link to AFSC home page

link to AFSC home page link to NMFS home page link to NOAA home page

Resource Assessment & Conservation Engineering (RACE) Division

AFSC Quarterly
Research Reports
Oct-Nov-Dec 2006
ABL Reports
FMA Reports
NMML Reports
RACE Reports
REFM Reports
Quarterly Index
Quarterly Home

Newport Laboratory: Fisheries Behavioral Ecology Program

Potential Biases in Fish Surveys Conducted With Underwater Vehicles: Evaluating the Role of Fish Behavior

Figure 5, see caption
Figure 6.  Basic responses of fishes to stimuli (i.e., light and sound) created by underwater survey vehicles shown as a function of stimulus intensity.  Four response types are shown: no response (A), increasing avoidance (B), increasing attraction (C), and attraction at low stimulus intensity followed by avoidance (D).  The table shows the likelihood of survey bias caused by responses occurring at two different distances (d) from the survey vehicle.  It is assumed that the observers or cameras on the vehicle have a range of view substantially shorter than the distances at which lights and sound can be detected by the fish.

Use of underwater vehicles including submersibles, ROVs and towed camera systems to assess the abundance and distribution of fishes has increased rapidly over the last several decades, particularly in deep water and in structurally complex seafloor habitats where surveys with traditional sampling gear are unsatisfactory. It is often assumed that visual survey data have less inherent bias than sampling with conventional extractive gear.

To evaluate the potential biases created by behavioral responses of fishes to underwater vehicles, Program members Allan Stoner and Cliff Ryer, along with Steve Parker (Oregon Department of Fish & Wildlife), Waldo Wakefield (NWFSC), and Peter Auster (University of Connecticut) compiled published information and personal observations on 46 demersal marine fish taxa.

Integration of the data showed that almost all of those fishes respond to underwater vehicles under certain circumstances. The responses are context specific, depending upon operational variables including vehicle type, speed, light and sound levels. Direct responses were common. Some fishes respond indirectly, by attraction to sediment disturbance and prey species gathered in artificial lights. Whether or not movements or changes in behavior affect survey bias is more difficult to assess.

A simple conceptual model (Fig. 6) was developed to evaluate relationships between stimulus intensity, distances from the vehicle where reactions occur, and survey bias. Largest bias is caused by attraction or avoidance that occurs outside the field of view provided by cameras or observers. While light level and vehicle speed have been explored experimentally in a few cases, much remains to be learned about how reactions and biases vary among species and age groups, among different vehicles, and under different operating conditions.

Given the poor understanding of survey bias, we recommend that surveys be conducted using methods that minimize variation in vehicle operation and that vehicle time is devoted specifically to manipulations of operating conditions to evaluate bias quantitatively. There is no good substitute for direct field observations on fish behavior, distribution and abundance; and survey design and accuracy can be improved through experimentation.

By Allan Stoner

<<< previous


            Home | FOIA | Privacy | | Accessibility      doc logo