Award: OCE-1820540

Predator-prey relationships are an important part of all ecological communities. In recent decades ecologists have discovered that predators can affect prey populations in two distinct ways. The first is by directly killing and consuming them, which is termed a consumptive effect. The second is when the prey adjusts their behavior or some physical trait to avoid being killed (such as spending less time feeding and more time being vigilant against predators, or growing a thicker shell), and those responses come at some cost to the prey, usually in the form of lost energy and slower growth. This is termed a non-consumptive effect. Knowing that predators can produce both consumptive and non-consumptive effects, the next important question is what is the relative importance of those two effects on the prey population? Specifically, when developing mathematical models to describe and predict fluctuations in a prey population, is it important to account for non-consumptive effects, or are consumptive effects the main driver of prey dynamics? This can be a difficult question to answer because consumptive effects can happen immediately (death) while the consequence of non-consumptive effects such as slow prey growth could take much longer to emerge. In this project we conducted a long-term field experiment to measure consumptive and non-consumptive effects on Eastern oysters in a Florida estuary. Oysters are an iconic species on the U.S. Atlantic and Gulf coasts; they create oyster reef habitat for many fish and invertebrate species, stabilize shorelines, improve water quality, and support fisheries. In our study location, oysters are preyed upon by several crab species as juveniles, and then by a conch when they are older are larger. We used data from the field experiment to inform a mathematical model of oyster populations. Using a model allowed us to combine information on mortality (the consumptive effect) and slower growth (the nonconsumptive effect; oysters clam up when around predators, reducing their feeding and slowing growth) and determine their effects on the expected long-term size of the population. Further, we could determine what the independent contributions of the two effects are, which is something that is difficult to measure in the field. Our key finding was that the effects of predator consumption on the long-term abundance and biomass of oysters is at least ten times greater than the non-consumptive effect of slower growth in the presence of predators. In fact, when we accounted for natural variability in growth and mortality rates, there was no meaningful difference between model predictions that included the non-consumptive effects and model predications that did not include those effects. That means that for these oyster reefs, there is actually very little long-term consequence of the non-consumptive effects of predators. We propose that the consequences of non-consumptive effects be examined using a similar approach in other species, because the methods used to detect non-consumptive effects in some short-term experiments may exaggerate their effects. A second important outcome of our work was that we were able to compare the spatial patterns of oyster demography (growth rates, mortality rates) within our study estuary (the Guana Tolomato Matanzas [GTM] estuary near St. Augustine, FL) to similar data from another estuary in the Gulf of Mexico (Apalachicola Bay, FL). The pattern in Apalachicola is the typical one, with oyster populations being most productive (lower mortality, faster growth) in regions of the mid to upper estuary where salinities are moderate, and less productive towards the mouth of the estuary, where salinity is higher and predators are more abundant. However there was not a gradient like that in the GTM, and areas of high oyster productivity were more difficult to predict. This makes it clear that the typical pattern is not universal, and careful sampling and modeling should precede restoration efforts to ensure it is targeted at locations with high potential productivity. We developed a modeling tool to help in that decisionmaking, in collaboration with the local National Estuarine Research Reserve (NERR). The broader impacts of this work include benefits to habitat management and conservation, and training early career scientists. This project contributed to the collection of a nearly eight-year time series on oyster populations in the GTM estuary, which is used by the GTM NERR to manage the long-term sustainability of that habitat. We also provided the GTM NERR with data analysis and modeling tools to use in restoration planning. We published our work in peer-reviewed journals and all of our data and model code is publicly available. During this project we trained a post-doctoral scholar at Oregon State. She is an applied mathematician who received training in ecology, ecological modeling, and scientific computing. Following our mentorship she took a position as faculty at Bates College. Last Modified: 07/09/2024 Submitted by: JamesWWhite