Permutation Entropy Variation of Seismic Noise prior to Eruptive Activity at Shinmoedake Volcano, Japan

NCU Institutional Repository > 地球科學學院 > 地球物理研究所 > 博碩士論文 > Item 987654321/80618

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/80618

题名:	Permutation Entropy Variation of Seismic Noise prior to Eruptive Activity at Shinmoedake Volcano, Japan
作者:	納莉娜;Nurfitriana, Izaina
贡献者:	地球科學學系
关键词:	排列熵;新燃岳火山;震顫;極化分析;Permutation entropy;Polarization analysis;Shinmoedake;Tremor;Complexity;Ambient noise
日期:	2019-07-17
上传时间:	2019-09-03 14:48:29 (UTC+8)
出版者:	國立中央大學
摘要:	在過去的二十年中,用於數據分析的排列熵(PE)的使用一直在迅速增長。衡量任意時間序列複雜性的能力在許多領域都有廣泛的應用。我們試圖在火山學中使用這種方法作為火山爆發預測的方法。我們在日本九州島研究了 2011 年 1 月的新燃岳火山爆發序列。 2011 年 1 月,新燃岳火山爆發了最大規模的火山爆發,其中包括持續兩週的子普林尼式噴氣,連續爆炸和熔岩侵位階段。在這項研究中,我們將 PE 計算應用於新燃岳火山周圍記錄的地震噪聲。我們還計算了譜圖和極化分析作為支持材料來分析 PE 的變化。結果顯示,PE 在 2011 年 1 月 1 日至 17 日呈現振盪模式,然後在 2011 年 1 月 18 日至 31 日期間下降並降低。2011 年 1 月 1 日至 17 日期間 PE 的振盪模式主要與天氣變化有關,且並未引起通過火山活動。 2011 年 1 月 18 日至 31 日期間 PE 值的下降與震顫發生同時發生。我們發現震顫中的確定性行為是導致一般 PE 值降低的主要因素。極化分析結果還表明, 在低 PE 值時期,地震源的方位角方向指向新燃岳火山口。我們還繪製了 Ichihara 和 Matsumoto(2017)的 PE 值和震顫深度位置,以檢查它們之間是否存在相關性。結果顯示 PE 與震顫深度位置成反比關係。當震顫源向上遷移並到達水層時,PE 減少。這種降低的 PE 可能是由於水層內沸騰產生的氣泡形成,蒸汽和高溫引起的衰減過程的結果。相反,當震顫源遷移到比水層更深時,PE 增加。這種增加的 PE 可能是由於沒有氣泡,蒸汽和更高的溫度。該系統不會衰減高頻並可能產生更隨機的信號。 PE 有可能在火山爆發前早期發現震顫。隨著震顫在主要噴發之前開始,PE 減少。在 2011 年 1 月的新燃岳噴發序列可觀察到 PE 值的增加和突然下降模式。 ;In the past two decades, the use of permutation entropy (PE) for data analysis has been growing rapidly. The ability to measure the complexity of an arbitrary time series has had a wide application in many fields. We attempt to use this method in volcanology as an approach to volcanic eruption forecasting. We studied the January 2011 Shinmoedake eruptions sequence, in Kyushu island, Japan. The Shinmoedake volcano had its largest eruption in January 2011 with phreatomagmatic, sub-plinian, continuous explosion and lava emplacement stages, that lasted for two weeks. In this study, we applied the PE calculation to seismic noise recorded around the Shinmoedake volcano. We also performed spectral and polarization analysis as supporting materials to analyze the variation of PE. The results show that PE exhibited an oscillatory pattern during 1-17 January 2011 and then decreased and became lower during 18-31 January 2011. The oscillatory pattern of PE during 1-17 January 2011 was related mostly to the weather change and was not caused by volcanic activities. The decrease of PE values during 18-31 January 2011 coincided with tremor occurrence. We found that the deterministic behavior in tremor was a causative factor that led to decreasing PE values in general. The polarization analysis results also showed that in the periods of low-PE value, the azimuth direction of seismic sources pointed toward the Shinmoedake crater. We also plotted the PE values and tremor depth locations from Ichihara and Matsumoto (2017) to check if there is a correlation between them. The results showed that PE has an inverse relationship with tremor depth locations. When tremor sources migrated upwards and reached the water layer, PE decreased. This decreasing PE was probably a result of attenuation processes due to bubble formation, steam, and high temperature that arise from boiling inside the water layer. On the contrary, when tremor sources migrated deeper than the water layer, PE increased. This increasing PE was probably due to the absence of bubbles, steam, and higher temperature. The system would not attenuate high frequencies and hence produce a more stochastic signal. PE has the potential to detect tremor early before the eruption. PE decreases as the tremor starts before the main eruption. PE also captured the eruption events in January 2011 Shinmoedake eruptions sequence. They were marked with an increasing and sudden drop pattern of PE values.
显示于类别:	[地球物理研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	174	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....