OUR LATEST NEWS
Whooshh Innovations Makes Available 2019 Fish Passage Data From Its FishL Recognition System At Bonneville Dam AFF
BONNEVILLE DAM, Oregon -- Whooshh Innovations releases a trove of data collected from its “FishL™ Recognition System” installed at the Bonneville dam AFF in 2019.
Whooshh Innovations “FishL™ Recognition” system, an autonomous, high definition, live fish imaging system, collected over 220,000 images from fish that bypassed through the Bonneville Dam Adult Fish Facility (AFF) between April 25th and October 18th of 2019.
Collecting high definition and infrared images without handling and delaying the fish in their upstream migration is a first on the main stem Columbia. Using highly sensitive and calibrated sensors to recognize entry and passage of fish through the system, 18 high definition images of each fish were captured along with logged date, time, fish length, girth, and fish speed, all in real time.
Twelve different species were imaged as they passed through the system, half of which were salmonids: chinook, coho, steelhead, sockeye, and pink salmon.
Other species included whitefish, American shad, large scale sucker, northern pikeminnow, peamouth chub, and smallmouth bass. Half of these non-salmonids are invasive to the Columbia River. The fish transferred through the Whooshh FishL™ Recognition System in approximately 1/3 of a second, traveling at a rate of about 5 m/s.
“The retained high quality FishL™ Recognition System images and the autonomously logged data, provide a wealth of valuable fisheries information and the retained data can be further analyzed and mined ” said CEO Vincent Bryan III.
“We want to make this data freely available to all of the biologists and scientists who have an interest in the recovery of the salmon in the Pacific Northwest. We believe better data, may well challenge current priorities, and will lead to the best fisheries management decisions for a changing environment ” he said.
examples of the type of data available.
Sealion caused injuries to fish were quite prevalent, and recorded in a range of injuries from minor scratches to major puncture wounds, to bites penetrating well below the epithelial surface, exposing the underlying flesh.
An estimated 15% of the Spring Chinook manifest pinniped injuries with the majority passing through in the last three weeks of May. 7% of Summer and 8% of Fall Chinook exhibited signs of pinniped injury. Pinniped injuries were not restricted to Chinook as pinniped injuries were also noted on sockeye, steelhead, and coho.
While counts of observed pinniped feeding has historically been the primary data resource, injured fish, not consumed, but which are then at greater risk of not reaching their spawning grounds, is a significant causal insight.
A large percentage of the total fish imaged were very small.
These were not smaller than usual adult salmonid as has been recently reported in the press, nor “jacks” returning after two years at sea, but rather 11% of the total fish imaged at the AFF were minijacks, juvenile salmonids <30 cm in forklength, which have reversed their outmigration and are heading back upriver having spent zero years in the ocean.
The fact that the number of mini-jacks were high, may have been an aberration and a one year event, or may be something which needs to be closely monitored each year to better understand future return numbers.
Forklength (length from nose to tail) was calculated on 1244 Sockeye.
The average fork length was 43.7 cm which is typical and consistent with historical size averages. However, when plotting the data collected on each fish of this large sample set, a different picture emerges.
Two almost equal populations, one much larger than 43.7 cm and one much smaller than 43.7 cm were plotted. The two populations may represent different age populations: four year old returning sockeye vs three year old returning sockeye or genetically different species.
This kind of macro insight can be very beneficial to fisheries managers, including when estimating the strength of the following years return run.