Numerical responses of aphid predators to varying prey density among Utah alfalfa fields

Abstract

Insect natural enemies are often hypothesized to exhibit spatial as well as temporal density dependence in responding to their prey. To evaluate the potential im portance of spatial density dependence in biological control of pea aphids (Acyrthosiphon pisum) in Utah alfalfa, we examined whether the distributions of aphid predators varied in relation to densities of their prey among fields. Sweep samples were taken for predators and prey during five periods over the growing season in fields scattered throughout a five km2 area. Densities of generalist predatory insects, damselbugs (Nabidae) and big-eyed bugs (Lygaeidae: Geocorinae), were not correlated with prey density on any occasion. In contrast, densities of the more narrowly aphidophagous ladybeetles (Coccinellidae) were strongly positively correlated with aphid densities among fields early (but not late) during the growing season. Alfalfa weevils (Hypera postica) were numerous at the study sites, and served as alternate prey for aphid predators. With the exception of damselbugs in early May, however, densities of ladybeetles, big-eyed bugs, and damselbugs were not correlated with alfalfa weevil densities among fields throughout the growing season. Differences in diet breadth best explain the contrast between the positive numerical response of specialist ladybeetles to spatial variation in aphid density, and the absence of such responses by generalist insect predators. Abundant evidence suggests that natural enemies often play major roles in the population dynamics of phytophagous insects (e.g., Strong et al., 1984; Luck et al., 1988). Of key interest are the degree to which these enemies (1) depress (determine) host population density, and (2) stabilize (regulate) long-term fluc tuations of host density (Hassell and Waage, 1984). Successful regulation of hosts at low densities has been conventionally the ultimate goal of biological control (Murdoch et al., 1985). Such regulation, furthermore, has generally been hypoth esized to depend on direct density dependent responses of natural enemies to their hosts, such that the intensity of natural enemy attack (e.g., percent parasitism) rises with increasing host density (Huffaker et al., 1976; Batra, 1982; Stiling, 1987). In addition to temporal density dependence, spatial density dependence has been widely considered to contribute instrumentally to successful biological con trol (see discussions in Readshaw, 1973; Hassell and Waage, 1984; Hassell, 1985; Walde and Murdoch, 1988). The potential importance of direct spatial density dependence in predator/prey interactions is highlighted by a diversity of math ematical models (e.g., Hassell and May, 1973, 1974; Beddington et al., 1978; Hassell, 1985,1987;KareivaandOdell, 1987; Walde and Murdoch, 1988). Recent summary reviews, however, indicate that such direct spatial density dependence occurs relatively infrequently (and about as frequently as inverse density depen dence) in host/parasitoid interactions in the field (Morrison and Strong, 1980; Lessells, 1985; Stiling, 1987; Walde and Murdoch, 1988). These findings call into Accepted for publication 10 September 1991. This content downloaded from 207.46.13.51 on Sun, 19 Jun 2016 06:18:54 UTC All use subject to http://about.jstor.org/terms VOLUME 65, NUMBER 1 31 question the general importance of spatial density dependence in insect biological control (Smith and Maelzer, 1986; see also Pacala et al., 1990). Spatial patterns in insect predator/prey (vs. parasitoid/host) interactions have been relatively little studied (but see Readshaw, 1973; Hull et al., 1976; Nachman, 1981; Kareiva, 1984, 1985, 1987; Riechert and Lockley, 1984; Nyrop, 1988; Turchin and Kareiva, 1989). The present study examines whether aphid predators exhibit a numerical response to spatial variation in the density of their prey among alfalfa fields. As adults, many of these predators are highly mobile insects with the potential to seek out and exploit large concentrations of prey over space and time (e.g., Hagen, 1962; Ewert and Chiang, 1966; Kieckhefer and Olson, 1974; Neuenschwander et al., 1975). Experimental evidence suggests that these predators are often instrumental in keeping aphid densities low in alfalfa and other crops (Way and Banks, 1968; Frazer et al., 1981; Kring et al., 1985; Rice and Wilde, 1988). Furthermore, at least some of these predators show strong positive nu merical responses over time to changing aphid densities (Neuenschwander et al., 1975; Wright and Laing, 1980). Frazer et al. (1981) suggested that spatial density dependence is exhibited by aphid predators (ladybeetles) within alfalfa fields, as these predators aggregate and thereby prevent aphid populations from increasing rapidly. In a series of inno vative field experiments, Kareiva (1984, 1985, 1987) demonstrated that habitat fragmentation disrupted the capacity of ladybeetles to aggregate, allowing local aphid populations within goldenrod fields to outbreak. Because it is important to understand how the degree of density dependence varies with spatial scale (Mor rison and Strong, 1980; Heads and Lawton, 1983), we have focused on an es pecially large spatial scale here by examining how the distributions of aphid predators vary in relation to aphid density among individual alfalfa fields.