MEPS 178:39-54 (1999)  -  doi:10.3354/meps178039

ABSTRACT: Low dissolved oxygen concentrations can affect growth rates, distributions and predator-prey interactions of aquatic organisms. Each of these potential effects is generally examined separately in laboratory and field studies. As a result, it is often difficult to determine the net influence of low oxygen on survival and which individual effect of low oxygen contributes most to mortality. We used a spatially explicit individual-based predation model to predict how effects of low dissolved oxygen on vertical distributions, predation rates, and larval growth combine to influence survival of estuarine fish larvae in a water column where the subpycnocline (bottom) and pycnocline layers are subject to oxygen depletion. We analyzed simulations involving 3 predators (scyphomedusae based on Chrysaora quinquecirrha, and fish that were relatively sensitive to, or tolerant of, low dissolved oxygen), water columns that differed in the relative thickness of the subpycnocline layer, and bottom dissolved oxygen concentrations ranging from <1 mg l^-1 to no-effect concentrations. The effect of dissolved oxygen on larval and predator vertical distributions, predator capture success and larval growth rates in simulations were based on previous experiments and field sampling in the Patuxent River, a tributary of the Chesapeake Bay, USA. Simulations indicated that bottom dissolved oxygen can strongly affect predation mortality of fish larvae. Thus, there is the potential for eutrophication to have a large effect on larval fish survival, and possibly recruitment, even in the absence of direct effects of low oxygen-induced mortality of larvae or the effects of nutrient enrichment on the abundance of larval prey or predators. Depending on predator characteristics and water column depth, lowest larval survival occurred when oxygen concentrations were either <1 mg l^-1 or sufficiently high that oxygen concentration had no effect on distributions or capture rates; highest survival generally occurred with hypoxic bottom layers (1 to 2 mg l^-1 dissolved oxygen). Bottom dissolved oxygen concentration also strongly affected the relative importance of fish and sea nettle predation in simulations that included both types of predators. Differences among predator types had important consequences for the magnitude and location of predation. Bottom-layer oxygen depletion shifted the focus of trophic interactions into the pycnocline and surface layers. Additional simulations indicated that distributional and capture success effects on larval survival were more important than growth rate effects, and that the direction of effects depended on the predator type and dissolved oxygen concentrations. Limitations of the model as well as implications of results for efforts to reduce nutrient loadings into estuaries are discussed.

KEY WORDS: Behavioral responses · Chesapeake Bay · Fish predator · Food web · Gelatinous zooplankton · Gobiosoma bosc · Hypoxia · Individual-based model