Computation of Self-recruitment in Fish Larvae using Forward- and Backward-in-Time Particle Tracking in a Lagrangian Model (SWIM-v2.0) of the Simulated Circulation of Lake Erie (AEM3D-v1.1.2)

Abstract. Accurately estimating self-recruitment (SR), the fraction of recruits in a location that originated locally, is critical for understanding population connectivity. Biophysical models have been typically applied to compute SR by releasing a certain number of larval particles from each assumed source location and tracking them forward in time. However, various strategies have been employed for releasing these larval particles: including randomly, consistently, or a number proportional to the location’s area or larval production, which causes ambiguous results. We demonstrate, using theoretical arguments and numerical simulations from Lake Whitefish (Coregonus clupeaformis) larvae in Lake Erie, that SR depends on larval production at each source location. This dependency suggests that SR may not be computed unambiguously in these models unless realistic larval production is released from all potential source locations. In contrast, parentage analysis studies typically computed SR by assessing the fraction of sampled juveniles that originate locally at a settlement location, instead of identifying larval production at all sources. Therefore, tracking larval particles backward from the settlement location is proposed as a straightforward approach for computing SR. Our findings demonstrate that SR is independent of the number of larval recruits at the settlement location, supporting the employment of backtracking models with randomly released larval particles. In this way, considerable effort and resources, that would otherwise be spent on identifying all potential sources and their larval output, in forward tracking can be saved. We believe this result will have important implications for studies on larval dispersal and recruitment in aquatic systems.