The golden mussel Limnoperna fortunei (Dunker, 1857) is a filter-feeding species of freshwater bivalve that is native
to southern China. The species readily invades water transfer tunnels, causing considerable damage. The
mussels typically attach to the walls of pipelines and gates, and are associated with serious biofouling. This
may lead to increased flow resistance of the tunnels and to corrosion of their walls, and may threaten their operation.
The mussels exhibit very high environmental adaptability and, once they become established in the
water transfer works, they are impossible to eliminate. Therefore, prevention of the entry of golden mussels
into the tunnels is considered to be the most efficient means of avoiding biofouling by this species. The main period
of invasion corresponds with their reproductive season, which therefore forms the focus of prevention
methods. Consequently, further study of reproduction of the golden mussel is essential. The process of attachment
of golden mussels to material with a solid surface was studied experimentally in the Xizhijiang River, a tributary
of the East River. It is the main water resource of the water transfer project supplying Shenzhen and
Dongguan, southern China. Experiments were conducted from March 2010 to April 2011. Fourteen sets of materials
were installed in the river to monitor attachment of the mussels. Shell-length frequency data of golden mussels
attached on the monitoring materials were determined monthly and analyzed for identification of cohorts of
individuals produced during successive reproductive seasons, and for measurement of their growth rates. The
Windows version of FiSAT (FiSAT II), a software package developed specifically for the analysis of lengthfrequency
data of aquatic biota, was applied in this study. The density of attachment of golden mussels to the
monitoring materials was related to the water depth and to the orientation of the attachment surface. Clay and
silt precipitation reduced the attachment density of the mussels. The attachment density on downward facing
surfaces with lower sediment precipitation was much higher than the density on the upward facing surfaces
with higher precipitation. A six-fold increase in precipitation of clay and silt was associated with a 90% decrease
in attachment density. The analysis of monthly shell-length frequency indicated that the mussels produced three
generations per year, each of which was associated with two reproductive peaks, resulting in six cohorts per year.
The growth rates of the six cohorts differed and varied from month to month. Mean growth rates were in the
range 0.4–3.5 mm/month. The average growth rate was logarithmically related to the water temperature, total
nitrogen and total phosphorus concentrations. Therefore, the time for the cohorts to reach maturity (shell length:
6–8 mm) varied with these environmental parameters. The feasibility of preventing the attachment of golden
mussels is discussed. Intensive prevention efforts should be performed during their peak reproductive seasons
(March to May, June to September, and October to December). Clay or silt precipitation is recommended as a potential
method for preventing the attachment of golden mussel to materials. However, because of the extended
and frequent reproductive activities of golden mussels, preventing their invasion of water transfer tunnels remains
a challenge; control strategies will need to be applied for long periods.