| 摘要: | 台灣屬亞熱帶季風氣候，每年有許多颱風和降雨。強降雨總是會造成大量的山體滑坡，並改變邊坡周圍的地貌。隨後的颱風和洪水事件以及淺層滑坡產生的土石流皆可能會將邊坡上大量不穩定的砂土沖刷至鄰近居民周遭。其後果將影響附近居民的安全，污染物也將影響到周遭環境。為了有效地驗證邊坡穩定性，土壤含水量是最重要的監測因素之一，因此，最好在降雨過程中能自動且方便地獲取邊坡的含水量。時域反射法（Time
Domain Reflectometry）可以同時滿足連續測量邊坡的土壤體積含水量（）和土壤導電度（Electrical conductivity, EC）的要求。以往的研究闡明淺層滑坡監測土壤含水量和導電度之間在不同的土壤濕潤-乾燥速率下的獨特關係。因此，本研究首先分析設置於曾文水庫流域的TDR 貫入器的觀測結果，驗證土壤含水量和導電度的關係，但由於既有TDR
裝置的不穩定性和當前TDR 貫入器的有限量測範圍，明顯地資料變異量導致了不甚滿意的結果。
基於上述的現地測量狀況，本研究利用COMSOL 數值模擬軟體進行電磁場、電能密度和應力及應變模擬，模擬各種不同的尺寸設計並考慮所有因素，進一步將現有的TDR 貫入器從單側外導體改良至雙側外導體。本研究初步提出TDR 貫入器的改良設計後，仍然進一步安排砂箱物理試驗以測試並確認其量測可行性，並旨在找出土壤含水量、導電度和基質吸力之間的關係。本研究之砂箱試驗不單使用既有的與改良後TDR 貫入器，還同時放入傳統的TDR 感測器，以觀察各個不同TDR 感測器的測量性能以及降雨
事件中可能發生的遲滯現象。
最後，本研究除了成功找出土壤含水量、導電度和基質吸力之間的緊密相依關係以外，也在實驗過程中發現不同土壤種類之特性，亦會影響TDR 貫入器的量測，並且以各TDR 感測器量測過程中所發現的遲滯現象，呼應了本研究前期提出的在不同降雨事件中，土壤含水量和導電度應會有不同關係之論點。;Taiwan is featured with subtropics monsoon climate and has many typhoons and rainfall
annually. Heavy rainfall always causes lots of landslides and changes landforms around the
slopes. Great quantities of unstable sand and soil on the slopes may be washed down by
subsequent typhoon and flood events, as well as the soil yields from the shallow landslides. The
consequence will affect the safety of residents nearby and raise the pollutant issue to
environment. To efficiently verify the safety of the slope, soil water content is one of the most
important factors for monitoring, therefore, it is better to automatically capture the water
content profile of the soil slope during rainfall events. Time Domain Reflectometry (TDR) can
achieve the demands of continuously measuring the volumetric water content () and the
electrical conductivity (EC) of the soil slope simultaneously. Previous study has reported the
unique relationship between the soil and EC for the shallow landslide monitoring. Therefore,
observations by TDR penetrometer at Zengwun Reservoir Watershed were firstly retrieved for
the relationship in different soil wetting-drying rates, but apparent variations led unsatisfied
results because of the instability of the TDR device and limited sampling volume of the current
TDR penetrometer.
Based on the aforementioned measurements in the field, this study further improved the
existed TDR penetrometer from one-side to dual-sides by using COMSOL numerical software
for electromagnetic field simulations, as well as the electrical energy density and the strain in
response to stress of each different design. Considering all factors, this study attempted to obtain
the optimal design of the TDR penetrometer. Although the improvement of the TDR
penetrometer is determined, a sandbox physical test was further arranged to confirm the
feasibility and aimed to find out the relationship between soil , EC and matric suction. The
existing TDR penetrator and the improved TDR penetrator were penetrated into the sandbox,
as well as the traditional TDR probe to observe the measurement performance with each different TDR sensors and the possible hysteresis effect during the rainfall event. At last,
transforming the electrical conductivity profile obtained by ERT to the water content profile,
trying to point out the importance of the water content monitoring in slope stability.
Finally, this study revealed the close relationship between soil water content, electrical
conductivity and matric suction, and this study also found the characteristics of different soil
may affect the measurement of TDR sensors, and the hysteresis effect found in the TDR sensor
measurement process echoes the issue that the soil water content and electrical conductivity
will have different relationship in different rainfall events. |
