and 5 UR "Ecologie et dynamique des syst emes anthropis es" (EDYSAN, FRE 3498 CNRSSummary 1. Dark diversity is a promising concept for prioritising management efforts as it focuses on species that are present in the regional pool, but locally absent even though environmental requirements are met. Currently, we lack knowledge of what characterises species belonging to the dark diversity more often than others, although this is important knowledge for restoration and conservation actions. 2. We applied the concept to a massive national (Danish) plant diversity data base, containing 236 923 records from 15 160 surveys involving 564 species. This enabled the first geographically comprehensive (43 000 km 2 ) assessment of dark diversity, at a spatial resolution relevant for conservation and restoration planning (78 m 2 ) across multiple terrestrial habitats, thereby maximising the practical applications of this concept. The probability for a given plant species to belong to the dark diversity was computed and logistically regressed against variables representing its ecological preferences (e.g. nutrient availability), strategies (competitor, stress tolerant, ruderal), mycorrhizal relationships, establishment capacities (seed mass) and dispersal abilities. 3. Forty-six percent of the species had a high probability (>95%) of being part of dark diversity, whereas for 7% of the species this probability was less than 60%. 4. Typical dark diversity plant species tended to depend on mycorrhiza, were mostly adapted to low light and low nutrient levels, had poor dispersal abilities and were ruderals and stress intolerant. 5. Synthesis and applications. Characterising species that are more often absent from suitable sites than others (dark diversity species) has important implications for the planning and management of natural ecosystems. From our study, practitioners gain insight into the factors triggering the absence of individual plant species in a seemingly suitable habitat. We highlight the need to carefully consider mycorrhizal inoculations with a suitable assemblage of fungi to promote the establishment success of dark diversity plants. Additionally, time-lags in plant species dispersal and establishment as well as spatial connectivity in fragmented habitats are central to consider in nature management although assisted migration might also aid poor dispersers. Finally, nutrient-poor localities are probably important 'islets' allowing nitrophobic dark diversity plant species to thrive within agricultural landscapes that are generally nutrient-rich.

No abstract

151 Dark diversity is a promising concept for prioritizing management efforts as it focuses on missing species, 16i.e., species present in the regional pool, but locally absent despite suitable environmental conditions. 2 We 17 applied the concept to a massive national plant diversity database (236,923 records from 15,160 surveys 18 involving 564 species) to provide the first geographically comprehensive assessment of dark diversity across 19 a large area (43,000 km 2 ), at a spatial scale (~75 m 2 ) relevant for conservation and restoration planning and 20 across multiple terrestrial habitats, thus maximising its practical application potential. The likelihood for a 21given plant species to belong to the dark diversity pool was computed and logistically regressed against its 22 ecological preferences (nutrient availability, pH etc.), strategies (competitor, stress tolerance, ruderal), 23 mycorrhizal dependence and infection percentage, seed mass and maximum dispersal distance. 3 Forty-six 24 percent of the species were absent in >95 % of the suitable sites, whereas 7 % of the species were absent in 25 less than 60 % of sites that were deemed suitable. 4 Species that were more likely to belong to the dark 26 diversity tended to depend on mycorrhiza, were mostly adapted to low light and nutrient levels, had poor 27 dispersal abilities, were ruderals and had a low stress tolerance. Synthesis and applications Our findings 28 have important implications for the planning and management of natural ecosystems requiring detailed 29 knowledge of what triggers the presence/absence of individual plant species in a seemingly suitable 30 habitat. We conclude that practitioners may need to carefully consider mycorrhizal inoculations with a 31 suitable assemblage of fungi for certain plant species to become established. Also assisted migration might 32 be necessary to help poor dispersers although spatial and temporal processes are also important to have in 33 mind. Finally, it is important to vary nutrient loads making room for plant species to colonise both nutrient-34 poor and nutrient-rich localities. 35

No abstract