Impact of habitat quality on metapopulation structure and distribution of two me litaeine butterfly species

CHEN Jie_Jun~1, WANG Yi_Fei~1, LEI Guang_Chun~2, WANG Rong_Jiang~2, XU Ru_Me i~(1) (1. Ministry of Education Key Laboratory for Biodiversity Science an d Ecological Engineering, Beijing Normal University, Beijing 100875, China; 2. College of Life Sciences, Peking University, Beijing 100871, China)

Melitaeine butterflies; Melitaea phoebe; Euphydryas aurinia; metapopu lation; habitat quality; host plant; nectar resources

We studied the impact of habitat quality factors on the metapopulation structure and distribution, and thus the persistence of two species of melitaei ne butterflies, Melitaea phoebe and Euphydryas aurinia, on a metapopulat ion scal e in Chicheng County, Hebei Province, China. These two species coexist in a netw ork of discrete habitat patches within a range of 10 km~2, larvae feeding on th ei r specific host plants Stemmacantha uniflora (Compositae) and Scabiosa tsc hiliensis (Dipsacaceae) respectively. However, their nectar resources are almo st the same during the adult flight period. The main flight period for E. auri nia is about one month earlier than M. phoebe, with an overlap of only abo ut seven days. T he host plants of these two melitaeine butterflies also have a great impact on t heir occurrence, but act in different ways. Density of the host plant has a grea t effect on the occurrence of M. phoebe, whereas the height of the host plan t is important for E. aurinia. The abundance and distribution of nectar resources ar e positively correlated to the abundance and distribution of the two butterfly s pecies. The vegetation diversity, evenness, and cover correlate negatively with the occurrence of E. aurinia, but not related to M. phoebe, while the ve getation height has a positive effect on the two butterfly species. We have also conside red and discussed the impact of slope aspect, gradient, and irradiation of the h abitat patches. Therefore, when discussing the factors affecting metapopulation structure and dynamics, we should pay attention not only to the patch size and i solation but also the quality differences among patches. Four major points can b e well concluded: 1) The specificity of larval host plants and the time segrega t ion of adult feeding permit the co_existence of the two species in this small an d patchy system. 2) The nectar resource is a major limiting factor and prone to being affected by weather stochasticity. The fluctuations in nectar resources ca n explain the dynamic rises and falls of these butterfly metapopulations from 19 98 to 2000. 3) The differences between flight and food searching ability of the two butterfly species, as well as the differences between the features and spati al distribution of the two host plants species determined the different metapopu lation structures of the two butterfly species. These results supported one of o ur former conjectures that E. aurinia is a classical metapopulation, and M . phoebe is a source_sink metapopulation. 4) It is the spatial pattern and qua lity of the habitat patches (as determined by other factors, e.g., topographic featu res) interacting with the insects’ behavior (e.g., plant preference and fli ght beha vior) that determined the metapopulation structure and distribution of these two butterfly species.