Relating species patterns and environmental variables by canonical ordination: An analysis of soft-bottom macrofauna in the region of Tromso, northern Norway

Relationships between species composition and environmental variables in soft-bottom faunal communities near the city of Tromso, northern Norway (69 degrees 40'N, 19 degrees 0'E), were examined using Canonical Correspondence Analysis (CCA), Quantitative data for 87 species and 14 environmental variables at 30 stations (21 sites) were considered. The variables used were depth, sediment parameters (grain size distribution, total organic carbon (TOC), C/N ratio), and an index representing organic effluent discharge. It was possible to explain roughly one-third of the species variance. Depth was the most important variable, but significant species patterns were also related to the effluents and to fine sand and gravel sediment fi actions. The species assemblage at enriched stations was strongly associated,vith organic discharge and TOC, but relationships between species and effluents were also detected in areas not adversely affected by the organic discharge. The amount of unexplained species variance suggests that various other factors also influence species assemblages. Thus the use of CCA indicates the need to characterise different ecological relationships in older to interpret the biological information more completely. (C) 1998 Elsevier Science Ltd. All rights reserved.