埤塘的經營管理與蜻蛉目生物多樣性的關係

Abstract

本研究旨在探討不同的埤塘經營管理方式，對於桃園埤塘的環境特性與蜻蛉目生物多樣性有何影響。將埤塘分為五大類群：(1) 純魚類養殖埤塘、(2) 步道劃設養殖埤塘、(3) 休閒垂釣養殖埤塘、(4) 規劃為生態公園的無養殖埤塘，以及 (5) 荒廢無用途的埤塘，共 24 個研究樣區，於 2014 年 6 月至 2015 年 7 月，調查蜻蛉目成蟲生物多樣性、天氣狀況及埤塘水質、埤塘結構和埤塘周圍土地利用與景觀參數等環境特性。計算辛普森指數 (Simpson’s index, 1-D)與夏儂指數(Shannon-Wiener’s index)等生物多樣性指數，並使用克瓦二氏檢定 (Kruskal-Wallis test)、曼惠特尼 U 檢定 (Mann-Whitney U test) 、一般線性模型 (general linear model)、斯皮爾曼等級相關 (Spearman rank correlation)分析、相似度檢定分析 (analysis of similarity)、生物與環境資料分析 (BIO-ENV analysis) 及典型對應分析(canonical correspondence analysis) 等統計方法，分析不同類別埤塘之間環境特性與蜻蛉目多樣性的差異，以及環境特性與蜻蛉目多樣性之間的關係。 由克瓦二氏檢定與曼惠特尼 U 檢定顯示，在不同經營管理類別下，埤塘的環境特性和蜻蛉目多樣性確實有差異。純養殖埤塘、步道劃設養殖埤塘和休閒垂釣養殖埤塘，其 pH 值、導電度與溶氧量都有較高的趨勢，主要與埤塘中過度施肥與飼料投入等魚類養殖行為，導致水中營養鹽過剩而產生優養化現象有關。而有魚類養殖行為之埤塘的堤岸材質皆以水泥堤岸為主，且僅有堤岸植被而無水生植被。生態公園埤塘的水質較佳，堤岸皆為自然草土材質，除堤岸植被以外更擁有豐富水生植被；荒廢無用途埤塘則環境變異較大。其中，以步道劃設養殖埤塘擁有最低的蜻蛉目多樣性，生態公園埤塘的蜻蛉目多樣性最高。另外，透過相似度檢定分析，發現生態公園埤塘的蜻蛉目群聚狀態明顯和其他類別有所區別，且蜻蛉目物種群聚之差別主要反應在堤岸材質與植被狀態的區別上。

探討環境特性與蜻蛉目多樣性之間的關係，透過一般線性模型發現埤塘周圍土地利用比例對於蜻蛉目多樣性的影響不大，僅在 1600 公尺半徑環域範圍下有影響。然而景觀尺度上地理位置的鄰近關係與環域範圍內埤塘的密度，可能是使得休閒垂釣養殖埤塘的蜻蛉目多樣性能夠高於其他魚類養殖埤塘的原因。再經由斯皮爾曼等級相關分析、BIO-ENV 分析及典型相關分析，發現在本研究中，面積最小的生態公園埤塘類別有著最豐富的蜻蛉目多樣性，而面積與蜻蛉目多樣性之間為顯著負相關，顯示並非埤塘面積越大就能擁有越多物種，而是受到其他由於不同經營管理方式所造成之相異的環境特性所影響。本研究結果也發現，影響蜻蛉目多樣性之最主要的環境因子為植被型態的差異，擁有挺水植物、浮水植物等水生植被的埤塘能具有較高的蜻蛉目生物多樣性，因為蜻蛉目昆蟲仰賴堤岸植被進行覓食、棲息等行為，更需要水生植被提供產卵的場所。次要的環境影響因子則為水質參數中的導電度。研究結果也發現，不同類別埤塘所出現的蜻蛉目物種也不同，在有魚類養殖行為的埤塘中多半為普遍的優勢種，對污染的耐受度較高；而出現於生態公園埤塘的物種則多為特別仰賴水生植被的物種。因而在擁有較多樣化植被型態，且植被狀態維護良好的生態公園埤塘中，能夠擁有最豐富的蜻蛉目生物多樣性。 本研究顯示，埤塘中的人為經營管理活動會影響埤塘的水質、植被型態等環境特性，進而影響池塘生態系中的生物多樣性，建議針對池塘生態系應有完善的保育規劃與投入，特別是自然堤岸材質的選擇，以及其周圍與水中植被狀態的維護；在魚類養殖行為的埤塘中即使無法栽植水生植被，也應盡量保留堤岸植被，並設置棲木供蜻蛉停棲，才能在人為活動與自然保育之間取得平衡。而除了埤塘本身的保育規劃，也應注意大的景觀尺度下埤塘數量的維護，以保持甚至提升埤塘之生物多樣性價值。

The objective of this study was to assess the relationship between farm pond management and odonate biodiversity. Twenty-four farm ponds are selected for this study and classified into five categories according to their pond management practices, including (1) fish farming ponds, (2) fish farming ponds with a path surrounding the pond, (3) fish farming ponds for recreational fishing, (4) ecology park ponds, and (5) abandoned ponds. Adult odonate biodiversity and environmental characteristics of farm ponds including water quality, pond structure and landscape variables were investigated during June 2014 to July 2015, and the weather condition of each farm pond was recorded during the investigation period. Simpson’s index (1-D) and Shannon-Wiener’s index were calculated for ondonate biodiversity. Kruskal-Wallis test, Mann-Whitney U test, general linear model, Spearman rank correlation, analysis of similarity, BIO-ENV analysis and canonical correspondence analysis were applied to examine effects of environmental factors of farm ponds on odonate biodiversity. Environmental characteristics of farm ponds and odonate biodiversity varied significantly with different farm pond management based on the results of Kruskal-Wallis test and Mann-Whitney U test. In particular, water qualities of farm ponds, such as pH, conductivity, and dissolved oxygen, are relatively high in all three fish farming ponds compared to ecology park ponds and abandoned ponds. These difference of pond water conditions may be due to aquaculture activitiesat fish farm ponds, because pond fertilization and input of fish feed resulted in eutrophication. As for pond physical structures, three categories of fish farming ponds all have concrete embankment with only terrestrial vegetation. Ecology park ponds have natural embankment with both aquatic and terrestrial vegetation. Abandoned ponds have mixed types of embankments. Among five classes of farm ponds, odonate biodiversity are the lowest in the fish farming ponds with a path surrounding the pond and are the highest in ecology park ponds. Landscape variables only showed significantly difference at the 1600-meter buffer range by using general linear model analysis. In addition, geographical location and number of ponds within the buffer range may be the reason to explain why fish farming ponds for recreational fishing can have higher odonate biodiversity than other fish farming ponds. Interestingly, pond sizes showed significant negative correlation with odonate biodiversity based on the results of Spearman rank correlation, BIO-ENV analysis and canonical correspondence analysis. The result of this study further suggested that pond size may be not as important as pond management to odonate biodiversity conservation. The most prominent factor contributing to odonate biodiversity and species composition is the presence of different vegetation types, because the survivals of odonates rely on not only riparian plants but also aquatic plants including emergent plants and floating plants. The second important factor is conductivity of pond water. Therefore, ecology park ponds with diverse vegetation types have the highest odonate biodiversity among five classes of ponds. This study shows that farm pond management can affect the environmental characteristics of pond ecosystems, and thus influence the odonate biodiversity. Farm ponds can host high biodiversity as long as pond management practices are considered both local pond habitat preservation and large landscape-scale pond conservation.