Portal de Periódicos da CAPES

Effects of Predatory Sunfish on the Density, Drift, and Refuge Use of Stream Salamander Larvae

1992; Wiley; Volume: 73; Issue: 4 Linguagem: Inglês

10.2307/1940687

1939-9170

Andrew Sih, Lee B. Kats, R. D. Moore,

Freshwater macroinvertebrate diversity and ecology

We addressed several controversial issues regarding the impact of predatory fish on prey in natural streams in Kentucky by experimentally adding predatory green sunfish, Lepomis cyanellus, to four stream pools that contained small—mouthed salamander larvae, Ambystoma barbouri, while leaving four other pools as fishless controls. We monitored the density, drift (in and out of pools), and habitat use of the larvae before and for 1 mo after fish addition (until larvae began to undergo metamorphosis). By quantifying larval density and drift, we were able to address two issues: (1) effect of prey exchange on predator impacts; and (2) relative contribution to total predator impact of predation per se vs. predator—induced emigration. Larval densities and behaviors were similar in experimental and control pools before fish addition. After fish addition: (1) density decreased more rapidly in fish pools than in control pools; (2) larvae in fish pools spent more time under rocks and less time in deep, central areas than did larvae in control pools; and (3) larval overnight drift rates (proportion of larvae in a pool, drifting out overnight) were higher in fish pools than in control pools. We calculated weekly average rates of change in larval density in fish and control pools and found that densities in control pools decreased very little except during a period with flooding; that is, flooding was a major cause of population reduction in fishless pools. In contrast, in fish pools, densities decreased steadily. Interestingly, the rate of decrease in larval density in fish pools was lower during the flood period than during other periods. This might be due to an increase in net drift into fish pools, or a decrease in fish predation during floods. Although net drift (drift in minus drift out) showed a tendency to be positive, it did not differ significantly from zero in any situation (fish or control pools, day or night). In particular, fish—induced increases in larval per capita emigration rates offset high immigration rates into fish pools. Because there was no significant net drift into fish pools, immigration could not swamp predator impacts. Because there was no significant net drift out of fish pools, the decrease in larval density in fish pools can be explained entirely by predation per se, rather than by predator—induced emigration. Our data suggested that, in the experiment year, °30—40% of all larvae in this population drifted through a fish pool at some time in their lives; this estimate is conservative because we were unable to get data on drift rates during floods. On average, only °6—8% of the larvae that drifted into a fish pool survived to drift out the downstream end. Thus fish predation should be a strong selective agent favoring behaviors that enhance larval survivorship and drift out of fish pools. Auxiliary experiments showed that: (1) larvae decrease their activity when they first drift into fish pools; this presumably decreases the conspicuousness of these larvae to visual predators; (2) the presence of fish is associated with a larval shift in diurnal activity towards greater activity at night rather than in the daytime; this is associated with drift out at night; and (3) larvae alter their refuge use and drift in response to caged fish, from which larvae probably only receive chemical cues. Overall, our study is one of the few to simultaneously monitor predator effects on prey population dynamics, dispersal, and refuge use. Information on each of these variables helped us to better understand predator effects on the other variables; we suggest that it should be generally useful to monitor a relatively full set of prey responses to predators.