Whole-lake acoustic telemetry to evaluate survival of stocked juvenile fish

Abstract

Estimates of juvenile survival are critical for informing population dynamics and the ecology of fish, yet these demographic parameters are difficult to measure. Here, we demonstrate that advances in animal tracking technology provide opportunities to evaluate survival of juvenile tagged fish. We implemented a whole-lake telemetry array in conjunction with small acoustic tags (including tags < 1.0 g) to track the fate of stocked juvenile cisco (Coregonus artedi) as part of a native species restoration effort in the Finger Lakes region of New York, USA. We used time-to-event modeling to characterize the survival function of stocked fish, where we infer mortality as the cessation of tag detections. Survival estimates revealed distinct stages of juvenile cisco mortality including high immediate post-release mortality, followed by a period of elevated mortality during an acclimation period. By characterizing mortality over time, the whole-lake biotelemetry effort provided information useful for adapting stocking practices that may improve survival of stocked fish, and ultimately the success of the species reintroduction effort. The combination of acoustic technology and time-to-event modeling to inform fish survival may have wide applicability across waterbodies where receiver arrays can be deployed at scale and where basic assumptions about population closure can be satisfied.

Figure 1

Study location. Study location of a whole-lake acoustic telemetry study to track survival rates of stocked juvenile fish in Keuka Lake, New York, USA. The solid black circles show acoustic receiver placement for whole-lake level array coverage and the transparent gray circles show the total number of ‘detection events’ (discrete events of individual tag detections) from October 2019 to August 2021. The inset map (bottom right) indicates the Finger Lakes region of New York State, USA.

Figure 2

Acoustic-tagged juvenile fish. Juvenile cisco (Coregonus artedi) re-introduced from extirpation in Keuka Lake, New York, USA. (a) Juvenile cisco with a surgically implanted acoustic transmitter. (b) A small (0.6 g) acoustic telemetry transmitter used for age-0 fish in this study. Photographs by M. Chalupnicki, USGS.

Figure 3

Kaplan–Meier survival curves. Kaplan Meier curves for tagged cisco (Coregonus artedi) with estimated survival probability (solid lines) and 95% confidence interval (dashed lines) for each cohort of stocked fish: (a) Fall 2019, (b) Fall 2020, and (c) Summer 2020. Blue lines indicate the age-0 release cohort (Fall 2019, Fall 2020) and orange lines indicate the age-1 release cohort (Fall 2019, Summer 2020). The first dates on the x-axes are the stocking date for each cohort of released fish.

Figure 4

Predicted survival from Cox proportional hazards models. Estimated Cox proportional hazard model survivorship curves for acoustic-tagged juvenile cisco stocked into Keuka Lake, New York, USA. (a) Predicted survival and 95% confidence limits (shaded areas) by age-at-release from the top ranked model (Survival ~ Age; 42% model weight) based on Akaike’s Information Criterion as adjusted for small sample size (AICc). (b) Predicted survival from the second AICc-ranked Cox model (Survival ~ Year + Age; 17% model weight).