台灣西南外海聚合板塊邊界動態氣水化合物及流體系統的三維震測及熱流分析

Abstract

台灣西南外海有特殊的地質架構、有高密度的反射震測資料以及廣泛分布的天然氣水合物，是一個研究聚合板塊邊界上的海底天然氣水合物系統以及流體移棲的絕佳地點。台灣位於南邊為隱沒構造、北邊為弧陸碰撞構造的板塊聚合邊界上，如此的板塊構造幾何關係使我們得以利用時空對等(time-space equivalence)的觀念，從南邊年輕的碰撞構造階段到北邊成熟的碰撞構造時期，來研究其地殼構造和流體移棲的變化。此外，臺灣陸地上快速的侵蝕作用造成海域沈積物中富含大量的碳氫化合物。過去研究顯示，台灣西南外海的天然氣水合物分布廣達15000平方公里以上，是具有可開發天然氣水合物儲集層的高潛能地區。\\ 我首先研究的是四方圈合海脊。這個位於增體岩體前緣的背斜海脊是天然氣水合物探測的主要好景區之一。本研究結合P-cable系統收集來的高解析三維反射震測資料、海底仿擬反射(Bottom-simulating reflection, BSR－為一反極性震波反射，可以用來指示天然氣水合物相位底界）在3維震測資料體中的分布、熱流探針資料以及地熱數值模型，來檢驗海床下一公里內地層中影響天然氣水合物生成的動態變化過程。天然氣水合物的穩定性會受不同的地質營力所造成溫壓擾動的影響。本研究中，我說明了集中式流體(focused fluid flow)沿著傾斜滲透層側翼、斜坡盆地、逆斷層和脊頂正斷層等構造向上遷移，會產生溫度擾動。此外，我亦對在海底沉積物滑移以及快速堆積這兩種現象其下方地層中觀察到的雙重海底仿擬反射(double BSR)特徵提出解釋，認為海底沉積物滑移或快速堆積所造成局部的壓力和溫度的急速變化會導致天然氣水合物穩定帶底界快速移動，這個天然氣水合物穩定帶底界向上或向下的急遽移動會分別導致沈積層中天然氣水合物的形成或解離，而在不同位置上形成新的BSR。因此，海床表面動態過程與淺部流體系統會影響double BSR之間地層天然氣水合物的分布及其飽和度，且若此現象廣泛出現於全球聚合板塊邊界，將可能會造成海裏大量甲烷的釋放。值得在氣候變遷研究中進一步探討。 \\ 在第二部分的研究中，我開發了一套程式來計算流體移棲的體積流率，並量化流體的收支、分別探討臺灣西南部海域增積岩體中沿斷層帶移動的集中式流體系統與擴散式流體系統的流體體積流率。我利用台灣西南外海廣泛分布的反射震測資料來快速計算不同地區流體移棲的體積流率。反射震測剖面上常顯示海底仿擬反射(BSR)在接近逆斷層的位置時會向海床偏移，反應出局部的熱流變化。我認為在這些位置，深部的較熱流體沿斷層所形成的通道向上流動，把熱傳送到較淺的位置，迫使海底仿擬反射(BSR)的位置變得較淺。針對此現象，我開發了一個二維穩定態的數值模型去量化造成此熱異常所需要的流體體積流率。這個方法結合了沿傾斜通道移動的集中式流體的二維模型之貝克勒數(Peclet number)解析解，以及計算地形變化對海床下溫度場影響的數值方法。在增積岩體海溝附近，流體沿著傾斜孔道移動的體積流率介於每年每公尺0.1 到16立方公尺。流體體積流率沿斷層通道一般較快，而沿斜坡盆地不整合面通道較慢。為了評估在造山演化下流體移棲的演化模式，我量化流體的收支，且分別討論在台灣隱沒帶和初始碰撞帶中，沿斷層移棲的集中式流體系統與以擴散方式移棲的流體系統之分配。本模擬結果顯示，在初始碰撞帶，流體有較強的集中排出現象，平均有總體積38.5\%的水在增積岩體壓實的過程中以集中式流體的方式排出。這個結果稍高於在隱沒帶中有25\%的水以集中式流體的方式被排出。\\ 本研究把時間維度加入三維震測資料，描述了天然氣水合物系統在壓力和溫度條件改變下的瞬時階段演化，此系統的空間範圍為數公里，解析度則為公尺等級。另外，藉由將時間維度加入區域性分布的二維震測資料，本研究為空間解析度為百公尺、規模大小為百公里的區域提供了一個量化的連結，建立隱沒－碰撞系統的演化、流體的收支系統以及體積流率間之關係。本研究的結果說明了集中式流體系統在聚合性板塊邊緣脫水過程中扮演的重要角色。

