Article:

Fischlin, A., 1982. Analyse eines Wald-Insekten-Systems: Der subalpine 
	Lärchen-Arvenwald und der graue Lärchenwickler Zeiraphera diniana Gn. (Lep., 
	Tortricidae). Diss. ETH No. 6977, Swiss Federal Institute of Technology: Zürich, 
	Switzerland, 294 pp.

Abstract:

Two hypotheses to explain the population dynamics of the larch bud moth Zeiraphera diniana Gn. in the subalpine larch- cembran pine forest of the European Alps have been modelled. They are: the larch-larch bud moth hypothesis and the migration hypothesis. The latter has two versions: the conglobation and the translocation hypothesis.

According to the larch-larch bud moth hypothesis the regular cyclic fluctuations in the density of the larch bud moth can be explained by the following mechanism: Heavy defoliation of larch trees by larch bud moth larvae results in the production of needles of poor nutritional value for the insects. Once stressed, host plants need several years to recover. When they produce normal needles again, the population of the herbivorous insect increases, and a new cycle starts.

According to the migration hypothesis the population density of larch bud moth is significantly influenced by immigrating and emigrating moths. In the conglobation hypothesis, outbreaks of larch bud moth are triggered and supported by the regular immigration of moths from a large area in which outbreaks rarely occur. In the translocation hypothesis outbreaks are caused by the immigration of moths originating from an outbreak area.

Simulations of the model for the Upper Engadine Valley indicate that the larch-larch bud moth relationship is sufficient to explain most of the dynamics of larch bud moth, especially the asymptotically stable cycles. Hence, in light of the available data, the larch-larch bud moth hypothesis is preferred over other hypotheses to explain the local dynamics, such as the antagonism and the polymorphism hypotheses. It has been concluded from the sensitivity analysis of the larch-larch bud moth model that the incorporation of these other hypotheses would explain the remaining, minor differences between the behaviour of the larch-larch bud moth model and the real system.

When migration is added to the model its output is slightly more realistic. The translocation hypothesis appears more plausible than the conglobation hypothesis. However, the flight behaviour of larch bud moth is such, that all migration results in nothing more than a mixing of animals within and between natural populations, without remarkable impact on the dynamics of its numbers. Thus migration is not considered to be a fundamental mechanism for the asymptotically stable cycles of larch bud moth. Nevertheless migration is essential for the explanation of the conspicuous resilience of the larch bud moth system. Further simulations were made to test the effects of Bacillus thuringiensis (BT) and Parapheromone treatments on the population dynamics of larch bud moth.

BT-treatments allow temporary avoidance of heavy defoliation. However, in spite of all treatments an outbreak occurs sooner or later. After persistent exposure to moderate stress, this leads to an additional hazard for the trees. Thus, not only is loss in wood production the same as without treatment, but also tree mortality even increases. Compared to the high costs of the treatments, these risks may not be outweighed by the small benefit of having suppressed for a few years nothing more than the visible aspects of defoliation.

Pheromone treatments look much more promising. However, success depends completely on the treatment of sufficiently large areas with low levels of moth immigration.

Despite the fact that it is impossible in reality to eradicate larch bud moth within a whole valley, this has been assumed in some simulations. They demonstrate how remarkable the resilience of the system is: Two years after extinction, the model system behaves as it did before. From this follows that eventual treatments can never be limited to a few years; they would have to be repeated all the time. If it should ever start, a pest management program with no end has to be faced.

Full text 11.6 MB Appendix Microfiches 0.9MB