| 摘要: | 近年來，大氣層中溫室效應氣體(greenhouse gases)含量逐年增加進而導致全球暖化及氣候變遷議的議題普遍受到國際間的重視。另一方面，人工溼地為一種低成本、省能源、容易操作、且兼具生態保育及景觀價值的廢水處理程序，然而在淨化水質的生化程序中，無可避免的也會產生甲烷(CH4)及氧化亞氮(N2O)兩種溫室效應氣體而引起關切。人工溼地在國內外應用的數目及面積規模均有逐年成長的趨勢，因此溫室氣體排放造成衝擊的相關研究有其迫切需要。
本研究初步選擇嘉南藥理科技大學校園人工溼地為研究對象，使用密閉罩法測量溼地不同位置點的溫室氣體釋放量，並同時監測溼地的廢水處理操作條件及水質淨化結果，自96年12月起每個月採樣分析一次。人工溼地系統由一個表面下流動式(subsurface flow, SSF)溼地連結一個表面流(free water surface, FWS)溼地所構成，總溼地面積約3,800 m2。主要研究目的為：(1)探討人工溼地溫室氣體釋放量的空間與時間變化；(2)比較不同類型人工溼地(FWS與SSF溼地)溫室氣體排放量的差異，並與國內已建立的天然溼地溫室氣體排放資料進行比較；(3)利用單變數線性回歸法分析溫室氣體排放與溼地水質環境條件的關係；(4)日夜間變化對人工溼地溫室效應氣體產量的影響。
調查結果顯示，不同位置及時間所測得的N2O及CH4釋放量範圍分別為-6.10 ~ 128.78 μg N2O m-2 h-1及-4.17 ~ 44.4 mg CH4 m-2 h-1。此數值落於文獻中有關人工溼地處理廢水所報導釋放量的範圍內(−46.3 ~ 150 μg N2O m−2 h−1及–15.63~ 72.46 mg CH4 m−2 h−1)，而似乎比國內天然溼地溫室氣體排放的文獻值稍微高。在時間變化的探討方面，SSF及FWS單元最高的N2O釋放量都發生在3月，而最低的釋放率則都在2月。SSF及FWS單元最高的CH4釋放量都發生在6月，而SSF單元最低的釋放量在2月，FWS單元最低的釋放量則在1月。春、夏季的N2O釋放量顯著高於冬季的釋放量(扣除FWS3)(P＝0.026)。春、夏季的CH4釋放量也顯著高於冬季的釋放量(P＝0.006)。此現象的可能原因為季節變化導致溼地環境溫度顯著變化，進而影響產生N2O及CH4的生物性作用。本研究發現N2O與CH4的釋放量在溼地的不同空間位置有明顯不同的釋放量大小，兩者都沿著人工溼地的流動距離而逐漸下降，進流端的釋放量明顯高於出流端(p<0.05)。
進一步將不同位置所調查的溫室氣體釋放量與水質結果，以單變數線性回歸進行交叉分析，結果發現在SSF單元不同採樣點的N2O釋放量以TN (R=0.953，P＜0.05)、TKN (R=0.950，P＜0.05)的相關性最好。而在SSF單元不同採樣點的CH4釋放量以TN (R=0.957，P＝0.04)的相關性最好，在FWS單元不同採樣點的CH4釋放量以BOD (R=0.999，P＜0.001)的相關性最好。此結果顯示人工溼地溫室氣體的產生是由釵h複雜的環境因子共同影響，加上本研究至目前所收集的數據有限，因此難以獲得單一變數對溫室氣體釋放的影響關係。雖然在本研究場址，FWS溼地配置於SSF溼地之後，但是統計分析結果顯示兩種不同類型溼地的N2O及CH4平均釋放量都無明顯差異(p>0.05)。初步結果顯示溼地的N2O及CH4的釋放量與氣溫(或水溫)的關係，與modified Arrhenius equation十分契合(p<0.05，R>0.796)。本研究還需要持續的長期監測調查，以累積足夠統計意義的數據方能做出一致性的結論。
Because global warming and climate changes are following upon an increase in atmospheric levels of greenhouse gases, there is intense concern with the sources and emissions of the gases. Constructed wetland technology is a natural treatment system for wastewater engineering and is characterized by the advantages of moderate capital costs and very low energy consumption and maintenance requirements. However, constructed wetlands are inherently the net source of greenhouse gases such as methane (CH4) and nitrous oxide (N2O). Wetland construction and the area covered by constructed wetlands are increasing domestically and globally, thus there is an urgent need to elucidate the impact of constructed wetlands on atmospheric burden of these gases.
In a study period from December 2007 to June 2008, emission rates of nitrous oxide and methane as well as several parameters of water quality were simultaneously and monthly monitored at various sampling locations of a practical-scale constructed wetland system in Chia-Nan University of Pharmacy and Science, which has been operated for tertiary treatment of campus wastewater for near three years. This wetland system consists of a subsurface flow (subsurface flow, SSF) wetland and followed by a free water surface flow (free water surface, FWS) wetland with a total surface area of 3,800 m2. The objectives of the study were to: (1) investigate the temporal and spatial variations in emission rate of greenhouse gases; (2) examine the differences in greenhouse gas emissions between various types of constructed wetlands, and compare these results with those of domestic natural wetlands from literature; (3) investigate the relationship between gas emissions and water quality of the constructed wetland to elucidate possible factors that can affect greenhouse gas emission; (4) investigate the day-night dynamic in greenhouse gas emission.
The results monitored in the first-half year study showed that emission rates ranged from -6.10 to 128.78 μg N2O m−2 h-1 and -4.17 ~ 44.4 mg CH4 m−2 h-1. These values fall within the emission rates reported by literature of constructed wetlands(−46.3 ~ 150 μg N2O m−2 h−1 and–15.63~ 72.46 mg CH4 m−2 h−1), but are slightly higher than that measured from domestic natural wetlands in other studies. The emission rate of either N2O or CH4 was found to be significantly different (p<0.05) between various sampling locations of FWS or SSF wetlands, significantly decreasing (p<0.05) along the flow path of either wetland. A variety of water quality parameters were further correlated with gas emission rates by using single variable linear regression method. TN (R=0.953，P＜0.05)、TKN(R=0.950，P＜0.05) of water was closely and significantly correlated with N2O emission in the SSF wetland. TN (R=0.957，P=0.04) of water was closely and significantly correlated with CH4 emission in the SSF wetland and BOD (R=0.999，P＜0.001) of water was closely and significantly correlated with CH4 emission in the FWS wetland.It imply the emissions of the greenhouse gases in constructed wetlands depend on various complex and interconnecting processes, in which their effects can not be simply and clearly elucidated by a statistic model and limited data. Although the FWS wetland is installed following the SSF wetland in this study, there was no significant difference (p<0.05) in N2O or CH4 emission between the two types of wetlands.
In temporal variation study, emission rate of either N2O or CH4 was found to vary with the month as gas sampling. Up to now, the greatest emission rates of N2O were recorded in March, while the lowest rate were noted in February.The greatest emission rates of CH4 were recorded in June, while the lowest rate were noted in February.The possible reasons for this phenomenon is temperature variation caused by seasonal change lead affect the biological processes responsible for gas emission in wetlands, resulting in the change of emission rate. The relationship between emission rate and temperature (either ambient or water) were found following the modified Arrhenius equation (p<0.05，R>0.796)。The above results and conclusions were made only based on the preliminary study of the first-half year project. To make a consistent conclusion, it is necessary to sustain the long-term study further so as to obtain statistically meaningful data. |
