Population level consequences of habitat-mediated density-dependent mortality: Finfish response to oyster population collapse

Gabrielle D. Love, Zachary Siders, Shirley M. Baker and Edward V. Camp, Fisheries and Aquatic Sciences, University of Florida, Gainesville, FL, David A. Gandy, Apalachicola Bay Laboratory, Florida Fish and Wildlife Research Institute, Eastpoint, FL

Background/Question/Methods
Habitat structural complexity is well-recognized for mediating population dynamics of associated fauna via multiple mechanisms, particularly juvenile density-dependent mortality. This infers that loss of key structural habitats, such as oyster reefs, may translate to diminished juvenile survival and recruitment to adult populations, population instability, and cascading trophic effects throughout the ecosystem. Many of these principles have been observed through small- or medium-scale experimentation, but there are rarely opportunities to empirically test the consequences of habitat shifts on fish populations at the ecosystem scale. Here we empirically assess the system-scale consequences that major alterations to structured habitat may have for finfish populations using the 2012 collapse of the oyster population in Apalachicola Bay, FL as a natural experiment. We evaluated if the sudden, large-scale decline of oyster reefs created cascading effects for abundance trends of 12 reef-associated taxa between 1998-2018. We used delta-log normal generalized linear models to standardize fisheries-independent monitoring data and created species-specific indices of relative abundance for (i) individuals of all life stages and (ii) recruits only. We statistically evaluated each time series (n=24) for evidence of shifts in abundance and variance through time using a change-point analysis (“changepoint” package, program R).
Results/Conclusions
Examination of all standardized abundance time series revealed no consistent pattern or response to the habitat change across taxa, though some showed marked periods of instability following 2012. Change-points occurring in 2012 or later were detected in 11 of the 24 total time series. Recruit time series account for 6 of those with change-points in or after 2012. Direction and magnitude of abundance mean and variance changes associated with these change-points were inconsistent. Responses to habitat loss were non-uniform and non-linear, varying by species and age. Obligate reef resident species and transients alike showed significant abundance trend changes after 2012, though fewer transient generalists were impacted. Although the loss of oyster reefs habitat demonstrated some cascading effects on the recruitment of species for which reefs provide refuge and food, including some higher trophic level species, no species exhibited clear and consistent declines in recruitment, nor did they translate to declines in overall population abundance. This suggests that oyster reef habitat may mediate but not fully drive density-dependent mortality, even for resident taxa.