Site icon Conservation news

Film that fish: Stereo-video speeds surveys of marine fish communities

  • Researchers use underwater visual surveys to assess the sizes of fish in marine communities and their associated habitats, but diver-based data collection is time-consuming and requires expertise, and results may vary among different data collectors.
  • A multinational research team recently published the first guide to help researchers using diver-operated stereo-video methods (stereo-DOVs) to standardize surveys of fish assemblages (species and their abundances) and their associated habitat.
  • The video provides a permanent, shareable record of each survey transect, including the species and numbers of fish seen, while the stereo option allows researchers to measure fish using overlapping images.
  • The guide provides information on appropriate equipment; designing a stereo‐DOV if needed; operating it during underwater studies; processing the video data after collection; and analyzing fish behavior, population features and habitat in the resulting video.

Marine biologists survey fish assemblages and their associated habitat to understand the ecosystem of a place, compare fish communities over time or in response to changes in management, and examine fish behavior.

These researchers typically survey fish communities through underwater visual censuses, in which a diver identifies and counts fishes within an area, usually determined by a predefined route, or transect. The method is a straightforward, non-destructive way to survey fish.

Yellow wrasse in the Florida Keys. While not fearful of divers, smaller fish can remain hidden behind sponges and corals during a survey. Image by Sue Palminteri/Mongabay.

However, studies have shown variability among different underwater data collectors and inaccuracies in estimating the length and numbers of fish in the sample area. Swimming a predefined route while simultaneously identifying, counting and estimating the size of fish is difficult and requires extensive training and experience. Moreover, researchers in different places conduct data collection transects in different ways.

All this variation among surveys has made it difficult to either synthesize or compare findings from various sites and time periods.

A multinational research team has recently examined how some of these limitations can be overcome or reduced by updating underwater surveys through the incorporation of diver‐operated stereo‐video, or stereo‐DOV. Their new scientific publication explains this method and provides a guide for researchers wanting to add the technology to their fish community transect surveys.

Surveying with video instead of pencil

The authors, from Australia, the U.S. and New Caledonia, explain that the size and cost of video cameras have decreased in recent years, allowing researchers to put together a rig that a single diver can maneuver as they follow a predetermined underwater route, or transect.

A scuba diver conducting a stereo-DOV survey at Mo’orea, French Polynesia. Image courtesy of Lauric Thiault.

Swimming along a transect with the stereo-video rig is the most common application of the technology, but as video quality has also improved over this period, researchers can better detect, measure and identify fish, as well as record and better determine fish behavior.

The authors include in their guide information on appropriate equipment; designing a stereo‐DOV if needed; operating it during underwater studies (keep that camera pointing forward!); processing the video data after collection; and analyzing fish behavior, population features and habitat in the resulting video.

No clear guidelines currently exist for surveying with stereo-video, so the authors’ aim in providing a standardized set of procedures for stereo-DOV surveys is to encourage survey and behavioral data collectors to follow the same process. Reducing variation among locations and data collectors allows researchers to more easily compare findings and synthesize data from different locations and time periods to help answer broad‐scale ecological questions.

The method works through stereophotogrammetry, which estimates the three-dimensional coordinates of points on an object (e.g. a fish) using measurements made in two or more images taken from different positions. The system uses two cameras attached to a bar, set 800 millimeters (31 inches) apart and turned inward just slightly so their visual fields overlap. Like our own visual system, this slight spatial difference allows stereo “vision.” The authors describe several software applications designed to help researchers identify common points on the object in each image, overlap the images based on these points, and measure the size of the fish or other object.

The authors recommend using a video camera with a full high-definition resolution of at least 1920 by 1080 pixels that can record at speeds of at least 60 frames per second. The camera must also have the option to disable auto focus and video stabilization, which can invalidate stereo-video measurements. While the authors state that most divers can easily learn to use a stereo-DOV, they must prepare and calibrate the system before each dive.

Filmed surveys allow repeat views

Recording an underwater survey with high-definition video enables divers to complete a transect more quickly than when they have to detect each individual or school of fish and take notes on the species, and their estimates of the numbers and size of the many fish they may spot.

While you focus on counting the jacks above, don’t overlook the shark below. Image courtesy of Mr. Cha, South Korea.

Recorded video shows species, abundance, diversity and size of fish along both sides of a transect, even with the lens pointed straight ahead. The stereo-video further permits measurement of fish and other objects. Comparing numbers and size of fish over time can indicate impact of management activities, such as a marine protected area or quota on the harvesting of algae-eating species.

The ever higher-definition video improves analysis of fish behavior, which can advance research, as well as show any fear fish have of humans in the water. Fish are usually relaxed, and fear suggests that spearfishing or other destructive practices are occurring in the area.

The footage provides a permanent searchable, shareable record of the site, including not only the fish species and habitat features but also the context, such as coral condition and water clarity. The video can be checked again at any time and allows multiple researchers to view and comment on the findings.

Filmed surveys require time and precise equipment

The stereo-DOV system must be stable enough to conduct stereo photogrammetry underwater. The authors recommend purchasing a commercial system from a recognized provider, though they offer detailed instructions on building one’s own system.

Stereo-DOV surveys require some basic equipment, including the stereo-DOV system (a) with a stabilizing arm (b), two video cameras, memory cards and reader (c, e), a cotton thread distance counter (d), marking tags (f), batteries and charger (h,i), a dive float (j), and spares of everything. Image © Goetze et al., 2019.

Adding stereo-DOV to surveys still requires one or more divers going to each site and spending time in the water, which can be relatively costly and time-consuming, depending on the location and availability of trained divers.

The diver’s immediate presence may frighten certain shy fish, but that’s true for visual censuses as well. Similarly, both standard and video-based surveys have trouble detecting fish that are cryptic or hide under corals during the day.

Environmental DNA (eDNA) has become increasingly useful for determining the presence and community composition of aquatic species, including corals. Although collecting species DNA from a water sample is quick, cheaper and easier than from a diver-based survey, even one collecting video data, it cannot measure everything. Fish in a given area don’t necessarily poop or shed scales or mucus at the same rate, for instance, and few studies have shown reliable links between fish abundance or biomass and presence of eDNA.

Could you count all these barracuda while sticking to the transect route? Recorded video of the school would enable repeat views and the chance to estimate numbers and sizes of the fish. Image courtesy of Mr. Cha, South Korea.

In their paper, the authors encourage divers surveying fish assemblages using stereo‐DOVs to adopt the guidelines, given the various advantages they describe. “An increased uptake of this methodology, following the standard procedures described herein,” they write, “will reduce variation in methodology, assist in the synthesis of data on continental and global scales and provide accurate information to improve fisheries management and conservation.


Goetze, J. S., Bond, T., Mclean, D. L., Saunders, B. J., Langlois, T. J., Lindfield, S., . . . Harvey, E. S. (2019). A field and video analysis guide for diver operated stereo‐video. Methods in Ecology and Evolution. doi:10.1111/2041-210x.13189

FEEDBACK: Use this form to send a message to the editor of this post. If you want to post a public comment, you can do that at the bottom of the page.

Exit mobile version