Plant Ecol (2015) 216:1103–1115 DOI 10.1007/s11258-015-0495-3 Asteraceae invaders have limited impacts on the pollination of common native annual species in SW Western Australia’s open woodland wildflower communities Xingwen Loy . Claire E. Wainwright . Margaret M. Mayfield Received: 2 April 2015 / Accepted: 2 July 2015 / Published online: 9 July 2015 Ó Springer Science+Business Media Dordrecht 2015 Abstract The York gum–jam woodlands of south- weed). Only two of the five native species examined west Western Australia support diverse annual wild- had significant seed set benefits attributable to insect flower communities despite extensive habitat pollination. One native plant species, Podotheca fragmentation, remnant isolation and the invasion of gnaphalioides, had pollinator assemblages that over- many exotic annual plant species. Few studies have lapped significantly with exotic A. calendula, with explored the pollinator–plant relationships maintain- some reduction in pollinator visitation evident. One ing these persistently species-rich ‘novel’ communi- species, Waitzia acuminata, was found to benefit from ties. We examine the pollination ecology of five native insect pollination only in the larger of two surveyed species common to York gum–jam woodland annual remnants, which may reflect emerging reproductive communities to determine whether native pollinators polymorphism among geographically isolated popu- may be mediating impacts of exotic annual plants on lations. We highlight two mechanisms in this system native wildflower species. We determined the polli- that may buffer pollinator-mediated impacts of exotic nation requirements of native focal species and the species on native species: autonomous seed produc- diversity and frequency of pollinator visitation to these tion, which may be increasingly prevalent in isolated focal plant species across invasion gradients. We also populations, and segregation of pollinator resources recorded the pollinator community of a dominant among species. Our findings illustrate the ways that exotic herb in this system: Arctotheca calendula (cape pollinator-mediated interactions can affect seed set within plant communities persisting in highly frag- mented and invaded agricultural landscapes. Communicated by Philip Ladd. Keywords Pollen limitation Á Biological invasion Á Electronic supplementary material The online version of this article (doi:10.1007/s11258-015-0495-3) contains supple- Annual plants Á Indirect interactions mentary material, which is available to authorised users. X. Loy (&) Á C. E. Wainwright Á M. M. Mayfield School of Biological Sciences, The University of Queensland, Brisbane, QLD 4072, Australia Introduction e-mail: [email protected] C. E. Wainwright Biological invasions are considered one of the largest e-mail: [email protected] threats to biodiversity worldwide (Gurevitch and M. M. Mayfield Padilla 2004; Wilcove et al. 1998). While negative e-mail: m.mayfi[email protected] impacts of exotic plant species on native plant 123 1104 Plant Ecol (2015) 216:1103–1115 communities are well-documented, the mechanisms impacts of pollinator-mediated interactions between driving these impacts are less clear. Though direct native and exotic species are clearly variable, and the competition with natives is often suggested to be key introduction of an exotic species into a native plant to mediating impacts of invasion on native commu- community can produce contrasting effects on differ- nities (Levine et al. 2003), interactions between native ent native species (Waters et al. 2014). Studying and exotic plant species can also be indirect and pollinator-mediated interactions between native and mediated by third-party mutualists, such as pollina- exotic plant species across many systems is key to tors. Though indirect interactions have been recog- gaining a more complete view of the role pollinating nised as important to the dynamics among native and insects play in driving changes in or maintaining exotic species, they have not been as widely studied as native biodiversity. direct impacts (White et al. 2006). As seed production In this study, we examined insect pollination in five in many plant species is pollen-limited (Joar Hegland native annual plant species from annual wildflower and Totland 2008; Kelly et al. 2007; Law et al. 2010; communities found in semi-arid woodlands of south- Pauw and Bond 2011; Rosenheim et al. 2014), west Western Australia. Using natural variation in invader-driven disruptions to pollinator–plant interac- plant community composition (and invasion extent tions that influence seed production may reduce the within communities) across two isolated remnants, we fitness of seed-limited native species and contribute to examine whether successful exotic