SMALL MESH FISHERY BYCATCH REDUCTION IN THE SOUTHERN NEW ENGLAND/MID-ATLANTIC WINDOWPANE STOCK AREA Cornell University Cooperative Extension Marine Program Emerson Hasbrouck John Scotti, Scott Curatolo-Wagemann, Tara Froehlich, Kristin Gerbino Jacqueline Wilson, Joseph Costanzo, Chris Mazzeo, Dan Kuehn Cornell University Funded By Dept. of Natural Resources – Ithaca, NY The New England Fishery Dr. Patrick Sullivan Management Council Groundfish Research Superior Trawl Inc. – Narragansett, RI Program Through the Jonathan Knight Northeast Consortium
Goals and Objectives The goal was to test the hypothesis that windowpane flounder bycatch levels can be significantly reduced in the inshore scup fishery when a trawl net is outfitted with a LMBP without significantly reducing the catch of scup. Project objectives included: Determine if a LMBP effectively reduces the catch of windowpane flounder in the SNE/MA small mesh scup fishery with existing gear and fishing practices in order to help avoid or reduce the impact of triggering AMs Determine the statistical level of difference (if any) between the control and experimental nets for the targeted scup catch and the windowpane flounder catch Complete an applied experiment across a range of strata and conditions including: areas, depths, bottom type, and times reflective of the small mesh scup fishery Validate these results for fishery managers and fishermen
Project Summary • The F/V Sea Breeze Too and F/V Elizabeth Katherine were chartered to conduct at-sea research trips • The vessels used identical nets and made adjustments to uncover the panel to switch from the control design to the experimental design • The two vessels towed side by side through the study area while targeting scup and windowpane • Comparisons of the control and experimental net were based on paired differences in catch by species
Project Summary At-sea research was conducted in October 2015 when there was the highest likelihood for co-occurrences of scup and windowpane flounder. • 4 trips (7 days) were conducted in SNE using an ABBA protocol • 84 total tows were conducted (42 tows each for the control and experimental nets)
Project Summary CCE tested and evaluated a large mesh belly panel (LMBP) to reduce windowpane flounder bycatch in the inshore small mesh scup fishery in SNE/MA Exceeding the Annual Catch Limit (ACL) for windowpane triggers Accountability Measures (AMs) that restrict groundfish and other trawl fishing activities • The AMs are gear modification areas that require the use of approved selective trawl gear in defined areas to minimize the catch of flatfish • The LMBP was also evaluated for its ability to maintain the catch efficiency of the target species (scup)
Northern & Southern Windowpane Flounder AM Areas
Southern New England Project Area
Sketch of Large Mesh Belly Panel Configuration The LMBP was made from 80cm (~32”) knot -center to knot-center diamond mesh. The panel was 2 meshes deep and was sewn into the standard 12cm (5”) mesh of the 1st bottom belly resulting in an effective area for fish escapement of 3 full 32” meshes, or an opening in the net belly that was ~8’ deep from front to back. The LMBP attached ~1 foot (2.5 meshes/12cm) behind the footrope and extended widthwise across the belly of the net (from gore to gore) for 30 meshes.
Project Results
Project Results Windowpane Flounder (by Weight) Boxplot Distribution of Windowpane Flounder Catch Weight in the Control and Experimental Net 50 40 30 Pounds 20 10 0 Control Experimental
Project Results Windowpane Flounder (by Weight) Total Catch Weight of Windowpane Flounder (lbs) in the Experimental and Control Nets for All Trips Combined
Project Results – Windowpane Flounder (by Weight) The reduction in windowpane flounder catch in the experimental net was 48.04% compared to the control net ( 52.5 % based on median) T- test results showed a highly significant difference in the catch weight between the control and experimental nets (p-value <0.0001). The Wilcoxon test yielded similar results (p-value<0.0001)
Project Results Windowpane Flounder (by Count) Boxplot Distribution of Windowpane Flounder Catch By Number of Fish in the Control and Experimental Net 150 100 Count 50 0 Control Experimental
Project Results Windowpane Flounder (by Count) Total Catch in Numbers of Windowpane Flounder in the Experimental and Control Nets for All Trips Combined 2500 2000 1500 1000 500 0 Control Experimental
Project Results Windowpane Flounder (by Count) The reduction in windowpane flounder catch in numbers of fish in the experimental net was 46.69% compared to the control net. T- test results showed a highly significant difference in the catch numbers between the control and experimental net (p-value = 0.00015) The Wilcoxon test yielded similar results (p-value <0.0001).
