以小琉球之基本控制點測量成果為例探討控制點位選擇之重要性

Assessing the Importance of Control Point Site Selection: A Case Study of Basic Control Points in Hsiaoliuchiu Island

本研究以小琉球地區之基本控制點為例，探討在地質活動的環境下基本控制點位選擇與穩定性評估的重要性。內政部國土測繪中心分別於2003至2004年及2017年完成小琉球15個基本控制點之GNSS與精密水準測量，其中10點為GNSS與水準共點。本研究將兩期成果重新約制於位於島嶼西側且環境相對穩定之點位LC06，求得小琉球相對於LC06之三維地表位移場，並結合2022至2025年間之野外調查，檢視各控制點樁位與周邊構造物、地形及地質間之關係。結果顯示，小琉球整體呈現數十mm等級之抬升，但LC02點位在13年間卻出現約百mm等級之沉陷及顯著水平位移，且方向與現地觀察到之邊坡滑動一致，周邊護坡與階梯亦持續產生破壞，顯示該點位實際坐落於邊坡滑動的區域中。位於小琉球中央谷地附近之LC03、LC04與LC07點位周邊則可見地形階地、局部擠壓抬升與裂隙發育，對應潛移逆斷層與小規模活動斷層之作用；在LC03、LC05與LC06點位附近亦可辨識泥貫入筒活動所造成的建物上拱與局部抬升。此外，潮位基準點TG74之觀測條件雖然良好，但所在堤防已整體向潮位站方向傾斜並產生大型滑移裂隙，顯示其作為高程及坐標基準點的穩定性亦存疑。綜合地表位移場與野外證據可知，小琉球多處基本控制點並非設置在真正穩定的基盤上，坐標較差反映的較可能是地表真實變形，而非單純測量誤差。因此本研究認為，在構造活動之區域規劃控制網時，控制點選擇不應僅依賴地形平坦與透空良好等環境條件，而必須整合區域地質構造、泥體構造分布、長期地表位移場及現地破壞跡象，排除潛在邊坡滑動區與局部構造活動帶，並透過定期檢測與野外查核維持點位穩定性，方能避免控制網在長期運作中發生難以察覺之網形扭曲，確保測量成果之可靠基準。

We investigated the importance of control point site selection and stability assessment in geologically active environments in this study, using the basic control network on Hsiaoliuchiu Island as a case example. The National Land Surveying and Mapping Center (NLSC) conducted GNSS and precise leveling surveys on 15 control points in 2003-2004 and 2017, including 10 points with collocated GNSS and leveling benchmarks. In this work, both survey epochs were minimally constrained to station LC06, located on the western side of the island in a relatively stable setting, to derive the three-dimensional displacement field of Hsiaoliuchiu. These geodetic results were integrated with systematic field investigations conducted between 2022 and 2025, focusing on the geological context, local structures, and deformation features surrounding each control point. The results show that the island has undergone several tens of millimeters of uplift, while LC02 experienced nearly 100 mm of subsidence and substantial horizontal displacement over 13 years. The displacement direction is consistent with on-site evidence of slope movement, and nearby retaining walls and stairways exhibit progressive structural damage, indicating that LC02 lies within an active sliding zone. Around LC03, LC04, and LC07 in the central valley, minor scarps, localized uplift, and surface cracking correspond to creeping reverse faults and small active structures. Additional deformation associated with mud diatremes is observed near stations LC03, LC05, and LC06, expressed as ground warping and building uplift. Although the benchmark TG74 provides excellent GNSS observing conditions, the host breakwater has tilted toward the station and developed large sliding fractures, calling into question its long-term stability as a vertical and horizontal reference. Overall, both geodetic and field evidence demonstrate that many control points on Hsiaoliuchiu are not situated on truly stable ground, and the observed coordinate discrepancies likely reflect genuine surface deformation rather than measurement error. We concluded that control point selection should not rely solely on flat terrain or adequate sky visibility in tectonically active regions. Instead, it must incorporate regional geological structures, the distribution of mobile shale features, long-term displacement fields, and direct field indicators of instability. Excluding potential landslide zones and local active structures, combined with regular re-surveys and field inspections, is essential to maintain control-point stability and to prevent subtle yet cumulative distortions of the control network, thereby ensuring the long-term reliability of geodetic reference frameworks.