Response of grazing impacts of an algivorous fish (Pseudogastromyzon myersi: Balitoridae) to seasonal disturbance in Hong Kong streams

1. Manipulative experiments were carried out in four Hong Kong streams (two shaded, two unshaded) to investigate the impact of grazing by an algivorous fish, Pseudogastromyzon myersi, on benthic algal biomass and assemblage composition. Experiments were conducted and repeated during both the dry and wet seasons to determine whether spate-induced disturbance modified any grazing effect. Treatments comprised fish exclusion and inclusion via closed and open cages, with a no-cage treatment used as a control for the cage effect. Treatments were maintained for 4 weeks in each experimental run. 2. Grazing by P. myersi reduced benthic algal biomass and the organic matter content of periphyton in open cages and the no-cage treatment relative to closed cages. The similarity between open-cage and no-cage treatments was evidence that the overall difference among treatments was caused by limiting fish access to closed cages and not merely an artifact of caging. Grazing effects were broadly similar in all streams, but there was a significant statistical interaction between treatments and seasons. 3. Analysis of dry-season data matched the overall trend in inter-treatment differences, confirming the effects of grazing by P. myersi on algal biomass and periphyton organic matter. Significant differences in algal assemblage composition between closed-cage and no-cage treatments during the dry season reflected reductions in the abundance of erect, stalked diatoms (Gomphonema) and filamentous cyanobacteria (Homeothrix). Removal of these vulnerable overstorey algae by P. myersi resulted in greater abundance of understorey diatoms (Achnanthes and Cocconeis) in the no-cage treatment in all streams during the dry season. The composition of algal assemblages in open cages was intermediate between the other two treatments. 4. Although fish densities were greater in all streams during the wet season, spate-induced disturbance obscured grazing effects and there were no significant differences among treatments attributable to fish grazing. Seasonal variation in impacts of P. myersi grazing provides support for the harsh-benign hypothesis, and confirms that biotic factors are less important controls of stream algal biomass and assemblage structure during periods (i.e. the wet season in Hong Kong) when abiotic disturbances are frequent or intense.