Empirical Evidence for Complex Source–sink Dynamics With Alternative States in a Butterfly Metapopulation

Abstract

In heterogeneous landscapes, a species’ habitat may be partitioned into sources and sinks. Conceptually, three kinds of habitat have been described: (1) “sources” are consistent net exporters of organisms; “true sinks” are net importers, and without immigration their populations go extinct; and (3) “pseudosinks” are also net importers, but without immigration they can sustain populations and sometimes can even become net exporters (sources). Previously, I have described sources and pseudosinks in a metapopulation of the butterfly Euphydryas editha and reported the extinction of populations in sources due to an unusual frost. Here, I describe the recolonization of the former sources by migrants from extant populations in the former pseudosinks. For comparison, a series of vacant patches was created in the former pseudosinks. Recolonization was tracked by sampling the population density in the vacant patches. Patches were sampled for 5 years. Contrary to theoretical expectations, the establishment rate of new populations was 10 times higher in outcrops, the former pseudosinks, than in clearings, the former sources. Initial population density was 150 times higher in outcrops (measured as number of larval webs per square meter). Yet in clearings, only the immigrants had poor reproductive success. On those occasions when a resident population was established, the residents had high reproductive success, and the population grew rapidly. The low recolonization rate of clearings could not be attributed to effects of patch size or spatial barriers. A temporal barrier was hypothesized, in which immigrants arrived too late each year to reproduce successfully on host plants in clearings. The host plant in clearings, Collinsia torreyi, typically senesced during the breeding season, but the host plant in outcrops, Pedicularis semibarbata, did not. The hypothesis proposed that immigrants oviposited later than residents because the immigrants originated in outcrops, which had a different microclimate that delayed adult eclosion. The hypothesis was supported: the newly established resident populations in clearings tended to oviposit 10 d earlier than populations in nearby outcrops. An experiment showed earlier eclosion times in clearings than in outcrops. Another experiment showed much higher larval survival in early clearings than in late clearings. Mortality was correlated with host‐plant senescence. The temporal barrier was strong enough to keep clearings as net importers of butterflies from outcrops, meaning that the net flow of butterflies reversed after the frost destroyed the resident populations in clearings. Thus, “early” clearings had been sources, but “late” clearings were true sinks, giving the system two kinds of source–sink relationships. After the frost, the metapopulation as a whole underwent a demographic source–sink inversion from one locally stable state to the other. The findings suggest that asymmetric constraints on dispersal, due to temporal structure, may be a mechanism for complex source–sink dynamics in h...

When Sources Become Sinks: Migrational Meltdown in Heterogeneous Habitats

“…Equilibria corresponding to a generalist and a specialist may be simultaneously stable. Random disturbances can then trigger rapid change in niche breadth, as observed in the case of the checkerspot butterfly (Boughton 1999). We show that small transient changes in the landscape structure may have irreversible effects on the evolution of specialization, illustrating a case of hysteresis.…”

“…The nature of the genetic and ecological factors that limit the niche breadth of a species in a heterogeneous habitat still remains largely an open question. In this paper, we suggest that host range evolution may be very sensitive to demographic disturbances affecting both population densities and migration patterns in a heterogeneous habitat.The present theoretical study is inspired by the dynamics and rapid evolution of a heterogeneous metapopulation of the checkerspot butterfly, Euphydryas editha Thomas et al 1996;Boughton 1999). Before 1989, larvae of E. editha were found on two host plants.…”

“…The present theoretical study is inspired by the dynamics and rapid evolution of a heterogeneous metapopulation of the checkerspot butterfly, Euphydryas editha Thomas et al 1996;Boughton 1999). Before 1989, larvae of E. editha were found on two host plants.…”

“…Populations on Collinsia torreyi behaved as source populations exporting migrants to pseudo-sink populations on Pedicularis semibarbata. The accidental extinction of populations on Collinsia, due to an unusual summer frost in the early 1990s, was followed by a complete reversal in source-sink dynamics and a dramatic narrowing of the butterfly diet (Boughton 1999). Since the disturbance, local populations have not regained the ability to use their former host, Collinsia.…”

Phenological asynchrony between herbivorous insects and their hosts: signal of climate change or pre-existing adaptive strategy?

“…A suite of populations feeding on the annual Collinsia tinctoria became extinct when host seeds responded to the California drought of [1977][1978] by not germinating at all (Singer & Ehrlich 1979;Ehrlich et al 1980). Taking these observations along with those of the asynchrony between bay checkerspot and its two annual hosts, it is not surprising that populations of Edith's checkerspot adapted to perennial hosts (especially the genus Pedicularis) have been dynamically more stable and suffered significantly lower extinction rates than populations feeding on annual plants Boughton 1999).…”