Sound pressure level weighting of the center of activity method to approximate sequential fish positions from acoustic telemetry

Proximity of acoustically tagged fish to a hydrophone correlates with detection probability, thus allowing fish center of activity (CA) estimation as weighted averages of hydrophones' positions over fixed listening intervals. Alternately, weighting by the detection sound pressure levels (SPL) would require only a single detection from each hydrophone. We tested SPL-weighted averaging performance relative to an independent and highly accurate positioning method, trilateration, for tagged fish. Positions calculated using SPL were similar to those using CA in the shape and sequence of the paths relative to the trilateration standard. Neither up-nor down-weighting transforms of SPL significantly affected solution error (209 m average, range 192-215 m) and did not differ significantly in error or shape from CA (194 m) over seven fish (1.9 million detections) even at the shortest of five tested averaging intervals (150, 300, 600, 1200, 2400 s). The method increased the stability of position estimates when a single moving receiver was used to create a synthetic array. Potentially, useful solutions can be calculated for tags outside array perimeters by extrapolating regressions to known source SPL.