Offshore southwestern Taiwan is among the best places to investigate complex gas hydrate systems and fluid flow in convergent margins due to its geological setting, dense coverage of reflection seismic data, and ample presence of hydrates in the region. Taiwan is located along a convergent boundary with a subduction to the south and an arc continent-collision to the north. This geometry makes it possible to apply the concept of time-space equivalence to study crustal structure and fluid flow from the younger stage in the south to a mature stage in the north. In addition, due to rapid erosion on land, high amounts of hydrocarbons have formed in the marine sediments. As gas hydrates have been inferred to cover an area of more than 15000 km\textsuperscript{2} off southwestern Taiwan, the region has high potentials for hosting exploitable gas hydrate reservoirs. \\ I first studied Four-Way-Closure Ridge, an anticlinal ridge near the toe of the accretionary prism that is a prime prospect site for gas hydrate exploration. I combined high-resolution P-Cable 3D seismic, with a bottom-simulating reflection (BSR – a reverse polarity seismic reflection indicating the base of the gas hydrate phase boundary) mapped in the cube, heat probe data and numerical geothermal modeling to examine the shallow sub-surface ($<$1 km meter below seafloor) dynamic processes affecting hydrate formation. There are different geological controls on hydrate stability, including temperature and pressure perturbation. I interpreted that temperature perturbation can be generated by focused fluid flow upward through dipping permeable strata in the limb, the slope basin, and along thrust faults and ridge-top normal faults. Furthermore, I have interpreted several double BSRs in seismic data that underlie seabed sedimentary sliding and depositional features. Abrupt changes in subsurface pressure and temperature due to the rapid seabed sedimentary processes can cause a rapid shift of the base of the gas hydrate stability zone. This shift may be either downward or upward and would result in the accumulation or dissociation of hydrate in sediments sandwiched by the double BSRs, respectively. I propose that dynamic surficial processes on the seafloor together with shallow focused fluid flow affect hydrate distribution and saturation at depth and may even result in methane expulsion into the ocean if such localized features are common along convergent plate boundaries. \\ For the second study, I wrote a code to calculate fluid migration rates and then quantified the fluid budget and partitioning of fluid flow between focused discharge along fault zones and diffusive flow across the Taiwan’s accretionary wedge. I developed a rapid method to estimate fluid migration rates using an extensive database of seismic reflection profiles offshore SW Taiwan. The region data show a BSR that sometimes deflects towards the seafloor near thrust faults, indicating localized heat flow variations. I interpreted that at these sites, advecting warm pore fluids transport heat to shallower depths and force the BSR to become shallower. I developed a 2-D steady-state numerical method to quantify the fluid flow rates required to cause such thermal anomalies. This method combines analytical Peclet number analysis for a 2D model of focused fluid flow along a dipping conduit, and a numerical scheme to account for bathymetric effects on the temperature field under the seabed. Fluid flow rates along dipping conduits near the trench of the accretionary wedge range between 0.1 to 16 m\textsuperscript{3} yr\textsuperscript{-1} m\textsuperscript{-1}, with faster and slower rates generally associated with faults and slope basin discontinuities, respectively. To evaluate the fluid pattern evolution during orogeny, I quantified the fluid budget and partitioning of fluid flow between focused discharge along fault zones and diffusive flow both in Taiwan’s subduction zone and initial collision zone. The model highlight stronger fluid advection from depth in the initial collision zone, where on average 38.5\% of the total volume of water released during compaction of the accretionary prism is expelled through focused fluid flow. This is slightly higher than the 25\% for the subduction zone. \\ By adding time dimension to the 3D seismic data, I have documented the transient stages of gas hydrate system when PT conditions have changed in a spatial resolution of meters and dimensions of kilometers. By adding time dimension to the regional 2D seismic data, I provided a quantitative link between subduction-collision system evolution, fluid budgets and flow rates in a spatial resolution of hundreds of meters and dimensions of hundreds of kilometers. These results illustrated the important role of focused fluid flow processes in convergent plate boundary dewatering.