台灣東北部宜蘭平原及龜山島之地熱研究

The geothermal study in Ilan Plain and Kueishantao in northeastern Taiwan

江協堂(Hsieh-Tang Chiang)

指導教授 ： 徐春田

國立臺灣大學/理學院/海洋研究所/博士(2010年)

https://doi.org/10.6342/NTU.2010.02087

摘要

本研究在台灣東北部宜蘭平原和海域地區收集許多磁力、地熱、水井溫度以及井下地溫監測資料，其中磁力和地熱資料經各種修正、處理後，再整編過去學者發表的相關資料，得到此區域較完整的磁力和地熱分佈圖。磁力異常圖顯示磁力高區落在東海陸棚、龜山島南側、宜蘭平原南側和沖繩海槽，其中宜蘭平原南側的高磁力異常可往外海延伸至宜蘭陸棚，該異常帶可能代表基盤存在一火成岩體。地熱分佈圖顯示研究區域內最高熱流值約734 mW/m2，位在龜山島東南側約8公里處，另外，龜山島附近海床底層水水溫變化很大，可能是受海床熱液噴發頻繁的影響，高熱流和底層水溫變化顯示龜山島的後火山活動還很強烈。

利用宜蘭平原磁力異常逆推的火成岩體深度和大小以及震測資料解釋的剖面構造，建立一個位在宜蘭平原南北向的二維地熱模型，模型中包含沈積層、基盤、斷層帶和入侵的火成岩脈等4個區域，模擬的結果，發現當火成岩體入侵的時間為2.3-2.5萬年或19.0-22.0萬時，其形成的地下溫度分佈可符合利澤國中水井目前的觀測結果，火成岩體入侵的年代與龜山島火成岩的定年同屬一數量級，兩者又都位於沖繩海槽的最西端邊緣，因此兩火成岩體的發生可能有關連。另外，利用區域性磁力異常推測居里溫度深度，以及海底地形資料，建立東北部海域三維地熱模型，當熱傳到達穩態時，地表熱流值約68-236 mW/m2，最高的兩個熱流區集中在沖繩海槽中心附近，其高熱流值可能跟露出海床的火山有關，但熱流高低是否可直接指示火山的活動性，需要再增加一些地熱量測做比對。比對實際53個測量點熱流測值之平均值為77.2 mW/m2，三維地熱模型之理論熱流平均值為91.0mW/m2，比實際測量值高約17.9%，其原因可能跟實際測量資料未考慮沈積速率修正量有關，如果整個研究區採用同一低沈積速率（0.09 cm/yr），則測量點的理論熱流平均值將增加到84.2mW/m2，若採用高沈積速率（0.52 cm/yr），則測量點的理論熱流平均值將增加到135.9 mW/m2，考慮沈積速率修正，測量值可能會較接近模型的理論值。

龜山島地溫井自2006年7月監測至今的結果，發現同一深度的溫度變化很小，年溫度變化約0.015℃至0.215℃，各深度溫度資料（除深度10公尺外）經頻率分析並無明顯的週期性，地溫的變化與氣候、季節無關，4年的監測資料顯示整口井的溫度有慢慢下降的趨勢，年平均下降率約0.048-0.116 ℃。另外，2007和2008年夏季發現5個熱脈衝事件，溫度在數十小時內變化約-0.022~0.15℃，這些熱脈衝事件與中央氣象局發佈龜山島附近的地震無關，但與颱風侵台時間非常吻合，然而颱風使地底下產生熱脈衝的機制，目前尚不十分清楚。

宜蘭平原上30口水井地溫測量結果，發現大部分水井受到地下水對流的影響地溫梯度很低或呈現負值，然而平原南邊有6口井的地溫梯度超過一般值（3℃/100公尺），初步估計地溫梯度大於6.0 ℃/100公尺以上的面積約36平方公里，假設各水井上部400公尺的地溫梯度皆如同淺部的測量值，則平原內第四紀地層35℃、40℃、45℃等溫線的範圍分別約95、53、21平方公里，厚度約60-100公尺，若地層的孔隙率介於20%-55%，則40℃以上的地熱水儲量約為3400-9260×106立方公尺，以礁溪溫泉冬季目前使用量每天約16000立方公尺計，這個儲量可提供相同使用量的使用時間大約580-1580年。

