A multivariate assessment of plant management impacts on macrophyte communities in a Scottish canal.
Submission of data collected from sites on the Union Canal, Scotland during 1977-84 to detrended correspondence analysis (DCA) and 2-way indicator species analysis (TWINSPAN) revealed plant community changes related to herbicide use, cutting and, in a few cases, shading. Dissolved oxygen status was significantly negatively correlated with one major axis of the DCA ordination. This axis appeared to reflect a trend of plant community change associated with efficient weed control operations. A second major axis was probably associated with a hydroseral gradient within the (open-water) early stages of vegetational change in the canal plant community. Timing and type of control operations appeared to play an important role in determining the hydroseral age of the plant community present. Sites recently subject to effective weed control generally showed increased dominance by rapid-growing filamentous algae or other opportunists. Untreated sites, sites sampled at long time intervals after control operations, and those where weed control measures had been inefficient, had a lower incidence of opportunist species in the long term. Floating-leaved rooted vegetation (e.g. Potamogeton natans), appeared to be naturally resistant to the weed control regimes used. Such species also competed well with Lemnaceae mats and became major problems in the canal. Repeated cutting, and early season diquat-alginate application gave acceptable weed control without altering macrophyte community structure to an unacceptable extent. Suboptimal diquat-alginate or terbutryn treatments, and late-season weed-cutting increased the abundance of opportunist algae and Lemna spp., and reduced the diversity of the canal macrophyte community.