Evaluation of a New Coded Electromyogram Transmitter for Studying Swimming Behavior and Energetics in Fish

Abstract

A new coded electromyogram (CEMG) transmitter was recently introduced to the market to allow broader application and greater flexibility of configurations than the conventional noncoded version. CEMG transmitters were implanted into 20 steelhead Oncorhynchus mykiss and calibrated to swimming speed in a respirometer to determine the relationship between swimming speed and the output of the transmitters and also to determine how the output of a single transmitter varied when implanted in multiple fish. Linear regression models showed a strong positive relationship between the output from CEMG transmitters and swimming speed. However, grouping signals from multiple transmitters produced a less accurate relationship between CEMG output and swimming speed than using signals from individual transmitters. The results, therefore, do not suggest that the CEMG transmitters acted similarly in all fish. Calibration data from one transmitter were not readily transferable among multiple fish implanted with the same transmitter, suggesting that the same transmitter implanted in multiple fish also performed dissimilarly. These results indicate that experimental designs that require more precise estimates of muscular activity should use individual fish–CEMG transmitter calibrations. Variation in fish length, fish weight, location of transmitter implantation (distance from snout), and distance between the electrode tips did not account for the variation in models. The smaller size of the new CEMG transmitters will allow them to be used in a larger range of fish species and sizes. The fact that the transmitter has a coded transmission reduces the likelihood of interference from outside signals and allows multiple fish to be continuously logged on a single receiver. This could lead to reduced project costs because fewer receivers may be needed. However, one downfall of this new transmitter is that it has a smaller range of output, which may lead to lower accuracy in estimating swimming speeds.