RELATIVE IMPACT OF INTERACTIONS WITHIN AND BETWEEN TROPHIC LEVELS DURING AN INSECT OUTBREAK

The aspen blotch miner, Phyllonorycter salicifoliella, has occurred at extremely high, "epidemic" densities on its quaking aspen host trees for at least 10 yr in Itasca State Park, Minnesota, USA. During the 1986 through 1988 growing seasons, I censused density, survival, and sources of mortality for eggs, larvae, and pupae on branches at two heights on four host trees. I constructed cohort life tables to determine: (1) dominant sources of mortality, and (2) which sources of mortality were spatially density dependent. The impact of parasitism on leaf miner success was also measured by excluding parasitoids with cages from four trees in 1986. Most mortality occurred during larval development and was caused by interference, parasitism, and unknown causes. Interference means that one larva kills another when mines coalesce during the first three instars. Egg mortality, predation, and death from premature leaf abscission were generally low, except in 1988 when a severe drought caused significant early leaf fall. I tested if dominant sources of mortality were density dependent at four spatial scales: per leaf, twig, branch, and tree. Larval interference was the only consistently density-dependent cause of death. No between-trophic-level interactions imposed density-dependent mortality. Although interference was spatially density dependent, it appears incapable of regulating P. salicifoliella population size. Instead, P. salicifoliella densities approached population "ceilings" delimited each year by availability of young leaves during oviposition. Cessation of the outbreak is unlikely until stochastic meteorological events increase overwintering mortality and/or severely disrupt synchrony between oviposition flights and availability of young foliage.