T. et al. Earth Planet Sp (2007) 59: 95. doi:10.1186/BF03352681.
[9] Narayanan, V. L., K. Shiokawa, Y. Otsuka, and S. Saito (2014), Airglow observations of nighttime mediumscale traveling ionospheric disturbances from Yonaguni: Statistical characteristics and lowlatitude limit, J. Geophys. Res. Space Physics, 119, 9268–9282, doi:10.1002/2014JA020368.
[10] Tsunoda, R. T. (2006), On the coupling of layer instabilities in the nighttime midlatitude ionosphere, J. Geophys. Res., 111, A11304, doi:10.1029/2006JA011630.
[11] Tsunoda, R. T., R. B. Cosgrove, and T. Ogawa (2004), Azimuthdependent Es layer instability: A missing link found, J. Geophys. Res., 109, A12303, doi:10.1029/2004JA010597.
[12] Tsugawa, T., Kotake, N., Otsuka, Y. et al. GPS Solut (2007) 11: 139. doi:10.1007/s1029100600455.
[13] Yeh, W.H., J.Y. Liu, C.Y. Huang, and S.P. Chen (2014), Explanation of the sporadicE layer formation by comparing FORMOSAT3/COSMIC data with meteor and wind shear information, J. Geophys. Res. Atmos., 119, 4568–4579, doi:10.1002/2013JD020798.
[14] Yokoyama, T., D. L. Hysell, Y. Otsuka, and M. Yamamoto (2009), Threedimen sional simulation of the coupled Perkins and Eslayer instabilities in the nighttime midlatitude ionosphere, J. Geophys. Res., 114, A03308, doi:10.1029/2008JA013789.
[15] 蔡和芳,「全球定位系統觀測電離層赤道異常之研究」, 國立中央大學,博士論文, 1999。
[16] 林佳廷,「平流層瞬間暖化對低緯度電離層的影響」, 國立成功大學, 碩士論文, 2012。
[17] 陳威翰,「利用GPS觀測地震和海嘯對電離層的影響」, 國立成功大學, 碩士論文, 2013。
[18] 陳柏成,「西元2012年Mount Tongariro火山噴發對電離層擾動之研究」, 國立成功大學, 碩士論文, 2014。
[19] 練川平,「數值模式研究地殼與低層大氣電流對電離層電子密度影響」, 國立成功大學, 碩士論文, 2015。
[20] 林宸生, 邱創乾, 陳德請, 「數位信號處理實務入門」, 高立圖書有限公司, 1999。

論文全文使用權限 
同意授權校內瀏覽/列印電子全文服務，於20190630起公開。 


