- 學位論文
河川渠道重金屬濃度污染源多變量統計追溯方法之探討 – 以桃園南崁溪與彰化渠道為例
Exploration of using Multivariate Statistical Methods for Source Tracing of Heavy Metal Contamination in River and Canals - The Nankan River in Taoyuan and Irrigation Canals in Changhua
摘要
隨著氣候變遷水資源的不穩定，提升可利用水資源的質與量就更顯重要，尤其是地表水資源。然而，地表水的污染事件近年仍層出不窮，包含河川、渠道等，需從受體現況資訊往回追溯污染源頭。本研究提出應用多變量統計分析（MSA）及釐清污染物傳輸載體等策略，建立工作框架以探討地表水污染溯源的可行性。 本研究「應用合成數據建立污染溯源的多變量統計分析工作框架」，將MSA應用於河川水質模式所模擬的合成數據，驗證MSA能否正確推論出河川水質模式中設定的污染情境。結果顯示，集群分析可以評估污染源數量與可能區位，但應同時考量支流排水與水量造成的影響；以主成分分析評估污染源特性時，要將第二主成分、Pearson相關係數及污染物的吸附係數都納入考量，以提高判斷污染源特性的準確性。另外，本研究「應用污染特徵與攔留器建立污染溯源的重金屬傳輸分析工作框架」，針對定期監測土壤重金屬有增量情形的農地，釐清灌溉水質中重金屬主要存在液相或固相。結果農田土壤與攔留器懸浮固體的重金屬污染特徵及皮爾森相關係數高，顯示水中懸浮固體是造成農田土壤重金屬累積的主因。而灌溉水中懸浮固體的濃度隨著水流速度減緩而增加，渠道底泥厚度亦有相同趨勢，推論懸浮固體來自引灌水源，應從源頭管制以減低重金屬造成的可能風險。 綜合兩項工作框架的探討結果，顯示在地表水污染溯源時，不可以忽略重金屬吸附於懸浮固體並以此作為載體的移動行為，而此，與重金屬於水體中吸附在懸浮固體的吸附係數有關。所以，對於事業偷排卻沒有明顯目視可見異常的地表水污染事件，建議可以本研究建立的工作框架縮限污染源區域與初步掌握污染源特性。
並列摘要
With climate change and unstable water resources, it is more important to enhance the quality and quantity of available water resources, especially in surface water resources. However, pollution incidents of surface water have continued to occur in recent years, including rivers, canals, etc. It must be traced from the available information (receptors) to the source of the pollution. In this study, I propose to apply two strategies, such as multivariate statistical analysis (MSA) and identification of pollutant transport carriers, to establish a working framework to explore the feasibility of surface water pollution traceability. This study "applies synthetic data to establish the working framework for multivariate statistical analysis of pollution traceability". I applied MSA to synthetic data simulated by a river water quality model to verify whether the MSA can correctly infer the pollution scenario assigned in the river water quality model. The results show that cluster analysis can infer the number and location of pollution sources, but it should be noted that the location and volume of tributary drainage may affect the assessment results. When applying principal component analysis to evaluate the source characteristics, the second principal component, Pearson correlation coefficient and pollutant adsorption coefficient should be taken into consideration in order to improve the accuracy of determining the source characteristics. In addition, this study "applied pollutant signature and interceptor to establish the working framework for heavy metal transport of pollution traceability." For agricultural land where regular monitoring results show an increment of heavy metal concentrations in agricultural soil. To prevent pollution recurrence, this study aims to identify transportation media and routes of heavy metals by soil pollutant signature and intercepted suspended solids (SS). The result shows the high Pearson correlation coefficients of heavy metal concentrations in agricultural soils and intercepted SS, implying that the heavy metals of SS in irrigation water is the main contribution. Regarding the origin of SS in an irrigation ditch, SS increases while the water velocity decreases. On the other hand, the ditch sediment and SS have the same trend. Sedimentation-resuspension process well explains the relation, which implies the SS originates from the irrigation source. In order to lower the potential risks caused by heavy metals, our results further emphasize the importance of controlling the SS from the source. Based on the results of the two working frameworks, it is shown that the transport behavior of heavy metals adsorbed on suspended solids and used as carriers cannot be ignored in the traceability of surface water pollution, which is related to the adsorption coefficient of heavy metals adsorbed on suspended solids in water. Therefore, the working framework established in this study can be applied to narrow down the source area and characteristics for pollution events caused by illegal discharges from factories to surface water bodies without obvious visual anomalies.