Project Results - Scup(by Weight) Boxplot Distribution of Scup Catch Weight in the Control and Experimental Net Scup Pounds 800 600 Pounds 400 200 0 Control Experimental
Project Results - Scup(by Weight) Total Catch Weight of Scup (lbs) in the Experimental and Control Nets for All Trips Combined 10000 9000 8000 7000 6000 Pounds 5000 4000 3000 2000 1000 0 Control Experimental
Results – Scup (by Weight) The overall reduction in scup catch weight in the experimental net due to the large mesh belly panel treatment was 26.14% compared to the control net. T- test results showed a marginally significant difference in the catch weight between the control and experimental net (p-value = 0.01341). The Wilcoxon test returned a similar result (p=0.001521). The significant difference in scup catch weight was attributed to the reduction in sub-legal sized fish.
Project Results - Scup(by Count) Boxplot Distribution of Scup Catch By Number in the Control and Experimental Nets Scup Count 2500 2000 1500 Count 1000 500 0 Control Experimental
Project Results - Scup(by Count) Total Catch in Numbers of Scup (lbs) in the Experimental and Control Nets for All Trips Combined 40000 35000 30000 25000 20000 15000 10000 5000 0 Control Experimental
Results (Scup by Count) The overall reduction in scup catch in numbers of fish in the experimental net due to the large mesh belly panel treatment was 47.60% compared to the control net. T- test results showed a highly significant difference in the catch numbers between the control and experimental net (p-value <0.0001). The Wilcoxon test yielded similar results (p-value <0.0001). The experimental net significantly reduced the catch of scup in numbers of fish compared to the control net. However, this significant difference can be attributed to the reduction in sub-legal sized fish.
Windowpane Flounder Length Frequency Distributions 300 Control 250 Number at Length 200 150 100 50 0 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839 Length (cm) 300 Experimental 250 Number at Length 200 150 100 50 0 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 Length (cm)
Length Frequency Distribution Results - Windowpane The peak of the windowpane flounder length distribution occurs in both nets at 27 cm. There was a 49.3% reduction in fish of this size as a result of the large mesh belly panel. A large reduction occurs at nearly all of the size intervals. The mean length of windowpane flounder in the control net was only 0.12 cm larger than the mean length in the experimental net. The large mesh belly panel is reducing the catch of all sizes of windowpane flounder.
Length Frequency Distribution - Scup The scup lengths that were collected were fork lengths. The commercial legal minimum size for scup is 9” (22.9 cm) measured in total length. We used a standard ASMFC conversion formula to determine that 22.9 cm (9”) total length equates to a 20.2 cm fork length.
Scup Length Frequency (Fork Length) Distributions Control 8000 7000 Number at Length 6000 5000 Minimum legal 4000 size 3000 2000 1000 0 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839 Length (cm) Experimental 8000 7000 Number at Length 6000 5000 Minimum legal 4000 size 3000 2000 1000 0 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839 Length (cm)
Length Frequency Results - Scup Peaks in the length frequency distribution occur in both nets at 9, 16 and 26 cm. The highest peak is at 9 cm in the control net. There was a 75% reduction in fish of this size in the experimental net. The experimental net reduced the catch of undersized, sublegal scup (20.2cm fork length) by 68%. The mean length of scup in the experimental net was 1.75 cm larger than the mean length in the control net indicating that the large mesh belly panel is allowing for escapement of smaller fish while retaining larger fish. The escapement of small scup may be attributed to scup stratifying by size within a school, with small fish at the bottom and larger fish at the top. Also smaller scup likely tire quicker than larger scup and may fall back into the net sooner while also being close to the bottom of the net. The reduction in undersized, sublegal scup observed in the experimental net is beneficial to the industry and to the scup resource.
Recommend
More recommend
