不同空間尺度下河川生態環境因子與無脊椎動物群落指標的關係

Abstract

溪流河川系統為一個階層的系統，受到不同空間尺度的交互作用所影響。河川棲地與水中生物群落都受到階層系統的影響，而受不同空間尺度其環境因子的作用。對於溪流河川系統管理而言，了解河川棲地和水中生物群落在不同空間尺度下之關係變化，可更深入的認識溪流河川生態系統的作用及模式。本研究目標是探討河川生態環境與生物群落之間的關係於不同空間尺度下之變化。研究的生物指標為具多樣化生態特性且可反映河川系統功能的底棲大型無脊椎動物群落。研究地點為美國馬里蘭州，分析的數據採用馬里蘭河川生物調查的資料(MBSS)，採樣期間為1995-1997，研究樣點共422個(18個流域)。分析資料包含23個環境因子以及82個生物因子。本次研究在空間尺度上針對state、subregion和catchment三種尺度分別作探討。研究方法使用多變量統計分析以探討不同空間尺度下其環境特性、底棲大型無脊椎動物群落結構組成及環境因子與底棲大型無脊椎動物群落結構之間的關係等。主成分分析研究結果發現高層對於不同空間尺度的環境下，對其環境的變異程度均有影響。另外，從多維度分析研究發現到搖蚊科(Chironomidae)不僅是任何空間尺度下的優勢種，對於不同空間尺度下的底棲大型無脊椎動物群落結構的變異程度也都有影響。對於底棲大型無脊椎動物群落與環境因子之間的關係，從典型相關分析以及生物和環境相接的逐步分析相關檢定中發現底棲大型無脊椎動物群落受到小尺度的環境因子淺瀨棲地品質優劣影響較大。淺瀨棲地品質是一個小尺度、綜合型的環境因子其探討淺瀨的棲地複雜程度，包含流速、水深及底質。在馬里蘭州，淺瀨棲地品質較高的樣點其底棲大型無脊椎動物群落的豐度以及物種數也較高。底棲大型無脊椎動物群落結構主要受到小尺度的淺瀨棲地品質因子的影響且其生物多樣性也會隨著淺瀨棲地品質增高而增加。研究結果還發現小蜉科(Ephemerellidae)以及紋石蛾科(Hydropsychidae)對於在subregion以及catchment尺度的環境因子具有較高的相關性(ρ=0.285-0.446; ρ=0.506-0.568)。此外，小蜉科對於有機溶解碳以及(DOC)以及水中硫酸鹽濃度也具有高度相關性，而紋石蛾科對於流速、流速-深度多樣性以及淺瀨棲地品質具有高度相關性。因此，本研究提出監測小尺度環境因子其變異程度對於了解生物在河川系統中的反應有很大的重要性，此外，我們亦可以利用小蜉科以及紋石蛾科以作為馬里蘭州的有機汙染與水力狀況之棲地品質的生態指標。

Stream systems are affected by complex processes interacting with each other at multiple scales due to the hierarchical systems. Identifying the relationships between the environment and aquatic communities at different spatial scales is an essential research issues for enhancing our understanding of the ecosystem patterns and processes. In this study, we investigated the relationships between environmental variables and benthic macroinvertebrate communities in river ecosystem at multiple spatial scales including state-, subregion-, and catchment-levels. The data we used were extracted from the Round One Survey of U.S. Maryland Biological Stream Survey (MBSS) database which was conducted during 1995-1997 and comprised of 422 sites in 18 catchments in Maryland. There were 23 environmental variables including 17 physical habitat variables and 6 water chemistry variables and 82 biological taxa containing 78 families and 4 orders of benthic macroinvertebrates. Multivariate statistical analysis was used to determine the variation of environmental characteristics, community compositions of benthic macroinvertebrates, and the relationships of benthic macroinvertebrate communities and environmental variables of multiple spatial scales. Principal component analysis (PCA) results indicated that the altitude was the most important environmental variables on characterizing the environment at multiple scales since altitude was a composite variable of environmental condition. Non-metric multidimensional scaling (NMDS) demonstrated that Chironomidae was the dominant family in Maryland at multiple scales and it also affected the benthic macroinvertebrate community structure across different spatial scales. Canonical correspondence analysis (CCA) and multidimensional overall pattern (BVSTEP) showed that the benthic macroinvertebrate community structure tended to be controlled by the environmental variable such as the riffle habitat quality at catchment-scale (5.03%-40.65%), after then at subregion-scale (2.57%-6.54%) or state-scale (1.86%) due to influences of the riffle habitat quality. Riffle habitat quality was a composite variable simultaneously influential by the velocity, stream depth, and substrate as quantifying the habitat complexity at catchment-scale. Also, higher abundance and richness of the benthic macroinvertebrate communities were found in sites with high riffle habitat quality in Maryland. Therefore, the benthic macroinvertebrate community structure showed higher dependence upon the environmental variables at catchment-scale. Moreover, the results of BVSTEP also presented that Ephemerellidae (ρ=0.285-0.446) and Hydropsychidae (ρ=0.506-0.568) were correlated with the environment at subregion- and catchment-scales. In addition, Ephemerellidae was highly correlated with DOC and SO4 concentration, and Hydropsychidae was particularly correlated with discharge, velocity/depth diversity, and riffle habitat quality. This research suggested that monitoring the variation of the small-scale environmental variables could be essential for understanding the biological response in river ecosystem and we could use Ephemerellidae and Hydropsychidae as the ecological indicators for habitat quality in response to organic pollutions and hydraulic comditions of Maryland freshwater ecosystems.