生物資源暨農學院: 生物環境系統工程學研究所指導教授: 林裕彬林韋志Lin, Wei-ChihWei-ChihLin2017-03-022018-06-292017-03-022018-06-292016http://ntur.lib.ntu.edu.tw//handle/246246/272603臺灣過去因受到全球工業化的衝擊，使得全台工廠林立，許多工業所排放之廢棄物，無經適當處理便直接進入農業環境中，最終污染灌溉區域水源與土壤；其中土壤重金屬污染具毒性高、持續時間長及污染物不易分解等特性，較容易造成農業環境與公眾健康較嚴重之負面影響。除此之外，土壤重金屬亦會對自然界某些物種造成影響，污染的土壤中重金屬會被較低階的消費者或是植物攝取入體內，再經由食物鏈傳遞到較高階的消費者中。因此，除了要建立完善之監控及管制程序外，同時在進行污染區域進行整治排序及決策時，應同時納入生態保育價值，並建立穩健性的整治決策。本研究建構一新的土壤種金屬整治及生態保育兼具的整治排序及決策分析方法，並以全台八種土壤重金屬以及八種保育等級鳥類為例，進行方法建構與驗證。土壤重金屬分布推估的部分，主要是根據現有調查資料，利用地理統計同時模擬八種重金屬，其在研究區內所有網格可能的空間分布，並根據環保署所制定的「台灣地區土壤重金屬含量及等級區分表」作為判定土壤重金屬污染區域之標準，並評估預測之污染地區之不確定性。本論文之鳥類空間分布推估的部分，主要是利用拔靴法及物種分布模式，模擬多組可能的物種分布，並結合生態熱點分析與系統性保育規劃模式，推估研究區內各網格之生態價值，並評估其不確定性，結合土壤重金屬模擬及不確定性分析，利用穩健的決策分析方法，分析不同整治決策的利弊，以指出在額外考量生態價值情況下，那些區域需優先進行整治。結果顯示，本研究所發展之環境汙染區域劃設及整治排序之方法，確實能夠提供決策者決定預期要達到的棲地無汙染指標(其越高代表棲地品質越好)、偽陽性率(無汙染區域卻被劃設為整治區域的比例)，以及穩健性(有多少組realization達到預設的棲地無汙染指標和偽陽性率)，協助決策者規劃出最佳的考量生態價值的整治排序與比例。本研究提供決策者考量生態價質的整治決策方法，且在受到資訊不確定性的影響下，以穩健性為評估指標進行環境整治之決策規劃。In the past, due to the global industrialization, various industries were set up in Taiwan. A ton of by-products or wastes from industries have been emitted into our environment without any proper treatments. These wastes eventually lead to water pollution in irrigation area and soil pollution. Because the heavy metals emitted from industrial sector into soil are highly toxic, long-lasting and persistent, they tend to adverse effects on public health and environment. In addition, soil heavy metals also affect species inhabiting in the contaminated area via food webs. Heavy metals tend to accumulate in plants and lower-level consumers inhabiting in contaminated soil, and further magnify into higher consumers level, such as birds, through unintentionally intaking heavy metal toxicities into their body when preying on plants or lower-level consumers. Therefore, in addition to setting up monitoring and controlling processes, a comprehensive remediation design focusing on soil heavy metal contamination needs to be proposed. In this study, we took eight soil heavy metals and eight birds of which the protected level is greater than conservation-deserving level as an example. Based on the survey data of soil heavy metals, we applied geostatistics to jointly simulate the possible distribution of eight soil heavy metals in Taiwan. After that, based on Taiwan’s Environmental Protection Administration (EPA) decided classification, in which the concentrations of soil heavy metals were classified into five classes, the area with concentrations within or above fourth range are regarded as polluted. Besides, the uncertainty analysis for simulated distributions of soil heavy metals is also completed. With respect to the simulation of bird distribution, we applied bootstrap resample technique and species distribution model to generate as many distributions as possible. The area with high ecological values are delineated using Zonation, a systematic conservation model, based on the possible bird distributions. Similar to soil heavy metals, the uncertainty analysis for simulated distributions of bird was also completed. Finally, based on the simulated maps of soil heavy metals and birds, a robust decision-making approach is used to evaluate the performance of each candidate solution and decide the area in need for remediation. The results show the approach proposed in this study can help decision-makers to set up the remediation area given the pollution-free habitat suitability (index of ecological value based on environmental pollution), false positive rate (dividing the number of non-polluted areas that are wrongly classified as contaminated by the total number of non-polluted areas) and robustness (the proportion of realizations in which the pollution-free habitat suitability and false positive rate both reach pre-defined standards). This study provides decision-makers with a structural approach in consideration of the influence of uncertainty which is applicable to the decision making regarding environmental remediation based on the quantified robustness.論文使用權限: 不同意授權整治優先順序重金屬污染生物保育穩健決策分析不確定性地理統計remediation prioritymultiple heavy metalspollutionbiological conservationrobust decisiongeostatistics[SDGs]SDG3[SDGs]SDG11[SDGs]SDG12thesis10.6342/NTU201600225