Antagonistic effects of large- and small-scale disturbances on exotic tree invasion in a native tussock grassland relict

Abstract

It is generally accepted that disturbances increase community invasibility. Yet the role of disturbance in plant invasions may be less predictable than often assumed, due to the influence of environmental stochasticity and interactions between disturbance regimes. We evaluated the single and interactive effects of prescribed burning (large-scale, infrequent event) and animal diggings (small-scale, frequent events) on the invasion success of Gleditsia triacanthos L. in a tussock grassland relict of the Inland Pampa, Argentina. Tree seedling emergence and survival were monitored over 4 years, after adjusting for propagule pressure through copious seed addition to all disturbance treatments. Burning altered community structure by suppressing tussock grasses and promoting exotic forbs, whereas simulated, armadillo-like diggings had little impact on herbaceous composition. Overall, seedling emergence rather than survival represented the main demographic bottleneck for tree invasion. Tree establishment success varied among seedling cohorts emerged in different climatic years. In a dry year, emergence was only slightly affected by disturbances. In contrast, for two consecutive wet years, initial burning and armadillo-like diggings exerted strong, antagonistic effects on tree recruitment. Whereas fire alone increased recruitment, the simulated burrowing regime prevented seedling emergence in both burned and unburned plots. The latter effect might be explained by reduced soil moisture, and increased seed burial and predation in excavated patches. Thus, the impact of a single, large-scale perturbation promoting woody plant invasion was overridden by a regime of small-scale, frequent disturbances. Our results show that grassland invasibility was contingent on inter-annual climatic variation as well as unexpected interactions between natural and anthropogenic disturbance agents.

Woody Plant Invasions in Pampa Grasslands: A Biogeographical and Community Assembly Perspective

“…These fi ndings show that different biotic factors may contribute to repel invasion by this native tree in both grasslands. In contrast, we found that even undisturbed grasslands were not fully immune to invasion by G. triacanthos (Mazía et al 2001(Mazía et al , 2010Chaneton et al 2004 ) .…”

“…5.6 ). Large-and small-scale disturbances thus promoted tree recruitment in the old fi eld, whereas they exerted antagonistic effects in the relict tussock community (Mazía et al 2010 ) . Final establishment of G. triacanthos was twice as high in the old fi eld as in the tussock grassland (Fig.…”

“…Tussock grasses strongly suppress tree seedling growth and survival in humid grasslands (Mazía et al 2001 ;Chaneton et al 2004 ) , and may indirectly prevent seedling emergence by providing cover to granivorous rodents (Mazía et al 2010 ) . These factors alone only rarely repel invasion, unless immigrating species are propagule-limited (Levine et al 2004 ;Lockwood et al 2005 ) , which is the case at increasing distance from seed sources.…”

“…Studies in the Inland pampa show that disturbance effects on tree establishment can be context-dependent. In a large-scale, 4-year experiment, we examined the in fl uence of two disturbance agents, burning and animal diggings, on G. triacanthos recruitment in a late-successional old fi eld and a relict of native tussock grassland (Chaneton et al 2004 ;Mazía et al 2010 ) . To adjust for propagule pressure, 40 seeds/m 2 were added in each of several sowing events to 12-m 2 plots, which were either burned or "burrowed" to simulate the digging activity of hairy armadillos ( Chaetrophractus villosus ) in a full factorial design (for details, see Mazía et al 2010 ) .…”

“…5.5 ). The greater resistance to tree invasion of the tussock grassland likely re fl ected the combined effects of grass competition and rodent granivory (Chaneton et al 2004 ;Mazía et al 2010 ) . Indeed, seed loss to predators was higher in the tussock than in the old-fi eld grassland (Fig.…”