Pollination insights for the conservation of a rare threatened plant species, Astragalus tragacantha (Fabaceae)

Abstract

The increase in habitat fragmentation impacts plant-pollinator interactions and threatens the sustainability of plant species. Astragalus tragacantha (Fabaceae), is a rare endangered plant species along the coastal habitats where the plant populations have undergone considerable fragmentation and decline of size. Controlled pollination treatments, the observation of pollinator activity, and pollinator captures, have been conducted to study: (1) the mating system of A. tragacantha and the potential for inbreeding depression and/or outbreeding depression based on controlled pollination treatments, (2) the pollinator composition among populations using a correspondence analysis and a hierarchical clustering, and (3) the link between pollinators and the plant reproductive success using a path-analysis model. In this study, we demonstrated that this plant was not autogamous self-pollinating and depended on pollinators for its reproduction. The absence of difference between manual and open pollinations regarding the reproductive success showed an absence of pollen limitation in our populations. We showed that populations differed in the composition of their pollinator guilds. Some pollinator species were predominant in certain populations. The pollination treatments revealed the existence of a mixed mating system in A. tragacantha populations. We showed an inbreeding depression potentially linked to a predominant pollinator-facilitated selfing, and the existence of outbreeding depression between some distant populations. These differences in pollinator guild and plant mating systems among populations must be considered during the restoration of populations along the Mediterranean coastal habitats in order to enhance the reproductive success and sustainability of A. tragacantha.

Unclusterable, underdispersed arrangement of insect‐pollinated plants in pollinator niche space

“…Clusterability tests could be profitably incorporated as a first step in analyses of ecological questions traditionally addressed by means of cluster analysis, so that the existence in multivariate spaces of objectively recognizable clusters of objects (e.g., species, plots, communities) can be rigorously evaluated before proceeding to clustering. This applies also to pollination studies, which have sometimes used clustering to identify plant groupings in multivariate spaces defined by floral traits or pollinator composition (Ollerton et al 2009, de Jager et al 2011, Abrahamczyk et al 2017, Schurr et al 2019, Vandelook et al 2019). As exemplified by this study, clusterability analysis allows tests of hypotheses in pollination ecology by providing an objective means of ascertaining whether groups defined by pollinator composition actually exist in a given plant species sample.…”

Pollinator Specific Richness and Their Interactions With Local Plant Species: 10 Years of Sampling in Mediterranean Habitats

“…Therefore, larger wild bee species are more able to forage on this species. A deeper knowledge of the pollinator guild has been indicated to be essential to provide management advice to build an efficient conservation programme (Schurr et al 2019). We also found 20 plant species that interacted with only one pollinator species.…”

“…In May 2016, a collective 3-d sampling campaign was completed as part of a training course on wild bees (Sampling 5) (Geslin et al 2018). Between 2017 and 2018 (Samplings 6 to 9), more intensive surveys were carried out every week for 2-5 mo (Schurr et al 2019, Ropars et al 2020. Finally, D. Genoud performed a one-time bee species survey in May 2018 (Sampling 10).…”

Unclusterable, underdispersed arrangement of insect-pollinated plants in pollinator niche space

“…Clusterability tests could be profitably incorporated as a first step in analyses of ecological questions traditionally addressed by means of cluster analysis, so that the existence in multivariate spaces of objectively recognizable clusters of objects (e.g., species, plots, communities) can be rigorously evaluated before proceeding to clustering. This applies also to pollination studies, which have sometimes used clustering to identify plant groupings in multivariate spaces defined by floral traits or pollinator composition (Ollerton et al 2009, de Jager et al 2011, Abrahamczyk et al 2017, Schurr et al 2019, Vandelook et al 2019). As exemplified by this study, clusterability analysis allows tests of hypotheses in pollination ecology by providing an objective means of ascertaining whether groups of plants defined by pollinators actually exist in a given species sample.…”