Comparison of two indexed gill-netting protocols for fish community surveys in Northern lakes

1) I compared the performance of two standard fish community assessment protocols, the NORDIC protocol and the Broad-scale Fish Community Monitoring (BsM) protocol (the latter consisting of two gear types). I utilized fish catch and attribute data collected from 21 Boreal Shield lakes (17 in Ontario, 4 in NWT) surveyed using both protocols, in a pairwise design. Fish community composition (species richness, diversity, and evenness), relative abundance (number and biomass per 100 m of net), and body size distributions were compared between NORDIC and BsM surveys, and among the three gill net gears - NORDIC (NRD), North American Standard (NA1), and Ontario Small Mesh Standard (ON2). The NORDIC protocol dedicates a higher proportion of total sampling effort to small-mesh gear compared to the BsM protocol, and the ranking of gears according to proportion of small mesh effort is ON2 > NRD > NA1. NORDIC surveys used 38% greater effort (total length of net deployed per area of lake) than BsM surveys over the 21 lakes examined. 2) Principal components analysis (PCA) of species relative abundances showed significant differences between surveys in community compositions, as well as a separation of communities between Ontario and NWT lakes. NORDIC surveys detected 19% more species per survey, with the additional species primarily belonging to smallbodied taxa (e.g., Cyprinidae, Gasterosteidae, Cottidae). Paired-comparisons of gears indicated that NA1 gangs (highest proportion of large mesh) yielded fewer species, lower diversity, and higher evenness at standardized levels of effort compared to NRD and ON2 gangs, but there were no significant differences between NRD and ON2 gangs. 3) NORDIC surveys tended to provide higher numeric catch per unit effort (NPUE) and lower biomass per unit effort (BPUE) estimates compared to BsM surveys for the whole community, but differences between surveys were stronger and more consistent for smallbodied species than large-bodied species. For most small-bodied species, both NPUE and BPUE were significantly higher in NORDIC than BsM surveys. In gear comparisons, differences generally followed mesh size compositions for both NPUE and BPUE; NA1 gear tended to provide higher estimates for large-bodied species, and lower estimates for small-bodied species compared to NRD and ON2 gears. NRD and ON2 gear provided comparable NPUE and BPUE estimates for small-bodied species, but NRD gear tended to provide higher estimates for large-bodied species. 4) Biomass size distributions of all captured fish differed significantly between surveys in most Ontario lakes, but not in most NWT lakes. Significant differences between the surveys were more consistent across lakes for large-bodied piscivores than for other taxa. Size distributions from NORDIC surveys generally had lower medians and higher CVs than distributions from BsM surveys. Both NRD and ON2 gears yielded size distributions that tended to be more multi-modal than distributions from NA1 gear. In gear comparisons, size distribution medians were NA1 > NRD > ON2, whereas size distribution CVs were NRD=ON2 > NA1. 5) Differences in fish community metrics between the surveys were not related to the physical characteristics of the survey lakes (area, depth, water clarity), with the exception that BPUE differences were weakly but significantly related to lake maximum depth. Differences between surveys appeared to be less distinct in NWT lakes than in Ontario lakes, presumably due to differences in fish community composition between regions. 6) Overall, BsM surveys tended to under-represent small-bodied fish and over-represent large-bodied fish relative to NORDIC surveys. Differences between survey results could likely be reduced by increasing the total sampling effort, and/or the relative amount of ON2 effort in BsM surveys.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.