Asteraceae species changes in local populations and community have pollinator-mediated effects on the reproductive dynamics. success of locally abundant native annual forbs. Exotic plant species can affect local native plant– Specifically, we asked: pollinator relationships in several ways. They may, for 1. Is there evidence that seed production in the native example, increase the diversity of available floral focal species is pollen-limited in this system? resources, leading to increases in local pollinator 2. Do the native focal species share pollinators with abundance (Feldman et al. 2004; Ghazoul 2006; the most common insect-pollinated exotic Aster- Jakobsson et al. 2009). Such pollinator increases may aceae (cape weed) in this system? or may not benefit native plants, however, as these 3. How does pollinator visitation to native plant new pollinators will not necessarily visit native species change between different densities of species and even if they do, may increase heterospeci- exotic Asteraceae within remnants? fic pollen exchange (transfer of pollen between different plant species) rather than intraspecific pollen transfer (Eaton et al. 2012). In addition, exotic plant species may reduce the pollination success of neigh- Methods bouring native plant species by outcompeting native plants for available pollination services (Dietzsch Study system et al. 2011; Knight et al. 2005; Waters et al. 2014). In spite of the often shared biogeographic and evolu- The York gum–jam woodlands of southwest Western tionary histories of native insect-pollinated plants and Australia experience a Mediterranean-type climate, their pollinators, pollinator interactions are more often with cool, wet winters and warm, dry summers. The opportunistic than oligolectic in nature and pollinator canopy of these woodlands is characterised by sparse infidelity is common (Kearns et al. 1998). Even where York gum (Eucalyptus loxophleba subsp. loxophleba) exotic and native species do not share pollinators, and jam trees (Acacia acuminata) with an understorey exotic species may indirectly impact native plant of sparse shrubs, perennial grasses, and diverse and pollination by reducing native species densities and abundant ephemeral annual flora dominated by Aster- thus floral density (through direct competition) (Kar- aceae species. The annual wildflowers in this system ron et al. 1995; Kirchner et al. 2005; Steffan- have a growing season that runs from July–November Dewenter et al. 2001) or by driving a reduction in (winter/spring), with flowering and seed production flowering effort per individual (driving a reallocation occurring between September and November. of resources away from flower production, Suter York gum–jam woodlands are found in the South- 2009; Wang et al. 2014). The mechanisms and west Australian Floristic Region, a hotspot of 123 Plant Ecol (2015) 216:1103–1115 1105 biodiversity and endemism, which has suffered severe 2.4 km2 of which only 0.2 km2 is York gum–jam population declines and extinctions due to widespread woodland. Due to variability in the composition of the land clearing for agricultural production over the last annual assemblages in these two reserves, data were 60 years (Beard 1990). Fertiliser application in farm- only collected in both reserves for W. acuminata and lands surrounding remaining York gum–jam wood- R. manglesii. P. lessonii was surveyed only in Kunjin, land fragments has increased nutrient availability while G. berardiana, P. gnaphalioides and the exotic along woodland borders, which has facilitated the A. calendula were surveyed only in Bendering. invasion of exotic annual plant species into remnant woodland plant communities (Hester and Hobbs Measuring invasion: landscape invasion levels 1992). These invasions have resulted in new plant and invasion parameters assemblages that support mixtures of native and exotic plants (Dwyer et al. 2015). These ‘novel’ communities The York gum–jam woodlands found in Bendering present ideal systems for studying the small-scale and Kunjin reserves span a gradient of mild to interactions between exotic and native plants in severe biotic invasion. To survey pollinator com- species-rich communities, particularly given their munities of native focal species across invasion close proximity to native-dominated reference gradients, we marked out four to eight 5 m 9 5m communities. study plots for each focal species spanning the natural variation in exotic cover in each reserve. Study species Each focal species was surveyed from at least four plots in each reserve where it was studied (Online Five native focal species and one common exotic Resource Table 1). Where possible, we surveyed species were selected for this study (Fig. 1): Goodenia multiple focal species within the