地下水污染源特徵(例如總量、位置和排放時間等)識別的精準程度，攸關整治策略的制定、責任歸屬與整治成本的高低。目前雖已發展有許多污染源特徵的推估與追蹤方法，但大多數僅能用於簡化模型，並在分析效率與準確性上仍有許多改進空間。相較於目前多數研究仍嘗試結合數學演算法增強推估的效能，本計劃主要目的為新技術的導入，將應用目前最新的量子計算進行污染溯源，同時評估此新技術的效能。為了達成此新技術驗證目標，本計畫將利用既有地下水數值模式產生假想例，產出數種不同的污染情境(如不同污染源數量、污染源分布、降雨或地質條件等變化)的案例。假想例中的點位觀測資料，將分別利用量子運算與傳統反推估法對模擬區域中的污染源特徵進行反推估。最後本工作將所有反推估結果進行量化與比較，並評估應用量子運算的污染溯源技術在不同情境下的效能。由於本計畫為將量子運算技術引入至土壤地下水領域的嘗試，希望未來能有許多量子運算的嶄新應用(如高異質性含水層模擬、高效整治藥劑開發、整治決策最佳化或由觀測濃度值進行污染預測等)能因本計畫的成果而有更多進展。 ;The accuracy of identifying groundwater contamination sources and transport characteristics (such as quantity, location, and released history) has significantly influenced remediation strategies, identifications of responsibilities, and budgets for remediation. There are numerous methods developed for characterizing contamination sources. However, most methods are available for simplified cases, and their efficiency and accuracy are under development. The study aims to develop and evaluate cutting-edge technology, quantum computing (QC), for source-tracking in soil and groundwater contamination. To achieve the research goal, this study employs traditional groundwater numerical models (MODFLOW and MT3D) to generate synthetic cases with several scenarios (such as different numbers or spatial distribution of pollution sources and precipitations or geological conditions). In addition, both QC and traditional inverse methods will be employed to estimate the characteristics of pollution sources. Finally, the study will evaluate the performance of QC on the source tracking with all of the results. This work is the first to introduce the QC for groundwater remediation tasks. The results are expected to extend the applications in the fields of soil and groundwater remediation for cases with highly heterogeneous aquifers and complex transport behavior.
