Monitoring small fish populations in streams: A comparison of four passive methods

We analysed the relative efficiencies and size-selectivities of four different passive capture methods in a small coastal stream. We used plastic minnow traps (PM), metal minnow traps (MM) and two types of fyke nets differing in mesh size (F1, small meshed; F2, large meshed) to capture over 12,000 fish belonging to 11 species. Over 97% of captured fish were Andalusian toothcarp (Aphanius baeticus), Iberian loach (Cobitis paludica) and sand smelt (Atherina boyeri). F1 was the most efficient trap type in capturing the three most abundant species. Catches by PM and F2 differed in taxonomic composition, the former being characterised by toothcarp and loach dominance and the latter by the catch of eel (Anguilla anguilla) and grey mullets (Fam. Mugilidae). There were large differences in the size of fish captured in each trap type, with fish size following the pattern F2 > MM > F1 > PM. Small juveniles of the three dominant species were captured only in PM, thus enabling us to follow their seasonal size variation. However, PM traps were inefficient for sand smelt sampling and failed to catch large individuals of this species. This schooling and mainly pelagic species was more accurately monitored through the use of F1. Our results suggest that a combination of PM and F1 traps could improve the efficacy of small fish sampling in streams.