Calculation of distance traveled by fishing vessels using GPS positional data: A theoretical evaluation of the sources of error

Using theoretical cruise paths of known distances, the errors associated with the use of polled GPS data to calculate the distance traveled by fishing vessels were explored. Specific attention was given to how errors vary in response to changes to the polling frequency, cruise path complexity, vessel speed, and GPS error and precision. Simulation results show that distance estimates derived from high-frequency polled GPS data are susceptible to both mapping (GPS precision) and measurement errors (GPS error), while estimates derived from low-frequency data are susceptible to interpolation errors. At high polling frequencies (>1 fix/40 s), GPS error and limitations of GPS precision can lead to overestimation of distance traveled exceeding 10% of the true distance with the effects inversely proportional to vessel speed. At low-polling frequencies (<1 fix/200 s) straight-line interpolation of sinuous cruise paths can cause distance underestimations >35%. The amount of underestimation is inversely related to the polling frequency, and increases with path sinuosity and vessel speed. Comparison of simulated path results to the distance estimates of actual fishing trips suggests that the conclusions are relevant to real-world applications. At speeds, spatial scales and cruise track complexities typical of most fishing operations, use of medium-frequency GPS polling (1 fix/40 s to 1 fix/200 s) will mitigate the factors affecting calculation of distance traveled; however, users should carefully consider their specific application prior to choosing a polling frequency.