Because the diffuser of the Barbers Point Ocean Outfall lies below safe diving depths, a remotely controlled

video camera system was used to determine the status of the marine fish communities and selected diurnally

exposed macroinvertebrate species residing on the diffuser. Video reconnaissance was completed over the

entire 534-m length. Three visual “transects,” which “sampled” approximately 31% of the total diffuser length,

were established on the diffuser pipe. Commencing in 1992, video samplings of the diffuser fish communities

have been carried annually except in 2000 when the equipment malfunctioned. The results of the thirteen

annual surveys to date indicate that the diffuser fish communities are dominated by species that are either small

as adults or juveniles of larger species, probably as a result of the presence of only small-scale shelter created

by small armor rock and gravel used in constructing the discharge pipe. Because of poor camera resolution,

differing angles of the camera, small fish sizes, and the fishes’ nature to flee from the approaching camera, the

fish census data are highly variable and should be viewed as more qualitative than quantitative in nature.

Despite this variability from transect to transect and year to year, the mean number of individual fish per

transect showed statistically significant changes over the fourteen-year period. Not unexpectedly, a related

parameter, the mean number of square meters examined to find an individual fish, also showed statistically

significant changes. The size of the area searched to find an individual fish is directly related only to the

number of fishes counted because the area searched on a given transect does not vary among the different

survey years. Thus, if the mean number of fishes censused per transect shows a statistically significant change,

the measure of the mean number of square meters examined to find an individual fish should show a similar

significant change, which it does. However, the application of another statistical test (Student–Newman–Keuls

test) showed neither a clear statistical separation among the mean number of individual fish per transect nor a

statistical separation for the area searched per fish over the thirteen sampling years. This lack of clear statistical

separation among the means for the different years, as well as the fact that the decrease in numbers of fishes

seen per transect does not follow any temporal trend, suggests that the changes are due to factors such as water

clarity or camera angle and resolution and not to any real change occurring in the diffuser fish communities.

The application of statistical procedures to the data derived using a video camera to census fish and

invertebrates is probably not appropriate because of a number of drawbacks inherent with the use of a remotely

operated video camera, including variability due to water clarity, camera angle relative to the substratum, and

camera resolution. Thus little significance should be attached to any change noted in this study of the fish or

macrobenthic communities residing on the Barbers Point diffuser because of the variable quality of the data

generated by use of the remotely controlled video system.