Functional diversity loss among pollinators has rapidly progressed across the globe and is expected to influence plant-pollinator interactions in natural communities. Although recent findings suggest that the disappearance of a certain pollinator functional group may cause niche expansions and/or shifts in other groups, no study has examined this prediction in natural communities with high plant and pollinator diversities. By comparing coastal pollination networks on continental and oceanic islands, we examined how community-level flower visit patterns are influenced by the relative biomass of long-tongued pollinators (RBLP). We found that RBLP significantly correlated with pollinator functional diversity and was lower in oceanic than in continental islands. Pollinator niches shifted with decreasing RBLP, such that diverse species with various proboscis lengths, especially shorttongued species, increasingly visited long-tubed flowers. However, we found no conspicuous negative impacts of low RBLP and the consequent niche shifts on pollinator visit frequencies to flowers in oceanic island communities. Notably, fruit set significantly decreased as RBLP decreased in a study plant species. These results suggest that niche shifts by other functional groups can generally compensate for a decline in long-tongued pollinators in natural communities, but there may be negative impacts on plant reproduction.

1. Although biodiversity loss is a critically important topic, our understanding of how both land abandonment and land-use intensification in semi-natural grasslands alters the community diversity and assembly mechanisms is very limited. Largescale economic drivers of land-use change might inadvertently result in the loss of vulnerable species and reduce ecosystem service provisioning.2. In this study, we assessed non-random community change by examining patterns of low-abundance species loss, and community assembly in semi-natural grasslands due to land abandonment and intensification in southwest Japan. We analysed relationships between evolutionary distinctiveness and abundance for each species. In addition, we used metrics of species and phylogenetic diversity and phylogenetic structure to assess patterns of non-random biodiversity loss due to both abandoned and intensified land-use.3. We demonstrated that low-abundance species were more evolutionarily distinct compared to high-abundance (i.e. dominant) species. Furthermore, land-use intensification resulted in further declines in low-abundance species, whereas land abandonment resulted in declines in all species, regardless of their initial abundance.4. We found that both forms of land-use change (abandonment and intensification) resulted in non-random patterns of community change, with traditional land-use maintaining the highest biodiversity and land-use change coinciding with decreased species and phylogenetic diversity measures. Intensified land-use caused phylogenetic community structure to be more closely related than expected by chance, whereas the metrics of how distant species are from one another on average based on phylogenetic tree (i.e. mean pairwise phylogenetic distance) did not change from land abandonment. Synthesis and applications.We show that the loss of phylogenetic diversity, and especially of low-abundance species with high value of phylogenetic distinctiveness, resulted in non-random community disassembly. Our results argue that in order to maintain biodiversity in these semi-natural grasslands, traditional management practices should be encouraged over intensification and simple abandonment. Government agencies should adopt policies or provide incentives that encourage the maintenance of traditional practices in rural Japan, and elsewhere where the combination of land consolidation and abandonment are important conservation issues.

No abstract

Palaemon paucidens has a large population and is an important food source for fish in Lake Biwa, Japan. They are abundant in shallow waters from spring to summer, after which most individuals migrate to offshore deep areas where they remain during autumn and winter. However, some individuals are nonmigratory, remaining in shallow waters over winter. It has been reported that P. paucidens individuals have declined in recent years; a better understanding of its seasonal distribution is needed to manage this species, and basic information on its seasonal distribution is indispensable. We tracked the environmental DNA (eDNA) distribution of P. paucidens in Lake Biwa over a year using a quantitative real-time polymerase chain reaction method.We collected water samples from offshore (both from the surface and from the benthic) and from shallow shore sites adjacent to the shorelines of the main lake and connecting freshwater lagoons. Offshore sampling took place in summer and winter, and shallow shore and lagoon sampling in all four seasons. During summer, eDNA concentrations were significantly higher in shallow and lagoon areas than offshore bottom sites. Conversely, during winter, eDNA concentrations were higher in offshore bottom sites, and relatively high and low eDNA concentrations in lagoons and shallow shore, respectively. These results most likely reflect the spatial and temporal distribution of this species in Lake Biwa. The eDNA concentrations peaked in early August at shallow shore sites in the main lake, with a significant decline in mid-October, while low eDNA concentrations were recorded at offshore bottom sites in late August. These results suggest that P. paucidens migrates from shallow waters to offshore bottom sites between early August and mid-October. These results provide important information for the management of this species. K E Y W O R D Senvironmental DNA (eDNA), Lake Biwa, migration, Palaemon paucidens, quantitative real-time PCR, spatial and temporal distribution | 55 WU et al.

No abstract