Determinants and effects of moth population dynamics in altitudinal gradients in northern Fennoscandia

Abstract

The aim of this PhD-project has been to increase our knowledge of trophic interactions along altitudinal gradients in a system of sympatric, cyclically outbreaking geometrid moth species in sub-arctic, coastal birch forest in northern Norway. The project resulted in four scientific publications; two sub-studies focused on the causal effect of natural enemies on the observed complex spatio-temporal dynamics of the moth species (Papers 1 & 2). The other two sub-studies investigated the ecological effects of variable moth larvae densities as a resource for predators in the sub-Arctic ecosystem (Papers 3 & 4). In paper 1, we rejected the hypothesis that larval parasitoids might terminate moth population peaks, one of the most important hypotheses on insect population regulation. We, however, found that larval parasitoids can influence local spatial variation in moth population density. Hence, we concluded that spatial replication of sampling sites is crucial to arrive at non-biased estimates for the temporal effect of larval parasitoids on host populations. The long enduring debate on the regulation potential of parasitoids on insect populations in the scientific literature may be viewed in light of our findings. Similarly, in Paper 2, we tested if the often-observed spatial outbreak patterns of moth populations might be a result of release from invertebrate predators. Along mountain slopes, moth outbreaks in northern Fennoscandia often occur close to the tree line, and we tested if the abundance of ground dwelling invertebrate predators was inversely correlated with altitude, testing the assumption that moth outbreaks may be a result of reduced predation impact at high altitudes. As our analyses could not reveal this predicted pattern, we rejected the hypothesis. Consequently, the first two sub-studies delimitated the number of candidate factors that might shape Fennoscandian moth dynamics by reevaluating the role of larval parasitoids and ground-dwelling invertebrate predators. In paper 3, we reanalyzed the data from Paper 1 from the parasitoid's perspective by determining the dependencies of individual parasitoid species to host densities. For comparison, we also estimated the strength of effects hidden in the spatial and temporal attributes of altitude and year on parasitoid prevalence rates. We found that altitude and year, rather than host density, explained by far most of the extra-binomial variation in the parasitoid prevalence data. Hence, we concluded that unknown factors, which co-vary with altitude and year, dominate the prevalence dynamics of the larval parasitoids in our study and, consequently, act to decouple parasitoid dynamics from the dynamics of their hosts. Moth larvae as a resource for predators were also the focal issue in Paper 4, which focused on the importance of prey availability and forest phenology for breeding parameters in the two passerine birds great tit and pied flycatcher. The study was based on the phenological conflict that spring arrives in mountain slopes first at low altitude, while the bird's main food resource during the breeding season (i.e. moth larvae) often dominates forest habitat at higher altitude. Thus, we tested if passerine birds select breeding habitat according to the onset of spring or the availability of moth larvae and if the two bird species differed in this respect. We found that the phenological timing was the main driving force for both species when selecting breeding habitat lowest in the gradients where spring arrived first. However, also the positive correlations with moth larvae density appeared to be significant in our analyses. The main difference between the two species was found in the timing strategy for egg laying and size of clutches, which might make the great tit more prone to mismatch during springs of lower temperature. In addition, while nest box occupancy decreased with altitude in both species, the same relationship with respect to clutch size was only found for pied flycatchers, but not for great tits. Finally, great tits were more prone to fail during their breeding attempt and we concluded that the pied flycatcher, by laying eggs up to two weeks after great tits, appears to be the better adapted species with respect to breedig at sub-arctic conditions. This PhD-thesis strengthens the view that so-called top-down interactions in the ecological moth system in northern Fennoscandia cannot explain complex moth population dynamics, but that moth larvae can be regarded as a valuable resource for species at other trophic levels. The consistent dominance of altitude and year in all our analyses, on the other hand, suggests that explanations for the complex spatio-temporal moth population dynamics ought to be looked for in mechanisms ruled by environmentally governed processes such as phenology.