We studied gillnet selectivity by using a direct method to compare fish-size distributions from gillnet catches (Nordic type, mesh size range 5–135 mm, knot-to-knot, factor between adjacent mesh sizes 1.25) with a reference distribution obtained using beach seines for fry and adult fish (net length 10, 40 and 50 m, mesh size 1, 6 and 10 mm, respectively) in five Czech reservoirs. Target species were roach Rutilus rutilus (L.), perch Perca fluviatilis L. and rudd Scardinius erythrophthalmus (L.). The most common pattern of difference in fish-size distributions between the two gears was that more 0+ and 1+ year old fish were caught by beach seines and slightly higher proportions of larger fish were observed in gillnet catches. For the three species, gillnets were unable to enmesh fish smaller than approximately 40 mm in standard length. We developed average species-specific corrections that increased proportions of 0+ and 1+ year old fish and decreased proportions of larger fish. However, the average correction gave corrected size distributions that were biased in about 20% of cases. We also applied Kurkilahti’s correction given in European Standard EN 14 757, but it had a negligible effect on reducing the bias in size distributions. Future research is needed to develop more accurate corrections for size distributions, otherwise interpreting results from gillnet sampling can be very misleading especially for those including data on 0+ and 1+ year old fish.
Read full abstract