Macroalgal assemblages as indicators of the broad-scale impacts of fish farms on temperate reef habitats

Intensive fish culture in open sea pens can deliver large amounts of nutrients to coastal ecosystems. Sheltered areas with high water quality are predominately chosen for this type of mariculture, and these systems may be adversely affected by the presence of the farms. Since macroalgal community composition has been shown to be a good indicator of environmental disturbance on reef, the present study investigated the effect of salmon farms on macroalgae in a semi-enclosed coastal waterway in southern Tasmania. Data on the macroalgal community were collected from two depths at 44 sites of varying distance from twelve active fish farm leases. This included reference sites at distances of 5 km or more. The sites were widely distributed throughout the study area, and varied in their exposure to wave action. The macroalgal community composition differed significantly between sites at 100 m from fish farms and sites at 5 km or more. Sites at 400 m varied in their response to farms, with some sites showing characteristics similar to 100 m sites. Impacts varied between swell exposed sites and sites only subjected to wind-generated waves. Chaetomorpha spp. and Ulva spp. were abundant near fish farms at exposed sites, whereas the abundance of filamentous green algae increased at sites near fish farms in sheltered sites. The percentage cover of indicator groups such as epiphytes and opportunistic algae in total provided the best indicators of fish farm impacts on a broad scale. The percent cover of canopy forming perennial algae did not decrease near fish farms indicating that their growth and recruitment has not been greatly affected by high levels of sedimentation from fish farms or prolonged fouling by opportunistic algal epiphytes to the present, however further study is needed to examine this in more detail. The above analysis utilised photographic quadrats to quantify community composition. Most other broad-scale sampling methods used to measure macroalgal composition require expertise to identify species in situ. However, this reduces the capacity of monitoring programs to collect large amounts of data. Using data collected from a subset of 36 sites, the photographic method was compared with a manual quadrat sampling method. The two methods produced similar multivariate results but manual quadrats had a slightly greater capacity to detect the impacts of the fish farms. This indicates that photographic quadrats are likely to be conservative in quantifying effects of fish farms, but still deliver appropriate resolution to detect major changes in dominant macroalgal cover and composition. Some adjustments to the photographic methods used will allow better resolution for algae obstructed by canopy or epiphytic overgrowth.