Editor’s note: Nick Calabrese grew up fishing for cod on the North Shore and witnessed a serious decline. Now a fisheries researcher, his work may herald good news for the Cape’s namesake.
Here is his story:
I am a research associate and PhD candidate at the School for Marine Science and Technology. SMAST is a graduate school at UMASS Dartmouth, located in New Bedford. I work in Dr. Kevin Stokesbury’s Marine Fisheries Field Research Group.
We collaborate with the fishing industry as well as state and federal officials on research projects including lobster surveys in wind farm lease areas, underwater camera surveys of scallop populations, maturity studies of whelks, and my project: a video trawl survey of Gulf of Maine cod.
We just completed our third video trawl survey in the western Gulf of Maine this May. For three years we’ve tracked a large year class of cod from 2019 that has continued to show healthy growth without major mortality.
The 2019 young of the year were observed in multiple surveys (they slip through our net’s bigger mesh). Again in 2020 we saw areas with high densities of fish that were approximately the size we would expect if they had been young in 2019.
We then saw these fish again in 2021 and 2022, growing as you would hope. Our analysis for 2022 shows a slight decrease in biomass, with some new signs of recruitment and a few larger fish.
What the future holds we don’t know yet, but it is very exciting to have some good news for a stock that so desperately needs it.
The Video Trawl Survey Explained
For years the fishing industry has observed high catch rates of cod in the Gulf of Maine that have not been reflected in the Northeast Fisheries Science Center survey indices. The opportunity to use our technology to survey the Gulf of Maine cod stock arose in 2016, and we began small scale surveys during the winter.
We had the privilege of working closely with commercial fishermen to design our video trawl survey. The survey uses a commercial trawl net fished from the F/V Justice in New Bedford.
Using video cameras to count fish means the net can be left open, and fish can pass through unharmed while we record them. This allows us to tow for hours, covering more ground than a traditional trawl survey and increasing the sample size dramatically. We hoped this would enable us to collect much more information, without killing more fish.
A circular plastic tub with bottom removed is sewn inside the codend of the net. Three cameras and two lights are angled toward the back of the net, and a winch holds our main wire which brings power to the lights and cameras as well as providing a live video feed to the wheelhouse.
A GoPro camera is used for high resolution wide-angle video to count all the fish passing through and a stereoscopic camera is used to measure a subsample.
Once the video is collected, it is uploaded to a website, where technicians count and measure the cod.
Video Survey Results & Future Plans
Starting in 2016 we began semi-annual surveys of Stellwagen Bank and Jeffrey’s Ledge in winter and spring, with the goal of proving our technology can survey cod and produce biomass estimates. During these surveys on Stellwagen we saw how densely populated the bank can get with baitfish.
In 2016 and 2017 we saw large numbers of cod aggregated on Stellwagen, but our catches declined rapidly after that. Surveying such a small area caused our catches to fluctuate as a result of fish movement rather than the true status of the stock. However, these surveys were critical to proving the effectiveness of our methods.
In 2018, my Master’s thesis showed that cod counts in the video matched the catch of traditional closed tows, which provided proof that our counts from video were accurate.
We then expanded our survey to a much larger portion of the Western Gulf of Maine in 2020. This area represented an average of 69 and 74 percent of commercial landings and Northeast Fisheries Science Center survey catch over the past 10 years. It was the maximum coverage we could get with our current resources.
Our next goal is to get the data used in the stock assessment of Gulf of Maine cod, but first there are a few obstacles to overcome. These obstacles are the basis for my PhD work. Here’s a quick background on the traditional assessment of Gulf of Maine cod we are working to improve.
How The Cod Stock Assessment Is Developed
A stock assessment uses all available data to describe the stock. These data sources include recreational catch, vessel and dealer reports for commercial catch, observer data for discards, and surveys of biological data.
The fisheries independent survey used in all the groundfish assessments in the Northeast is the NEFSC bottom trawl survey, where random stations are surveyed every spring and fall. This survey creates what is called an index of abundance, where an indicator of stock size (catch per tow) is tracked through time.
To produce an index of abundance from the video trawl we would need to have a long series of consistent surveys. Since we don’t have many years of data, we produce absolute estimates of abundance. Using sensors in the net, our speed, and the duration of a tow, we know the exact surface area we cover. From the area covered, and the cod caught, we estimate a density of cod per square kilometer.
By extrapolating this density to the survey area, we can estimate the biomass of cod in that survey area. The problem with this method is that our estimates assume that all fish that were on the seafloor in front of our net were captured. We know this is not true, thus our estimates become extremely conservative.
For my PhD dissertation I have been working on a way to estimate the percentage of cod in the path of our net that are captured. This is called the efficiency of the net, and it can then be used to make more accurate estimates of cod population.
There are many ways in which a cod can escape the net before being captured. We used a mark-recapture experiment to estimate overall efficiency. In a mark-recapture experiment fish are tagged and released. The tags function like the chip in your credit card, and are read by a scanner built into the codend of the net.
Population size can then be calculated from the rate of recaptures. By comparing this estimate of population size to the estimate we calculated previously, we can then estimate our efficiency. We tested this system in January and were able to tag more than 1000 cod with eight recaptures. The data is still being analyzed but we hope to have an estimate of efficiency soon.
We have seen some promising results in recent years and with our improvements to the survey it is our hope that the data will be used in future assessments.
Hopefully someday soon we can be targeting Gulf of Maine cod once again.