Fluctuations in abundance of a stem-boring moth damaging shoots of Phragmites australis: causes and effects of overexploitation of food in a late-successional grass monoculture.
Fluctuations in the abundance of the noctuid Archanara geminipuncta, which causes damage to the silicate-rich shoots of common reed (Phragmites australis), were studied in the German Federal Republic. The pest showed a non-equilibrium nature of population fluctuations in this late-successional, apparent, and undisturbed grass monoculture. The striking similarities to pest outbreaks in crops (early successional plants in man-made pure stands) are discussed. The larvae needed thick shoots for stem-boring and pupation, so high population densities occurred only within optimal habitats (reed growing in wet habitats), with up to 96% of the shoots damaged. Thick shoots contained more silicate, minerals and nitrogen than thin shoots. Contrary to expectations, not only wet reed, but also dry reed, had a high coefficient of variation to moth abundance between habitats and years. Larval survival was positively correlated with the mean shoot diameter (= availability of suitable food) and negatively correlated with the number of damaged shoots (= intraspecific competition). In a multiple regression, these easily measurable shoot characters explained 82% of the variation in the abundance of pupae and 65% of the variance of larval survival. Larvae needed only 2-3 shoots before pupation, but even in wet reed, the ratio of damaged shoots to pupae could be as high as 25. Neither chemical plant defences nor predation prevented outbreaks, and population dynamics could not be explained by these factors. Population collapse appeared to be due to the overexploitation of food resources (when larvae damaged nearly all thick reed shoots) with a subsequent scramble competition resulting in larvae starving to death. Larval damage induced the production of narrow side shoots. The number of side shoots per damaged stem was positively correlated with average damage level in the stand, and this indicated that the response of reed to damage depended on properties of the clone (due to the high physiological integration) and habitat (light-induction of dormant buds due to the thinned canopy), but not on properties of the ramet. Reed responses were conflicting or suboptimal in that regrowth tissue (side shoots) not only rebuilt assimilation and competition ability, but also increased plant susceptibility to gall makers.