關鍵字

宜蘭平原 ； 地熱 ； 龜山島 ； 地溫監測

並列摘要

We have collected magnetic, heat flow, temperature data in water wells and monitored the temperature in a well offshore of northeastern Taiwan. After processing and integrating of the previous studies, we completed magnetic anomaly and geothermal data set. The magnetic data show that the high anomaly areas are located in the East Sea, the south of Kueishantao, the south of Ilan Plain and the Okinawa Trough. The elongated magnetic anomaly in southern Ilan Plain implied possibly the existence of volcanic having extended to the Ilan shelf. Geothermal data show the maximum heat flow was located in the south of Kueishantao. Also the significant variation of bottom water temperature was caused by the frequent hot fluid erupted from the sea floor near Kueishantao. Both the high heat flow and the variation of bottom water temperature indicate the strong post volcanic activity in the vicinity of Kueishantao.

A 2-D geothermal model across Ilan Plain in NS direction was established from the interpretation of magnetic inversion and seismic profiles. A geothermal model including sedimentary formation, basement, fault zone and intrusion body of which the general quantities for the thermal parameters were derived. The simulated temperature distribution could fit the observation of Litsu well while the age of the intrusion body is estimated to be of 23-25 thousand years or 190-220 thousand years. The age of the intrusion body is in the same order of that dated from Kueishantao rocks. Therefore, the occurrence of the two igneous rocks could be correlated. On the other hands, a 3-D geothermal model in northeastern Taiwan was established from the Curie point depth inversion and the topography. The simulation result shows the equilibrium heat flow is 68- 236 mW/m2. Two maximum heat flows located in the middle of Okinawa Trough may be correlated to the surrounding extrusion volcanoes. However, further geothermal investigation is necessary to confirm their relationship. The heat flow value derived from 3-D model is 17.9% higher than that averaged from 53 field measurements. The heat flow difference of the two data sets could be resulted from the sedimentation effects. Adapting sedimentation correction for the measured data could decrease the difference effectively.

A temperature observation well in Kueishantao has been monitored since 2006. The results show the annual temperature variations at different depths of the well are from 0.015℃ to 0.215℃. No significant period related to the seasonal variation was found from the spectrum analysis. However, a temperature declined rate of 0.048-0.116 ℃/yr has been found from the 4-year continuously monitored data. Five heat pulses with -0.022~0.150℃ in amplitude surging in tens of hours coincided with the time of typhoon visited Taiwan are of great interesting. However, it is not clear for the mechanism of borehole temperature variation triggerring by a typhoon.

The thermal gradients in the 30 water wells are low or even negative. However, there are six wells the thermal gradients are greater than 3℃/100 m (about an average value of the world). We estimated the area with thermal gradient greater than 6℃/100 m is about 36 km2. Assuming a layer of 400 m in thickness with the same thermal gradient as that measured in shallow water wells, the area of equal temperature of 35℃, 40℃,45℃are 95 km2, 53 km2, 21 km2, respectively. If the porosity of rock is roughly estimated 20-55%, the reserve volume of 40℃hot water will be 3400-9260×106 m3. According to the data issued by the Chiaohsi Hot Spring Center in northern Ialn Plain, the reserved quantity of the hot water could provide 16000 m3 consumption daily for several hunred years.

並列關鍵字

Ilan plain ； geothermal ； Kueishantao ； temperature monitoring

參考文獻

陳愛筑，2005，三維重力分析與台灣區域地殼構造之大尺度變化，國立台灣大學海洋研究所碩士論文，共58頁。

陳志豪，2006，TCDP鑽井岩心之熱學參數直接量測與相關物理特性研究，國立中央大學應用地質研究所碩士論文，共58頁。

Angelier, J. 2005, Does extrusion occurs at both tips of the Taiwan collision belt? Insights from active deformation studies in the Ilan Plain and Pingtung Plain regions. Geodynamics and Environment in East Asia International Conference & 5th Taiwan-France Earth Science Symposium, 35-37.

Ansari, A.H. and Alamdar, K., 2009, Reduction to the pole of magnetic anomalies using analytical signal. World Appl. Sci. J., 7(4), 405-409.

Arculus, R.J., 1994, Aspects of magma genesis in arcs. Lithos 33, 189–208.